On 10/18/10 9:13 AM, Jeff Squyres wrote:
2. Marty: It is worthwhile to be able to profile the impact of MPIT itself -- e.g., see if MPIT itself is performing poorly. I extrapolate this to mean that it can be useful to help quantify the at least part of the MPIT's own overhead (as I typed that sentence, I'm not sure how much of MPIT's overhead will be measurable by simply timing how long its individual function calls...? I guess there's other metrics that can be measured, too...?).
Statistical clock-profiling could do it without any interface in the MPIT APIs. So it is probably not worth putting it in. It's a trade-off between observability and performance and complexity. Even having argued that it's nice to be able to observe, I'm happy with not adding the extra complexity. Marty