RE: comments on draft-ietf-sming-reqs-02.txt: Arrays

["Jason, Jamie" <jamie.jason@intel.com>]

Apologies to Juergen.  I forgot to hit "reply all", so you'll get two copies
of the reply.

Jamie

> Jamie> Having said that, I do believe that just because an underlying
> Jamie> protocol cannot support them directly should not be a reason
> Jamie> for hamstringing SMIng.  Currently, workarounds for arrays
> Jamie> suffer from the same lack of atomicty.  So, in my opinion,
> Jamie> adding array support does not deviate from the status quo of
> Jamie> today.
> 
> So you are saying that issue 4.1.25 implies atomic access or not? I
> think this must be spelled out very clearly.
> 
> Despite this important little detail, I believe that this new feature
> is at best a nice-to-have.

I think that if we are to do arrays, then it needs to be made explicitly
clear that this means atomic access.  And, this should probably be added to
the text for the requirement.  If this prompts a change to a protocol, then
this may be a good thing.

As I said earlier, I accept the WG consensus if it is to make it
nice-to-have.

Jamie