[ofiwg] libfabric network atomic operations, processor atomic ops, and coherence

[Hefty, Sean]

> > There is no guarantee that NIC/network based atomics will be coherent with
> CPU based atomics, or that they will be coherent between NICs, or the final
> result will even be atomic.  [...]
> 
> Would I be correct in reading that last clause as “or [that] the [visibility
> of the] final result will even be atomic”, meaning that visibility of one
> subpart (byte, for example) of the result should not be taken as evidence of
> visibility of the whole?  Or in other words, that the paragraph about CPU
> visibility provides the only guarantees of visibility.  (So if tearing occurs,
> it will have been dealt with before that guaranteed-visible point.)

I wasn't trying to write spec language. :)

The point I was making above was related to data correctness.  If 2 or more 'actors' are both performing atomic operations on the same target memory, the result is undefined.  An actor can be a NIC or CPU.

E.g. An atomic through NIC A adds 1 to each element.  An atomic through NIC B subtracts 1 to each element.  The results may end up with each element unchanged, incremented by 1, or decremented by 1.  And the change may not be the same for each element -- some may be +1, some -1, some unchanged.

The visibility discussion in the man page is describing when another actor can see the results of an atomic operation.  Before a second actor can perform atomic operations on a target region, the results of the first actor must first be visible.  This would ensure that the target region is updated in a consistent manner.

- Sean