It would have to be for 1.8.5 as there is no way to change that configure.m4 without re-releasing.
We still apparently have a thread-related performance issue as reported by Intel. It appears that we didn’t completely manage to fix the blasted thread locks, leaving some still on by default, thereby causing a roughly 10-15% loss of performance relative to earlier in the 1.8 series. So a 1.8.5 is going to be required fairly soon anyway. Sigh > On Dec 22, 2014, at 9:46 AM, Howard Pritchard <hpprit...@gmail.com> wrote: > > I opened an issue 322 <https://github.com/open-mpi/ompi/issues/322> about > this and gave put it on 1.8.5 milestone. > I'll submit a PR to remove the sn/xpmem.h usage in the vader > config file. > > I think to do justice to SGI UV, someone would have to put in time > to figure out how to use their GRU api. I'm pretty sure that's the way the > sgi mpi delivers small messages efficiently. > > Howard > > > > 2014-12-22 8:43 GMT-07:00 Nathan Hjelm <hje...@lanl.gov > <mailto:hje...@lanl.gov>>: > > Yeah, I figured out why XPMEM is failing on SGI UV but have not figured > out a fix. If possible can we remove the check for sn/xpmem.h in the > ompi/mca/btl/vader/configure.m4. I hopefully will have a better fix for > 1.8.5. > > -Nathan > > On Fri, Dec 19, 2014 at 11:59:29PM -0800, Paul Hargrove wrote: > > Sorry to rain on the parade, but SGI UV is still broken by default. > > I reported this as present in 1.8.4rc5 and Nathan had claimed to be > > working on it. > > A reminder that all it takes is a 1-line change in > > ompi/mca/btl/vader/configure.m4 to not search for sn/xpmem.h > > -Paul > > On Fri, Dec 19, 2014 at 7:26 PM, Ralph Castain <r...@open-mpi.org > > <mailto:r...@open-mpi.org>> wrote: > > > > The Open MPI Team, representing a consortium of research, academic, and > > industry partners, is pleased to announce the release of Open MPI > > version 1.8.4. > > > > v1.8.4 is a bug fix release. All users are encouraged to upgrade to > > v1.8.4 when possible. > > > > Version 1.8.4 can be downloaded from the main Open MPI web site or any > > of its mirrors (mirrors will be updating shortly). > > > > Here is a list of changes in v1.8.4 as compared to v1.8.3: > > > > - Fix MPI_SIZEOF; now available in mpif.h for modern Fortran compilers > > (see README for more details). Also fixed various compiler/linker > > errors. > > - Fixed inadvertant Fortran ABI break between v1.8.1 and v1.8.2 in the > > mpi interface module when compiled with gfortran >= v4.9. > > - Fix various MPI_THREAD_MULTIPLE issues in the TCP BTL. > > - mpirun no longer requires the --hetero-nodes switch; it will > > automatically detect when running in heterogeneous scenarios. > > - Update LSF support, to include revamped affinity functionality. > > - Update embedded hwloc to v1.9.1. > > - Fixed max registerable memory computation in the openib BTL. > > - Updated error message when debuggers are unable to find various > > symbols/types to be more clear. Thanks to Dave Love for raising the > > issue. > > - Added proper support for LSF and PBS/Torque libraries in static > > builds. > > - Rankfiles now support physical processor IDs. > > - Fixed potential hang in MPI_ABORT. > > - Fixed problems with the PSM MTL and "re-connect" scenarios, such as > > MPI_INTERCOMM_CREATE. > > - Fix MPI_IREDUCE_SCATTER with a single process. > > - Fix (rare) race condition in stdout/stderr funneling to mpirun where > > some trailing output could get lost when a process terminated. > > - Removed inadvertent change that set --enable-mpi-thread-multiple "on" > > by default, thus impacting performance for non-threaded apps. > > - Significantly reduced startup time by optimizing internal hash table > > implementation. > > - Fixed OS X linking with the Fortran mpi module when used with > > gfortran >= 4.9. Thanks to Github user yafshar for raising the > > issue. > > - Fixed memory leak on Cygwin platforms. Thanks for Marco Atzeri for > > reporting the issue. > > - Fixed seg fault in neighborhood collectives when the degree of the > > topology is higher than the communicator size. Thanks to Lisandro > > Dalcin for reporting the issue. > > - Fixed segfault in neighborhood collectives under certain use-cases. > > - Fixed various issues regarding Solaris support. Thanks to Siegmar > > Gross for patiently identifying all the issues. > > - Fixed PMI configure tests for certain Slurm installation patterns. > > - Fixed param registration issue in Java bindings. Thanks to Takahiro > > Kawashima and Siegmar Gross for identifying the issue. > > - Several man page fixes. > > - Silence several warnings and close some memory leaks (more remain, > > but it's better than it was). > > - Re-enabled the use of CMA and knem in the shared memory BTL. > > - Updated mpirun manpage to correctly explain new map/rank/binding > > options. > > - Fixed MPI_IALLGATHER problem with intercommunicators. Thanks for > > Takahiro Kawashima for the patch. > > - Numerous updates and performance improvements to OpenSHMEM. > > - Turned off message coalescing in the openib BTL until a proper fix > > for that capability can be provided (tentatively expected for 1.8.5) > > - Fix a bug in iof output that dates back to the dinosaurs which would > > output extra bytes if the system was very heavily loaded > > - Fix a bug where specifying mca_component_show_load_errors=0 could > > cause ompi_info to segfault > > - Updated valgrind suppression file > > > > _______________________________________________ > > announce mailing list > > annou...@open-mpi.org <mailto:annou...@open-mpi.org> > > Subscription: http://www.open-mpi.org/mailman/listinfo.cgi/announce > > <http://www.open-mpi.org/mailman/listinfo.cgi/announce> > > Searchable archives: > > http://www.open-mpi.org/community/lists/announce/2014/12/index.php > > <http://www.open-mpi.org/community/lists/announce/2014/12/index.php> > > > > -- > > Paul H. Hargrove phhargr...@lbl.gov > > <mailto:phhargr...@lbl.gov> > > Computer Languages & Systems Software (CLaSS) Group > > Computer Science Department Tel: +1-510-495-2352 > > <tel:%2B1-510-495-2352> > > Lawrence Berkeley National Laboratory Fax: +1-510-486-6900 > > <tel:%2B1-510-486-6900> > > > _______________________________________________ > > devel mailing list > > de...@open-mpi.org <mailto:de...@open-mpi.org> > > Subscription: http://www.open-mpi.org/mailman/listinfo.cgi/devel > > <http://www.open-mpi.org/mailman/listinfo.cgi/devel> > > Link to this post: > > http://www.open-mpi.org/community/lists/devel/2014/12/16704.php > > <http://www.open-mpi.org/community/lists/devel/2014/12/16704.php> > > > _______________________________________________ > devel mailing list > de...@open-mpi.org <mailto:de...@open-mpi.org> > Subscription: http://www.open-mpi.org/mailman/listinfo.cgi/devel > <http://www.open-mpi.org/mailman/listinfo.cgi/devel> > Link to this post: > http://www.open-mpi.org/community/lists/devel/2014/12/16710.php > <http://www.open-mpi.org/community/lists/devel/2014/12/16710.php> > > _______________________________________________ > devel mailing list > de...@open-mpi.org > Subscription: http://www.open-mpi.org/mailman/listinfo.cgi/devel > Link to this post: > http://www.open-mpi.org/community/lists/devel/2014/12/16715.php
