1
1
openmpi/ompi/mca/coll/tuned
George Bosilca c5b1bdd57c Correctly deal with the error case. The problem is tricky: the MPI standard doesn't allow
MPI_ERR_IN_STATUS to be returned from any functions that return only one completed request
(few exception here: wait_some and wait_all and the test versions). As we use an wait_all
in these send_receive functions we should convert the MPI_ERR_IN_STATUS to the real
error, i.e. the one comming from the MPI_ERROR field in the status corresponding to the
failed request.

This commit was SVN r20907.
2009-03-31 23:44:59 +00:00
..
coll_tuned_allgather.c - Last of intrusive commits (promised)... err for now. 2009-03-04 17:06:51 +00:00
coll_tuned_allgatherv.c - Last of intrusive commits (promised)... err for now. 2009-03-04 17:06:51 +00:00
coll_tuned_allreduce.c Correctly forward the right module if we call another collective function. Kudos to 2009-01-14 03:22:54 +00:00
coll_tuned_alltoall.c CIDs 1080-1090: assert() checks were not sufficient to check for 2009-01-03 15:56:25 +00:00
coll_tuned_alltoallv.c CIDs 1080-1090: assert() checks were not sufficient to check for 2009-01-03 15:56:25 +00:00
coll_tuned_barrier.c CIDs 1080-1090: assert() checks were not sufficient to check for 2009-01-03 15:56:25 +00:00
coll_tuned_bcast.c CIDs 1080-1090: assert() checks were not sufficient to check for 2009-01-03 15:56:25 +00:00
coll_tuned_component.c - On the way to get the BTLs split out and lessen dependency on orte: 2009-02-14 02:26:12 +00:00
coll_tuned_decision_dynamic.c - Last of intrusive commits (promised)... err for now. 2009-03-04 17:06:51 +00:00
coll_tuned_decision_fixed.c - Last of intrusive commits (promised)... err for now. 2009-03-04 17:06:51 +00:00
coll_tuned_dynamic_file.c - Last of intrusive commits (promised)... err for now. 2009-03-04 17:06:51 +00:00
coll_tuned_dynamic_file.h Update to have public symbols be consistent; use new prefix rule 2005-12-22 13:49:33 +00:00
coll_tuned_dynamic_rules.c Effectively revert the orte_output system and return to direct use of opal_output at all levels. Retain the orte_show_help subsystem to allow aggregation of show_help messages at the HNP. 2008-06-09 14:53:58 +00:00
coll_tuned_dynamic_rules.h Adding variant of linear alltoall algorithm where the number of 2007-02-20 04:25:00 +00:00
coll_tuned_forced.c Adding variant of linear alltoall algorithm where the number of 2007-02-20 04:25:00 +00:00
coll_tuned_forced.h Adding variant of linear alltoall algorithm where the number of 2007-02-20 04:25:00 +00:00
coll_tuned_gather.c CIDs 1080-1090: assert() checks were not sufficient to check for 2009-01-03 15:56:25 +00:00
coll_tuned_module.c Added padding to predefined handle structures to maintain library version to 2009-02-24 17:17:33 +00:00
coll_tuned_reduce_scatter.c Correctly forward the right module if we call another collective function. Kudos to 2009-01-14 03:22:54 +00:00
coll_tuned_reduce.c CIDs 1080-1090: assert() checks were not sufficient to check for 2009-01-03 15:56:25 +00:00
coll_tuned_scatter.c CIDs 1080-1090: assert() checks were not sufficient to check for 2009-01-03 15:56:25 +00:00
coll_tuned_topo.c - Last of intrusive commits (promised)... err for now. 2009-03-04 17:06:51 +00:00
coll_tuned_topo.h added binomial tree based Gather alogrithm, passed IBM and Intel tests 2007-02-28 01:11:01 +00:00
coll_tuned_util.c Correctly deal with the error case. The problem is tricky: the MPI standard doesn't allow 2009-03-31 23:44:59 +00:00
coll_tuned_util.h Effectively revert the orte_output system and return to direct use of opal_output at all levels. Retain the orte_show_help subsystem to allow aggregation of show_help messages at the HNP. 2008-06-09 14:53:58 +00:00
coll_tuned.h - Last of intrusive commits (promised)... err for now. 2009-03-04 17:06:51 +00:00
configure.params Remove unneeded PARAM_INIT_FILE variable in configure.params files used by 2007-01-08 03:44:22 +00:00
Makefile.am Add MPI_Alltoallv to tuned collectives and add a pairwise implementation of MPI_Alltoallv. However, do not change the default behavior for now. The only way to use new pairwise implementation is via mca parameters. 2008-05-07 02:31:24 +00:00