1
1
Граф коммитов

4654 Коммитов

Автор SHA1 Сообщение Дата
Rainer Keller
1bd94f2d98 - When calling ompi_mtl_portals_finalize, when then pml/ob1 is used
(aka w/o  --mca pml cm), make sure PtlEQGet will actually work
   on ompi_mtl_portals.ptl_eq_h -- do so without adding code to 
   ompi_mtl_portals_progress.

   Otherwise we abort() with
[nid09979:32503] ompi_mtl_portals_finalize: Going to call ompi_mtl_portals_progress
[nid09979:32503]  Error returned from PtlEQGet.  Error code - 14
[nid09979:32502] Signal: Aborted (6)
[nid09979:32502] Signal code:  (-6) 

This commit was SVN r21761.
2009-08-04 22:48:07 +00:00
George Bosilca
98bdf5d17b Remove a compiler warning about missing braces around
initializer.

This commit was SVN r21760.
2009-08-04 21:41:14 +00:00
Jeff Squyres
846e6e4bad Fix borkedness from r21755:
* No need for OPAL_SIZEOF_BOOL and OPAL_SIZEOF_INT in comm_inln.h --
   just use sizeof()
 * Fix logic in ompi_setup_cxx.m4 to account for the case where we
   ''do'' have a C++ compiler (duh!!)
 * Fix spelling error in a shell variable that ended up making a
   bad/empty #define

This should bring the trunk back to being functional.  Sorry for the
interruption, folks...

This commit was SVN r21758.

The following SVN revision numbers were found above:
  r21755 --> open-mpi/ompi@90d6491737
2009-08-04 18:45:11 +00:00
Rainer Keller
5a4c6434e7 - Single request, so use single wait
Technically we should have used MPI_STATUSES_IGNORE, anyhow.
   (however both MPI_STATUS*_IGNORE are ((MPI_Status*)0) ;-)

This commit was SVN r21757.
2009-08-04 14:51:15 +00:00
Rainer Keller
2989b65e68 - Initialize the MPI_COMPLEX, to enable the pre-split behaviour.
This however may not be what we finally want to support MPI_COMPLEX:
   As with a possible difference of C99 _Bool and C++ Bool, we may want
   to have a base opal_datatype_complex (as was initially in the
   ompi-ddt branch), instead of assuming a struct of
   opal_datatype_float... Think of stricter alignment.

   The configure checks for (float/double/long double) complex have
   been added with the ompi-ddt branch.

This commit was SVN r21756.
2009-08-04 14:44:42 +00:00
Jeff Squyres
90d6491737 Since ompi_info is no longer written in C++, we no longer require a
C++ compiler in configure.  If we have a C++ compiler, then the MPI
C++ bindings are built by default.  If we don't have a C++ compiler,
then the MPI C++ bindings are not built by default.

--enable-mpi-cxx will now force an error if there is no C++ compiler
available.  --disable-mpi-cxx (or the lack of a C++ compiler) will now
disable many of the C++ compiler checks in configure.

Note that there are a few items to clean up regarding the difference
between C's _Bool type and C++'s bool type.  Right now, we assume that
they are the same.  But they aren't, and they shouldn't be treated as
such.  This cleanup will be forced in MPI-2.2 with the introduction of
the MPI_C_BOOL MPI datatype.

This commit was SVN r21755.
2009-08-04 11:54:01 +00:00
George Bosilca
32416761ce Make the open/close symmetric with regards to the local variable
contruction/destruction.

This commit was SVN r21753.
2009-08-03 16:45:18 +00:00
George Bosilca
b3ae54fb46 As some of our components do not play nicely with the open/close functionality
in the case of errors, close the opened components if and only if the whole
opening process complete succesfully.

This commit was SVN r21752.
2009-08-03 16:44:12 +00:00
Avneesh Pant
af09e7678c Convert a few opal_output() calls to instead use orte_show_help() as well as do some minor cosmetic changes dealing with tab spacing and c-blocks being enclosed with \{\}. There was also a long standing bug with the PSM mtl if the number of hardware contexts on adapter were less than the number of cores on a node (The default case is they are the same hence no issues were reported). For completeness we take care of this case as well but it requires us to tell PSM how many local processes are running on a node and the local rank of the process on a node so it can allocate the available hardware contexts appropriately.
This commit was SVN r21745.
2009-07-30 02:55:20 +00:00
Jeff Squyres
bff6a1be93 Gah! "vis", not "visi"
This commit was SVN r21741.
2009-07-29 14:22:46 +00:00
Jeff Squyres
3b9ca5a084 Reduce the label name so that it's not longer than 24 chars
This commit was SVN r21740.
2009-07-29 14:19:44 +00:00
George Bosilca
0bf381e931 This patch try to solve a issue on Leopard. The supposedly global
variables that are not initialized and are declared in a file that
doesn't export any globally visible function are marked as
non-initialized constants, i.e. uninitialized common symbols. For some
obscure reasons, they get removed from the object files on Mac OS X.

