File openmpi.changes of Package openmpi

Mon Aug  4 13:59:53 CEST 2014 -

- Move plugins to main package (bnc#887130).

Tue Jun  3 11:24:34 CEST 2014 -

- Update to OFED 3.12 final.

Fri May  9 15:03:47 CEST 2014 -

- Support for udapl was dropped beginning with 1.7 so remove
  all traces in the spec file.

Mon May  5 16:33:30 CEST 2014 -

- Include a file only for x86 and x86_64.

Mon May  5 11:06:54 CEST 2014 -

- Fix source URL.
- Fix file list for devel package.

Wed Apr 30 14:09:05 CEST 2014 -

- Fix file list for devel package.

Tue Apr 29 18:00:19 CEST 2014 -

- Split off everything that is only needed for developing code
  using OpenMPI into openmpi-devel, including stuff like the
  wrapper scripts for the compilers.
- The openmpi-testsuite package now creates a package that only
  contains the test log.
- Add openmpi-no_network_in_build.patch to not run the nfs based
  test inside the buildroot.

Wed Apr 23 17:59:27 CEST 2014 -

- Update to 1.8.1:
  - Fix serious bug where running mpi bugs as root deletes files
    in /.

Thu Apr  3 16:58:45 CEST 2014 -

- Update to 1.8. Changes since 1.7.5:
  * Commit upstream ROMIO fix for mixed NFS+local filesystem environments.
  * Several fixes for MPI-3 one-sided support.  For example,
    arbitrary-length datatypes are now supported.
  * Add config support for the Mellanox ConnectX 4 card.
  * Add missing MPI_COMM_GET|SET_INFO functions, and missing
    MPI_WEIGHTS_EMPTY and MPI_ERR_RMA_SHARED constants.  Thanks to
    Lisandro Dalcin for pointing out the issue.
  * Update some help messages in OSHMEM, the usnic BTL, the TCP BTL, and
    ORTE, and update documentation about ompi_info's --level option.
  * Fix some compiler warnings.
  * Ensure that ORTE daemons are not bound to a single processor
    if TaskAffinity is set on by default in Slurm. Thanks to Artem Polyakov
    for identifying the problem and providing a patch

- fix ifarch test for intel/amd.

Tue Apr  1 09:51:22 UTC 2014 -

- fixed specfile, adopted version to 1.8 rc1  

Mon Mar 31 12:04:41 CEST 2014 -

- Update to 1.8 rc1.

Wed Mar  5 13:38:04 CET 2014 -

- Compile with PSM support on i686 and c86_64 (fate#315889)

Wed Feb 12 17:38:09 CET 2014 -

- Update to 1.7.4, i.e. a version with support for MPI-3
  (fate#316375, fate#314327, fate#314835):
  The list of features and bug fixes since 1.7.2 is far too long
  to list here. See the file NEWS in the package documentation for
  a detailed listing.  
  as ackagwe documentation.
- Create openmpi-testsuite.spec to run the opnmpi testsuite.
- Because of one yet unfixed bug running the test suite will always

Mon Feb 10 20:02:28 UTC 2014 -

- Added infinipath-psm-devel to BuildRequires

Thu Jan 16 12:16:43 CET 2014 -

- Create openmpi-testsuite.spec to run the openmpi testsuite.
- Because of one yet unfixed bug running the test suite will always

Sun Oct 20 12:04:38 UTC 2013 -

- Split libraries into subpackage to satisfy program dependencies
  without requiring the development environment

Sat Oct 12 14:16:27 UTC 2013 -

- Enable use of GCC intrinsics for atomics instead of custom assembler
- Remove remove-assembler-for-armv5.diff

