that a) STATUS[ES]_IGNORE *is* NULL, and b) ROMIO blindly sends its
status through to STATUS_SET_ELEMENTS, even if the status is IGNORE.
So we have some legal cases where IGNORE can be passed through here.
Well, that's what we get for trying to do good error checking. :-(
This commit was SVN r12999.
- Fix some fpritnf's in ompi_mpi_abort() that incorrectly assumed that
we were always being invoked from MPI_ABORT (ompi_mpi_abort() may be
invoked from a bunch of different places)
- Also try to opal_backtrace_print() if opal_bactrace_buffer() is not
supported.
- Print a message in MPI_ABORT if we're aborting.
This commit was SVN r12998.
This is somewhat limited currently for expample, if you have 3 ports on Node A and 5 ports
on Node B then the peers will use 3 ports to communicate with each other.
This is on a subnet basis, so for any pair of nodes we take the
intersection of the available ports within a subnet.
We use subnets to determine reachability for lazy connection establishment. So
if Node A and Node B each have two HCA's (on seperate networks) then the
subnet's must be distinct, otherwise we will try to wire up HCA's on seperate
networks.
This commit was SVN r12978.
* Make sure that the pval always writes to the correct portion of the
lval. This only matters on 32 bit big endian machines.
* On 32 bit machines when assigning to pval, the other 4 bytes of lval
weren't being written, which could lead to bogus data
We use macros so that there aren't casts all over the code and the pval
assignment can occur to the correct 4 bytes. Refs trac:587
This commit was SVN r12974.
The following Trac tickets were found above:
Ticket 587 --> https://svn.open-mpi.org/trac/ompi/ticket/587
the value of __n holds. This is not a problem in the first case
because sizeof(int) == sizeof(MPI_Flogical), so no alloc is actually
performed (which is most compilers, and why we haven't been bitten by
this yet). But the second case -- where sizeof(int) !=
sizeof(MPI_Flogical) -- is definitely a problem and needs the "+1" in
the alloc, or Bad Things will happen.
This commit was SVN r12953.
The following SVN revision numbers were found above:
r12712 --> open-mpi/ompi@3e11c76d4c
is allocated on a per comm_world instance, with the lowest rank
in comm_world on the given host creating and initializing the file,
and then notifying the remaining files via the OOB.
Reviewed: Ralph Castain, Brian Barrett
Addressing ticket #674.
This commit was SVN r12949.
completion of the RDMA operation associated with the fragment. The
PML will call the BML free which in turn will call the BTL free. The MX
BTL will not release the fragment if it not tagged with 0xff.
This commit was SVN r12947.
* Added Create_errhandler for MPI::File
* Make errors_throw_exceptions a first-class predefined exception
handler, and make it work for Comm, File, and Win
* Deal with error handlers and attributes for Files, Types, and Wins
like we do with Comms - can't just cast the callbacks from C++
signatures to C signatures. Callbacks will then fire with the
C object, not the C++ object. That's bad.
Refs trac:455
This commit was SVN r12945.
The following Trac tickets were found above:
Ticket 455 --> https://svn.open-mpi.org/trac/ompi/ticket/455
* Add line about heterogeneous support to ompi_info output
* Print warning and abort if heterogeneous detected and
no heterogeneous support available.
Refs trac:587
This commit was SVN r12943.
The following Trac tickets were found above:
Ticket 587 --> https://svn.open-mpi.org/trac/ompi/ticket/587
inuse decrement for the increment that was at the start of the procs loop.
Otherwise, the inuse count can end up higher than it actually is and a btl
can end up in the progress loop when it isn't active to any peer.
Refs trac:543
This commit was SVN r12938.
The following Trac tickets were found above:
Ticket 543 --> https://svn.open-mpi.org/trac/ompi/ticket/543
the data was buffered by the MX library. If it's the case then we declare
the send as completed and disable the completion event for the mx request.
This commit was SVN r12935.
protocol over the MX BTL. Now, we have only one matching, the one in Open
MPI.
The problem is that when the unexpected handler is triggered, not all the
message is on the host memory. In the best case we get one MX fragment (internal
MX fragment), in the worst we get NULL. The only way to fit this with the
design of the PML is to force the eager protocol at the MX internal fragment
size, and to limit the send/receive protocol at the same size. Tests show
the outcome is not far from optimal (if the pipeline depth is increased
a little bit).
Set MX_PIPELINE_LOG in order to allow MX to use internal fragments of 4K.
This commit was SVN r12930.
performance on a 2G Myrinet card, as it look like pipelining the messages
by 1M is faster than a simple send/receive. However, when using a 10G card
the send/receive will limit the maximum bandwidth to 2.5Gbs. The reason is
the scarce bus resources that have to be shared between the Myrinet hardware
and the memcpy operation. The PUT protocol remove the memcpy, we now have a
true zero-copy mechanism. But, there is no pipelining yet as it look like the
RDMA pipeline somehow disappeared from the OB1 PML ...
This commit was SVN r12925.
add_error_class and add_error_code files. Also fixed the update of the
lastusedcode attribute, all of work according to my tests pretty fine.
Please note: the testcode attached to the bug 683 still reports some bugs. I
am however pretty sure that the testcode is wrong at that points:
- the standard says that the attribute MPI_LASTUSEDCODE has to be updated for
a new error_class or a new error_code. The test currently assumes, that only
the add_error_code call changes the attribute value.
- you have to comment out the two lines 73 and 74 in order to make the
test finish, since these lines check for the error string of non-existent
codes.
- line 126 the error-string of MPI_ERR_ARG is not "invalid argument" but a
little bit more, so the test thinks the output is wrong. So probably the test
has to be update to match the according error string of MPI_ERR_ARG.
Fixes trac:682
This commit was SVN r12913.
The following Trac tickets were found above:
Ticket 682 --> https://svn.open-mpi.org/trac/ompi/ticket/682
It contains four algorithms:
Bruck (ciel(logP) steps), Recursive Doubling (log(P) for power-of-2 processes), Ring (P-1 steps),
and Neighbor Exchange (P/2 steps for even number of processes).
All algorithms passed occ, IMB-2.3, and intel verification tests from ompi-tests/ for up to 56 processes.
The fixed decision function is based on results collected over MX on the Grig cluster at
the University of Tennessee at Knoxville.
I have also added (and commented out) copy of MPICH2 decision function for allgather
(from their IJHPCA 2005 paper).
