1
1
openmpi/ompi/mca/coll/ml/help-mpi-coll-ml.txt
Gilles Gouaillardet edfbeba7bf coll/ml: better error handling
when CHECK_AND_RECYCLE detects an error, a message is displayed
if the error occurs on an intrinsic communicator, then abort
the program (instead of trying to free the communicator)

cmr=v1.8.3:reviewer=hjelmn

This commit was SVN r32659.
2014-09-01 10:00:49 +00:00

65 строки
1.9 KiB
Plaintext

# -*- text -*-
#
# Copyright (c) 2009-2014 Oak Ridge National Laboratory. All rights reserved.
# Copyright (c) 2014 Research Organization for Information Science
# and Technology (RIST). All rights reserved.
# $COPYRIGHT$
#
# Additional copyrights may follow
#
# $HEADER$
#
# This is the US/English help file for Open MPI's Hierarchical Collective
# Component (coll/ml).
#
[empty-sub-group]
ML topology configuration explicitly requested for this subgroup:
%s
Such configuration results in a creation of empty groups. As a result, ML
framework cannot configure requested collective operations. ML framework will be
disabled. One configuration that might enable ML component is --mca bcol_base_string basesmuma,ptpcoll
--mca sbgp_base_subgroups_string basesmuma,p2p
[allreduce-not-supported]
This BCOL is configured in one of the hierarchy :
%s
The BCOL does not support Allreduce for all
operations and datatype combination. In addition, you did not suggest
alternate topology building configurations.
[allreduce-alt-nosupport]
The hierarchy is configured with alternate BCOL:
%s
Both the original topology and alternate topology not support Allreduce for all
operations and datatype combination. In addition, you did not suggest
alternate topology building configurations.
[fragmentation-disabled]
ML could not be used because the mca param coll_ml_enable_fragmentation
was set to zero and there is a bcol that does not support
zero copy method.
[static-bcast-disabled]
ML could not be used because the mca param coll_ml_bcast_algorithm
was not set to static and other broadcast implementation was available.
[coll-ml-check-error]
ML detected an error on communicator %s
This communicator cannot be used any more
[coll-ml-check-fatal-error]
ML detected an unrecoverable error on intrinsic communicator %s
The program will now abort