On Thu, Mar 05, 2020 at 02:52:44PM -0500, Tom Lane wrote:
>Tomas Vondra <tomas.vondra@2ndquadrant.com> writes:
>> FWIW I'm not suggesting you go and implement #1 or #2 for me, that'd be
>> up to me I guess. But I disagree there's no use case for it, and #3
>> makes this featuer useless for me.
>
>OK, then we need to do something else. Do you have ideas for other
>alternatives?
>
I don't have any other ideas, unfortunately. And I think if I had one,
it'd probably be some sort of ugly hack anyway :-/
>If not, we probably should bite the bullet and go for #1, since
>I have little doubt that we'll need that someday anyway.
>The trick will be to keep down the cache invalidation overhead...
>
Yeah, I agree #1 seems like the cleanest/best option. Are you worried
about the overhead due to the extra complexity, or overhead due to
cache getting invalidated for this particular reason?
regards
--
Tomas Vondra http://www.2ndQuadrant.com
PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services