Hello Raymond,
Sure, it definitely makes sense to reduce the overhead when the extension is
disabled. I wanted to understand the source of performance issue, and your
explanations where not enough for reproducing it.
Thanks again Fabien. I am attaching the patch to this email in the hope of
getting it approved during the next commit fest.
I will continue trying to find a simple performance test to exemplify the
performance degradation that I have seen with more complex workloads.
Patch applies and compiles cleanly. Global and local make check ok.
The patch adds an early exit in one of the hook when pgss is not enabled
on a given query. This seems to be a long time oversight of some earlier
additions which only had some (small or large depending) performance
impact.
About the comment "...and" -> "... and" (add a space)
Otherwise all is well.
--
Fabien.
