jeff.science at gmail.com
Fri Mar 14 12:04:24 CDT 2014
M**LOC doesn't act on standard types so it's really not relevant. I said this already.
Sent from my iPhone
> On Mar 14, 2014, at 11:49 AM, Jed Brown <jedbrown at mcs.anl.gov> wrote:
> Jeff Hammond <jeff.science at gmail.com> writes:
>> There's an easy solution for "bad" ops: fall back to C&S
>> implementation the way one does on any shared memory arch when a
>> particular atomic isn't supported but C&S is.
>> This means that no mutexes are required, NICs that have HW support for
>> NO_OP, REPLACE, SUM, XOR, etc. can use them, and when a user asks for
>> something silly like PROD, the performance is degraded for those ops
>> alone and not for anything else.
> 1. How does the above not justify the same for M**LOC?
> 2. Now we expect every system to have the equivalent of CMPXCHG16B and
> that basic types will never be longer than 16B (unless a longer atomic
> is made available)?
More information about the mpiwg-rma