[Mpi3-tools] MPI_T session question

[Pavan Balaji]

On 08/17/2013 02:11 PM, William Gropp wrote:
> MPI, where it implements the performance variables, doesn't need to
> store more than one - only in the session structure would there be
> copies.

I think you are saying the same thing I wrote in my email.

>   I view (without having participated) the design of the session as a
> place to put all of the overhead out of the critical path, including
> potential thread safety overheads.  My comment about a high quality
> implementation is that it is possible, with good and careful design, to
> avoid most of the extra costs in the case of a single active session
> (probably one indirection and a branch).

Yes, of course.  The single session case is easy.

In fact, depending on how many variables we have, the multi-session case 
might not be too expensive either.  I was just checking to make sure my 
understanding was correct.

  -- Pavan

-- 
Pavan Balaji
http://www.mcs.anl.gov/~balaji