Re: Transactional sequence stuff breaks pg_upgrade

I wrote:
> I believe I've identified the reason why skink and some other buildfarm
> members have been failing the pg_upgrade test recently.
> ...
> Not sure what we want to do about it. One idea is to make
> ALTER SEQUENCE not so transactional when in binary-upgrade mode.

On closer inspection, the only thing that pg_upgrade needs is to be
able to do ALTER SEQUENCE OWNED BY without a relfilenode bump. PFA
two versions of a patch that fixes this problem, at least to the
extent that it gets through check-world without triggering the Assert
I added to GetNewRelFileNode (which HEAD doesn't). The first one
is a minimally-invasive hack; the second one puts the responsibility
for deciding if a sequence rewrite is needed into init_params. That's
bulkier but might be useful if we ever grow additional ALTER SEQUENCE
options that don't need a rewrite. OTOH, I'm not very clear on what
such options might look like, so maybe the extra flexibility is useless.
Thoughts?

regards, tom lane