So far I found two solution to this problem. One require the addition
of "-c" to the linker command, the second one (corresponding to this
patch) force them to became a common initialized symbol.

This commit was SVN r21739.
2009-07-28 17:06:16 +00:00
Avneesh Pant
38e48d4e2f Add support for MCA parameters for PSM MTL to specify IB unit, port, IB service level and PSM debug level to use. Also specify in the openib btl params file that QLogic hardware supports a max inlined messages size of 0 only.
This commit was SVN r21734.
2009-07-24 20:09:39 +00:00
Jeff Squyres
c7376ae053 Start using Libtool's shared library versioning scheme. See lengthy
note in VERSION file.

NOTE: the versions will ''always'' be 0:0:0 on the SVN trunk and
developer branches.  They will only have meaningful values (starting
with 0:0:0 in 1.3.4) on release branches.  Only RM's will modify these
values immediately preceeding a release.

This commit was SVN r21729.
2009-07-23 21:35:17 +00:00
Edgar Gabriel
9a369d0fc1 - we accidentally decreased the counter for the number of dynamic
communicators twice, once in dpm.disconnect_wait, and once in
comm_free. The second location seems to be the right place for that (since a
communicator could be freed, and not disconnected), remove the instance in
disconnect_wait.

- add some error messages in case something goes wrong.

This commit was SVN r21720.
2009-07-20 19:54:24 +00:00
Terry Dontje
d432c9fdbc Add asserts to catch when btl_eager_limit is smaller than the pml headers.
This commit was SVN r21707.
2009-07-17 14:54:18 +00:00
George Bosilca
8275120656 Get rid of the ompi_convertor.h header file. Replace all references to ompi_convertor
by opal_convertor.
Cleanup the pcie BTL.

This commit was SVN r21703.
2009-07-16 19:13:30 +00:00
George Bosilca
2b57fc3835 Add the missing headers.
This commit was SVN r21701.
2009-07-16 18:42:14 +00:00
George Bosilca
3e971e61f3 The system headers are supposed to be protected by #ifdef and not by #if.
This commit was SVN r21700.
2009-07-16 18:27:33 +00:00
George Bosilca
d07ffedc54 No opal datatype functions in the BTL. The datatype attached to the
convertor is an ompi_datatype_t so calling the ompi level functions
is the way to go.

This commit was SVN r21698.
2009-07-16 18:25:08 +00:00
Jeff Squyres
9dc7f884b2 Fix yet another compile error from the great DDT split (r21641). Sigh.
This commit was SVN r21697.

The following SVN revision numbers were found above:
  r21641 --> open-mpi/ompi@6c5532072a
2009-07-16 18:08:03 +00:00
George Bosilca
e1383027e1 Correct a comment and cleanup/reorder the code.
This commit was SVN r21696.
2009-07-16 17:41:32 +00:00
George Bosilca
33e7cc864c The ops should be indexed based on the MPI datatype index (which is
actually the same as the Fortran value of the MPI type).

This commit was SVN r21689.
2009-07-15 21:30:09 +00:00
Ralph Castain
e75d9b8296 Use orte_notifier to alert sys admins to checksum violations in the csum pml.
Add ability to store the RM's jobid string to tag the notifier message so that the sys admin knows what job had the problem.

This commit was SVN r21687.
2009-07-15 19:43:26 +00:00
Rainer Keller
8243831d76 - Get OpenIB BTL to work with old libibverbs installation
Tested on smoky.

This commit was SVN r21685.
2009-07-15 16:12:47 +00:00
Ralph Castain
d908bbe4a8 Add missing header file
This commit was SVN r21684.
2009-07-15 15:09:59 +00:00
Matthias Jurenz
459784fc7b The meaning of the VT configure option '--with-openmpi' was changed (see r21679), so that the options MPICC, --with-mpi-inc-dir, and --with-mpi-status-size are not longer necessary.
This commit was SVN r21683.

