Kisung Kim <kskim80(at)gmail(dot)com> writes:
> Tom Lane-2 wrote
>> This is an empirical decision based on the
>> frequency with which such a deduction would help versus the planner
>> cycles that would be spent looking for such cases.
> Is there any thing referring to this decision, like mail threads or
> conference materials?
[ shrug... ] It's doubtless been discussed on pgsql-hackers in the past,
but I don't plan to go digging through those archives for you. Looking
for constraint exclusion via
https://www.postgresql.org/search/?m=1
or your favorite search engine might help you.
regards, tom lane