1
1
Граф коммитов

2089 Коммитов

Автор SHA1 Сообщение Дата
George Bosilca
3fc0e969b7 Cast in order to return the expected bool.
This commit was SVN r11992.
2006-10-05 05:11:05 +00:00
George Bosilca
b27f1814c6 If the function is expected to return a bool then let's return only
true or false.

This commit was SVN r11991.
2006-10-05 05:10:34 +00:00
George Bosilca
4689c56210 Always cast the return of malloc.
This commit was SVN r11990.
2006-10-05 05:07:43 +00:00
George Bosilca
5fb0c63959 Keep the ompi_config.h in the begining of the include list. This fixes
the #365. I found another way around on Windows.

This commit was SVN r11989.
2006-10-05 05:02:27 +00:00
George Bosilca
bf1d4ee259 Cast to the uint64_t before playing with the << operator in order to make sure we keep
all usefull informations.

This commit was SVN r11986.
2006-10-04 23:58:59 +00:00
George Bosilca
9e4b64ce81 Cleaning up the file, removing useless asserts. Commenting all calls to
opal_output.

This commit was SVN r11985.
2006-10-04 23:57:05 +00:00
George Bosilca
54476382e6 Remove all useless asserts and correct one of the comments.
Correctly compute the length to be unpacked.
Protect all outputs (only on debug mode and only if mpi_ddt_unpack is set).

This commit was SVN r11984.
2006-10-04 23:56:10 +00:00
George Bosilca
66e039496d This is the minimal patch for #365 (BLACS problem). Correctly compute the displacement
of the elements when we unroll a loop.

This commit was SVN r11983.
2006-10-04 23:54:22 +00:00
Jeff Squyres
e08c6e81f5 * Fixes trac:338: Only look at root-significant values at the root (e.g.,
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
2006-10-04 22:36:01 +00:00
George Bosilca
e4df4285b1 Reorder the enum in order to allow some compilers to optimize the big switch in
the header analisys.

This commit was SVN r11975.
2006-10-04 20:03:28 +00:00
George Bosilca
4d18d208d4 Getting closer to a full Total View support.
This commit was SVN r11974.
2006-10-04 20:01:33 +00:00
George Bosilca
e9572c7aee Mea culpa ... that was an weird error. When playing with 64 bits types we have to
the constant before using the << operator ... otherwise everything is 32 bits ...
which does not give the right answer.

This commit was SVN r11972.
2006-10-04 16:23:51 +00:00
George Bosilca
d05f492901 With the lastest changes for heterogeneous environments we have to
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.
2006-10-03 20:25:30 +00:00
George Bosilca
432659a2d8 Fix some typos.
This commit was SVN r11961.
2006-10-03 20:21:24 +00:00
Andrew Friedley
836261b85a Fixes ticket 186.
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.
2006-10-03 18:15:26 +00:00
Andrew Friedley
1177844d7a Fixes trac:183.
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
2006-10-03 18:08:48 +00:00
George Bosilca
f4da7a80bd Fine grain selection for heterogeneous environments. The hetero version of the
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.
2006-10-03 08:13:16 +00:00
Jeff Squyres
9ccbab464d Remove unused variable.
This commit was SVN r11918.
2006-10-01 11:51:41 +00:00
Dan Lacher
ba0389723e Ticket: #346
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.
2006-09-29 23:58:58 +00:00
Jeff Squyres
17539dc154 Fixes trac:430. Fix a few places where optimization checking conflicted
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
2006-09-29 22:49:04 +00:00
Edgar Gabriel
ec55acd8f4 orte_rml.send_buffer returns the number of bytes sent or a negative value if
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.
2006-09-29 20:28:45 +00:00
Jeff Squyres
785a2e1c90 Move the man page installs to install-data-hook. Putting them in
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.
2006-09-29 14:34:39 +00:00
Ralph Castain
0411f9772e Begin instrumenting for scalability tests.
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.
2006-09-29 13:19:44 +00:00
Jeff Squyres
7b9b90f10a openmpi.3 was deleted; OpenMPI.3 was the one that was left.
This commit was SVN r11886.
2006-09-29 11:59:27 +00:00
Jeff Squyres
b3a35b7d1c Remove entries for files that are no longer there.
This commit was SVN r11885.
2006-09-29 11:57:08 +00:00
Andrew Lumsdaine
e187f058b7 Remove mpi.3 to fix naming conflict
This commit was SVN r11883.
2006-09-29 01:49:21 +00:00
Andrew Lumsdaine
a1ef0067e0 Remove openmpi.3 to fix naming conflict
This commit was SVN r11882.
2006-09-29 01:48:13 +00:00
George Bosilca
7c8c8d6a46 Keep the critical path as short as possible.
This commit was SVN r11881.
2006-09-28 23:59:24 +00:00
George Bosilca
8d2a8229bb We don't use the send and receive request destructor.
This commit was SVN r11880.
2006-09-28 23:57:49 +00:00
George Bosilca
7f2fd41ace Make sure we trigger the PERUSE event before releasing the request.
This commit was SVN r11879.
2006-09-28 23:54:38 +00:00
George Bosilca
9ae37e474b Force the initialization of the convertor if we detect truncation of messages.
This commit was SVN r11877.
2006-09-28 23:42:56 +00:00
George Bosilca
e5ccc1aece Keep the loop as short as possible. And specialize the search for ANY_TAG.
This commit was SVN r11874.
2006-09-28 22:47:40 +00:00
Jeff Squyres
9c37afc265 Add Makefile mojo to install the MPI API man pages and include them in
the distribution tarball.

This commit was SVN r11872.
2006-09-28 19:59:28 +00:00
Rolf vandeVaart
aa7317d635 Initial man pages for all the MPI APIs. There are 304 of them. Also a general man page for mpi,openmpi,MPI, and OpenMPI. Extra work is needed so they get installed properly.
This commit was SVN r11871.
2006-09-28 18:39:36 +00:00
Brian Barrett
451d362296 MPI_Fence's stupid "I might be in a fence epoch but you can't tell until
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
2006-09-28 15:11:11 +00:00
Gleb Natapov
26660d657a Openib BTL supports PUT and GET protocol, communicate it to PML. We get better
bandwidth with GET protocol and mpi_leave_pinned option.

