Noah Misch <noah@leadboat.com> wrote:
> I'd say, run them with this patch alone. The important thing is
> to not penalize existing COPY users.
Sounds good.
> Incidentally, the "did you want ... ?" was a genuine question. I
> see very little performance risk here, so the tests could be quite
> cursory, even absent entirely.
From what I've seen, I tend to agree. If there's a committer ready
to go over this, I would say that it might be worth waiting for the
benchmark results against the patch from the day before yesterday to
be run before "pulling the trigger" on it; but proceed on the basis
that it's a near-certainty that it will test out OK.
-Kevin