This commit was SVN r12910.
which can cause segfaults on shutdown. Calling mx_finalize() isn't enough
to shutdown the thread, so must close endpoints as well.
Refs trac:513
This commit was SVN r12908.
The following Trac tickets were found above:
Ticket 513 --> https://svn.open-mpi.org/trac/ompi/ticket/513
if the remote architecture differs from the local architecture and the
btl doesn't support heterogeneous transport.
Refs trac:587
This commit was SVN r12879.
The following Trac tickets were found above:
Ticket 587 --> https://svn.open-mpi.org/trac/ompi/ticket/587
udapl/openib/vapi/gm mpools a deprecated. rdma mpool has parameter that allows
to limit its size mpool_rdma_rcache_size_limit (default is 0 - unlimited).
This commit was SVN r12878.
process when creating a datatype from an internal description.
Refs trac:640
This commit was SVN r12877.
The following Trac tickets were found above:
Ticket 640 --> https://svn.open-mpi.org/trac/ompi/ticket/640
Move the req_mtl structure back to the end of each of the structures in
the CM PML. The req_mtl structure is cast into a mtl_*_request_structure
for each MTL, which is larger than the req_mtl itself. The cast will cause
the *_request to overwrite parts of the heavy requests if the req_mtl
isn't the *LAST* thing on each structure (hence the comment). This was
moved as an optimization at some point, which caused buffer sends to fail...
Refs trac:669
This commit was SVN r12873.
The following SVN revision numbers were found above:
r12871 --> open-mpi/ompi@597598b712
The following Trac tickets were found above:
Ticket 669 --> https://svn.open-mpi.org/trac/ompi/ticket/669
CM PML. The req_mtl structure is cast into a mtl_*_request_structure for
each MTL, which is larger than the req_mtl itself. The cast will cause
the *_request to overwrite parts of the heavy requests if the req_mtl
isn't the *LAST* thing on each structure (hence the comment). This was
moved as an optimization at some point, which caused buffer sends to
fail...
Refs trac:669
This commit was SVN r12871.
The following Trac tickets were found above:
Ticket 669 --> https://svn.open-mpi.org/trac/ompi/ticket/669
Add ability for ini files to recognize "use_eager_rdma" flag. Set the
default to "no" (because we should assume that HCAs cannot support the
property necessary for using RDMA for eager messages -- that the last
byte of the message is guaranteed to be written to memory last --
unless proven otherwise. For example, iWARP cards apparently do not
provide this guarantee), and then set all Mellanox and IBM HCAs to
override the default to enable this behavior on these cards.
This commit was SVN r12851.
The following Trac tickets were found above:
Ticket 366 --> https://svn.open-mpi.org/trac/ompi/ticket/366
I found only two places that were looking at the tokens:
1. the odls - we used the tokens to separately process the globals container data from everything else. In this case, I left the subscription that returned the globals data alone, but "stripped" the subscription that returned the launch data for the procs. These subscriptions have nothing to do with the xcast message.
2. the pml_base_modex - the callback function was getting process names from the returned tokens. Actually, this function was doing a very bad thing - it was assuming that the first token returned was *always* the process name. This is currently true, but is one of those assumptions that someone could have easily changed - and suddenly found the system inexplicably failing. I modified the function to (a) get the name sent back to us, (b) "stripped" the value structures of tokens and segment strings, and (c) correctly obtained process names from the returned values. I also reindented the heck out of the code so it was legible (at least, to my old eyes).
This commit was SVN r12813.
This commit fixes several aspects regarding MPI conformance of requests.
* Eliminate the last argument of ompi_errhandler_request_invoke(); we
''always'' want to invoke the back-end exception handler with the
real error code.
* Make it clear in comments that we only invoke the ''first''
exception in a given array of requests, even if there's more than
one request with a non-MPI_SUCCESS value for MPI_ERROR.
* Defer the freeing of requests upon exception in the back-end
functions to MPI_WAIT* and MPI_TEST* until later; the requests are
kept so that we know what handler to invoke when we actually invoke
the exception. After figuring that out, ''then'' we free requests
with pending exceptions on them.
* Clean up return codes from the back-end MPI_TEST* and MPI_WAIT*
functions.
* Slightly modify ompi_errcode_get_mpi_code() to return unity if it
receives an MPI error code (vs. an OMPI error code).
This commit was SVN r12810.
The following Trac tickets were found above:
Ticket 659 --> https://svn.open-mpi.org/trac/ompi/ticket/659
usually is ok on little-endian systems, as the upper 32 bits will likely
be ignored, but on 32-bit big-endian systems, lval is complete junk.
Use ival if 32 bit mode, lval if 64.
Mixing of 32 and 64 bit architectures won't work without more changes.
This commit was SVN r12802.
* Do not add new procs to the global list during modex callback or
when sharing orte names during accept/connect. For modex, we
cache the modex info for later, in case that proc ever does get
added to the global proc list. For accept/connect orte name
exchange between the roots, we only need the orte name, so no
need to add a proc structure anyway. The procs will be added
to the global process list during the proc exchange later in
the wireup process
* Rename proc_get_namebuf and proc_get_proclist to proc_pack
and proc_unpack and extend them to include all information
needed to build that proc struct on a remote node (which
includes ORTE name, architecture, and hostname). Change
unpack to call pml_add_procs for the entire list of new
procs at once, rather than one at a time.
* Remove ompi_proc_find_and_add from the public proc
interface and make it a private function. This function
would add a half-created proc to the global proc list, so
making it harder to call is a good thing.
This means that there's only two ways to add new procs into the global proc list at this time: During MPI_INIT via the call to ompi_proc_init, where my job is added to the list and via ompi_proc_unpack using a buffer from a packed proc list sent to us by someone else. Currently, this is enough to implement MPI semantics. We can extend the interface more if we like, but that may require HNP communication to get the remote proc information and I wanted to avoid that if at all possible.
Refs trac:564
This commit was SVN r12798.
The following Trac tickets were found above:
Ticket 564 --> https://svn.open-mpi.org/trac/ompi/ticket/564
* don't load data into a buffer until we have the data, as
the data contains some header information needed to
properly load the data
This commit was SVN r12792.
Obviously, people like bproc will have to get the app_num via another avenue...but that's a problem for another day. Several options are easily available.
This commit was SVN r12788.
- we have to be able to attach a string to an error class, not just to an
error code
- according to MPI-2 the attribute MPI_LASTUSEDCODE has to be updated
everytime you add a new code or a new class. Thus, you have to have single
list for both.