This commit was SVN r11862.
2006-09-28 11:48:04 +00:00
Gleb Natapov
30608c5a50 Add separate queues for pending rget and rput frags. Process only limited
number of pending packets at once, otherwise we can spin forever.

This commit was SVN r11861.
2006-09-28 11:41:45 +00:00
Jeff Squyres
a5928a2a89 Ralf W. swears to me that this is a product of a very, very old AM bug
and is no longer necessary (especially since we enforce pretty recent
Gnu tools).

This commit was SVN r11855.
2006-09-27 21:33:26 +00:00
Terry Dontje
d228784fa8 Fixed trac #409 by adding a new MCA parameter named
btl_base_warn_component_unused.

This commit was SVN r11844.
2006-09-27 13:24:42 +00:00
Jeff Squyres
860788ff8f Fixes trac:389
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
2006-09-27 11:24:36 +00:00
Edgar Gabriel
4dafe7ce0d The length of the ranks-arrays can be longer than the size of the according
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
2006-09-27 11:02:47 +00:00
Jeff Squyres
afa7f53d43 Refs trac:429.
Tweak a comment to match the code.

This commit was SVN r11839.

The following Trac tickets were found above:
  Ticket 429 --> https://svn.open-mpi.org/trac/ompi/ticket/429
2006-09-27 00:04:24 +00:00
Jeff Squyres
13d5e5aab4 Refs trac:429
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
2006-09-26 23:36:22 +00:00
Brian Barrett
85cc94f634 Change mpool shared memory file size from 512MB to a scaling factor based
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
2006-09-26 16:02:31 +00:00
Brian Barrett
58dce54513 don't use macro or it will try to use the (now destroyed) MPI_COMM_WORLD,
which isn't going to work well

Refs trac:419

This commit was SVN r11822.

The following Trac tickets were found above:
  Ticket 419 --> https://svn.open-mpi.org/trac/ompi/ticket/419
2006-09-26 15:56:51 +00:00
Terry Dontje
bc93adee26 Fixed connection inversion bug by putting in sequence checking for
first sendrecv exchanges for each connection.  This was to fix Trac
#390.

This commit was SVN r11821.
2006-09-26 13:53:00 +00:00
Gleb Natapov
7b1b4f95e3 Local GID table contains not what I thought it contains. It contains local HCA
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.
2006-09-26 12:12:33 +00:00
Brian Barrett
a546b6834b Only add the ROMIO source directory into DIST_SUBDIRS and SUBDIRS if
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
2006-09-25 23:51:13 +00:00
Brian Barrett
c3306f7073 * don't abort if we get a status error - just pass it on to the next layer up
This commit was SVN r11791.
2006-09-25 17:28:24 +00:00
Brian Barrett
b18055451d * need to include errno.h to get errno
This commit was SVN r11790.
2006-09-25 17:17:43 +00:00
Jeff Squyres
a822c34ebf Ralf W. categorically told me that this kind of statement in
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.
2006-09-25 14:28:04 +00:00
Jeff Squyres
c5cc1f0c1a Add man page for wrapper compilers.
Fixes trac:358.

This commit was SVN r11773.

The following Trac tickets were found above:
  Ticket 358 --> https://svn.open-mpi.org/trac/ompi/ticket/358
2006-09-25 14:11:21 +00:00
Gleb Natapov
9cd25158d6 Fix btl_openib_max_btls parameter handling.
This commit was SVN r11772.
2006-09-25 11:18:20 +00:00
Terry Dontje
d636db5832 Fixed bug trac #213 by moving the udapl btl header to being a footer.
Also fixed bug trac #346.

This commit was SVN r11760.
2006-09-22 19:28:09 +00:00
Brian Barrett
ab6cbb2359 * update ompi_mpi_abort to call abort_procs_request on the processes that
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.
2006-09-22 15:04:04 +00:00
Gleb Natapov
601a6ca17a Use real subnet prefix instead of sm_lid.
This commit was SVN r11749.
2006-09-22 10:27:12 +00:00
Brian Barrett
8fc278c3a3 Rest of the fix for #325. It uses a bit more space, but now we can reasonably
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
2006-09-21 20:49:15 +00:00
Brian Barrett
2ec0c4f593 * Fix race condition in post/wait/start/complete synchronization where one
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
2006-09-21 19:57:57 +00:00
George Bosilca
383c4e8c18 Don't print the optimized description if there is nothing.
This commit was SVN r11733.
2006-09-21 05:40:51 +00:00
George Bosilca
645790dd9c Pedantic...
This commit was SVN r11731.
2006-09-20 22:20:10 +00:00
George Bosilca
688a16ea78 A long time waiting patch. Get rid of the comm->c_pml_procs. It was (and that was
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.
2006-09-20 22:14:46 +00:00
George Bosilca
20459bd982 Remove the HIDDEN flag. It is not used anywhere.
This commit was SVN r11729.
2006-09-20 20:57:10 +00:00
George Bosilca
1c464d340c Do not increase the reference count for the datatype if it is not required. Plus
some typos.

This commit was SVN r11728.
2006-09-20 20:14:15 +00:00
Brian Barrett
ddffc8748f * remove unneeded files that were included in the mpich2 tarball
This commit was SVN r11726.
2006-09-20 18:27:28 +00:00
George Bosilca
ff7e670c6a Sends and receives are now fully implemented. Correctly handle the reference counts
on groups. Correct the rearrangement of the communicators. Other minors improvements.

This commit was SVN r11722.
2006-09-20 07:51:30 +00:00
Brian Barrett
10a230373b Add a number of missing constants to the C++ bindings
refs trac:322

This commit was SVN r11720.

The following Trac tickets were found above:
  Ticket 322 --> https://svn.open-mpi.org/trac/ompi/ticket/322
2006-09-19 22:28:21 +00:00
Andrew Friedley
e776b01811 This assert fails if -mca pml_dr_enable_csum 0 is set, which isn't what we want..
This commit was SVN r11719.
2006-09-19 19:57:33 +00:00
George Bosilca
6f3782bbd7 When we succesfully cancel a request we have to set it's pml_complete flag to true
if we want to be able to reuse the request. If not, the request will never be freed
even if the user call MPI_Request_free.

