The following review has been posted through the commitfest application: make installcheck-world: tested, failed Implements feature: tested, failed Spec compliant: not tested Documentation: not tested
I just run a few test on your latest patch. It works well. I agree with you, I think just tracking generic_calls is enough to get the reason of performance change from pg_stat_statements. I mean if you need more detailed information about the plan and execution of prepared statements, it is better to store this information in a separate view. It seems more reasonable to me.