Thus, we got rid of the error_class structure. In the error-code structure, we
can distinguish whether we are dealing with an error code or an error class by
looking at the err->code element of the structure. In case its value is
MPI_UNDEFINED, the according entry is a class, else it is an error code. All
predefined error codes have the code and the class field set to the same
value.
The test MPI_Add_error_class1 passes now.
Fixes trac:418
This commit was SVN r12764.
The following Trac tickets were found above:
Ticket 418 --> https://svn.open-mpi.org/trac/ompi/ticket/418
r12714) for supporting compilers / architectures with different
padding rules.
This commit was SVN r12749.
The following SVN revisions from the original message are invalid or
inconsistent and therefore were not cross-referenced:
r12491
r12714
hits the buffer on the other side. For this kind of BTLs we need to send
FIN through the same BTL, PUT was performed with so network will handle
ordering for us. If we will use another BTL, receiver can get FIN before
data will hit the buffer and complete request prematurely. We mark such
problematic BTLs with MCA_BTL_FLAGS_FAKE_RDMA flag (this kind of RDMA
is really fake, because the real one guaranties that sender will see the
completion only after receiver's NIC confirmed that all the data was
received).
This commit was SVN r12732.
It calls mca_pml_ob1_send_fin_btl() which may fail and doesn't check return
code. This breaks all RDMA transports event when only one BTL is used. Revert
it for now, I am working on a real fix for the problem (I hope).
This commit was SVN r12731.
The following SVN revision numbers were found above:
r12720 --> open-mpi/ompi@3e3689320b
regresion from v1.1 was reviewed and put to v1.2 branch. So revert this part
of r12721 back.
This commit was SVN r12730.
The following SVN revision numbers were found above:
r12433 --> open-mpi/ompi@82f7c0dd69
r12721 --> open-mpi/ompi@3edd850d2e
protocol when multiple NICS are available between 2 peers. The fix force
the FIN message to take exactly the same path as the fragment it describe
(i.e. same path means same BTL). Otherwise, the FIN can be received by
the peer before the RDMA complete and the request will get freed
too early.
This commit was SVN r12720.
- consistent error message when something fails (via BTL_ERROR macro)
- decrease the number of jumps.
- cleanup some parts of the code.
This commit was SVN r12719.
* Always invoke the error handler on MPI_COMM_WORLD for
invalid windows (except in win_create, which should
instead be on the given communicator).
* Allow get_errhandler in addition to set_errhandler
on MPI_WIN_NULL
Refs trac:647
This commit was SVN r12718.
The following Trac tickets were found above:
Ticket 647 --> https://svn.open-mpi.org/trac/ompi/ticket/647
The temporary solution is to switch into EV_NONBLOCK mode earlier (right after the mx_connect loop) so that there isn't a giant slowdown when processes enter the stage gate 2 barrier before other proesses. They will now not block in the event library for any period of time, which appears to have a 50% speedup when running at > 64 procs.
Refs trac:645
This commit was SVN r12713.
The following Trac tickets were found above:
Ticket 645 --> https://svn.open-mpi.org/trac/ompi/ticket/645
OMPI_ARRAY_INT_2_LOGICAL had an array bounds error - fixed this and the
analogous error in OMPI_ARRAY_LOGICAL_2_INT.
This commit was SVN r12712.
The following Trac tickets were found above:
Ticket 482 --> https://svn.open-mpi.org/trac/ompi/ticket/482
case where sizeof(INTEGER) > sizeof(int).
This commit was SVN r12707.
The following SVN revision numbers were found above:
r12684 --> open-mpi/ompi@e2c605f32a
so this isn't an issue there either. Refs trac:488
This commit was SVN r12675.
The following Trac tickets were found above:
Ticket 488 --> https://svn.open-mpi.org/trac/ompi/ticket/488
* Fix a counter roll-over issue that could result from a large (but
not excessive) number of outstanding put/get/accumulate calls
during a single synchronization issues (Refs trac:506)
* Fix epoch issue with rdma component that would effect PWSC
synchronization (Refs trac:507)
This commit was SVN r12673.
The following Trac tickets were found above:
Ticket 506 --> https://svn.open-mpi.org/trac/ompi/ticket/506
Ticket 507 --> https://svn.open-mpi.org/trac/ompi/ticket/507
* use one-sided datatype check instead of send/receive and check both
the origin and target datatypes
* allow error handler to be set on MPI_WIN_NULL, per standard
* Allow recursive calls into the pt2pt osc component's progress
function
* Fix an uninitialized variable problem in the unlock header
This commit was SVN r12667.
because they are in ORTE, not OMPI. Also, remove the ORTE_PROCESS_NAME macros
in iof base as they are duplicates of the ones that were in ns_types, which
meant that bad things happened if you changed what an orte_process_name_t
looked like.
This commit was SVN r12646.
the same time, remove some of the MPI-related options from OPAL:
- provide mechanism to change at runtime whether sched_yield() should
be called when the progress engine is idle
- provide mechanism for changing the rate at which the event engine
is called when there are "no" users of the event engine (ie, when
using MPI but not TCP)
- fix some function names in the progress engine to better match
their intended use (and remove MPI naming scheme)
- remove progress_mpi_enable / progress_mpi_disable because
we can now use the functions to set the sched_yield and
tick rate interfaces
- rename opal_progress_events() to opal_progress_set_event_flag()
because the first really isn't descriptive of what the function
does and I always got confused by it
This commit was SVN r12645.
Ensure that the new predefined MPI-2 attribute callback functions take
the proper types (INTEGER, kind=MPI_ADDRESS_KIND instead of just
INTEGER).
This commit was SVN r12639.
The following Trac tickets were found above:
Ticket 624 --> https://svn.open-mpi.org/trac/ompi/ticket/624
Accordingly, there are new APIs to the name service to support the ability to get a job's parent, root, immediate children, and all its descendants. In addition, the terminate_job, terminate_orted, and signal_job APIs for the PLS have been modified to accept attributes that define the extent of their actions. For example, doing a "terminate_job" with an attribute of ORTE_NS_INCLUDE_DESCENDANTS will terminate the given jobid AND all jobs that descended from it.
I have tested this capability on a MacBook under rsh, Odin under SLURM, and LANL's Flash (bproc). It worked successfully on non-MPI jobs (both simple and including a spawn), and MPI jobs (again, both simple and with a spawn).
This commit was SVN r12597.