This commit was SVN r11717.
2006-09-19 18:04:09 +00:00
Brian Barrett
d1402cf8f5 * Update ROMIO release to the one included with MPICH2-1.0.4p1, tagged in
vendor/romio as mpich2-1.0.4p1.

This commit was SVN r11715.
2006-09-19 16:13:46 +00:00
Gleb Natapov
18c54f574f add rmb()
This commit was SVN r11710.
2006-09-19 13:27:05 +00:00
Gleb Natapov
ac42284c16 Print more helpful message in case we can't find active port.
This commit was SVN r11706.
2006-09-19 08:56:32 +00:00
Rainer Keller
80166a9516 - fix typos
This commit was SVN r11703.
2006-09-19 07:55:41 +00:00
George Bosilca
b49ff3a3b7 We are now able to provide full information about the requests. And we can
parse all requests (sends or receives), and show them based on the communicator.

This commit was SVN r11702.
2006-09-19 06:31:42 +00:00
George Bosilca
139a9df741 Next step forward. Now we have the send and receive queue as well. There
is still a small problem with the groups but it clearly getting better ...

This commit was SVN r11701.
2006-09-19 02:07:06 +00:00
Ralph Castain
0ad0d84afd Add two new API functions to the RMGR, and modify the "spawn" API to support the enhanced MPI-2 functionality.
No implementation backs these new APIs - just placeholders for now.

This commit was SVN r11699.
2006-09-19 01:45:05 +00:00
George Bosilca
e679b923a2 Don't let the compiler copy a struct into another location. Do it by hand only
on the required fields.

This commit was SVN r11696.
2006-09-18 18:00:16 +00:00
George Bosilca
2029284820 Typo.
This commit was SVN r11695.
2006-09-18 17:57:55 +00:00
George Bosilca
67336f90b0 First cut for the Total View support for communicators, groups and requests. Not yet
included on the conpilation step.

This commit was SVN r11690.
2006-09-18 17:32:04 +00:00
Rainer Keller
40cb5d3e30 - Fix peruse compilation
This commit was SVN r11685.
2006-09-18 07:41:09 +00:00
Rainer Keller
611f3ba408 - Fix buglet found by Feng Sheng's IO tests;
Correctly initialize the status for non-blocking io.

This commit was SVN r11672.
2006-09-15 12:37:29 +00:00
Jeff Squyres
8226dab86c Fixes trac:377
Add --enable-orterun-prefix-by-default (and a synonym:
--enable-mpirun-prefix-by-default) to make orterun always behave as if
"--prefix $prefix" was given on the command line (where $prefix is the
value given to the --prefix option to configure).  This prevents many
rsh/ssh users from needing to modify their shell startup files to set
the LD_LIBRARY_PATH for Open MPI (they will still need to set PATH or
otherwise find the OMPI executables to mpicc/mpirun/etc. their MPI
applications).

Also added --noprefix option to orterun to disable this behavior.
Finally, note that even if --enable-orterun-prefix-by-default is
specified, if the user specifies --prefix or /path/to/mpirun, these
options will override the default value of the prefix ($prefix).

This commit was SVN r11669.

The following Trac tickets were found above:
  Ticket 377 --> https://svn.open-mpi.org/trac/ompi/ticket/377
2006-09-15 02:52:08 +00:00
Jeff Squyres
d30b1ad61a Remove a file that should not be in SVN
This commit was SVN r11667.
2006-09-15 02:37:34 +00:00
George Bosilca
a3ad4a7fc8 The visibility flags (and/or Windows friendly export) is now on for all BTLs.
This commit was SVN r11662.
2006-09-14 22:19:39 +00:00
Ralph Castain
37dfdb76eb Here is the major MAD-cure commit. I have written plenty about it, so I refer you here to those messages for a description of everything that was done.
This commit was SVN r11661.
2006-09-14 21:29:51 +00:00
Galen Shipman
877b819ddb Initial commit of QLogic PSM MTL.
This provides support for the Infinipath interconnect using the PSM API. 

Of note: 
This version has a "hackaround" we always return 1 or greater from
the MTL PSM progress function, this should be examined further.

This commit was SVN r11655.
2006-09-14 16:44:02 +00:00
Galen Shipman
0be42b22b9 Fix connection timeout error handling..
This commit was SVN r11651.
2006-09-14 14:02:14 +00:00
Sven Stork
09e0d9d0d9 - revert commit 11632
- Real fix for pgi compilers and missing asprintf declaration.
  The problem was that C++ headers has been included before "ompi_config.h".
  This lead to the problem that "stdio.h" has been source without _GNU_SOURCE.

This commit was SVN r11649.
2006-09-14 08:10:02 +00:00
Jeff Squyres
a8e9fa09da Fix some compiler warnings introduced in r11619. I checked with
George: ompi_ddt_type_size() returns a signed int only because of the
MPI spec; it will never return a negative value.  So casting the
return value out of it to a (uint32_t) is safe, and makes the
comparisons be between two unsigned values. 

This commit was SVN r11639.

The following SVN revision numbers were found above:
  r11619 --> open-mpi/ompi@8667648a1b
2006-09-13 16:42:31 +00:00
Sven Stork
0fd414b8e3 - use opal version, will satisfy some pgi compilers
This commit was SVN r11632.
2006-09-13 14:30:54 +00:00
Gleb Natapov
7999c08107 consolidate credit management and CQ polling code.
This commit was SVN r11622.
2006-09-12 09:17:59 +00:00
Graham Fagg
8667648a1b Simple fix (for ticket 363). We push segment size to type size. In other algorithms we switch of segementing altogether. But really the DDT can probably handle partial types so we could really keep the segsize constant (for all but reduce ops) and treat it just as byte arrays..
todos: macroize it as we do it 10 different ways, add mca params to control handling (push up size, no change, switch off segmenting)

This commit was SVN r11619.
2006-09-12 00:01:27 +00:00
Jeff Squyres
713206d4f9 Updates to r11563:
* Consolidate everything inside of the same AM_CONDITIONAL that is
   used to suck in the glue convenience library in ompi/Makefile.am:
   OMPI_WANT_F77_BINDINGS.  This AM conditional is set to true if we
   want (and can support) the F77 MPI API bindings at all (And does
   not say anything about whether we are compiling the top-level or
   bottom-level f77 directory to get the bindings).
 * Clarify all the comments surrounding the [confusing!] issue.
 * The problem with r11563 was that it used the wrong AM_CONDITIONAL
   to decide whether to build the separate F77 library or not; it
   would do so only if the top-level library was being built (e.g., on
   systems like OSX where weak symbols don't work the way we need them
   to).  This patch somewhat simplifies the situation by encapsulating
   everything in one large conditional (OMPI_WANT_F77_BINDINGS, as
   described above).  Hence, libmpi_f77 will exist (and be installed)
   if F77 support is enabled overall, regardless of whether you're on
   a system with insufficient weak symbol support (e.g., OSX) or not
   (e.g., Linux).