The following SVN revision numbers were found above:
  r21679 --> open-mpi/ompi@77b397895c
2009-07-15 11:52:48 +00:00
Matthias Jurenz
77b397895c - fixed bug in Fortran MPI wrappers: added conversion for the MPI Fortran constants MPI_BOTTOM and MPI_IN_PLACE
- reduced list of supported MPI implementations to Open MPI
  (changed meaning of configure option '--with-openmpi': this option indicates an Open MPI internal build)
- updated version number of integrated VT to 5.4.10

This commit was SVN r21679.
2009-07-15 10:38:10 +00:00
Ralph Castain
2a1cf2ae81 Create two new modex functions for send/recv of specific OPAL data types - basically, ensure that the values are packed and unpacked appropriately. Provides a convenience function as otherwise anyone wanting to send a value had to do this themselves.
Use it to send the arch.

This commit was SVN r21677.
2009-07-15 04:54:53 +00:00
Ralph Castain
dbac602be5 Add support for the add-host and add-hostfile MPI Info keys to allow Comm_spawn users to add new hosts to those already known by mpirun.
Requires full testing once comm_spawn is fixed (Edgar is working that now).

This commit was SVN r21664.
2009-07-14 14:34:11 +00:00
Rainer Keller
e37959c168 - Need to name all the headers to get make distcheck to compile.
This commit was SVN r21663.
2009-07-14 13:54:01 +00:00
Matthias Jurenz
ea501084db Added support for cycle counter timer on Alpha and Sparc platforms
(thanks to George Bosilca)

This commit was SVN r21662.
2009-07-14 07:18:18 +00:00
Shiqing Fan
f991c87c6a A few structures that need to be exported for Windows.
This commit was SVN r21658.
2009-07-14 06:53:42 +00:00
Ralph Castain
60edbc7220 Fix hetero operations and comm_spawn (to a point).
Remove all architecture references from ORTE and put them back in the modex using modex_send/recv calls.

Hetero operations are now fully supported again. Comm_spawn now works up to the point where it segfaults due to an error in the CID code - which now allows Edgar to dig further! :-)

This commit was SVN r21655.
2009-07-13 20:03:41 +00:00
George Bosilca
2143424eb5 The MCA parameter should always be taken into account, independent on
how many networks are available on the node.

This commit was SVN r21652.
2009-07-13 19:40:00 +00:00
Rainer Keller
1b9448581b - Keep the size of the predefined as before...
This commit was SVN r21651.
2009-07-13 18:05:08 +00:00
Josh Hursey
8d9d2ba7d1 Fix the datatype usage in CRCP Bkmrk. as a result of the great datatype shift in r21641
This commit was SVN r21650.

The following SVN revision numbers were found above:
  r21641 --> open-mpi/ompi@6c5532072a
2009-07-13 17:54:26 +00:00
Shiqing Fan
503f2817b3 Corresponding changes to r21641 and r21642 for Windows.
- Add a CMake macro for checking OPAL_MAX_XXX values, re-written from OPAL_WITH_OPTION_MIN_MAX_VALUE m4 function. 
- Definition prefix changes and additional datatype alignments checking.
- Finish the datatype splitting on Windows too. :-)

This commit was SVN r21649.

