openmpi

dfs

Per the PR and discussion on today's telecon, extend the process name definition as a two-field struct of uint32_t's down to the OPAL layer. This resolves issues created by prior commits that impacted both heterogeneous and SPARC support. This also simplifies the OMPI code base by removing the need for frequent memcpy's when transitioning between the OMPI/ORTE layers and OPAL.

2014-11-11 17:00:42 -08:00

errmgr

Per the PR and discussion on today's telecon, extend the process name definition as a two-field struct of uint32_t's down to the OPAL layer. This resolves issues created by prior commits that impacted both heterogeneous and SPARC support. This also simplifies the OMPI code base by removing the need for frequent memcpy's when transitioning between the OMPI/ORTE layers and OPAL.

2014-11-11 17:00:42 -08:00

ess

Per the PR and discussion on today's telecon, extend the process name definition as a two-field struct of uint32_t's down to the OPAL layer. This resolves issues created by prior commits that impacted both heterogeneous and SPARC support. This also simplifies the OMPI code base by removing the need for frequent memcpy's when transitioning between the OMPI/ORTE layers and OPAL.

2014-11-11 17:00:42 -08:00

filem

Per the PMIx RFC:

2014-08-21 18:56:47 +00:00

grpcomm

Update the grpcomm direct component's priority so it sits at the bottom of the list, as it should now that the other components are active. Cleanup up the signature print function a touch to make it more readable. Remove the unneeded xcast functions in brks and rcd components as we will just fall thru to using the "direct" one

2014-11-03 14:43:17 -08:00

iof

Per the PR and discussion on today's telecon, extend the process name definition as a two-field struct of uint32_t's down to the OPAL layer. This resolves issues created by prior commits that impacted both heterogeneous and SPARC support. This also simplifies the OMPI code base by removing the need for frequent memcpy's when transitioning between the OMPI/ORTE layers and OPAL.

2014-11-11 17:00:42 -08:00

odls

Be more selective about closing fd's for alps/odls

2014-11-19 11:21:30 -07:00

oob

OOB UD: opal_ignore removed from oob ud component: component is compilable. Added support of new RML API, support of opal_buffer as input data. Added usage of routed component.

2014-11-20 10:20:35 +02:00

plm

Ensure that mpirun correctly selects the HNP ess component without attempting to init the PMI subsystem as mpirun won't be supported anyway, so let's avoid the error message. Also, daemons launched by the plm/slurm component must use the ess/slurm module as we cannot trust the Slurm PMI_Init functions to correctly tell us when PMI support is available.

2014-11-03 21:35:42 -08:00

ras

fix some compiler warnings in ras/alps

2014-11-18 11:32:37 -07:00

rmaps

Support physical processor ids in rankfile

2014-11-10 14:00:40 -08:00

rml

Per the PR and discussion on today's telecon, extend the process name definition as a two-field struct of uint32_t's down to the OPAL layer. This resolves issues created by prior commits that impacted both heterogeneous and SPARC support. This also simplifies the OMPI code base by removing the need for frequent memcpy's when transitioning between the OMPI/ORTE layers and OPAL.

2014-11-11 17:00:42 -08:00

routed

Fix C++ builds by removing no-longer-needed type declaration

2014-11-14 11:44:24 -08:00

rtc

check-help-strings cleanup

2014-08-11 03:26:21 +00:00

snapc

Per the PMIx RFC:

2014-08-21 18:56:47 +00:00

sstore

use the orte attributes in the FT code to fix compile errors

2014-06-26 03:19:17 +00:00

state

Fix a race condition caused by a bad attribute flag that created an OR instead of an AND condition check

2014-08-22 22:48:16 +00:00