This commit was SVN r11618.

The following SVN revision numbers were found above:
  r11563 --> open-mpi/ompi@c8f3ff71b1
2006-09-11 23:18:24 +00:00
Jeff Squyres
6173e952a2 Remove these files from SVN; they should not be committed.
This will force everyone to re-autogen.  Sorry!

This commit was SVN r11616.
2006-09-11 22:43:20 +00:00
Gleb Natapov
03cda61302 Fix hang in receiving into MPI_alloced area.
This code hangs with openib BTL:

int size = 4000000;
sbuf = malloc(size);  
MPI_Alloc_mem(size, MPI_INFO_NULL, &rbuf);

if (rank == 0)
{
    MPI_Recv(rbuf, size, MPI_CHAR, 1, 1, MPI_COMM_WORLD, &stat);
}else{
    MPI_Send(sbuf, size, MPI_CHAR, 0, 1, MPI_COMM_WORLD);
}

This commit was SVN r11613.
2006-09-11 12:18:59 +00:00
Sven Stork
72bf1e4a25 - add parameter checks for standard compliance
This commit was SVN r11610.
2006-09-11 10:23:35 +00:00
Gleb Natapov
fa17445384 fix compilation warning.
This commit was SVN r11601.
2006-09-10 06:17:33 +00:00
Jeff Squyres
fb4d7ab268 * Fix svn:ignore
* Remove files that should not be in SVN

This commit was SVN r11565.
2006-09-08 10:35:45 +00:00
Brian Barrett
c8f3ff71b1 Install Fortran 77 bindings as a stand-alone library, rather than as part of
libmpi.

Refs trac:248

This commit was SVN r11563.

The following Trac tickets were found above:
  Ticket 248 --> https://svn.open-mpi.org/trac/ompi/ticket/248
2006-09-08 01:35:49 +00:00
Brian Barrett
6b5a126d0b * Only install Fortran 77 headers if we want Fortran 77 bindings
This commit was SVN r11562.
2006-09-08 01:33:21 +00:00
Gleb Natapov
9b93f48e22 fix compile warnings in previous commit.
This commit was SVN r11554.
2006-09-07 13:31:50 +00:00
Gleb Natapov
d0caffa0aa Consolidate receive buffers prepost code for HP/LP QPs.
This commit was SVN r11552.
2006-09-07 13:05:41 +00:00
Gleb Natapov
e7650ff48a Bad things happen if min_rdma_size is smaller then data delivered in the RNDV
packet. Fix this.

This commit was SVN r11548.
2006-09-07 10:42:35 +00:00
Gleb Natapov
298c825592 Remove #if OMPI_MCA_BTL_OPENIB_HAVE_SRQ. Always compile SRQ.
This commit was SVN r11537.
2006-09-06 05:45:37 +00:00
Gleb Natapov
4f05da21a1 remove unused macro
This commit was SVN r11536.
2006-09-06 05:36:26 +00:00
Gleb Natapov
424e412391 Make eager rdma work with SRQ enabled.
This commit was SVN r11530.
2006-09-05 16:04:04 +00:00
Gleb Natapov
c13240a1d1 remove rdma_credits from openib BTL header. Use one field for regular and rdma credits.
This commit was SVN r11529.
2006-09-05 16:02:09 +00:00
Gleb Natapov
fe932ca7bf consolidate part of HP/LP fields.
This commit was SVN r11528.
2006-09-05 16:00:18 +00:00
Gleb Natapov
b6bac100b0 Move error path out of the way.
This commit was SVN r11527.
2006-09-05 15:59:02 +00:00
Gleb Natapov
ffe7051488 fix compilation warnings.
This commit was SVN r11524.
2006-09-05 09:16:22 +00:00
Jeff Squyres
8406746854 Fixes trac:330
Had the wrong type for one of the arguments of MPI_TYPE_GET_CONTENTS
(MPI_Fint should have been MPI_Aint).

This commit was SVN r11517.

The following Trac tickets were found above:
  Ticket 330 --> https://svn.open-mpi.org/trac/ompi/ticket/330
2006-09-01 19:58:04 +00:00
Jeff Squyres
7fe337ce3b Yoinks -- remove some debugging output.
This commit was SVN r11515.
2006-09-01 11:48:26 +00:00
Brian Barrett
e0555889a9 * RMA_SYNC is a more appropriate error message for these than RmA_CONFLICT
* Print a warning error message if a target is not in an exposure epoch
    and an update is received.  This results in the app continuing with
    that call having never happened, rather than evil hangs.

refs trac:325

This commit was SVN r11514.

The following Trac tickets were found above:
  Ticket 325 --> https://svn.open-mpi.org/trac/ompi/ticket/325
2006-08-31 21:07:52 +00:00
Jeff Squyres
6c2e938d31 Update the comment to reflect that this can now be a comma-delimited list.
This commit was SVN r11507.
2006-08-30 20:28:48 +00:00
Jeff Squyres
91bdbc0673 This commit fixes a few things. It looks bigger than it is because a
bunch of code changed indenting level and some code got moved out of
one function and made into its own subroutine.