The following SVN revision numbers were found above:
  r21641 --> open-mpi/ompi@6c5532072a
  r21642 --> open-mpi/ompi@c971c09eb6
2009-07-13 17:39:41 +00:00
Pavel Shamis
e6b340d7ad Fixing compilation failure - adding missing output.h file.
This commit was SVN r21643.
2009-07-13 08:59:13 +00:00
Rainer Keller
c971c09eb6 - Runtime and include files missed in last commit
This commit was SVN r21642.
2009-07-13 04:59:13 +00:00
Rainer Keller
6c5532072a - Split the datatype engine into two parts: an MPI specific part in
OMPI
   and a language agnostic part in OPAL. The convertor is completely
   moved into OPAL.  This offers several benefits as described in RFC
   http://www.open-mpi.org/community/lists/devel/2009/07/6387.php
   namely:
    - Fewer basic types (int* and float* types, boolean and wchar
    - Fixing naming scheme to ompi-nomenclature.
    - Usability outside of the ompi-layer.
 - Due to the fixed nature of simple opal types, their information is
   completely
   known at compile time and therefore constified
 - With fewer datatypes (22), the actual sizes of bit-field types may be
   reduced
   from 64 to 32 bits, allowing reorganizing the opal_datatype
   structure, eliminating holes and keeping data required in convertor
   (upon send/recv) in one cacheline...
   This has implications to the convertor-datastructure and other parts
   of the code.
 - Several performance tests have been run, the netpipe latency does not
   change with
   this patch on Linux/x86-64 on the smoky cluster.
 - Extensive tests have been done to verify correctness (no new
   regressions) using:
   1. mpi_test_suite on linux/x86-64 using clean ompi-trunk and
    ompi-ddt:
    a. running both trunk and ompi-ddt resulted in no differences
       (except for MPI_SHORT_INT and MPI_TYPE_MIX_LB_UB do now run
       correctly).
    b. with --enable-memchecker and running under valgrind (one buglet
       when run with static found in test-suite, commited)
   2. ibm testsuite on linux/x86-64 using clean ompi-trunk and ompi-ddt:
      all passed (except for the dynamic/ tests failed!! as trunk/MTT)
   3. compilation and usage of HDF5 tests on Jaguar using PGI and
      PathScale compilers.
   4. compilation and usage on Scicortex.
 - Please note, that for the heterogeneous case, (-m32 compiled
   binaries/ompi), neither
   ompi-trunk, nor ompi-ddt branch would successfully launch.

This commit was SVN r21641.
2009-07-13 04:56:31 +00:00
Ralph Castain
3561880546 Silence compiler warning about comparing signed and unsigned values
This commit was SVN r21637.
2009-07-11 18:36:43 +00:00
Brian Barrett
2f3c0b4fcf Drain pipe from service thread to main thread during shutdown. By this
point, the event engine has been shut down until btl finalization is
done, so opal_progress in the wait loop is not an option - we have
to drain from inside the btl.

Clean up the looping structure for the finalize routine

Update copyrights.

This commit was SVN r21620.
2009-07-09 22:13:10 +00:00
Brian Barrett
ac34b1de69 RDMA CM doesn't retry if a packet is dropped, just timesout during route
discovery, which results in a timeout and we don't recover.  Instead,
try to recover a couple of times by retrying.

This commit was SVN r21619.
2009-07-09 22:10:06 +00:00
Edgar Gabriel
b6f292f794 add a uint8_t to the startup modex which allows us to recognize whether
different processes have requested different levels of thread support. This
verification is restricted to MPI_COMM_WORLD.

In case one ore more processes have requested support for MPI_THREAD_MULTIPLE,
the cid selection algorithm will fall back to the original, thread safe
approach. Else, it uses the block-algorithm.

For dynamic communicators, we always fall back now to the original algorithm. 
This has been tested for homogeneous and heterogeneous settings for
MCW. However, I could not test yet the dynamic comm scenario for technical
reasons, and that's why I don't close yet ticket 1949.

This commit was SVN r21613.
2009-07-07 18:32:14 +00:00
George Bosilca
311e27b42f Pretty print an error message when the specified range of ports (for both
IPv4 and IPv6) is outside the legal boundaries. This fixes trac:1869.

This commit was SVN r21612.

The following Trac tickets were found above:
  Ticket 1869 --> https://svn.open-mpi.org/trac/ompi/ticket/1869
2009-07-07 17:52:30 +00:00
George Bosilca
4038834dfb Convert the port number in network order before binding the socket.
Thanks to Mariusz Mamonski (mamonski@man.poznan.pl) for the bug
report and patch.

This commit was SVN r21610.
2009-07-07 17:21:28 +00:00
George Bosilca
90d719125c If the file interface is disabled don't try to print
a string as an int.

This commit was SVN r21609.
2009-07-07 15:28:34 +00:00
Matthias Jurenz
03f94ae066 Fixed initialization of new hash node
This commit was SVN r21605.
2009-07-06 14:03:01 +00:00
Matthias Jurenz
b4026c40ad - added tool 'otfinfo' which can be used to get basic information of a trace
(also usable to check validation of a trace)
- removed OTF python bindings
  (a working version of the OTF python bindings are available in the latest stand-alone release;
  see http://www.tu-dresden.de/zih/otf/)
- incremented OTF version number

This commit was SVN r21601.
2009-07-06 11:55:09 +00:00
Jeff Squyres
92e40cb20a Enable the coll sync component to barrier before each 1000th collective.
This commit was SVN r21594.
2009-07-02 20:16:45 +00:00
Ralph Castain
e30826c6e1 Quiet some compiler warnings
This commit was SVN r21591.
2009-07-02 17:48:36 +00:00
Brian Barrett
3b410b0200 Increase context ref count and push on list before calling rdma_resolve_addr,
in case the event returns before rdma_resolve_addr returns.

