ports/openmpi
ports/openmpi
1
1
Форкнуть 0
Код Релизы Активность
7 516 коммитов 3 Ветки 0 Теги
Граф коммитов

7516 Коммитов

This Branch
This Branch
Все ветки
Автор SHA1 Сообщение Дата
Jeff Squyres
858612fd06 Face the possibilty that the child may have already died. 
This commit was SVN r9508.
 2006-04-01 02:23:10 +00:00
Jeff Squyres
bd6825f513 Add svn:ignore 
This commit was SVN r9507.
 2006-04-01 02:11:34 +00:00
Brian Barrett
c1578808e8 * commit configuration file for building OMPI against the Portals reference 
implementation when using the (small) run-time support provided by the
  Portals reference implementation instead of our own

This commit was SVN r9505.
 2006-03-31 21:11:21 +00:00
George Bosilca
ae47cbfe02 Accept an array of request set to NULL if the count is set to ZERO. 
This commit was SVN r9504.
 2006-03-31 20:32:56 +00:00
Brian Barrett
4b1b99d7c9 * better support platforms without any concept of a terminal at all 
This commit was SVN r9503.
 2006-03-31 18:51:06 +00:00
George Bosilca
65058d65e8 Update the NEWS to include PERUSE. 
This commit was SVN r9502.
 2006-03-31 17:20:11 +00:00
George Bosilca
58cd591d3b PERUSE support for OB1. There we go, now the trunk has a partial peruse implementation. 
We support all the events in the PERUSE specifications, but right now only one event
of each type can be attached to a communicator. This will be worked out in the future.
The events were places in such a way, that we will be able to measure the overhead
for our threading implementation (the cost of the synchronization objects).

This commit was SVN r9500.
 2006-03-31 17:09:09 +00:00
George Bosilca
1226d452bf Add a base _START macro that will do the base initialization. Additinaly, that allow me to 
add the PERUSE event is a more homogeneous manner (all PML's will have them).

This commit was SVN r9499.
 2006-03-31 17:05:09 +00:00
George Bosilca
988ba2af24 Cleanup the PERUSE code to match the latest specifications. All the unused code is 
still in the obsolete files, just in case they rollback the specification in order to
allow event for anything else than communicators.

This commit was SVN r9498.
 2006-03-31 17:03:09 +00:00
George Bosilca
551a0b11e3 Correct an ifdef issue and add one more PERUSE events. It is called 
PERUSE_COMM_REQ_XFER_CONTINUE and it's purpose is to show how we segment the
messages (across several networks or not).

This commit was SVN r9497.
 2006-03-31 16:34:23 +00:00
Andrew Friedley
74b2f77a4c The expected cleanup/refactoring commit.. 
Not much got tested that wasn't already - I've uncovered a connection
establishment deadlock and wanted to get these changes committed before I
attack it.

The big changes:
 - Moved much of the connection code from btl_udapl_component.c to
   btl_udapl_endpoint.c.
 - Cleaned up initialization of various fragment members.
 - MCA_BTL_UDAPL_ERROR macro, which is compiled in/out appropriately.

This commit was SVN r9496.
 2006-03-31 16:25:19 +00:00
George Bosilca
f09a6f50df The real name is ompi_mpi_abort_print_stack. 
This commit was SVN r9495.
 2006-03-31 04:21:09 +00:00
George Bosilca
0c58d0f519 Mixing declarations and code is not allowed by the ISO C90. 
This commit was SVN r9494.
 2006-03-31 03:21:28 +00:00
Brian Barrett
becc55abf6 * add missing extern in header file 
This commit was SVN r9493.
 2006-03-31 02:45:06 +00:00
Sushant Sharma
8feab23fdc updated acinclude.m4 to include config/ompi_check_xcpu.m4 
This commit was SVN r9492.
 2006-03-31 02:25:04 +00:00
Sushant Sharma
46f84b1e8e Added xcpu component in pls and soh. 
This commit was SVN r9491.
 2006-03-31 02:19:52 +00:00
Brian Barrett
c2b6e86766 back out r9489, as the rest of the changeset clearly didn't come with this 
piece

This commit was SVN r9490.

