Re: pg_upgrade < 9.3 -> >=9.3 misses a step around multixacts

Andres Freund <andres@2ndquadrant.com> writes:
> On 2014-07-21 13:03:21 -0400, Tom Lane wrote:
>> I don't see any point in being picky about it.  What we want is to reject
>> values that are conclusively bogus; things that are just close to bogus
>> are not so interesting.

> It'd just have been about optimizing away repeated calls to ReadNew*...

Oh.  I think the existing code is probably efficient enough about that.
vac_update_datfrozenxid() only does it once anyway, and the calls in
vac_update_relstats() are arranged so that we don't expect them to be hit
at all in normal cases.

            regards, tom lane