This commit was SVN r21588.
2009-07-02 16:12:19 +00:00
Shiqing Fan
0b56a8a4d5 Enable IPv6 on Windows by default, and fix two type casts for IPv6 operations.
This commit was SVN r21586.
2009-07-02 14:41:03 +00:00
Ralph Castain
f832352b45 Clean up some compiler warnings
This commit was SVN r21577.
2009-07-01 16:51:11 +00:00
Ralph Castain
1408559645 Weird - restore Shiqing's typecast fixes. No idea how they got reverted in that last commit
This commit was SVN r21573.
2009-07-01 13:26:07 +00:00
Ralph Castain
de90bd8285 Add missing include file so that Absoft compile tests will pass
This commit was SVN r21572.
2009-07-01 13:19:06 +00:00
Shiqing Fan
da6c7e4518 Just two type casts for Windows.
This commit was SVN r21571.
2009-07-01 09:06:59 +00:00
Ralph Castain
e72346bf9d Cleanup a define to separate it from the info.h file
This commit was SVN r21569.
2009-06-30 22:56:25 +00:00
Ralph Castain
9e524a41cf Re-implement ompi_info in C, replacing the prior C++ code. This removes the last C++ code from the OMPI code base, thus relieving the requirement for C++ support unless the user requests the C++ bindings.
Jeff will update the build system later to remove the C++ checks per the above caveat.

This commit was SVN r21568.
2009-06-30 21:41:28 +00:00
Jeff Squyres
cad12fda5f * Remove an extra blank line from the help file
* Add the help file to the Makefile.am so that it gets installed

This commit was SVN r21567.
2009-06-30 18:58:09 +00:00
Eugene Loh
fcd9fabae9 Minor cleanup in the sm BTL. http://www.open-mpi.org/community/lists/devel/2009/06/6363.php
This commit was SVN r21556.
2009-06-27 23:42:09 +00:00
Nysal Jan
5b13fd004a Bump up default MTU for eHCA 2. Improves peak unidirectional bandwidth by around 14%
This commit was SVN r21553.
2009-06-27 07:39:30 +00:00
Ralph Castain
ee18838e2f Remove svn conflict lines due to commit r21551 in the sm btl. I #if 0'd out the offending line that cause the conflict just in case it was the correct one. However, this now compiles cleanly, minus the following warnings that I wasn't sure which way to resolve:
btl_sm.c: In function ‘mca_btl_sm_sendi’:
btl_sm.c:734: warning: comparison between signed and unsigned
btl_sm.c: In function ‘mca_btl_sm_send’:
btl_sm.c:812: warning: comparison between signed and unsigned

This commit was SVN r21552.

The following SVN revision numbers were found above:
  r21551 --> open-mpi/ompi@bd995d26b4
2009-06-27 01:39:15 +00:00
Eugene Loh
bd995d26b4 Try to improve flow control in the sm BTL:
- poll FIFO occasionally even if just sending messages
- retry pending sends more often
  - just before trying a new send
  - as part of mca_btl_sm_component_progress
Maintain two new mca_btl_sm_component variables, num_outstanding_frags
and num_pending_sends, to keep overhead low.


Drain only one message fragment from the FIFO per btl_sm_component_progress
call (rather than drain until empty, which in retrospect everyone considers
to have been a mistake).

This commit was SVN r21551.
2009-06-27 00:12:56 +00:00
George Bosilca
24e74922ce Not yet the right time for this to be there.
This commit was SVN r21540.
2009-06-26 16:00:55 +00:00
Shiqing Fan
5b4b8f8899 Get rid of a bunch of compiler warnings on Windows.
This commit was SVN r21536.
2009-06-26 07:51:39 +00:00
Lenny Verkhovsky
7f8dc7c8b8 fix for r21524, mispell fix HAVE_IBV_FORK_INIT
This commit was SVN r21533.