Thu Jun 27 13:30:32 CEST 2013 -

- Update to 1.7.2:
  * Updated default SRQ parameters for the openib BTL.
  * Major VampirTrace update to
  * Fix to set flag==1 when MPI_IPROBE is called with MPI_PROC_NULL.
  * Set the Intel Phi device to be ignored by default by the openib BTL.
  * Decrease the internal memory storage used by intrinsic MPI datatypes
    for Fortran types.  Thanks to Takahiro Kawashima for the initial
  * Fix total registered memory calculation for Mellanox ConnectIB and
    OFED 2.0.
  * Fix possible data corruption in the MXM MTL component.
  * Remove extraneous -L from hwloc's embedding.  Thanks to Stefan
    Friedel for reporting the issue.
  * Fix contiguous datatype memory check.  Thanks to Eric Chamberland
    for reporting the issue.
  * Make the openib BTL more friendly to ignoring verbs devices that are
    not RC-capable.
  * Fix some MPI datatype engine issues.  Thanks to Thomas Jahns for
    reporting the issue.
  * Add INI information for Chelsio T5 device.
  * Integrate MXM STREAM support for MPI_ISEND and MPI_IRECV, and other
    minor MXM fixes.
  * Improved alignment for OpenFabrics buffers.
  * Fix to not show amorphous "MPI was already finalized" error when
    failing to MPI_File_close an open file.  Thanks to Brian Smith for
    reporting the issue.
- Modify spec so that openmpi can be compiled without UDAPL support
  but with support for IB verbs.
- Simplify macros so that the build service parser understands them.
- Add openmpi-no_date_and_time.patch to not use __DATE__.

Mon Apr  8 12:20:50 CEST 2013 -

- Update to 1.7.
  The following is an abbreviated list of changes in v1.7 (note that
  countless other smaller improvements and enhancements are not shown

  - Added MPI-3 functionality:
      - Matched probe
      - Non-blocking collectives
      - MPI_INFO_ENV support
      - Fortran '08 bindings (see below)

  - Dropped support for checkpoint/restart due to loss of maintainer :-(
  - Enabled compile-time warning of deprecated MPI functions by default
    (in supported compilers).
  - Revamped Fortran MPI bindings (see the README for details):
    - "mpifort" is now the preferred wrapper compiler for Fortran
    - Added "use mpi_f08" bindings (for compilers that support it)
    - Added better "use mpi" support (for compilers that support it)
    - Removed incorrect MPI_SCATTERV interface from "mpi" module that
      was added in the 1.5.x series for ABI reasons.
  - Lots of VampirTrace upgrades and fixes; upgrade to v5.14.3.
  - Modified process affinity system to provide warning when bindings
    result in being "bound to all", which is equivalent to not being
  - Removed maffinity, paffinity, and carto frameworks (and associated
    MCA params).
  - Upgraded to hwloc v1.5.1.
  - Added performance improvements to the OpenIB (OpenFabrics) BTL.
  - Made malloc hooks more friendly to IO interprosers. Thanks to the
    bug report and suggested fix from Darshan maintainer Phil Carns.
  - Added support for the DMTCP checkpoint/restart system.
  - Added support for the Cray uGNI interconnect.
  - Fixed header file problems on OpenBSD.
  - Fixed issue with MPI_TYPE_CREATE_F90_REAL.
  - Wrapper compilers now explicitly list/link all Open MPI libraries if
    they detect static linking CLI arguments.
  - Open MPI now requires a C99 compiler to build. Please upgrade your
    C compiler if you do not have a C99-compliant compiler.
  - Fix MPI_GET_PROCESSOR_NAME Fortran binding to set ierr properly.
    Thanks to LANL for spotting the error.
  - Many MXM and FCA updates.
  - Fixed erroneous free of putenv'ed string that showed up in Valgrind
  - Fixed MPI_IN_PLACE case for MPI_ALLGATHER.
  - Fixed a bug that prevented MCA params from being forwarded to
    daemons upon launch.
  - Fixed issues with VT and CUDA --with-cuda[-libdir] configuration CLI
  - Entirely new implementation of many MPI collective routines focused
    on better performance.
  - Revamped autogen / build system.
  - Add new sensor framework to ORTE that includes modules for detecting
    stalled applications and processes that consume too much memory.
  - Added new state machine framework to ORTE that converts ORTE into an
    event-driven state machine using the event library.
  - Added a new MCA parameter (ess_base_stream_buffering) that allows the user
    to override the system default for buffering of stdout/stderr streams
    (via setvbuf). Parameter is not visible via ompi_info.
  - Revamped the launch system to allow consideration of node hardware
    in assigning process locations and bindings.
  - Added the -novm option to preserve the prior launch behavior.
  - Revamped the process mapping system to utilize node hardware by adding
    new map-by, rank-by, and bind-to cmd line options.
  - Added new MCA parameter to provide protection against IO forwarding
  - Dropped support for native Windows due to loss of maintainers. :-(
  - Added a new parallel I/O component and multiple new frameworks to
    support parallel I/O operations.
  - Fix typo in orte_setup_hadoop.m4. Thanks to Aleksej Saushev for
    reporting it
  - Fix a very old error in opal_path_access(). Thanks to Marco Atzeri
    for chasing it down. 
- Remove openmpi-sync-builtins.patch as it's not needed anymore.

