Re: pgsql: Fix assorted inconsistencies in GIN opclass support function dec

On 2016-05-02 17:09:48 -0400, Tom Lane wrote:
> But this is likely moot anyway, because of the need to bump all the
> contrib modules' versions in order to install parallel-safety labels on
> their functions.  (I wonder why that isn't on the open-items list.)

I think the concensus when discussing that was that that'd essentially
be a "feature", and thus too late for 9.6.  I'm a bit doubtful about
that position, but I also don't have a strong opinion the other way round.