On 05/23/2013 03:59 AM US Central Time, Marc-Andre Hermanns wrote:

we had a similar reports from Scalasca users on systems with MPICH3.

It's "MPICH", whose latest version is 3.0.4. Please don't advertise "MPICH3" anywhere :-).

We solved it for us with a preprocessor macro that is prepended to all relevant buffer parameters. Depending on the MPI version the macro resolves to a 'const' or not. Quite low-tech, I know. ;-)

I'm not saying it cannot be solved. In fact, we proposed a similar solution to the Tau folks. I'm just saying it's not backward compatible. But maybe I was the only person surprised by that. -- Pavan -- Pavan Balaji http://www.mcs.anl.gov/~balaji