Hi,
On 2020-06-03 21:31:01 +0200, Tomas Vondra wrote:
> So there seems to be +40% between 9.6 and 10, and further +25% between
> 10 and master. However, plain hashagg, measured e.g. like this:
Ugh.
Since I am a likely culprit, I'm taking a look.
> Not sure what to think about this. Seems slot_deform_tuple got way more
> expensive between 9.6 and 10, for some reason.
Might indicate too many calls instead. Or it could just be the fact that
we have expressions deform all columns once, instead of deforming
columns one-by-one now.
Greetings,
Andres Freund