The Open MPI Team, representing a consortium of research, academic, and 
industry partners, is pleased to announce the release of Open MPI version 2.1.0.

v2.1.0 is a new stable series that builds on the v2.0.x series.  It includes 
some new features and the usual assortment of bug fixes (many of the same bug 
fixes will be included in v2.0.3 when it is released).  v2.1.0 should be 
backwards compatible with v2.0.x releases per Open MPI's definition of 
"backwards compatible" (which includes ABI:

Open MPI v2.1.0 can be downloaded from the main web site (mirrors will be 
updating shortly):

Major new features:

- The main focus of the Open MPI v2.1.0 release was to update to PMIx
  v1.2.1.  When using PMIx (e.g., via mpirun-based launches, or via
  direct launches with recent versions of popular resource managers),
  launch time scalability is improved, and the run time memory
  footprint is greatly decreased when launching large numbers of MPI /
  OpenSHMEM processes.
- Update OpenSHMEM API conformance to v1.3.
- The usnic BTL now supports MPI_THREAD_MULTIPLE.
- General/overall performance improvements to MPI_THREAD_MULTIPLE.
- Add a summary message at the bottom of configure that tells you many
  of the configuration options specified and/or discovered by Open

Changes in behavior compared to prior versions:

- None.

Removed legacy support:

- The ptmalloc2 hooks have been removed from the Open MPI code base.
  This is not really a user-noticable change; it is only mentioned
  here because there was much rejoycing in the Open MPI developer

Bug fixes/minor improvements:

- New MCA parameters:
  - iof_base_redirect_app_stderr_to_stdout: as its name implies, it
    combines MPI / OpenSHMEM applications' stderr into its stdout
  - opal_event_include: allow the user to specify which FD selection
    mechanism is used by the underlying event engine.
  - opal_stacktrace_output: indicate where stacktraces should be sent
    upon MPI / OpenSHMEM process crashes ("none", "stdout", "stderr",
  - orte_timeout_for_stack_trace: number of seconds to wait for stack
    traces to be reported (or <=0 to wait forever).
  - mtl_ofi_control_prog_type/mtl_ofi_data_prog_type: specify libfabric
    progress model to be used for control and data.
- Fix MPI_WTICK regression where the time reported may be inaccurate
  on systems with processor frequency scalaing enabled.
- Fix regression that lowered the memory maximum message bandwidth for
  large messages on some BTL network transports, such as openib, sm,
  and vader.
- Fix a name collision in the shared file pointer MPI IO file locking
  scheme.  Thanks to Nicolas Joly for reporting the issue.
- Fix datatype extent/offset errors in MPI_PUT and MPI_RACCUMULATE
  when using the Portals 4 one-sided component.
- Add support for non-contiguous datatypes to the Portals 4 one-sided
- Various updates for the UCX PML.
- Updates to the following man pages:
  - mpirun(1)
  - MPI_WIN_GET_NAME(3). Thanks to Nicolas Joly for reporting the
  - MPI_INFO_GET_[NKEYS|NTHKEY](3). Thanks to Nicolas Joly for
    reporting the typo.
- Fixed a problem in the TCP BTL when using MPI_THREAD_MULTIPLE.
  Thanks to Evgueni Petrov for reporting.
- Fixed external32 representation in the romio314 module.  Note that
  for now, external32 representation is not correctly supported by the
  ompio module.  Thanks to Thomas Gastine for bringing this to our
- Add note how to disable a warning message about when a high-speed
  MPI transport is not found.  Thanks to Susan Schwarz for reporting
  the issue.
- Ensure that sending SIGINT when using the rsh/ssh launcher does not
  orphan children nodes in the launch tree.
- Fix the help message when showing deprecated MCA param names to show
  the correct (i.e., deprecated) name.
- Enable support for the openib BTL to use multiple different
  InfiniBand subnets.
- Fix a minor error in MPI_AINT_DIFF.
- Fix bugs with MPI_IN_PLACE handling in:
  - Thanks to all the users who helped diagnose these issues.
- Allow qrsh to tree spawn (if the back-end system supports it).
- Fix MPI_T_PVAR_GET_INDEX to return the correct index.
- Correctly position the shared file pointer in append mode in the
  OMPIO component.
- Add some deprecated names into shmem.h for backwards compatibility
  with legacy codes.
- Fix MPI_MODE_NOCHECK support.
- Fix a regression in PowerPC atomics support.  Thanks to Orion
  Poplawski for reporting the issue.
- Fixes for assembly code with aggressively-optimized compilers on
  x86_64/AMD64 platforms.
- Fix one more place where configure was mangling custom CFLAGS.
  Thanks to Phil Tooley (@Telemin) for reporting the issue.
- Better handle builds with external installations of hwloc.
- Fixed a hang with MPI_PUT and MPI_WIN_LOCK_ALL.
- Fixed a bug when using MPI_GET on non-contiguous datatypes and
- Fixed a bug when using POST/START/COMPLETE/WAIT after a fence.
- Fix configure portability by cleaning up a few uses of "==" with
  "test".  Thanks to Kevin Buckley for pointing out the issue.
- Fix bug when using darrays with lib and extent of darray datatypes.
- Updates to make Open MPI binary builds more bit-for-bit
  reproducable.  Thanks to Alastair McKinstry for the suggestion.
- Fix issues regarding persistent request handling.
- Ensure that shmemx.h is a standalone OpenSHMEM header file.  Thanks
  to Nick Park (@nspark) for the report.
- Ensure that we always send SIGTERM prior to SIGKILL.  Thanks to Noel
  Rycroft for the report.
- Added ConnectX-5 and Chelsio T6 device defaults for the openib BTL.
- OpenSHMEM no longer supports MXM less than v2.0.
- Plug a memory leak in ompi_osc_sm_free.  Thanks to Joseph Schuchart
  for the report.
- The "self" BTL now uses less memory.
- The vader BTL is now more efficient in terms of memory usage when
  using XPMEM.
- Removed the --enable-openib-failover configure option.  This is not
  considered backwards-incompatible because this option was stale and
  had long-since stopped working, anyway.
- Allow jobs launched under Cray aprun to use hyperthreads if
  opal_hwloc_base_hwthreads_as_cpus MCA parameter is set.
- Add support for 32-bit and floating point Cray Aries atomic
- Add support for network AMOs for MPI_ACCUMULATE, MPI_FETCH_AND_OP,
  and MPI_COMPARE_AND_SWAP if the "ompi_single_intrinsic" info key is
  set on the window or the "acc_single_intrinsic" MCA param is set.
- Automatically disqualify RDMA CM support in the openib BTL if
- Make configure smarter/better about auto-detecting Linux CMA
- Improve the scalability of MPI_COMM_SPLIT_TYPE.
- Fix the mixing of C99 and C++ header files with the MPI C++
  bindings.  Thanks to Alastair McKinstry for the bug report.
- Add support for ARM v8.
- Several MCA parameters now directly support MPI_T enumerator
  semantics (i.e., they accept a limited set of values -- e.g., MCA
  parameters that accept boolean values).
- Added --with-libmpi-name=STRING configure option for vendor releases
  of Open MPI.  See the README for more detail.
- Fix a problem with Open MPI's internal memory checker.  Thanks to Yvan
  Fournier for reporting.
- Fix a multi-threaded issue with MPI_WAIT.  Thanks to Pascal Deveze for

Known issues (to be addressed in v2.1.1):

- See the list of fixes slated for v2.1.1 here:

Jeff Squyres

announce mailing list

Reply via email to