On Thu, Apr 10, 2014 at 9:55 AM, Tom Lane <tgl@sss.pgh.pa.us> wrote:
> I'm pretty sure David Rowley did some benchmarking. The results should be > in this thread somewhere I think, but they currently evade me... Maybe David > can re-post, if he's following this...
I saw benchmarks addressing window aggregation, but none looking for side-effects on plain aggregation.
I think it was only around SUM(numeric), and you'll need to scroll down a bit to find it as it's mixed in with a load of window agg tests. At the time it was the only forward transition function that I had added any overhead to, so I think it was the only one I bothered to test at the time.
However, I do remember benchmarking the bool_and and bool_or changes after I rewrote the way they worked and also found the same as you... no difference, I just can't find the post with my results... Though it sounds like Tom found the same as me so I don't think it's worth me looking any more for them.