It would be desirable to add FFM micros to regular performance testing thereby creating a standardized set of FFM benchmarks that are run upon each build. This would allow us to track the evolution of the FFM library and also increase the probability that we see regressions.