Mon Apr  8 11:46:59 CEST 2013 -

- Define sysconfdir to be below the version specific openmpi
  directory (bnc#805244).

Tue Apr  2 14:13:19 CEST 2013 -

- Unset the system default in addition to unregistering the
  openmpi version (bnc#810598).
- Unconditionally set the just installed openmpi version
  to be the default.

Sun Mar  3 12:40:46 UTC 2013 -

- Add patch for support of sync builtins

Thu Feb 21 20:27:23 UTC 2013 -

- Update to version 1.6.4
  - Fix to obtaining the correct available nodes when a rankfile is
    providing the allocation.  Thanks to Siegmar Gross for
    reporting the problem.
  - Updates for MXM 2.0.
  - Major VT update to
  - Automatically provide compiler flags that compile properly on
    some types of ARM systems.
  - Fix slot_list behavior when multiple sockets are specified.
    Thanks to Siegmar Gross for reporting the problem.
  - Fixed memory leak in one-sided operations.  Thanks to Victor
    Vysotskiy for letting us know about this one.
  - Added performance improvements to the OpenIB (OpenFabrics) BTL.
  - Improved error message when process affinity fails.
  - Fixed MPI_MINLOC on man pages for MPI_REDUCE(_LOCAL).  Thanks
    to Jed Brown for noticing the problem and supplying a fix.
  - Made malloc hooks more friendly to IO interprosers.  Thanks to
    the bug report and suggested fix from Darshan maintainer Phil
  - Restored ability to direct launch under SLURM without PMI
  - Major VT update to
  - Support FCA v3.0+.
  - Fixed issue with MPI_TYPE_CREATE_F90_REAL.
  - Fix an issue with using external libltdl installations. 
    Thanks to opolawski for identifying the problem.
  - Fixed MPI_IN_PLACE case for MPI_ALLGATHER for FCA.
  - Allow SLURM PMI support to look in lib64 directories.  Thanks
    to Guillaume Papaure for the patch.
  - Restore "use mpi" ABI compatibility with the rest of the
    1.5/1.6 series (except for v1.6.3, where it was accidentally
  - Fix a very old error in opal_path_access(). Thanks to Marco
    Atzeri for chasing it down.
- Update devel package files list