- Gleb pointed out that I wasn't taking into account values from the
  default section of the INI file (and not finding values in the INI
  file is not an error).
- I incorrectly thought that 0x5ad was Mellanox's vendor ID.  Turns
  out that 0x5ad is Cisco's ID, while 0x2c9 is Mellanox.
  Specifically, Cisco burns its own firmware into the HCA which
  replaces the vendor ID, although the part ID stays the same.  So
  it's Mellanox hardware with Cisco firmware.  And apparently several
  of us do that.  :-)  So I expanded the concept of the vendor_id in
  the INI file to allow for lists of vendor IDs.  
- Along with that, I updated the default INI file to list all the IB
  vendors (that I am aware of -- certainly open to putting more data
  in there from other vendors) who overwrite Mellanox's vendor_id with
  their own for the part numbers that we have on file.

This commit was SVN r11506.
2006-08-30 20:21:47 +00:00
Jeff Squyres
aadc7e24ef Also check for zero-length strings.
This commit was SVN r11499.
2006-08-30 11:59:02 +00:00
George Bosilca
ee27c497f9 Allow duplication of MPI_LB and MPI_UB. A very usefull feature :)
This commit was SVN r11496.
2006-08-29 20:19:58 +00:00
George Bosilca
3312aa4b0a The pack/unpack will return 1 only if all data has been packed/unpacked. We have to make
sure we provide exactly the amount of data these functions expect, otherwise they will
return 0.

This commit was SVN r11484.
2006-08-29 17:17:35 +00:00
Gleb Natapov
338134b535 run dos2unix on wtime.c and make MPI_Wtime work as it did before.
This commit was SVN r11482.
2006-08-29 10:11:48 +00:00
George Bosilca
e479951b3b And now the correct version of the timers. In fact, MPI_Wtime is supposed
to return the value on seconds not some other unit based on the resolution
of MPI_Wtick. Which I think it's the wrong solution, as instead of forcing
the user to do additional computations in order to convert when he needs
the result in seconds, force us to convert every time. Unfortunately,
converting requires a division with a double which is a costly
operation. But, MPI is a standard and we have to follow it ...

This commit was SVN r11481.
2006-08-29 04:30:33 +00:00
George Bosilca
e33c35112b Correct the conversion between int and bool. Apply it on all files except
the one that will be modified by Ralph for the ORTE 2.0. The missing ones
are in the rsh PLS.

This commit was SVN r11476.
2006-08-28 18:59:16 +00:00
George Bosilca
8673c83578 The bool type on Windows is not an integer. Therefore just casting an
int to bool is not allowed. We have to make something cleaner.

This commit was SVN r11475.
2006-08-28 18:51:09 +00:00
Jeff Squyres
9e2488bfe9 George found a great way to avoid warnings from flex for that unused
function.  Woo hoo!

This commit was SVN r11469.
2006-08-28 13:44:37 +00:00
Gleb Natapov
c70eb43e43 Align eager RDMA buffer so that last byte of the buffer is on the last byte of
the CPU cache line. Improves zero byte latency a little bit because of L1 cache
miss reduction.

This commit was SVN r11465.
2006-08-28 11:03:56 +00:00
Gleb Natapov
40ca1dd2d4 ran dos2unix on it
This commit was SVN r11463.
2006-08-28 10:27:27 +00:00
George Bosilca
07b8d3c72c On Windows we can now deliver Open MPI on several flavors:
- everything statically built (dynamically opened).
- OPAL, ORTE and OMPI static libraries and all the components
  as dynamic files(DLL).
- everything as dynamic files (DLL).

This commit was SVN r11461.
2006-08-28 04:19:42 +00:00
George Bosilca
36be7bf204 Unix friendly EOL.
This commit was SVN r11457.
2006-08-28 04:09:43 +00:00
George Bosilca
bbc2a1e4b8 The Fortran prototypes should be handled with care. OMPI_DECLSPEC
they should become.

This commit was SVN r11456.
2006-08-28 04:07:57 +00:00
Jeff Squyres
7515e8b234 Fix a typo caught by MTT.
This commit was SVN r11450.
2006-08-27 15:43:27 +00:00
Jeff Squyres
129f8a9eb8 Add a newline at the end of the file (squelch a compiler warning).
This commit was SVN r11445.
2006-08-27 12:45:54 +00:00
George Bosilca
ee75c45ec5 Use the OPAL timers toi report timers as accurately as possible.
This commit was SVN r11442.
2006-08-27 04:58:02 +00:00
George Bosilca
b645dbdf10 First load the C++ header files.
This commit was SVN r11438.
2006-08-27 04:49:48 +00:00
George Bosilca
3b39df8ae1 More protection around what we really want to get exported.
This commit was SVN r11437.
2006-08-27 04:49:02 +00:00
Jeff Squyres
196decaf08 * Remove redundant include
* Add missing include

This commit was SVN r11432.
2006-08-26 20:14:19 +00:00
Brian Barrett
c1a77b921d Need to add/subtract modes when dealing with PWSC, since a given window
can be in both a Post and Start state.  Also, the asserts were only
correct assuming that we were never in the post and start state at the
same time, which was obviously silly.

refs trac:303

This commit was SVN r11428.

The following Trac tickets were found above:
  Ticket 303 --> https://svn.open-mpi.org/trac/ompi/ticket/303
2006-08-25 20:39:33 +00:00
Sven Stork
d09863926c - fix dist target
This commit was SVN r11421.
2006-08-25 09:45:08 +00:00
George Bosilca
c8daf226b6 Export the RDMA OSC.
This commit was SVN r11417.
2006-08-24 21:17:31 +00:00
Brian Barrett
6e390155a1 * Add constants related to one-sided support that were missing from the
fortran include files

This commit was SVN r11416.
2006-08-24 21:14:05 +00:00
Jeff Squyres
17d313f6e1 It turns out that Fortran has some specific rules about copying
strings.  Here's one: no matter how much of the string you copy, the
destination string must be space-padded for the entire remaining area.
Specifically, even if you call MPI_INFO_GET and tell MPI to only copy
a max of N characters of the value into the result string, if the
Fortran string is M characters (where M > N), MPI must space-pad the
remaining (M-N) characters to be spaces.  So you're supposed to obey
the argument to MPI_INFO_GET... sorta.