Same sort of problem and fix as described in r12323 - mca_pml_ob1_recv_frag_progress() was segfaulting due to a NULL req_proc pointer. The path leading to this was through the mca_pml_ob1_check_cantmatch_for_match() function, where we can match a frag using the same macros as mca_pml_ob1_frag_match() and never initialize the req_proc pointer.
This commit was SVN r12582.
The following SVN revision numbers were found above:
r12323 --> open-mpi/ompi@c752502dee
The fix is to set opcode to SEND at the entrance to the send function before
checking credits and putting fragment to the pending list. We do the same thing
in put/get functions i.e setting opcode at the entrance to the function.
This commit was SVN r12559.
1. Fix the "hang" condition when an application isn't found. It turned out that the ODLS had some difficulty with the process actually not having been started - hence, it never called the waitpid callback. As a result, the "terminated" trigger didn't fire, and so mpirun didn't wake up. With this change, the HNP's errmgr forces the issue by causing the trigger to fire itself when an abort condition occurs.
2. Shift the recording of the pid and the nodename from mpi_init to the orted launcher. This allows programs such as Eclipse PTP to get the pids even for non-MPI applications. In the case of bproc, the pls handles this chore since we don't use orteds in that system.
This commit was SVN r12558.
- consistent arguments checking (not allowing to select an algorithm which
is not available)
- consistent way of computing the segcount (number of datatypes by segment).
- small cleanups.
- more informative debugging messages.
This commit was SVN r12545.
description. Most of the bcast algorithms can be completed using this
generic function once we create the tree structure. Add all kind of
trees.
There are 2 versions of the generic bcast function. One using overlapping
between receives (for intermediary nodes) and then blocking sends to all
childs and another where all sends are non blocking. I still have to
figure out which one give the smallest overhead.
This commit was SVN r12530.
is that if one add "pml=" to the configuration file, really bad things
happen. All PMLs will get initialize, and each of them will initialize
all BTLs. This patch force the mca_pml_base_pml to get initialized in
all cases before we go out of the mca_pml_base_open function.
This commit was SVN r12527.
(all versions up to and including 20060925). The issue has been
reported to Intel, along with a small [non-MPI] test program that
reproduces the problem (the test program and the OMPI C++ bindings
work fine with Intel C++ 9.0 and many other C++ compilers).
In short, a static initializer for a global variable (i.e., its
constructor is fired before main()) that takes as an argument a
reference to a typedef'd type will simply get the wrong value in the
argument. Specifically:
{{{
namespace MPI {
Intracomm COMM_WORLD(MPI_COMM_WORLD);
}
}}}
The constructor for MPI::Intracomm should get the value of
&ompi_mpi_comm_world. It does not; it seems to get a random value.
As mandated by MPI-2, annex B.13.4, for C/C++ interoperability, the
prototype for this constructor is:
{{{
class Intracomm {
public:
Intracomm(const MPI_Comm& data);
};
}}}
Experiments with icpc 9.1/20060925 have shown that removing the
reference from the prototype makes it work (!). After lots of
discussions about this issue with a C++ expert (Doug Gregor from IU),
we decided the following (cut-n-paste from an e-mail):
-----
> So here's my question: given that OMPI's MPI_<CLASS> types are all
> pointers, is there any legal MPI program that adheres to the above
> bindings that would fail to compile or work properly if we simply
> removed the "&" from the second binding, above?
I don't know of any way that a program could detect this change. FWIW,
the C++ committee has agreed that implementation of the C++ standard
library are allowed to decide arbitrarily between const& and by-value.
If they don't care, MPI users won't care.
When you remove the '&', I suggest also removing the "const". It is
redundant, but can trigger some strange name mangling in Sun's C++
compiler.
-----
So with this change:
* we now work again with the Intel 9.1 compiler
* our C++ bindings do not exactly conform to the MPI-2 spec, but
valid/legal MPI C++ apps cannot tell the difference (i.e., the
functionality is the same)
This commit was SVN r12514.
r12486 but somehow I miss it.
Update the pack and unpack functions for contigusous datatypes to minimize their
impact on the performance. Keep them as condensed as possible.
This commit was SVN r12513.
The following SVN revision numbers were found above:
r12486 --> open-mpi/ompi@8746369338
usefull things are now done without the call to any external function,
directly in the macros in this file.
Create a new flag for the convertor to help figure out when we have
to do anything special for pack/unpack. A convertor with the flag
CONVERTOR_NO_OP is a basic convertor, designed for contiguous types.
This commit was SVN r12485.
set it up before the match when we know the peer, saving some
time on the critical path. If the receive is ANY_SOURCE then
we initialize the convertor on _MATCHED. Anyway, we will set it
up only once per receive.
This commit was SVN r12484.
N gatherv's:
for (i = 0 ... size)
MPI_Gatherv(..., root = i, ...)
The new algorithm simply does (effectively):
MPI_Gatherv(..., root = 0, ...)
MPI_Bcast(..., root = 0, ...)
This commit was SVN r12469.
* Add MPI::Status methods Set_elements() and Set_cancelled()
* Added a bunch of comments in various places in the MPI C++ bindings
implementatio just to explain what's going on (because C++ can hide
a lot from you)
* Insert C++ callbacks for the MPI_Grequest callback functions
registered by MPI::Grequest::Start(). These callbacks keep a
little meta-data (created by Grequest::Start()) that allow the
proper callback signatures from C (i.e., from ompi_grequest_<foo>()
in libmpi.a -- C code), translate arguments as required, and then
invoke the callbacks with proper C++ signatures (i.e., call
user-defined callbacks with C++ function signatures).
This commit was SVN r12446.
The following Trac tickets were found above:
Ticket 580 --> https://svn.open-mpi.org/trac/ompi/ticket/580
1. Added reporting points around the xcasts in MPI_Init. Note that these times will include time spent waiting for a trigger to fire, which is why the times between stage gates did NOT include these times initially. The inter-stage-gate times still do NOT include the xcast time - the xcast time is reported separately.
2. Added the process vpid on the MPI_Init timing reports for clarity.
3. Added a report from the xcast function on the HNP that outputs the number of bytes in the message being sent to the processes.
This commit was SVN r12422.
size of the complex type as determined by configure; not the size of
the next larger complex type (i.e., a complex*N is 2 real*(N/2)'s, not
2 real*N's).
This commit was SVN r12421.
