- Which topics is advanced enough to benefit from plenary discussion?
MPI_T semantics need, IIRC, some concrete answers from implementors (and other interested users):
* Can peruse-like events be collected in MPICH without terrible pain?
* If so, can MPICH and OMPI agree on common semantics that are implementable in both?
* Will reserving, e.g., the “MPI_T_*” namespace for events/pvars/cvars with common semantics specified in a side doc be unreasonably limiting?
* Are there things besides peruse that count as low-hanging fruit here?
QMPI needs, ultimately, the static/dynamic tool list decision (from MPI’s perspective). Everything else is IMO _mostly_ settled, but we really can’t write a reasonable chapter without knowing how much flexibility we’re requiring implementations to support.
- Which timeslots are best suited?
Earlier in the day is better for me, but my afternoons/evenings (CEST) are pretty clear that week.
As I basically just returned from sick leave I did not get any work done, so would not have anything further to discuss today.
Does anyone of you have a topic to discuss?
Nothing beyond the above.
Cheers,
Marc-Andre
--
Dr. rer. nat. Marc-André Hermanns
[er/sein, he/his]
IT Center
Group: High Performance Computing
Division: Computational Science and Engineering
RWTH Aachen University
Seffenter Weg 23
52074 Aachen
Phone: +49 241 80-24381
hermanns@itc.rwth-aachen.de
www.itc.rwth-aachen.de
Social Media Kanäle des IT Centers:
https://blog.rwth-aachen.de/itc/
https://www.facebook.com/itcenterrwth
https://www.linkedin.com/company/itcenterrwth
https://twitter.com/ITCenterRWTH
https://www.youtube.com/@ITCenterRWTHAachen
_______________________________________________
mpiwg-tools mailing list
mpiwg-tools@lists.mpi-forum.org
https://lists.mpi-forum.org/mailman/listinfo/mpiwg-tools