Tue Jan 22 08:32:17 UTC 2013 -

- fixed Source0 URL

Sun Jan 20 12:20:07 UTC 2013 -

- Update to version 1.6.3
  - Fix mpirun --launch-agent behavior when a prefix is specified.
    Thanks to Reuti for identifying the issue.
  - Fixed memchecker configury.
  - Brought over some compiler warning squashes from the development trunk.
  - Fix spawning from a singleton to multiple hosts when the "add-host"
    MPI_Info key is used.  Thanks to Brian Budge for pointing out the
  - Add Mellanox ConnextIB IDs and max inline value.
  - Fix rankfile when no -np is given.
  - FreeBSD detection improvement.  Thanks to Brooks Davis for the
  - Removed TCP warnings on Windows.
  - Improved collective algorithm selection for very large messages.
  - Fix PSM MTL affinity settings.
  - Fix issue with MPI_OP_COMMUTATIVE in the mpif.h bindings.  Thanks to
    Ake Sandgren for providing a patch to fix the issue.
  - Fix issue with MPI_SIZEOF when using CHARACTER and LOGICAL types in
    the mpi module.  Thanks to Ake Sandgren for providing a patch to fix
    the issue.
- Update to version 1.6.2
  - Fix issue with MX MTL.  Thanks to Doug Eadline for raising the issue.
  - Fix singleton MPI_COMM_SPAWN when the result job spans multiple nodes.
  - Fix MXM hang, and update for latest version of MXM.
  - Update to support Mellanox FCA 2.5.
  - Fix startup hang for large jobs.
  - Ensure MPI_TESTANY / MPI_WAITANY properly set the empty status when
  - Fix MPI_CART_SUB behavior of not copying periods to the new
    communicator properly.  Thanks to John Craske for the bug report.
  - Add btl_openib_abort_not_enough_reg_mem MCA parameter to cause Open
    MPI to abort MPI jobs if there is not enough registered memory
    available on the system (vs. just printing a warning).  Thanks to
    Brock Palen for raising the issue.
  - Minor fix to Fortran MPI_INFO_GET: only copy a value back to the
    user's buffer if the flag is .TRUE.
  - Fix VampirTrace compilation issue with the PGI compiler suite.
- Update to version 1.6.1
  - A bunch of changes to eliminate hangs on OpenFabrics-based networks.
    Users with Mellanox hardware are ***STRONGLY ENCOURAGED*** to check
    their registered memory kernel module settings to ensure that the OS
    will allow registering more than 8GB of memory.  See this FAQ item
    for details:
  - Fall back to send/receive semantics if registered memory is
    unavilable for RDMA.
  - Fix two fragment leaks when registered memory is exhausted.
  - Hueristically determine how much registered memory is available
    and warn if it's significantly less than all of RAM.
  - Artifically limit the amount of registered memory each MPI process
    can use to about 1/Nth to total registered memory available.
  - Improve error messages when events occur that are likely due to
    unexpected registered memory exhaustion.
  - Fix double semicolon error in the C++ in <mpi.h>.  Thanks to John
    Foster for pointing out the issue.
  - Allow -Xclang to be specified multiple times in CFLAGS.  Thanks to
    P. Martin for raising the issue.
  - Break up a giant "print *" statement in the ABI-preserving incorrect
    MPI_SCATTER interface in the "large" Fortran "mpi" module.  Thanks
    to Juan Escobar for the initial patch.
  - Switch the MPI_ALLTOALLV default algorithm to a pairwise exchange.
  - Increase the openib BTL default CQ length to handle more types of
    OpenFabrics devices.
  - Lots of VampirTrace fixes; upgrade to v5.13.0.4.
  - Map MPI_2INTEGER to underlying MPI_INTEGERs, not MPI_INTs.
  - Ensure that the OMPI version number is toleant of handling spaces.
    Thanks to dragonboy for identifying the issue.
  - Fixed IN parameter marking on Fortran "mpi" module
    MPI_COMM_TEST_INTER interface.
  - Various MXM improvements.
  - Make the output of "mpirun --report-bindings" much more friendly /
  - Properly handle MPI_COMPLEX8|16|32.
  - More fixes for mpirun's processor affinity options (--bind-to-core
    and friends).
  - Use aligned memory for OpenFabrics registered memory.
  - Multiple fixes for parameter checking in MPI_ALLGATHERV,
    mpi4py community (Bennet Fauber, Lisandro Dalcin, Jonathan Dursi).
  - Fixed file positioning overflows in MPI_FILE_GET_POSITION,
  - Removed the broken --cpu-set mpirun option.
  - Fix cleanup of MPI errorcodes.  Thanks to Alexey Bayduraev for the
  - Fix default hostfile location.  Thanks to G�tz Waschk for noticing
    the issue.
  - Improve several error messages.
- Use explicit macros so it is clear what files go where

Fri Jan 18 19:58:06 CET 2013 -

- Fix typo in the description of devel-static and clean up the
  spec file a bit.
- Don't remove the libtool archive but rather include it in the
  devel-static package.

Thu Jan 17 16:41:49 UTC 2013 -

- Removed circular dependency on -devel package. 
- Moved static library files to separate -devel-static package per openSUSE 
  packaging guidelines.

