On 28/12/2023 10:03, Richard Guo wrote:
> During that time, join removal was only performed for left joins, so it
> was not possible for 'ojrelid' to be negative. However, with the
> introduction of the SJE feature, inner joins can also be removed, and
> 'ojrelid' is set to -1 in the inner join case. That's how we see this
> error.
Agreed, I have come to the same conclusion.
> if (bms_is_subset(phinfo->ph_needed, joinrelids) &&
> bms_is_member(relid, phinfo->ph_eval_at) &&
> - !bms_is_member(ojrelid, phinfo->ph_eval_at))
> + (sjinfo == NULL || !bms_is_member(ojrelid, phinfo->ph_eval_at)))
It would be better to check "ojrelid >= 0" as a sign. But why do you
think SJE could need to delete PlaceHolderVar at all?
>
> Alternatively, we can modify bms_is_member() to return false for
> negative numbers instead of emitting an error, as suggested by the
> comment there.
> - /* XXX better to just return false for x<0 ? */
> + /* negative number cannot be a member of the bitmapset */
> if (x < 0)
> - elog(ERROR, "negative bitmapset member not allowed");
> + return false;
>
> I prefer the second option, but I'm open to other thoughts.
I don't like this option. It could mask some blunders somewhere like a
current one.
> FWIW, here is a simplified repro for this error.
>
> create table t (a int primary key, b int);
>
> explain (verbose, costs off)
> select 1 from t t1 left join
> (lateral (select 1 as x, * from t t2) s1 inner join
> (select * from t t3) s2 on s1.a = s2.a)
> on true
> where s1.x = 1;
> ERROR: negative bitmapset member not allowed
It can be simplified even more (without LATERAL):
explain (verbose, costs off)
select 1 from t t1 left join
(select 1 as x, * from t t2) s1 inner join
(select * from t t3) using (a)
on true
where s1.x = 1;
--
regards,
Andrei Lepikhov
Postgres Professional