Re: BUG #18477: A specific SQL query with "ORDER BY ... NULLS FIRST" is performing poorly if an ordering column is n - Mailing list pgsql-bugs

From David Rowley
Subject Re: BUG #18477: A specific SQL query with "ORDER BY ... NULLS FIRST" is performing poorly if an ordering column is n
Date
Msg-id CAApHDvoQmbesAm93i6mC_kzi_6x3ng-XT2Na9OFfSnw=eGisag@mail.gmail.com
Whole thread Raw
In response to Re: BUG #18477: A specific SQL query with "ORDER BY ... NULLS FIRST" is performing poorly if an ordering column is n  (Laurenz Albe <laurenz.albe@cybertec.at>)
List pgsql-bugs
On Thu, 6 Jun 2024 at 20:10, Laurenz Albe <laurenz.albe@cybertec.at> wrote:
>
> On Thu, 2024-06-06 at 17:04 +1200, David Rowley wrote:
> > If the user has some genuine reason for creating a NULLS FIRST index
> > for some other query, then it might be nice if we were able to use
> > that index for Merge Joins instead of them having to create another
> > index to speed up the join query.
>
> I cannot imagine a genuine reason for creating a NULLS FIRST index
> on a NOT NULL column, but perhaps I am too narrow-minded.

You didn't quote the part where I said "if you had two nullable
columns". It might be worth rereading what I wrote and taking the time
to understand it.

David



pgsql-bugs by date:

Previous
From: Laurenz Albe
Date:
Subject: Re: BUG #18477: A specific SQL query with "ORDER BY ... NULLS FIRST" is performing poorly if an ordering column is n
Next
From: PG Bug reporting form
Date:
Subject: BUG #18497: Heap-use-after-free in plpgsql