* Add some more error checking to GREQUEST_START
* Move the error checking in GREQUEST_COMPLETE up to inside the
MPI_PARAM_CHECK block, where it belongs
* Invoke the gen request query_fn in all the Right spots (per MPI-2:8.2)
* Distinguish between grequests created from C and Fortran
* Use the OBJ system to reference count to release the grequest at
the Right time and invoke the grequest free_fn properly (see
lengthy comment in grequest.c above the destructor)
* Have ompi_grequest_complete() call ompi_request_complete() rather
than [poorly] copy the contents of ompi_request_complete()
* Fix Fortran function callback pointer typedefs to use proper
Fortran types
* Edit ompi_request_test* and ompi_request_wait* to properly handle
generalized requests. This adds an "if" statement in the critical
path for all the back-end test* and wait* functions :-(,
but fortunately George took out two "if" statements from the
critical path last week. So we're still ahead. :-)
* Move ompi_request_test() out of request.h and into request.c (all
other test* and wait* functions were already in the .c file -- and
ompi_request_test() was too long to be statically inlined anyway)
This commit was SVN r12402.
The following Trac tickets were found above:
Ticket 496 --> https://svn.open-mpi.org/trac/ompi/ticket/496
Had group discussion about this on the weekly call. The decision was
that we should pass the real error code to the back-end exception
handler because it's pretty useless to pass MPI_ERR_IN_STATUS to the
back-end exception handler (because exception handlers don't have
access to the request or the status - this has potential issues for
fault tolerance kinds of scenarios). So in TESTALL, TESTSOME,
WAITALL, and WAITSOME, we examine the error code and if it's not
MPI_SUCCESS, return MPI_ERR_IN_STATUS.
This commit was SVN r12389.
The following Trac tickets were found above:
Ticket 549 --> https://svn.open-mpi.org/trac/ompi/ticket/549
mca_btl_openib_endpoint_connect_eager_rdma() is called recursively. He also
noticed that orte_pointer_array_add() can't fail because we allocate max number
of elements at init time. So just remove error handling and locking. No locking
- no deadlocks.
This commit was SVN r12388.
something is going wrong down in the code it is removed from the array. So add
mutex to prevent concurrent access to the array from different threads.
This commit was SVN r12385.
Setup subscriptions to correctly return the MPI_APPNUM attribute.
Fix an unreported bug that was found. The universe size was incorrectly defined in the attributes code. As coded, it looked for size_t values and based its size computation on those numbers. Unfortunately, the node_slots value had been changed to an orte_std_cntr_t awhile back! So the universe size was never updated.
Update the hello_nodename test to check for MPI_APPNUM.
Add a definition to ns_types for ORTE_PROC_MY_NAME - just a shortcut for orte_process_info.my_name. Brought over from ORTE 2.0 as it will be used extensively there.
This commit was SVN r12377.
What's happening is that we're holding openib_btl->eager_rdma_lock when
we call mca_btl_openib_endpoint_send_eager_rdma() on
btl_openib_endpoint.c:1227. This in turn calls
mca_btl_openib_endpoint_send() on line 1179. Then, if the endpoint
state isn't MCA_BTL_IB_CONNECTED or MCA_BTL_IB_FAILED, we call
opal_progress(), where we eventually try to lock
openib_btl->eager_rdma_lock at btl_openib_component.c:997.
The fix removes this lock altogether. Instead we atomically set local RDMA
pointer to prevent other threads to create rdma buffer for the same endpoint.
And we increment eager_rdma_buffers_count atomically thus polling thread doesn't
need lock around it.
This commit was SVN r12369.
Have no idea why this function always returns a failure. It should
always return SUCCESS (provided the status is value).
This commit was SVN r12364.
The following Trac tickets were found above:
Ticket 496 --> https://svn.open-mpi.org/trac/ompi/ticket/496
* For MPI_TEST, MPI_TESTANY, MPI_WAIT, and MPI_WAITANY (i.e., the
TEST/WAIT functions that return up to exactly one completed
request), return the actual error code.
* For MPI_TESTALL, MPI_TESTSOME, MPI_WAITALL, MPI_WAITSOME, (i.e.,
the TEST/WAIT functions that can return more than one completed
request), return MPI_ERR_IN_STATUS.
This commit was SVN r12355.
The following Trac tickets were found above:
Ticket 549 --> https://svn.open-mpi.org/trac/ompi/ticket/549
This commit essentially caches the invoking comm/win/file on the
ompi_request_t. This, paired with the req_type field, allows us to
retrieve the invoking MPI object and invoke the proper errhandler.
The patch is missing most updates for the MPI-2 one-sided stuff (i.e.,
the patch mainly fixes comms and files); I didn't really understand
that code and didn't want to hazard trying to figure it out when Brian
can probably do it much more quickly.
So #250 will still stay open, pending MPI-2 one-sided updates for this
stuff.
This commit was SVN r12339.
The following Trac tickets were found above:
Ticket 250 --> https://svn.open-mpi.org/trac/ompi/ticket/250
* Create a new request type: NOOP (described below)
* For all MPI_*_INIT functions, OBJ_NEW an ompi_request_t and set its
type to NOOP
* Ensure that the NOOP requests are OBJ_RELEASE'd when they are done
* MPI_START looks at the request type; if NOOP, just return success. If
not, call the PML start() function
* MPI_STARTALL always pass the entire array of requests back to the PML
(see next point)
* Make the PMLs only process PML requests (i.e., ignore/skip anything
that isn't of type PML -- such as the NOOP requests)
* Add a little more param error checking in STARTALL
This commit was SVN r12338.
The following Trac tickets were found above:
Ticket 529 --> https://svn.open-mpi.org/trac/ompi/ticket/529
* Remove an extra OMPI_REQUEST_INIT() from the grequest constructor
(it was already invoked by the parent MPI_Request constructor)
* Set the state of the generalized request to ACTIVE (because this is
invoked from MPI_GREQUEST_START -- analogous to MPI_START)
* Before invoking the query function in MPI_REQUEST_COMPLETE, set the
status on the base request to ompi_status_empty. This gives a set
of default values for the request, including one for
status.MPI_ERROR = MPI_SUCCESS (because we check the value of
MPI_ERROR in MPI_TEST* and MPI_WAIT* processing, and use it to
determine whether the upper-level API call should raise an MPI
exception or not).
This commit was SVN r12337.