Mon Aug  6 15:55:24 UTC 2012 -

- fix build on ARMv5/v6 platforms as well

Thu Jul 26 23:44:59 UTC 2012 -
- updated to version 1.6 changelog:
- Fix some process affinity issues.  When binding a process, Open MPI
  will now bind to all available hyperthreads in a core (or socket,
  depending on the binding options specified).
  --> Note that "mpirun --bind-to-socket ..." does not work on POWER6-
      and POWER7-based systems with some Linux kernel versions.  See
      the FAQ on the Open MPI web site for more information.
- Add support for ARM5 and ARM6 (in addition to the existing ARM7
  support).  Thanks to Evan Clinton for the patch.
- Minor Mellanox MXM fixes.
- Properly detect FDR10, FDR, and EDR OpenFabrics devices.
- Minor fixes to the mpirun(1) and MPI_Comm_create(3) man pages.
- Prevent segv if COMM_SPAWN_MULTIPLE fails.  Thanks to Fujitsu for
  the patch.
- Disable interposed memory management in fakeroot environments.  This
  fixes a problem in some build environments.
- Minor hwloc updates.
- Array versions of MPI_TEST and MPI_WAIT with a count==0 will now
  return immediately with MPI_SUCCESS.  Thanks to Jeremiah Willcock
  for the suggestion.
- Update VampirTrace to v5.12.2.
- Properly handle forwarding stdin to all processes when "mpirun
  --stdin all" is used.
- Workaround XLC assembly bug.
- OS X Tiger (10.4) has not been supported for a while, so forcibly
  abort configure if we detect it.
- Fix segv in the openib BTL when running on SPARC 64 systems.
- Fix some include file ordering issues on some BSD-based platforms.
  Thanks to Paul Hargove for this (and many, many other) fixes.
- Properly handle .FALSE. return parameter value to attribute copy
  callback functions.
- Fix a bunch of minor C++ API issues; thanks to Fujitsu for the patch.
- Fixed the default hostfile MCA parameter behavior.
- Per the MPI spec, ensure not to touch the port_name parameter to
  MPI_CLOSE_PORT (it's an IN parameter). 

Thu Jul 26 23:34:31 UTC 2012 -

- no need for this patch change-otf-soname.diff 

Thu Apr 19 17:29:51 UTC 2012 -

- Require mpi-selector in the "preun" instead of the "post" section as needed
  by the spec file.

Thu Apr 19 15:33:20 UTC 2012 -

- change the soname of the internal libotf to avoid mismatches between
  the font library and this one

Sat Apr 14 08:33:37 UTC 2012 -

- fix ARM support again

Tue Mar 20 19:05:25 CET 2012 -

- Update to 1.5.4:
  * Add support for the (as yet unreleased) Mellanox MXM transport.
  * Add support for dynamic service levels (SLs) in the openib BTL.
  * Fixed C++ bindings cosmetic/warnings issue with
  * Also allow the word "slots" in rankfiles (i.e., not just "slot").
  * Add Mellanox ConnectX 3 device IDs to the openib BTL defaults.
  * Various FCA updates.
  * Add missing ARM assembly code files.
  * Update to allow more than 128 entries in an appfile.
  * Various VT updates and bug fixes.
  * Update description of btl_openib_cq_size to be more accurate.
  * Various assembly "clobber" fixes.
  * Fix a hang in carto selection in obscure situations.
  * Updated to Automake 1.11.1
  * Fix compilation of LSF support.
  * Update MPI_Comm_spawn_multiple.3 man page to reflect what it
    actually does.
  * Fix for possible corruption of the environment.
  * Enable use of PSM on direct-launch SLURM jobs.
  * Update paffinity hwloc to v1.2, and to fix minor bugs affinity
    assignment bugs on PPC64/Linux platforms.
  * Let the openib BTL auto-detect its bandwidth.
  * Support new MPI-2.2 datatypes.
  * Updates to support more datatypes in MPI one-sided communication.
  * Fix recursive locking bug when MPI-IO was used with
  * Fix mpirun handling of prefix conflicts.
  * Ensure mpirun's --xterm options leaves sessions attached.
  * Fixed type of sendcounts and displs in the "use mpi" F90 module.
    ABI is preserved, but applications may well be broken.  See the
    README for more details.
  * Fix indexed datatype leaks.
  * Fix debugger mapping when mpirun's -npernode option is used.
  * Fix segv associated with MPI_Comm_create with MPI_GROUP_EMPTY.
  * Improved LoadLeveler ORTE support.
  * Add new btl_openib_gid_index MCA parameter to allow selecting which
    GID to use on an OpenFabrics device's GID table.
  * Add support for PCI relaxed ordering in the OpenFabrics BTL (when
  * Update rsh logic to allow correct SGE operation.
  * Ensure that the mca_paffinity_alone MCA parameter only appears once
    in the ompi_info output.
  * Fixed return codes from MPI_PROBE and MPI_IPROBE.
  * If something goes wrong during MPI_INIT, fix the error
    message to say that it's illegal to invoke MPI_INIT before

