On Thu, May 2, 2019 at 7:36 AM John Naylor <john.naylor@2ndquadrant.com> wrote:
>
> On Wed, May 1, 2019 at 11:24 PM Andres Freund <andres@anarazel.de> wrote:
> >
> > Hi,
> >
> > On 2019-04-18 14:10:29 -0700, Andres Freund wrote:
> > > My compromise suggestion would be to try to give John and Amit ~2 weeks
> > > to come up with a cleanup proposal, and then decide whether to 1) revert
> > > 2) apply the new patch, 3) decide to live with the warts for 12, and
> > > apply the patch in 13. As we would already have a patch, 3) seems like
> > > it'd be more tenable than without.
> >
> > I think decision time has come. My tentative impression is that we're
> > not there yet,
You are right that patch is not in committable shape, but the patch to
move the map to relcache is presented and the main work left there is
to review/test and add the invalidation calls as per discussion. It
is just that I don't want to that in haste leading to some other
problems. So, that patch should not take too much time and will
resolve the main complaint. Basically, I was planning to re-post that
patch as the discussion concludes between me and Alvaro and then
probably you can also look into it once to see if that addresses the
main complaint. There are a few other points for which John has
prepared a patch and that might need some work based on your inputs.
>> and should revert the improvements in v12, and apply the
> > improved version early in v13. As a second choice, we should live with
> > the current approach, if John and Amit "promise" further effort to clean
> > this up for v13.
>
> Yes, the revised approach is not currently as mature as the one in
> HEAD. It's not ready. Not wanting to attempt Promise Driven
> Development, I'd rather revert, and only try again if there's enough
> time and interest.
>
I can certainly help with moving patch (for cleanup) forward.
--
With Regards,
Amit Kapila.
EnterpriseDB: http://www.enterprisedb.com