This looks fine. There is one point we may want to clarify/highlight in some manner. When we reach 1.3.4, the SLURM support changes so things won't work for LANL's interactive login method (a rather unique one) unless they upgrade to slurm xxx (forget number - can look it up when I get back).

I'm not sure how to handle that point, though - outside of the tri-labs, who already know about the restriction, nobody else is likely to notice or care. So perhaps nothing need be said? I fear the language may get complex, and wind up confusing people who shouldn't care.

Thoughts?
Ralph


On Wed, Oct 28, 2009 at 2:02 PM, Jeff Squyres <jsquyres@cisco.com> wrote:
Folks -- here's what I put in the NEWS section for 1.3.4rc2.  Please send me any corrections / additions / etc. for 1.3.4 final ASAP:

- Fix some issues in OMPI's SRPM with regard to shell_scripts_basename
 and its use with mpi-selector.  Thanks to Bill Johnstone for
 pointing out the problem.
- Added many new MPI job process affinity options to mpirun.  See the
 newly-updated mpirun(1) man page for details.
- Several updates to mpirun's XML output.
- Update to fix a few Valgrind warnings with regards to the ptmalloc2
 allocator and Open MPI's use of PLPA.
- Many updates and fixes to the (non-default) "sm" collective
 component (i.e., native shared memory MPI collective operations).
- Updates and fixes to some MPI_COMM_SPAWN_MULTIPLE corner cases.
- Fix some internal copying functions in Open MPI's use of PLPA.
- Correct some SLURM nodelist parsing logic that may have interfered
 with large jobs.  Additionally, per advice from the SLURM team,
 change the environment variable that we use for obtaining the job's
 allocation.
- Revert to an older, safer (but slower) communicator ID allocation
 algorithm.
- Fixed minimum distance finding for OpenFabrics devices in the openib
 BTL.
- Relax the parameter checking MPI_CART_CREATE a bit.
- Fix MPI_COMM_SPAWN[_MULTIPLE] to only error-check the info arguments
 on the root process.  Thanks to Federico Golfre Andreasi for
 reporting the problem.
- Fixed some BLCR configure issues.
- Fixed a potential deadlock when the openib BTL was used with
 MPI_THREAD_MULTIPLE.
- Fixed dynamic rules selection for the "tuned" coll component.
- Added a launch progress meter to mpirun (useful for large jobs; set
 the orte_report_launch_progress MCA parameter to 1 to see it).
- Reduced the number of file descriptors consumed by each MPI process.
- Add new device IDs for Chelsio T3 RNICs to the openib BTL config file.
- Fix some CRS self component issues.
- Added some MCA parameters to the PSM MTL to tune its run-time
 behavior.
- Fix some VT issues with MPI_BOTTOM/MPI_IN_PLACE.
- Man page updates from the Debain Open MPI package maintainers.
- Add cycle counter support for the Alpha and Sparc platforms.
- Pass visibility flags to libltdl's configure script, resulting in
 those symbols being hidden.  This appears to mainly solve the
 problem of applications attempting to use different versions of
 libltdl from that used to build Open MPI.


--
Jeff Squyres
jsquyres@cisco.com

_______________________________________________
devel mailing list
devel@open-mpi.org
http://www.open-mpi.org/mailman/listinfo.cgi/devel