The following Trac tickets were found above:
Ticket 532 --> https://svn.open-mpi.org/trac/ompi/ticket/532
allocation logic is completely done outside the data-type engine (in the PML) there is
no need for any special case inside the data-type engine. There is less arguments for
the ompi_convertor_pack and ompi_convertor_unpack as well (the last field free_after is
not required anymore as there is no memory allocated in the engine itself). This change
affect all components using datatypes. I test most of them, but it might happens that I
miss some ... If it's the case please let me know (don't shoot the pianist!!).
This commit was SVN r12331.
the default decision functions (for broadcast, reduce and barrier) are based on a
high performance network (not TCP). It should give good performance (really good) for
any network having the following caracteristics: small latency (5 microseconds) and good
bandwidth (more than 1Gb/s).
+ Cleanup of the reduce algorithms, plus 2 new algorithms (binary and binomial). Now most
of the reduce algorithms use a generic tree based function for completing the reduce.
+ Added macros for computing the trees (they are used for bcast and reduce right now).
+ Allow the usage of all 5 topologies.
+ Jelena's implementation of a binary tree that can be used for non commutative operations.
Right now only the tree building function is there, it will get activated soon.
+ Some others minor cleanups.
This commit was SVN r12326.
A segfault would occur in mca_pml_ob1_recv_request_progress() when trying to prepare the convertor for unpacking, because the request's req_proc field was NULL.
Turns out that we weren't setting the req_proc field in the MCA_PML_OB1_CHECK_SPECIFIC_AND_WILD_RECEIVES_FOR_MATCH macro. Instead of just setting it there I removed the other place req_proc was being set correctly, and instead took care of all the cases at once in mca_pml_ob1_recv_frag_match().
This commit was SVN r12323.
parameter. For optimisation purpose only this BTL is used to send packet
through instead of trying to send packets through all BTLs. But actually the
code was wrong. It simply used provided bml_btl and it may represent different
endpoint from packet's destination. The fixed code checks if packet's
destination is reachable through the BTL, finds appropriate bml_btl and only
then tries to send it through correct bml_btl.
This commit was SVN r12319.
is done to assure alignment so strictly aligned CPUs (like SPARC) do not
sigbus. This also may benefit other platforms too.
This commit fixes trac:494.
This commit was SVN r12312.
The following Trac tickets were found above:
Ticket 494 --> https://svn.open-mpi.org/trac/ompi/ticket/494
If you want to look at our launch and MPI process startup times, you can do so with two MCA params:
OMPI_MCA_orte_timing: set it to anything non-zero and you will get the launch time for different steps in the job launch procedure. The degree of detail depends on the launch environment. rsh will provide you with the average, min, and max launch time for the daemons. SLURM block launches the daemon, so you only get the time to launch the daemons and the total time to launch the job. Ditto for bproc. TM looks more like rsh. Only those four environments are currently supported - anyone interested in extending this capability to other environs is welcome to do so. In all cases, you also get the time to setup the job for launch.
OMPI_MCA_ompi_timing: set it to anything non-zero and you will get the time for mpi_init to reach the compound registry command, the time to execute that command, the time to go from our stage1 barrier to the stage2 barrier, and the time to go from the stage2 barrier to the end of mpi_init. This will be output for each process, so you'll have to compile any statistics on your own. Note: if someone develops a nice parser to do so, it would be really appreciated if you could/would share!
This commit was SVN r12302.
"this is bogus" kind of answer. Passing in bad error codes should
only happen in erroneous sections of the OMPI code base, but still,
it's far more social to print a message saying, "hey, you messed up!"
rather than seg faulting.
Reviewed by Edgar.
This commit was SVN r12295.
mentioned in the comment the completion/callback of the triggered
send operation can happen before the call returns. If this happens and
if the pipeline depth is 0 before we triggered the send operation and
this is the last send operation of the request then the completion detection
code will decrement the pipeline depth and check it for equality to 0.
Because (0-1) != 0 the pml completion function for this request will
*not* be called.
This part 2 of the fix for ticket #246.
This commit was SVN r12292.
MPI_PROC_NULL translates to MPI_PROC_NULL, and an MPI_GROUP_EMPTY as one of
the groups doesn't cause a segmentation fault, but returns MPI_UNDEFINED for
all ranks to be translated.
This commit was SVN r12233.
possible things contained in the conditional like other rules are (for
example, a SOURCES rule in a conditional automatically has its files
added to the dist rules, even if that conditional isn't tru when
make dist occurs). So the man files weren't in the tarball.
Put the EXTRA_DIST with the files explicitly listed outside any conditionals
so the man pages always end up in the tarball.
This commit was SVN r12220.
all platforms. The only exceptions (and I will not deal with them
anytime soon) are on Windows:
- the write functions which require the length to be an int when it's
a size_t on all UNIX variants.
- all iovec manipulation functions where the iov_len is again an int
when it's a size_t on most of the UNIXes.
As these only happens on Windows, so I think we're set for now :)
This commit was SVN r12215.
provide one for the internals of Open MPI. For mpi.h, typedef MPI_Aint
either to ptrdiff_t or whatever we used as ptrdiff_t if that type doesn't
actually exist.
This commit was SVN r12212.
The following SVN revision numbers were found above:
r12146 --> open-mpi/ompi@8852c00c36
they might require special tools (not sure if sed with multiple -e
arguments is totally portable)
- ignore the opalcc.1 man page. Couldn't do this in the previous
man page commit (r12192) because I was removing opalcc.1 in that
commit.
This commit was SVN r12194.
The following SVN revision numbers were found above:
r12192 --> open-mpi/ompi@581a4b0a4e
- Only install opal{cc,c++} and orte{cc,c++} if configured with
--with-devel-headers. Right now, they are always installed, but
there are no header files installed for either project, so there's
really not much way for a user to actually compile an OPAL / ORTE
application.
- Drop support for opalCC and orteCC. It's a pain to setup all the
symlinks (indeed, they are currently done wrong for opalCC) and
there's no history like there is for mpiCC.
- Change what is currently opalcc.1 to opal_wrapper.1 and add some
macros that get sed'ed so that the man pages appear to be
customized for the given command.
- Install the wrapper data files even if we compiled with
--disable-binaries. This is for the use case of doing multi-lib
builds, where one word size will only have the library built, but
we need both set of wrapper data files to piece together to
activate the multi-lib support in the wrapper compilers.
This commit was SVN r12192.
Modify the mapper to better bookmark its stopping place each time, and to pick up the next time from there. This needs to be validated on a multi-node system.
