On Thu, Jul 22, 2021 at 9:42 AM Tom Lane <tgl@sss.pgh.pa.us> wrote:
> Right. The point here is that before v13, hash aggregation was not
> subject to the work_mem limit, nor any related limit. If you did an
> aggregation requiring more than 2GB-plus-slop, it would work just fine
> as long as your machine had enough RAM. Now, the performance sucks and
> there is no knob you can turn to fix it. That's unacceptable in my book.
Oh! That makes way more sense.
I suspect David's theory about hash_agg_set_limits()'s ngroup limit is
correct. It certainly seems like a good starting point.
--
Peter Geoghegan
