Anyone interested, please add your name to doodle, and we'll find a time that everyone can meet.
--- Agenda ---
In the Open MPI call today we discussed a few aspects of
https://github.com/open-mpi/ompi/issues/398
1) Moving ompi_info_t down to opal_info_t to allow lower level components access to this functionality
2) Implementing OMPI_Comm_Info get/set in a way that can be reused for Windows and Files also. There are a number of issues around how the standard words the return values from get() that are left up to the implementation, for example:
- values for non-explicitly set keys that the MPI layer is using.
- values for non-explicitly set keys that the MPI layer is not using.
- values for explicitly overwritten values.
- Communication (to user via docs??) of what hints Open MPI recognizes.
- Communication (to user via docs??) of what values are required to be the same/different on all ranks of Comm.
- additional consistancy checking of values in debugging mode?
- ability to print/log unrecognized hints.
https://github.com/open-mpi/ompi/issues/398
1) Moving ompi_info_t down to opal_info_t to allow lower level components access to this functionality
2) Implementing OMPI_Comm_Info get/set in a way that can be reused for Windows and Files also. There are a number of issues around how the standard words the return values from get() that are left up to the implementation, for example:
- values for non-explicitly set keys that the MPI layer is using.
- values for non-explicitly set keys that the MPI layer is not using.
- values for explicitly overwritten values.
- Communication (to user via docs??) of what hints Open MPI recognizes.
- Communication (to user via docs??) of what values are required to be the same/different on all ranks of Comm.
- additional consistancy checking of values in debugging mode?
- ability to print/log unrecognized hints.
---
Geoffrey Paulsen
Software Engineer, IBM Platform MPI
IBM Platform-MPI
Phone: 720-349-2832
Email: gpaul...@us.ibm.com
www.ibm.com