The following SVN revisions from the original message are invalid or
inconsistent and therefore were not cross-referenced:
  r21524
2009-06-25 17:45:38 +00:00
Terry Dontje
efac1b73fb Surround use of want_fork_support structure field with ifdef instead of c conditional.
This commit was SVN r21526.
2009-06-25 12:03:30 +00:00
Matthias Jurenz
626799ee0f Fixed detection of Sun Studio compiler on Linux for automatic instrumentation
This commit was SVN r21525.
2009-06-25 11:11:41 +00:00
Nysal Jan
938599cb2d Fix build failure with latest IBM XL C/C++ v10.1 compiler. Also this seems like cleaner code.
This commit was SVN r21497.
2009-06-23 14:08:04 +00:00
George Bosilca
7f24b41051 This function doesn't have to be globally visible.
This commit was SVN r21494.
2009-06-22 17:14:54 +00:00
Ralph Castain
110c95fb1c Make the match for the "env" option to mpi_show_mca_params to be less strict - match it if the string at least starts with "env" so "enviro" and "environ" all match.
This commit was SVN r21478.
2009-06-19 03:43:53 +00:00
Matthias Jurenz
edce15d08f Changes in VT documentation:
- added note for BFD license issues
- minor cleanups

This commit was SVN r21472.
2009-06-18 14:33:04 +00:00
Jeff Squyres
c39998db17 Also show the "you might not have enough registered memory" warning
message earlier in the openib BTL startup sequence

This commit was SVN r21469.
2009-06-18 12:24:39 +00:00
Matthias Jurenz
d7aa9abc4e Added configure option '--with[out]-bfd' to control usage of BFD library to get symbol information for GNU, Intel, and Pathscale compiler instrumentation
This commit was SVN r21458.
2009-06-17 08:51:26 +00:00
Rolf vandeVaart
36a560506c Fix error message to match code default.
This commit was SVN r21452.
2009-06-16 20:59:53 +00:00
Rainer Keller
5e6061af02 - Few fixes and comments
This commit was SVN r21443.
2009-06-15 21:12:04 +00:00
Matthias Jurenz
24afd66352 Reverting previous commit and going back to r21178
This commit was SVN r21433.

The following SVN revision numbers were found above:
  r21178 --> open-mpi/ompi@fb3bbb8021
2009-06-15 11:41:33 +00:00
Ralph Castain
44bb265a52 Add a new MPI_Info key to preposition OMPI libraries - implementation underway, but this just defines and passes the new key
This commit was SVN r21425.
2009-06-12 17:53:13 +00:00
Jeff Squyres
814a8f5e0f * Fix #1916: endian problems in iwarp wireup on big endian machines
(now works on both big and little endian machines)
 * Be a little more flexible when looking for active devices in
   btl_openib_component.c
 * Add device name and port number to lots of verbose and help
   messages
 * Add a bunch of verbose messages to give insight into what is
   occurring during all the CPC wireups

This commit was SVN r21418.
2009-06-11 17:30:30 +00:00
Ralph Castain
4881cd0df3 Revert the prior change out from the individual .h files - the problem was in the Makefile.am's, causing the make dist to fail.
This commit was SVN r21414.
2009-06-11 03:15:47 +00:00
Ralph Castain
91ab2b3e4f Specify complete path to included header files so it compiles in all environments
This commit was SVN r21412.
2009-06-11 02:46:30 +00:00
Avneesh Pant
295c0c36bc Add new QLogic adapter to device parameter file.
This commit was SVN r21409.
2009-06-10 23:37:38 +00:00
Terry Dontje
fa9f356c8c This commit fixes trac:1931. By separating out structures and defines used by
the debugger plugins into files suffixed by _dbg.h.

This commit was SVN r21404.

The following Trac tickets were found above:
  Ticket 1931 --> https://svn.open-mpi.org/trac/ompi/ticket/1931
2009-06-10 14:57:28 +00:00
Ralph Castain
c3c1ab1337 Correct a comment in paffinity.h about what paffinity_get returns - it was inaccurate.
Revamp the affinity detection/set procedure in mpi_init to correctly detect when we have already been bound to processors, given the revised understanding of paffinity_get. Add a new paffinity macro to make checking for already bound a little nicer.