The following SVN revision numbers were found above:
  r9489 --> open-mpi/ompi@a4cfbbcac3
 2006-03-31 02:05:36 +00:00
Sushant Sharma
a4cfbbcac3 Added xcpu component in pls and soh. 
This commit was SVN r9489.
 2006-03-31 01:28:07 +00:00
George Bosilca
994959345a The if outside the loop not inside as we test for a "constant" thing. 
This commit was SVN r9488.
 2006-03-31 00:38:11 +00:00
Jeff Squyres
fd61d78599 Add two MCA parameters to the MPI level to control behavior during 
MPI_ABORT.  From the ompi_info output:

       MCA mpi: parameter "mpi_abort_delay" (current value: "0")
                If nonzero, print out an identifying message when
                MPI_ABORT is invoked (hostname, PID of the process
                that called MPI_ABORT) and delay for that many seconds
                before exiting (a negative delay value means to never
                abort).  This allows attaching of a debugger before
                quitting the job.
       MCA mpi: parameter "mpi_abort_print_stack" (current value: "0")
                If nonzero, print out a stack trace when MPI_ABORT is
                invoked

This commit was SVN r9487.
 2006-03-31 00:31:15 +00:00
Galen Shipman
1d67917b69 must handle header validation correctly for each case, not enough in common 
for the MACRO 

This commit was SVN r9486.
 2006-03-30 21:27:21 +00:00
Ralph Castain
8ba453b866 Modify the rmgr_proxy component so it includes the automatic wire-up of stdio. 
This commit was SVN r9483.
 2006-03-30 19:44:28 +00:00
Tim Woodall
9a73fe8beb check for valid sequence number before attempting to use communicator 
This commit was SVN r9482.
 2006-03-30 19:36:15 +00:00
George Bosilca
2b3779cd6e Correct some of the casting issues. By default the compilers attach an signed type 
to the defines. As our internal types (job_id and co.) are unsigned that generate
several errors (integer overflow in expression and comparison between signed and
unsigned). Casting the defines to the correct type solve these problems.

This commit was SVN r9481.
 2006-03-30 19:28:17 +00:00
Gleb Natapov
256bf70530 Forgot to add file to previous commit 
This commit was SVN r9480.
 2006-03-30 17:37:52 +00:00
Tim Woodall
637511e759 correct cleanup of callbacks 
This commit was SVN r9479.
 2006-03-30 16:55:02 +00:00
Gleb Natapov
79bcfb096f Add type to frag. Sometimes we need to know that a frag is from short rdma area. 
I used hack for this that doesn't work for mvapi, so changing it to something more sane.

This commit was SVN r9477.
 2006-03-30 15:26:21 +00:00
Brian Barrett
e0eb9a19e7 * make orte_process_name_t be of fixed size (rather than depending on the size of 
size_t).  This should be the last piece of the puzzle required to get 32/64
  interoperability working for ORTE.

This commit was SVN r9476.
 2006-03-30 14:59:41 +00:00
Brian Barrett
6be35fb604 * Use the ORTE_<type> constants instead of internal DSS_TYPE_<type>_T constants 
for the type to be packed / unpacked when dealing with sized types (like
  size_t) so that the dss_unpack code to deal with types of different sizes is
  activated.  Necessary for proper 32/64 interoperability.

This commit was SVN r9475.
 2006-03-30 14:33:25 +00:00
Gleb Natapov
ea11582191 Porting of short message RDMA from openib BTL. Endpoint registers circular buffer and sends its address and rkey to the peer. Peer uses this buffer to eagerly RDMA small message into it. Endpoint polls the buffer for message arrival before checking HP/LP QPs. Set btl_mvapi_use_eager_rdma to 1 to enable it. 
This commit was SVN r9474.
 2006-03-30 12:55:31 +00:00
George Bosilca
99ba9bea10 Correct the CRAY define (here is a interesting webpage with all the compilers defines 
http://predef.sourceforge.net/precomp.html).

