[mpiwg-rma] Atomicity of Accumulates
maikpeterson at googlemail.com
Thu Aug 17 04:05:32 CDT 2017
simon, on most systems today there is no way to sync userlevel libs like mpi
with isa instructions in terms of atomicity. if you want perf. do it by
i) separate them
ii) merge them with hand-crafted methods within threads
several order faster than mpi sequentializing...
2017-08-16 11:24 GMT+02:00 Simon Schwitanski <
simon.schwitanski at rwth-aachen.de>:
> I have got a question about the atomicity of accumulate operations as
> described in 11.7.1. It says (page 461, lines 8 - 11):
> > The previous correctness conditions imply that a location updated by a
> > call to an accumulate operation cannot be accessed by a load or an RMA
> > call other than accumulate until the accumulate operation has completed
> > (at the target).
> I am a little bit confused about what this actually means: Does it mean
> that local load accesses on the target are (somehow) delayed if there is
> an ongoing accumulate operation to the same location? Or does it just
> mean that concurrent load accesses are forbidden and lead to undefined
> behavior? And why does this not hold for store accesses?
> mpiwg-rma mailing list
> mpiwg-rma at lists.mpi-forum.org
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the mpiwg-rma