Precedents:

 * http://www.ibiblio.org/pub/languages/fortran/ch2-13.html
 * LAM/MPI
 * Sun CT MPI

This commit was SVN r11412.
2006-08-24 19:11:39 +00:00
George Bosilca
6f4aa36dfc Add this header file in order to allow the export of the component structure.
This commit was SVN r11404.
2006-08-24 17:40:22 +00:00
Sami Ayyorgun
aa8cd63418 changed some barrier variables for shared-memory to volatile
This commit was SVN r11403.
2006-08-24 16:53:10 +00:00
Torsten Hoefler
6b22641669 added LibNBC (http://www.unixer.de/NBC) as collv1 (blocking) component.
I know it does not make much sense but one can play around with the
performance. Numbers are available at http://www.unixer.de/research/nbcoll/perf/.
This is the first step towards collv2. Next step includes the addition
of non-blocking functions to the MPI-Layer and the collv1 interface.

It implements all MPI-1 collective algorithms in a non-blocking manner.
However, the collv1 interface does not allow non-blocking collectives so
that all collectives are used blocking by the ompi-glue layer.

I wanted to add LibNBC as a separate subdirectory, but I could not
convince the buildsystem (and had not the time). So the component looks
pretty messy. It would be great if somebody could explain me how to move
all nbc*{c,h}, and {hb,dict}*{c,h} to a seperate subdirectory.

It's .ompi_ignored because I did not test it exhaustively yet.

This commit was SVN r11401.
2006-08-24 16:47:18 +00:00
George Bosilca
3f0a7cad9e The last patch for Windows support. Mostly casting and conversion to C++ friendly headers.
This commit was SVN r11400.
2006-08-24 16:38:08 +00:00
George Bosilca
b8f54ae381 The ompi_group_table is never referenced. So wipe it out.
This commit was SVN r11399.
2006-08-24 16:27:08 +00:00
Gleb Natapov
21e99cd334 init mtu parameter when no warn is set.
This commit was SVN r11388.
2006-08-24 10:42:42 +00:00
Jeff Squyres
918d1e018f Un-friggen' believable. The values for MPI_WIN_BASE, MPI_WIN_SIZE,
and MPI_WIN_DISP_UNIT were off by one from their C counterparts.

This fixes trac:304.

This commit was SVN r11385.

The following Trac tickets were found above:
  Ticket 304 --> https://svn.open-mpi.org/trac/ompi/ticket/304
2006-08-24 03:33:35 +00:00
George Bosilca
d6b6f465b6 Cast everything to make the microsoft C++ compiler happy.
This commit was SVN r11373.
2006-08-23 16:35:16 +00:00
George Bosilca
f8f2dd8e03 As class is a reserved keyword we are not supposed to have any variables
with this name.

This commit was SVN r11372.
2006-08-23 16:34:00 +00:00
Jeff Squyres
8b4b9b9a8e Oops -- in MPI_INFO_GET_NTHKEY, the key argument is an '''out'''
value, not an '''in''' value.  So the string needs to be converted
c2f, not f2c.

This commit was SVN r11367.
2006-08-23 14:43:47 +00:00
Jeff Squyres
fd9a94434d Update to r11332. This seems like a slightly safer fix.
This commit was SVN r11358.

The following SVN revision numbers were found above:
  r11332 --> open-mpi/ompi@4f984056ac
2006-08-23 13:46:05 +00:00
Jeff Squyres
523128100e A bunch of fixes for Fortran string issues. In general, ensure to
convert between fortran and C string representations properly.  In
doing so, we properly adhere to the MPI spec stating that MPI_Info
keys and values must be whitespace-trimmed when coming in from
Fortran.  Hence, this fixes bug #241.

This commit was SVN r11356.
2006-08-23 13:10:44 +00:00
George Bosilca
b621313653 Remove all references to the removed files.
This commit was SVN r11344.
2006-08-23 02:18:24 +00:00
George Bosilca
858dba5f13 These files are not usefull anymore.
This commit was SVN r11343.
2006-08-23 01:26:58 +00:00
Brian Barrett
4f984056ac MPI_STATUS_SIZE in Fortran is 5, so we need to jump by 5 integers instead
of 4 when we are finding the next MPI_STATUS in the array.

Refs trac:236

This commit was SVN r11332.

The following Trac tickets were found above:
  Ticket 236 --> https://svn.open-mpi.org/trac/ompi/ticket/236
2006-08-22 20:20:46 +00:00
Brian Barrett
1c4de419cc * fix Galen's typo ;)
This commit was SVN r11331.
2006-08-22 20:18:53 +00:00
Galen Shipman
73e9ef46fc use int32_t not size_t (ORTE interface change)..
This commit was SVN r11323.
2006-08-22 17:13:10 +00:00
Galen Shipman
fbf7e9cf1c use int32_t's not size_t's (interface change in ORTE)..
This commit was SVN r11322.
2006-08-22 16:26:36 +00:00
Galen Shipman
99d526a07c Add support for 1_0_0 source btl components
This commit was SVN r11321.
2006-08-22 16:25:36 +00:00
George Bosilca
6fc16516cc orte_std_cntr_t vs. size_t round 3. I back up this one as it
wasn't suppose to be committed (and because it's wrong).

This commit was SVN r11318.
2006-08-22 15:15:09 +00:00
George Bosilca
0417d27f46 orte_std_cntr_t vs. size_t round 2. Advantage for size_t ...
This commit was SVN r11317.
2006-08-22 14:58:31 +00:00
Jeff Squyres
fbb484dea2 MPI_COMM_GET_NAME had the compiler-added extra string length parameter
(but didn't use it), but MPI_TYPE_GET_NAME and MPI_WIN_GET_NAME did
not.

This commit changes all three functions to pass the compile-added
string length parameter to clear out the remainder of the string with
spaces (i.e., the rest of the string that was not set with the name).
This is what was done in LAM/MPI, and apparently what was done in
Sun's MPI, because the test that Rolf attached now passes.

Fixes trac:274.

This commit was SVN r11301.

The following Trac tickets were found above:
  Ticket 274 --> https://svn.open-mpi.org/trac/ompi/ticket/274
2006-08-21 19:35:33 +00:00
Ralph Castain
6d27fee3a2 Silence Cyrador...who had a valid complaint.
This commit was SVN r11282.
2006-08-21 14:26:11 +00:00
Brian Barrett
880730fcf6 * Make sure to add the FCFLAGS_f90 variable to FCFLAGS. This is the magic
that the compiler might need to inform the compiler that a .f90 extension
  means "this is Fortran 90 code".  Fortran compilers are so weird.

  refs trac:284

This commit was SVN r11280.