  For further changes see NEWS in the package documentation.

Fri Mar 16 17:52:23 UTC 2012 -

- updated to 1.4.5
- added files to make openmpi available via Modules 
Changes since 1.4.4
- Fixed the --disable-memory-manager configure switch.
  (** also to appear in 1.5.5)
- Fix typos in code and man pages.  Thanks to Fujitsu for these fixes.
  (** also to appear in 1.5.5)
- Improve management of the registration cache; when full, try freeing
  old entries and attempt to re-register.
- Fixed a data packing pointer alignment issue.  Thanks to Fujitsu
  for the patch.
  (** also to appear in 1.5.5)
- Add ability to turn off warning about having the shared memory backing
  store over a networked filesystem.  Thanks to Chris Samuel for this
  (** also to appear in 1.5.5)
- Removed an unnecessary memmove() and plugged a couple of small memory leaks
  in the openib OOB connection setup code.
- Fixed some QLogic bugs. Thanks to Mark Debbage from QLogic for the patches. 
- Fixed problem with MPI_IN_PLACE and other sentinel Fortran constants
  on OS X.
  (** also to appear in 1.5.5)
- Fix SLURM cpus-per-task allocation.
  (** also to appear in 1.5.5)
- Fix the datatype engine for when data left over from the previous
  pack was larger than the allowed space in the pack buffer. Thanks to
  Yuki Matsumoto and Takahiro Kawashima for the bug report and the
- Fix Fortran value for MPI_MAX_PORT_NAME.  Thanks to Enzo Dari for
  raising the issue.
- Workaround an Intel compiler v12.1.0 2011.6.233 vector optimization
- Fix issues on Solaris with the openib BTL.
- Fixes for the Oracle Studio 12.2 Fortran compiler.
- Update iWARP parameters for the Intel NICs.
  (** also to appear in 1.5.5)
- Fix obscure cases where MPI_ALLGATHER could crash.  Thanks to Andrew
  Senin for reporting the problem.
  (** also to appear in 1.5.5)
Sun Jan 29 17:55:52 UTC 2012 -

- Remove redundant tags/sections per specfile guideline suggestions
- Parallel building using %_smp_mflags
- Package missing files to fix build error
- Remove unexplained s390 exclude, use ExclusiveArch instead
  based upon asm files list

Thu Jan 12 17:18:06 CET 2012 -

- Include the symlink so that mpicc can be run.

