huttered40 / critter

Critical path analysis of MPI parallel programs
BSD 2-Clause "Simplified" License
2 stars 1 forks source link

Parameterize critical path propagation #23

Closed huttered40 closed 4 years ago

huttered40 commented 5 years ago

Motivation described in #20.

1) Remove any possibility for overlap with blocking communication (although technically blocking send can return before communication ever occurs, do we cover this with a barrier? Should we use a barrier? 2) Remove any possibility for idle time with nonblocking communication. Do we have any extra barriers here that shouldn't be?

huttered40 commented 4 years ago

No longer relevant.