[mpich2-dev] some comments on attribute catching, create/free() keyvals and all that.

[Robert Latham]

On Fri, Mar 13, 2009 at 04:34:09PM -0200, Lisandro Dalcin wrote:
> Dave, this is not related at all with the ticket... but with a
> behavior of MPICH2 (and likely others MPI's out there, just confirmed
> this on Open MPI) about how create()/free() keyvals work...
> 
> Please, try to read my mail and run the code example... I would lilke
> to know your opinions about this behaviour, this could be really
> dangerous :-)

Well, it's a *little* related :> 

Dave did some work to make keyval freeing more robust to address those
tickets, and as a result, your testcase passes with MPICH2 as of r4059

It's surprising that freeing one keyval would mess up your ability to
create valid keyvals.   

==rob

-- 
Rob Latham
Mathematics and Computer Science Division    A215 0178 EA2D B059 8CDF
Argonne National Lab, IL USA                 B29D F333 664A 4280 315B