Wed Oct 26 17:29:32 CEST 2011 -

- Update to OpenMPI 1.4.4 to fix problems with f90 code (bnc#724941):
  * Modified a memcpy() call in the openib btl connection setup to use
    memmove() instead because of the possibility of an overlapping
    copy (as identified by valgrind).
  * Changed use of sys_timer_get_cycles() to the more appropriate
    wrapper: opal_timer_base_get_cycles().  Thanks to Jani Monoses
    for this fix.
  * Corrected the reported default value of btl_openib_ib_timeout
    in the "IB retries exceeded" error message.  Thanks to Kevin Buckley
    for this correction.
  * Increase rdmacm adress resolution timeout from 1s to 30s & 
    update Chelsio T4 openib BTL defaults.  Thanks to Steve Wise
    for these updates.
    (** also to appear in 1.5.5)
  * Ensure that MPI_Accumulate error return in 1.4 is consistent with
    1.5.x and trunk.
  * Allow the word "slots" in rankfiles (i.e., not just "slot").
    (** also appeared in 1.5.4)
  * Add Mellanox ConnectX 3 device IDs to the openib BTL defaults.
    (** also appeared in 1.5.4)
  * Update description of btl_openib_cq_size to be more accurate.
  * Ensure mpirun's --xterm options leaves sessions attached.
    (** also appeared in 1.5.4)
  * Update to allow more than 128 entries in an appfile.
    (** also appeared in 1.5.4)
  * Update description of btl_openib_cq_size to be more accurate.
    (** also appeared in 1.5.4)
  * Fix for deadlock when handling recursive attribute keyval deletions
    (e.g., when using ROMIO with MPI_THREAD_MULTIPLE).
  * Fix indexed datatype leaks.  Thanks to Pascal Deveze for supplying
    the initial patch.  (** also appeared in 1.5.4)
  * Fixed the F90 types of the sendcounts and displs parameters to
    MPI_SCATTERV.  Thanks to Stanislav Sazykin for identifying the issue.
    (** also appeared in 1.5.4)
  * Exclude opal/libltdl from "make distclean" when --disable-dlopen is
    used.  Thanks to David Gunter for reporting the issue.
  * Fixed a segv in MPI_Comm_create when called with GROUP_EMPTY.
    Thanks to Dominik Goeddeke for finding this.
    (** also appeared in 1.5.4)
  * Fixed return codes from MPI_PROBE and MPI_IPROBE.
    (** also appeared in 1.5.4)
  * Fixed undefined symbol error when using the vtf90 profiling tool.
  * Fix for referencing an uninitialized variable in DPM ORTE.  Thanks
    to Avinash Malik for reporting the issue.
  * Fix for correctly handling multi-token args when using debuggers.
  * Eliminated the unneeded u_int*_t datatype definitions.
  * Change in ORTE DPM to get around gcc 4.[45].x compiler wanrings
    about possibly calling free() on a non-heap variable, even though it
    will never happen because the refcount will never go to zero.

  * Fix for libtool issue when using pgcc to compile ompi in conjunction
    with the -tp option.
  * Fixed a race condition in osc_rdma_sync.  Thanks to Guillaume
    Thouvenin for finding this issue.
  * Clarification of MPI_Init_thread man page.
  * Fixed an indexing problem in precondition_transports.
  * Fixed a problem in which duplicated libs were being specified for
    linking.  Thanks to Hicham Mouline for noticing it.
  * Various fixes.
  * Fix for memchecking buffers during MPI_*INIT.
  * Man page cleanups.  Thanks to Jeremiah Willcock and Jed Brown.
  * Fix for VT rpmbuild on RHEL5.
  * Support Solaris legacy munmap prototype changes.
    (** also appeared in 1.5.4)
  * Expands app_idx to int32_t to allow more than 127 app_contexts.
  * Guard the inclusion of execinfo.h since not all platforms have it.  Thanks
    to Aleksej Saushev for identifying this issue.
    (** also appeared in 1.5.4)
  * Fix to avoid possible environment corruption.  Thanks to Peter Thompson
    for identifying the issue and supplying a patch.
    (** also appeared in 1.5.4)
  * Fixed paffinity base MCA duplicate registrations.  Thanks to Gus
    Correa for bringing this to our attention.
  * Fix recursive locking bug when MPI-IO was used with
    MPI_THREAD_MULTIPLE.  (** also appeared in 1.5.4)
  * F90 MPI API fixes.
  * Fixed a misleading MPI_Bcast error message.  Thanks to Jeremiah
    Willcock for reporting this.
  * Added <sys/stat.h> to ptmalloc's hooks.c (it's not always included
    by default on some systems).
  * Libtool patch to get around a build problem when using the IBM XL
  * Fix to detect and avoid overlapping memcpy().  Thanks to
    Francis Pellegrini for identifying the issue.
  * Fix to allow ompi to work on top of RoCE vLANs.
  * Restored a missing debugger flag to support TotalView.  Thanks to
    David Turner and the TV folks for supplying the fix.
  * Updated SLURM support to 1.5.1.
  * Removed an extraneous #include from the TCP BTL.
  * When specifying OOB ports, fix to convert the ports into network
    byte order before binding.
  * Fixed use of memory barriers in the SM BTL.  This fixed segv's when
    compiling with Intel 10.0.025 or PGI 9.0-3.
  * Fix to prevent the SM BTL from creating its mmap'd file in
    directories that are remotely mounted.

