does it exist in 1.3.4? if so, it was a trivial change - can't see why
it couldn't be added.
On Oct 20, 2009, at 3:53 AM, Terry Dontje wrote:
Not that I want to delay 1.3.4 anymore than it has been but this
seems like a possible blocker. Can this be CMR'd for 1.3.4 or is
this truly a 1.4 CMR?
--td
Ralph Castain wrote:
Fixed in r22111
Thanks
On Oct 19, 2009, at 8:13 PM, Eugene Loh wrote:
Somewhere between r22090 and r22109, ompi_info started
(erroneously) requesting CPUs.
E.g., r22090 is good:
% ompi_info | grep Open MPI:
Open MPI: 1.7a1r22090
But r22109 is bad:
% ompi_info | grep Open MPI:
--------------------------------------------------------------------------
Your job has requested more cpus per process(rank) than there
are cpus in a socket:
Cpus/rank: 1
#cpus/socket: 0
Please correct one or both of these values and try again.
--------------------------------------------------------------------------
A component framework failed to open properly.
ompi_info will likely not display all configuration information
Open MPI: 1.7a1r22109
_______________________________________________
devel mailing list
de...@open-mpi.org
http://www.open-mpi.org/mailman/listinfo.cgi/devel
_______________________________________________
devel mailing list
de...@open-mpi.org
http://www.open-mpi.org/mailman/listinfo.cgi/devel
_______________________________________________
devel mailing list
de...@open-mpi.org
http://www.open-mpi.org/mailman/listinfo.cgi/devel