Fix a major memory corruption problem in the registry put/get functions that was doing multiple free's. Not sure how valgrind missed this one, though it only occurred in specific circumstances (such as comm_spawn).
This commit was SVN r12179.
This change does a couple of things:
1. Since the USE_PARENT_ALLOC attribute is a directive about regarding allocation of resources to a job, it more properly should be an attribute of the RAS. Change the name to reflect that and move the attribute define to the ras_types.h file.
2. Add the attributes list to the RMAPS map_job interface. This provides us with the desired flexibility to dynamically specify directives for mapping. The system will - in the absence of any attribute-based directive - default to the values provided in the MCA parameters (either from environment or command-line interface).
This commit was SVN r12164.
size and diplacement of data-type. After this patch all data can contain size_t bytes
and the displacements are defined as ptrdiff_t. All of the files I was able to compile
have been modified to match this requirement.
This commit was SVN r12146.
The UD BTL isn't gone - the latest version is in my afriedle-ud branch. This version on the trunk was very old, ompi_ignore'd, lacked performance, and probably contained bugs. The maintained version on my branch is working solid, and will eventually come back, but not for v1.2.
This commit was SVN r12144.
Fix the problem observed by multiple people that comm_spawned children were (once again) being mapped onto the same nodes as their parents. This was caused by going through the RAS a second time, thus overwriting the mapper's bookkeeping that told RMAPS where it had left off.
To solve this - and to continue moving forward on the ORTE development - we introduce the concept of attributes to control the behavior of the RM frameworks. I defined the attributes and a list of attributes as new ORTE data types to make it easier for people to pass them around (since they are now fundamental to the system, and therefore we will be packing and unpacking them frequently). Thus, all the functions to manipulate attributes can be implemented and debugged in one place.
I used those capabilities in two places:
1. Added an attribute list to the rmgr.spawn interface.
2. Added an attribute list to the ras.allocate interface. At the moment, the only attribute I modified the various RAS components to recognize is the USE_PARENT_ALLOCATION one (as defined in rmgr_types.h).
So the RAS components now know how to reuse an allocation. I have debugged this under rsh, but it now needs to be tested on a wider set of platforms.
This commit was SVN r12138.
* Update comments in some MPI_FILE_* functions to reflect that the
MPI specs have different page numbers in the ps and pdf (woof!).
* Update comments to say "Retain" where we meant retain (not "return)
* Add a check in MPI_ERRHANDLER_FREE to raise an MPI exception if the
user attempts to free an intrinsic errhandler *and* the refcount is
1 (meaning that it would actually free the intrinsic). This
protects erroneous programs from segv'ing.
* Remove lengthy comment from comm_get_errhandler.c which is no
longer valid (because of the MPI-2 errata that says that users *do*
have to call MPI_ERRHANDLER_FREE).
This commit was SVN r12128.
The following SVN revision numbers were found above:
r12122 --> open-mpi/ompi@407b3cb788
The following Trac tickets were found above:
Ticket 502 --> https://svn.open-mpi.org/trac/ompi/ticket/502
MPI::SEEK_* because iostreams (well, ios_base, but I don't think that
should be included directly) can use SEEK_* as values in an enum, which
means that 'const int' is bad for them.
* Remove now useless comments in the cxx example programs
* include iostream after mpi.h so that our examples work with other MPI
implementations that don't try to be as friendly with the constants.
Refs trac:387
This commit was SVN r12125.
The following Trac tickets were found above:
Ticket 387 --> https://svn.open-mpi.org/trac/ompi/ticket/387
* Fix MPI-2 page number in comments for a specific reference in the
spec
* Allow getting/setting the errhandler on MPI_FILE_NULL
* Allow freeing of intrinsic errhandlers, per MPI-2 errata (if you GET
an errhandler on a communicator, you must be able to FREE it, even
if it's an intrinsic).
Thanks to Lisandro Dalcin for reporting these problems.
This commit was SVN r12122.
some issues with the C #defines SEEK_{SET, END, POS}. The workaround
involves some hackery that should work in almost every common use case
for the C stdio constants (and all the legal issues of the MPI constants).
The one issue is that the C stdio constants are now const ints instead
of #defines, which means that #ifdef checks will fail for the constants.
Behavior can be disabled at either configure time or build time.
Refs trac:387
This commit was SVN r12121.
The following Trac tickets were found above:
Ticket 387 --> https://svn.open-mpi.org/trac/ompi/ticket/387
where a window was in both the passive and active side of a lock sequence.
Refs trac:488
This commit was SVN r12112.
The following Trac tickets were found above:
Ticket 488 --> https://svn.open-mpi.org/trac/ompi/ticket/488
Just follow inc_num and you will understand. Now _resize will grow the list to match
the required number of elements as described in the comment in the .h file.
This commit was SVN r12074.
constrained:
* Make sure we always have a number of eager fragments available
that scales with the number of processes communicating with
a given proc over shared memory
* Use FREE_LIST_GET instead of FREE_LIST_WAIT to return an
error to the PML when resource exhaustion occurs
* Don't dereference the frag during alloc unless we're sure
it's not NULL
Reviewed by: Galen
Refs trac:413
This commit was SVN r12053.
The following Trac tickets were found above:
Ticket 413 --> https://svn.open-mpi.org/trac/ompi/ticket/413
then use broadcast in order to wake them up. If there is only one then use signal
(which is supposed to be faster) and of course if there are no threads
waiting then just continue.
This commit was SVN r12049.
Doing pointer math properly (e.g., incrementing by the right amount)
helps you not overflow buffers, cause random chaos, and contribute to
the heat death of the universe. Sigh.
This commit was SVN r12015.
The following Trac tickets were found above:
Ticket 236 --> https://svn.open-mpi.org/trac/ompi/ticket/236
recvbuf in MPI_GATHER).
* Minor style updates (constants on the left of == and !=)
* Fix a minor buglet that crept in r11904: had a recvbuf where it
should have been recvcount. Thankfully, this would have only
affected erroneous programs. ;-)
This commit was SVN r11980.
The following SVN revision numbers were found above:
r11904 --> open-mpi/ompi@17539dc154
The following Trac tickets were found above:
Ticket 338 --> https://svn.open-mpi.org/trac/ompi/ticket/338
compute the hetero flag a little bit different. The hetero flag is now
attached to a convertor if and only if we can use the optimized conversion
functions. It's a little bit broader than before (the 2 architectures
has to be identical).
This commit was SVN r11962.
