Nicolas Barbier <nicolas.barbier@gmail.com> writes:
> 2010/8/25 Simon Riggs <simon@2ndquadrant.com>:
>> You're exactly correct and I now understand Markus' comment. Do you
>> think that exact meaning prevents my proposal from being useful?
> Not at all, because I guess that updates to non-UNIQUE columns are way
> more common that updates to UNIQUE columns.
In particular, HOT updates are known useful even though they have
that restriction and more.
It strikes me that a possibly useful simplification of the idea is a
lock type that allows HOT updates and not non-HOT ones; or more
precisely not ones that change any indexed columns --- if the row ends
up having to go off-page for lack of space, that need not concern us.
regards, tom lane