The following Trac tickets were found above:
  Ticket 284 --> https://svn.open-mpi.org/trac/ompi/ticket/284
2006-08-21 14:15:55 +00:00
Brian Barrett
75d32b4013 * It appears that AIX provides a mallopt, but not the options to disable
sbrk and the use of mmap().  So rather than checking just for mallopt(),
  we should also be checking for those defines when determining if we can
  disable giving memory back to the OS or not.

This commit was SVN r11279.
2006-08-21 14:00:30 +00:00
George Bosilca
392178b1a8 Move it in the right place.
This commit was SVN r11276.
2006-08-21 04:05:19 +00:00
George Bosilca
ebd7a49428 Create shared memory "files" on Windows. These "files" will be
allocated directly by the OS in the paging file (the HUGE file
that cannot be defragmented with any tools). Unlike UNIX, they
do not have physical existence as files.

This commit was SVN r11273.
2006-08-20 19:45:28 +00:00
George Bosilca
6afa4c6c64 Windows friendly version. We have to split the OMPI_DECLSPEC in at least 3
different macros, one for each project. Therefore, now we have OPAL_DECLSPEC,
ORTE_DECLSPEC and OMPI_DECLSPEC. Please use them based on the sub-project.

This commit was SVN r11270.
2006-08-20 15:54:04 +00:00
George Bosilca
9b4bab7d34 One step toward the create array completion.
This commit was SVN r11269.
2006-08-20 15:51:54 +00:00
Brian Barrett
1daa21e1e3 It appears that most versions of the IBM XL compiler (including the latest
releases on Linux and OS X) don't handle const_cast<> of 2-dimensional 
arrays properly.  If we're using one of the compilers that isn't friendly
to such casts, fall back to a standard C-style cast.

refs: #271

This commit was SVN r11263.
2006-08-19 22:55:59 +00:00
Brian Barrett
943e7dcfba * use a temporary to avoid passing pointers to size_t-sized structures into
the mca param functions, which expect poinrters to integers

This commit was SVN r11262.
2006-08-18 21:36:07 +00:00
Ralph Castain
6bf06d4602 Fix connect-accept by cleaning up two minor bugs.
This commit was SVN r11260.
2006-08-18 21:12:03 +00:00
Dan Lacher
f2526d60ed Minor fix for a dropped comma.
This commit was SVN r11259.
2006-08-18 17:55:57 +00:00
Galen Shipman
e5c594c211 More updates for the async error handler for btl's
In order to provide backwards compatability the framework versions are bumped
and the handler registeration function is at the end of the btl struct.
Testing done on sm, openib, and gm.. 

This commit was SVN r11256.
2006-08-17 22:02:01 +00:00
Brian Barrett
f1bfd174da * need to set SUCCESS when completing a request
This commit was SVN r11255.
2006-08-17 20:03:10 +00:00
Brian Barrett
f0afe38293 * Need to retain / release datatype and communicator so that the MPI layer
handles can be freed before communication completes.

This commit was SVN r11248.
2006-08-17 16:30:03 +00:00
Brian Barrett
31a51c73d9 * fix dumb typo
This commit was SVN r11246.
2006-08-17 15:57:49 +00:00
Brian Barrett
df84dbad00 * use the osc base debugging stream for all output, and do the whole
verbose MCA param thing instead of changing -1 to 0 and back in the
  output stream param.

This commit was SVN r11245.
2006-08-17 14:52:20 +00:00
Brian Barrett
f31bf10b98 If we're ANY_SOURCE, don't grab the ompi_proc to get the convertor by going
through the c_pml_procs, as it might be an intercommunicator and therefore
c_my_rank might not be a valid index.

Fixes trac:266.

This commit was SVN r11238.

The following Trac tickets were found above:
  Ticket 266 --> https://svn.open-mpi.org/trac/ompi/ticket/266
2006-08-16 22:35:53 +00:00
Galen Shipman
7473d04a9a Simple failover is working.. ;-)
This commit was SVN r11237.
2006-08-16 22:32:18 +00:00
Brian Barrett
292068b34b * check return status of module init
This commit was SVN r11235.
2006-08-16 21:27:57 +00:00
Galen Shipman
e809a442e7 add the error handler registration to OB1..
This commit was SVN r11234.
2006-08-16 20:56:22 +00:00
Brian Barrett
74e95bc65f * more fixes for ticket #264. We need to keep the original address around,
so use the req_buff field for keeping track of the bsend buffer and the
  req_addr field for the user buffer, the way the comments suggested we
  were doing it

This commit was SVN r11233.
2006-08-16 20:24:28 +00:00
Galen Shipman
3b49953ce2 Add error callback to the btl interface, this allows error to be delivered to
the upperlayer assynchronously although there are some issues with this.. such
as there are multiple consumers of the btl's.. who get's the

This commit was SVN r11232.
2006-08-16 20:21:38 +00:00
Brian Barrett
0f47949703 * partial fix for #264... We need to return an MPI_ERR_BUFFER if we've run
out of buffer space

This commit was SVN r11229.
2006-08-16 17:32:31 +00:00
Ralph Castain
8c7f0ed9ae Change the SOH to the new State Monitoring and Reporting (SMR) framework. New API's will be appearing in the new framework shortly - this just gets the name change into the system.
Other changes:

1. Remove the old xcpu components as they are not functional.

2. Fix a "bug" in orterun whereby we called dump_aborted_procs even when we normally terminated. There is still some kind of bug in this procedure, however, as we appear to be calling the orterun job_state_callback function every time a process terminates (instead of only once when they have all terminated). I'll continue digging into that one.

This will require an autogen/configure, I'm afraid.