First, move the OPAL_THREAD_LOCK out to the same level as its corresponding UNLOCK. It was possible to hit the UNLOCK without ever acquiring the lock.
Since the OPAL_THREAD_ADD64() is now protected by this lock, we can just do the decrement non-atomically.
This commit was SVN r11958.
Don't try to acquire ompi_request_lock here, which in all cases is already held. Avoids deadlock that occurs even when threads are enabled and we're running a THREAD_SINGLE app.
Reviewed by Galen.
This commit was SVN r11957.
The following Trac tickets were found above:
Ticket 183 --> https://svn.open-mpi.org/trac/ompi/ticket/183
conversion function are more complex and costly than a simple memcpy. Therefore,
we want to decrease as much as possible the usage of these functions.
We now check not only th HOMOGENEOUS flag on the datatype or convertor, but the
bits indicating a type is in use. If a communication transfert a type having the
same representation on both peers we can use the optimized version of the conversion.
In same time we build a more accurate conversion table for each master convertor,
based on the minimum differences between the 2 architectures.
This commit was SVN r11945.
remove requirements on .la files on wrapper scripts
Ticket: #374
extend compilers to support 32 bit and 64 bit in one version of the wrapper
Submitted by: Dan Lacher
Reviewed by: Rolf Vandevaart
This commit was SVN r11908.
with the use of MPI_IN_PLACE, and make some optimization checks more
correct. Thanks to Lisandro Dalcin for reporting the problems.
This commit was SVN r11904.
The following Trac tickets were found above:
Ticket 430 --> https://svn.open-mpi.org/trac/ompi/ticket/430
something went wrong. A positiv number > 0 is however a correct value (in
contrary to orte_rml.recv_buffer, which really returns ORTE_SUCCESS or an
error code).
Note: this part of the code is correct on 1.1 and 1.2 branch, no need to move
this change patch to the release branches.
This commit was SVN r11897.
install-exec-hook is not only wrong, it can cause ordering issues such
as trying to put sym links to man pages in directories that do not yet
exist.
This commit was SVN r11893.
I have added a new MCA param (hey, you can't have too many!) called OMPI_MCA_orte_timing. If set to anything other than zero, the system will report out critical timing loops. At the moment, this includes three measurements:
1. Time spent going through the RDS->RAS->RMAPS, setting up triggers, etc. prior to calling the actual PLS launch function. This is reported out as time to setup job.
2. Time spent in MPI_Init from start of that function (well, right after opal_init) to the place where we send all of our info the registry. Reported out as time from start to exec_compound_cmd
3. Time actually spent executing the compound cmd. Reported out as time to exec_compound_cmd.
A few additional timing points will be added shortly.
These may eventually be removed or (better) setup with a conditional compile flag.
This commit was SVN r11892.
I do something else" rule screws me up again. If we're in a FENCE, but
not in ACCESS | EXPOSE, put us in ACCESS|EXPOSE, as we are now known we
now in a real Fence epoch. Yay silly MPI standards
Refs trac:441
This commit was SVN r11865.
The following Trac tickets were found above:
Ticket 441 --> https://svn.open-mpi.org/trac/ompi/ticket/441
Make the Fortran MPI_MAX_DATAREP_STRING follow the same convention as
the rest of the Fortran constants -- be one less than the C constant
of the same name.
This commit was SVN r11842.
The following Trac tickets were found above:
Ticket 389 --> https://svn.open-mpi.org/trac/ompi/ticket/389
groups. And zero is also an acceptable value according to the MPI spec.
Fixes trac:428
This commit was SVN r11841.
The following Trac tickets were found above:
Ticket 428 --> https://svn.open-mpi.org/trac/ompi/ticket/428
Fixes simple off-by-one error in the error check for
MPI_INFO_GET_NTHKEY.
This commit was SVN r11838.
The following Trac tickets were found above:
Ticket 429 --> https://svn.open-mpi.org/trac/ompi/ticket/429
on the number of known local procs, with a high and low watermark. Right
now, we default to a low watermark point of 64MB, a per-proc scaling
factor of 32MB, and a high watermark point of 512MB
Refs trac:212
This commit was SVN r11824.
The following Trac tickets were found above:
Ticket 212 --> https://svn.open-mpi.org/trac/ompi/ticket/212
GIDs (there can be more than one) and not GIDs of the HCA on the network. Entry
zero always have to be initialized so we use it, and warn user if there is more
then one port active and default subnet is configured on at least one of them.
This commit was SVN r11815.
the component is configured successfully. Otherwise, we can end up
trying to run make in the romio directory without any Makefiles. This
really only happens on the targets that recurse into DIST_SUBDIRS - ie
dist, maintainer-clean, and distclean
refs trac:411
This commit was SVN r11807.
The following Trac tickets were found above:
Ticket 411 --> https://svn.open-mpi.org/trac/ompi/ticket/411
Makefile.am's is from a very old Automake bug which has long-since
been fixed. Since we require very recent versions of AM, we don't
need these anymore.
This commit was SVN r11774.
should die, according to the MPI standard. It's possible that the
ORTE layer may kill additional processes, but that's beyond our
control and seems to be allowed by the standard (ie, it might also
end up killing all the procs in all the jobs covered by the
communicator).
* update the stack trace printing code to use the framework rather
than calling execinfo directly, so that we should be able to get
stack traces on all the platforms we support stack tracing on
(if the user wants stack traces on abort, of course)
This commit was SVN r11753.
tell if the remote proc should be in an exposure epoch or not.
Refs trac:325
This commit was SVN r11746.
The following Trac tickets were found above:
Ticket 325 --> https://svn.open-mpi.org/trac/ompi/ticket/325
epoch's control data could overwrite the previous epoch's data because
we were reusing data structures between PW and SC. Instead, we now
have explicit post_msg and complete_msg counters for completion.
refs trac:354
* Only register the rdma osc callback once, as it turns out that some
btls (MX) do somethng more than update a table during the register
call, and each register call sucks up valuable fragments...
This commit was SVN r11745.
The following Trac tickets were found above:
Ticket 354 --> https://svn.open-mpi.org/trac/ompi/ticket/354
long ago) supposed to be used as a cache for accessing the PML procs. But in
all of the PMLs the PML proc contain only one field i.e. a pointer to the ompi_proc.
This pointer can be accessed using the c_remote_group easily. Therefore, there is no
meaning of keeping the PML procs around. Slim fast commit ...
This commit was SVN r11730.