This commit was SVN r9472.
 2006-03-30 00:20:43 +00:00
Jeff Squyres
1597ee94b7 Update notes on the Portland compiler 
This commit was SVN r9471.
 2006-03-29 23:09:02 +00:00
Galen Shipman
641fa6c0d2 more fixes, reset state on completion.. 
This commit was SVN r9469.
 2006-03-29 22:21:35 +00:00
Galen Shipman
2945f77f9e randomly drop fragments without local completion, currently commented out as 
we must handle the other cases first.. 

This commit was SVN r9468.
 2006-03-29 22:19:58 +00:00
Andrew Friedley
0eba366b07 Various pieces all over to make basic small message send/recv work. Next step 
is clean up the code.. it is in need of refactoring and testing.

Thanks to Brian for help in troubleshooting!

This commit was SVN r9466.
 2006-03-29 21:55:41 +00:00
Jeff Squyres
142f59a435 Add release note about PGI compilers. 
This commit was SVN r9459.
 2006-03-29 19:45:45 +00:00
Brian Barrett
02c8a51b76 * fix endian encoding for 64 bit numbers to use hton64 
* cleanup the unpack_size_mismatch macros a little bit
* ad comment about endianness of size_mismatch cleanup code so that I don't
  think I've found a bug that really isn't and lose an hour tracking it down
  again...

This commit was SVN r9458.
 2006-03-29 18:58:02 +00:00
Galen Shipman
5271948ec0 --- opal object changes 
add object size to opal class
no longer need the size when allocating a new object as this is stored in
the class structure

--- dr changes 
Previous rev. maintained state on the communicator used for acking duplicate
fragments, but the communicator may be destroyed prior to successfull
delivery of an ack to the peer. We must therefore maintain this state
globally on a per peer, not a per peer, per communicator basis. 
This requires that we use a global rank on the wire and translate this as
appropriate to a local rank within the communicator. 

This commit was SVN r9454.
 2006-03-29 16:19:17 +00:00
Gleb Natapov
2d9757e81b Fix deadlock in opal_condition. If broadcast is called on condition with more than one thread waiting on it the first awaken thread may consume all signals so other threads will never run. 
This commit was SVN r9451.
 2006-03-29 14:54:58 +00:00
Sven Stork
711b30fbac - introduction of user buffer checking for pt2pt operations 
This commit was SVN r9449.
 2006-03-29 09:26:27 +00:00
Brian Barrett
cf425f6289 * trick the stupid compiler (GCC in this case) into shutting up about not 
being able to convert from object pointer to function pointer. 

This commit was SVN r9445.
 2006-03-29 01:26:16 +00:00
Brian Barrett
99e4c89183 * some typo fixes for orterun manpage 
* Install orterun manpage as mpirun.1 and mpiexec.1 as well as orterun.1

This commit was SVN r9444.
 2006-03-29 01:04:43 +00:00
Jeff Squyres
07b0e559f2 Fix copyright 
This commit was SVN r9443.
 2006-03-29 00:53:11 +00:00
George Bosilca
50b5a02f8b Let the oob to call opal_progress instead of opal_progress_event. Now, the MPI 
communications will be advanced in MPI_Finalize.

This commit was SVN r9442.
 2006-03-28 22:09:40 +00:00
Brian Barrett
ae85aae6e5 * add some debugging if conftestval the file doesn't exist 
* Try fclose()ing the file to see if that helps with the errors reported
  with Absoft 8.2

This commit was SVN r9439.
 2006-03-28 13:55:39 +00:00
George Bosilca
5d465cf118 Call the constructor on the DR lock. 
This commit was SVN r9438.
 2006-03-28 07:34:02 +00:00
George Bosilca
af7930a498 Don't forget to add the new .h file to the distribution list. 
This commit was SVN r9437.
 2006-03-28 07:16:39 +00:00
Graham Fagg
19906e66dc missing lock? 
This commit was SVN r9436.
 2006-03-28 06:15:48 +00:00
George Bosilca
c93680074f Enable the master convertor and add a function to clean them all up. Without talking 
about the comments :)

This commit was SVN r9435.
 2006-03-28 05:11:48 +00:00
George Bosilca
46c442fe0d We do not have direct access to the module. Grab the one attached to the 
window instead.

This commit was SVN r9434.
 2006-03-28 05:06:40 +00:00