Thu Oct  6 15:00:51 UTC 2011 -

- Run ./ during ARM build to ensure relevant changes are
  applied prior to the ./configure step

Mon Oct  3 12:53:19 CEST 2011 -

- update to 1.4.3:
  * latest upstream release with many fixes and features
  * changelog is too long to list here (> 200kb, read included NEWS
   for details)
- fix support for ARM platforms

Fri Sep 30 13:33:07 UTC 2011 -

- exclude arm from built, assembler routines are missing and
  current upstream patch is broken

Mon Aug 30 22:33:21 UTC 2010 -

- Do not include build date and build host info on binaries
  this should save a significant number of rebuilds/publish
  all over aprox 120 packages and its dependencies. 

Thu Jun 17 09:08:24 UTC 2010 -

- add baselibs.conf to make boost-32bit installable

Tue Feb 23 10:31:16 CET 2010 -

- chain rest of ofed build-/requires to master dapl define

Mon Feb 15 12:19:36 CET 2010 -

- set a master define for dapl and disable it 

Mon Jun 15 00:21:01 CEST 2009 -

- drop unused buildrequires for sysfsutils 

Wed Mar  4 16:47:46 CET 2009 -

- gzip man pages 

Fri Dec 26 11:13:46 CET 2008 -

- mpi-selector is used in %post - prereq it

Fri Dec 19 15:22:31 CET 2008 -

- giving up: exclude s390 and s390x

Fri Nov 14 13:37:28 CET 2008 -

- Fix setting of LD_LIBRARY_PATH.

Tue Oct 28 06:57:11 CET 2008 -

- Updated to version 1.2.8 from OFED 1.4 rc3
  - Tweaked one memory barrier in the openib component to be more conservative.
    May fix a problem observed on PPC machines.  See ticket #1532.
  - Fix OpenFabrics IB partition support. See ticket #1557.
  - Restore v1.1 feature that sourced .profile on remote nodes if the default
    shell will not do so (e.g. /bin/sh and /bin/ksh).  See ticket #1560.
  - Fix segfault in MPI_Init_thread() if ompi_mpi_init() fails. See ticket #1562.
  - Adjust SLURM support to first look for $SLURM_JOB_CPUS_PER_NODE instead of
    the deprecated $SLURM_TASKS_PER_NODE environment variable.  This change
    may be *required* when using SLURM v1.2 and above.  See ticket #1536.
  - Fix the MPIR_Proctable to be in process rank order. See ticket #1529.
  - Fix a regession introduced in 1.2.6 for the IBM eHCA. See ticket #1526.

Mon Oct 27 21:51:35 CET 2008 -

- buildreq compat-dapl-devel only on non-s390* 

Thu Sep 18 10:03:17 CEST 2008 -

- Moved mpi-select unregister to %preun in spec (bnc#344966)

Thu Sep 11 19:29:49 CEST 2008 -

- Updated to version 1.2.7 from OFED 1.4 RC1

Mon May 12 17:12:41 CEST 2008 -

- Updated to v1.2.5 from OFED 1.3 GA
- Set %config to files within the /etc directory (bnc#376940)
- Set libpath in mpivars scripts (bnc#376940)
- Set openMPI as the system MPI if none other is set (bnc#376940)
- Removed fix for open call (now found in base package)

Tue Jan 29 21:05:13 CET 2008 -

- Fix open call.

Tue Oct  2 22:20:45 CEST 2007 -

- openmpi components should be in the openmpi package 

Sat Jul 14 16:49:57 CEST 2007 -

- libnuma is x86-64 only.

Fri Jul 13 10:56:59 CEST 2007 -

- build also against libnuma

Thu Jul 12 15:55:03 CEST 2007 -

- build against dapl

Wed Jul 11 14:59:10 CEST 2007 -

- Initial Package, Version 1.2.2