This commit was SVN r21402.
2009-06-09 14:33:35 +00:00
George Bosilca
f9a510fd8a There is no need for an atomic read if we are not in a threaded case.
This commit was SVN r21394.
2009-06-08 23:55:52 +00:00
Matthias Jurenz
fac893838f Added configure option '--with-bfd-objects' to include the object files from the static BFD library into the VT libraries
(it's strongly recommend for RPM builds of OMPI to avoid BFD incompatibilities)

This commit was SVN r21383.
2009-06-05 11:28:55 +00:00
Ralph Castain
4d3aa5a8a4 Once again, into the breach!
Yes, friends, our favorite PCIE BTL has resurfaced as mgmt vacillates over its existence. This is an updated version that actually mostly works, in its final stages of debugging.

Some generalization still remains to be done...

This commit was SVN r21358.
2009-06-02 22:26:36 +00:00
Jeff Squyres
b363fc2a6b Per the discussion yesterday, remove all use of MPI_Flogical -- it's
not an official MPI type, so it shouldn't be named that way.  Instead,
use ompi_fortran_logical_t.

This commit was SVN r21353.
2009-06-02 12:04:17 +00:00
Shiqing Fan
e46bf10efd Correctly include win32 util header.
This commit was SVN r21343.
2009-06-01 19:16:00 +00:00
Rainer Keller
b572dc3591 - As discussed revert r21330, Fortran-configure info should
not end up in OPAL
 - Will post an updated patch for the OMPI_ALIGNMENT_ parts (within C).

This commit was SVN r21342.

The following SVN revision numbers were found above:
  r21330 --> open-mpi/ompi@95596d1814
2009-06-01 19:02:34 +00:00
Jeff Squyres
2f810a1e5e Fix a compiler warning and do some slightly-smarter unsigned int
checks.

This commit was SVN r21341.
2009-06-01 14:12:46 +00:00
Jeff Squyres
055b50213b A slightly better way to fix the problem noted in r21137.
This commit was SVN r21339.

The following SVN revision numbers were found above:
  r21137 --> open-mpi/ompi@f6da7d86a2
2009-06-01 13:58:37 +00:00
Josh Hursey
1fef44085b As a result of r21308, if we are building with "--disable-mpi-io " make sure that MPI_MAX_DATAREP_STRING is defined in mpi.h
MTT caught builds failing with the following error which flagged the problem:
{{{
param.cc:802: error: ‘MPI_MAX_DATAREP_STRING’ was not declared in this scope
}}}

This commit was SVN r21337.

The following SVN revision numbers were found above:
  r21308 --> open-mpi/ompi@2f9765926e
2009-06-01 12:44:53 +00:00
Rainer Keller
95596d1814 - Move alignment and size output generated by configure-tests
into the OPAL namespace, eliminating cases like opal/util/arch.c
   testing for ompi_fortran_logical_t.
   As this is processor- and compiler-related information
   (e.g. does the compiler/architecture support REAL*16)
   this should have been on the OPAL layer.
 - Unifies f77 code using MPI_Flogical instead of opal_fortran_logical_t

 - Tested locally (Linux/x86-64) with mpich and intel testsuite
   but would like to get this week-ends MTT output


 - PLEASE NOTE: configure-internal macro-names and
   ompi_cv_ variables have not been changed, so that
   external platform (not in contrib/) files still work.

This commit was SVN r21330.
2009-05-30 15:54:29 +00:00
Nysal Jan
85773de539 Move 'multi-frag receive' fixes to csum PML
This commit was SVN r21323.
2009-05-29 08:07:33 +00:00
Rainer Keller
9ef87898e0 (originally, did not want to do this during business hours, but
well..)
 - As Jeff suggested, for m4 macros, dont use _ OPAL, but
   rather OPAL_ prefix
 - Set the variable before AC_SUBST, so that replacement happens
   in f77 header-file, too.

This commit was SVN r21316.
2009-05-28 20:28:43 +00:00
Jeff Squyres
07c7a7a255 Make ompi_info behave a bit better if STL decides not to play with 0
length string constructors.  Print a developer warning if we have a
too-long field (in developer builds only).

This commit was SVN r21315.
2009-05-28 16:15:06 +00:00
Jeff Squyres
590fbcff6c Shorten the labels so that we don't throw C++ exceptions for 0-length
strings (true fix for that coming after this).  Also rename so that
its names that MPI users will care about.

This commit was SVN r21314.
2009-05-28 15:51:46 +00:00