See the wiki here<http://wiki.mpich.org/mpich/index.php/Tool_interfaces_(MPI-T),_MPICH_parameters_and_instrumentation#New_MPI_T_Control_Variable_Design> for how to add a cvar in MPICH. See MPI 3.0 Chapter 14.3.6 for how to access cvars in MPI. --Junchao Zhang On Sat, Feb 22, 2014 at 7:56 AM, William Gropp <[email protected]> wrote:

Yes, CVAR already reads env variables, and PETSc could use the MPI_T interface to set the value at any time.

Bill

William Gropp Director, Parallel Computing Institute Deputy Director for Research Institute for Advanced Computing Applications and Technologies Thomas M. Siebel Chair in Computer Science University of Illinois Urbana-Champaign

On Feb 22, 2014, at 5:41 AM, Jeff Hammond wrote:

I'll look into a general setenv-like mechanism for CVARs rather than an abort-specific one-off. Might be there already through MPI_T interface (which is standard, unlike MPIX).

Jeff

_______________________________________________ discuss mailing list [email protected] To manage subscription options or unsubscribe: https://lists.mpich.org/mailman/listinfo/discuss