[mpiwg-tools] PMPI and sessions init

Wesley Bland work at wesbland.com
Tue Nov 10 09:02:13 CST 2020


> @Wesley: Why exactly did we come up with the restriction of "no
> registration after init"? And did we push off the callback solution to
> a later time?

There’s a number of reasons, but the main one was that it makes the tool stack unpredictable and prevents the tool from caching the next functions in the call stack. It also gives us a point at which we can "turn on” interception of the QMPI_Register_function call so that meta-tools can intercept registration and start creating their own call stacks.

Thanks,
Wes
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mpi-forum.org/pipermail/mpiwg-tools/attachments/20201110/854718bf/attachment.html>


More information about the mpiwg-tools mailing list