This commit was SVN r11228.
2006-08-16 16:35:09 +00:00
Brian Barrett
6d414f2d44 * use the MTL-specific output stream for all error messages
* use OPAL_OUTPUT_VERBOSE rather than printfs for debugging messages

This commit was SVN r11227.
2006-08-16 16:28:58 +00:00
Brian Barrett
3d75d828d3 * Work on ticket #259 -- Add MCA parameters to allow for tuning of the
usual suspects in the CM PML

This commit was SVN r11225.
2006-08-16 14:57:13 +00:00
Brian Barrett
139348647a * remove odd character
This commit was SVN r11212.
2006-08-15 23:58:00 +00:00
Brian Barrett
dc74a6a8e1 * implement iprobe for the MX MTL
This commit was SVN r11211.
2006-08-15 22:16:50 +00:00
Galen Shipman
259b9f04b9 GM needs ack+csum by default for PML DR..
This commit was SVN r11210.
2006-08-15 22:15:22 +00:00
Brian Barrett
0d218c6bdc * implement cancel for MX
This commit was SVN r11209.
2006-08-15 21:59:37 +00:00
Galen Shipman
84e7b90a19 Fix DR PML after the great MTL crusade.. Added a bit of debugging while I was
in there trying to track things down.. 

This commit was SVN r11208.
2006-08-15 21:44:55 +00:00
Brian Barrett
1cf4d0bd18 * Start of fix for #258 -- implement cancel so that we pass down to the
MTL layer.  Needed to include more knowledge of which fragment was
  which since both thin and heavy requests can be canceled

This commit was SVN r11207.
2006-08-15 21:12:03 +00:00
Ralph Castain
5dfd54c778 With the branch to 1.2 made....
Clean up the remainder of the size_t references in the runtime itself. Convert to orte_std_cntr_t wherever it makes sense (only avoid those places where the actual memory size is referenced).

Remove the obsolete oob barrier function (we actually obsoleted it a long time ago - just never bothered to clean it up).

I have done my best to go through all the components and catch everything, even if I couldn't test compile them since I wasn't on that type of system. Still, I cannot guarantee that problems won't show up when you test this on specific systems. Usually, these will just show as "warning: comparison between signed and unsigned" notes which are easily fixed (just change a size_t to orte_std_cntr_t).

In some places, people didn't use size_t, but instead used some other variant (e.g., I found several places with uint32_t). I tried to catch all of them, but...

Once we get all the instances caught and fixed, this should once and for all resolve many of the heterogeneity problems.

This commit was SVN r11204.
2006-08-15 19:54:10 +00:00
David Daniel
59f2d86c36 * Move Gleb's rcache work from the gleb-rcache branch to the trunk
This commit was SVN r11198.
2006-08-15 18:40:08 +00:00
Brian Barrett
cdffc3158d * only set threads if not running at thread single
This commit was SVN r11193.
2006-08-15 15:55:53 +00:00
Sven Stork
556d73085d - Introduce the info key "mpool" to specify which mpool to use for the
allocation. This is necessary to detect if the user requests a specific
  mpool for the allocationi. Searching the key values for a specific mpool
  name does not work for the case that the user provides an info object
  without mpool specific information (see Ticket #254).
- In the case that the user provides a info object without requesting a
  specific mpool we use malloc to allocate buffer instead of returning
  NULL (fix for Ticket #254 )

This commit was SVN r11188.
2006-08-15 09:21:40 +00:00
Ralph Castain
d2912f03e0 Cleanup a historical naming convention problem. Move the socket_errno definitions to the OPAL layer and change the name accordingly. This cleans up some interrelationship issues as well as removing a name confusion.
This commit was SVN r11186.
2006-08-14 20:14:44 +00:00
Jeff Squyres
474564a6b1 Bring over all the work from the /tmp/ib-hw-detect branch. In
addition to my design and testing, it was conceptually approved by
Gil, Gleb, Pasha, Brad, and Galen.  Functionally [probably somewhat
lightly] tested by Galen.  We may still have to shake out some bugs
during the next few months, but it seems to be working for all the
cases that I can throw at it.

Here's a summary of the changes from that branch: 

* Move MCA parameter registration to a new file (btl_openib_mca.c):
   * Properly check the retun status of registering MCA params
   * Check for valid values of MCA parameters
   * Make help strings better
   * Otherwise, the only default value of an MCA param that was
     changed was max_btls; it went from 4 to -1 (meaning: use all
     available)
 * Properly prototyped internal functions in _component.c
   * Made a bunch of functions static that didn't need to be public
   * Renamed to remove "mca_" prefix from static functions
   * Call new MCA param registration function
   * Call new INI file read/lookup/finalize functions
   * Updated a bunch of macros to be "BTL_" instead of "ORTE_"
   * Be a little more consistent with return values
   * Handle -1 for the max_btls MCA param
   * Fixed a free() that should have been an OBJ_RELEASE()
   * Some re-indenting
 * Added INI-file parsing
   * New flex file: btl_openib_ini.l
   * New default HCA params .ini file (probably to be expanded over
     time by other HCA vendors)
   * Added more show_help messages for parsing problems
   * Read in INI files and cache the values for later lookup
   * When component opens an HCA, lookup to see if any corresponding
     values were found in the INI files (ID'ed by the HCA vendor_id
     and vendor_part_id)
   * Added btl_openib_verbose MCA param that shows what the INI-file
     stuff does (e.g., shows which MTU your HCA ends up using)
   * Added btl_openib_hca_param_files as a colon-delimited list of INI
     files to check for values during startup (in order,
     left-to-right, just like the MCA base directory param).
   * MTU is currently the only value supported in this framework.
   * It is not a fatal error if we don't find params for the HCA in
     the INI file(s).  Instead, just print a warning.  New MCA param
     btl_openib_warn_no_hca_params_found can be used to disable
     printing the warning.
 * Add MTU to peer negotiation when making a connection
   * Exchange maximum MTU; select the lesser of the two

This commit was SVN r11182.
2006-08-14 19:30:37 +00:00
Brian Barrett
dd6fa1da2a * Fix for ticket #242, print a friendly error message if we can't reach
a particular peer.  Will now fail during MPI_INIT.  Printing of the
  error messages about no endpoints can be turned off.

This commit was SVN r11181.
2006-08-14 19:17:36 +00:00
Brian Barrett
24de93e2d4 * fix some compile errors with having OB1 direct called
This commit was SVN r11170.
2006-08-14 15:43:03 +00:00