Re: Postgres Optimizer ignores information about foreign key relationship, severly misestimating number of returned rows in join

From: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
To: "Ehrenreich, Sigrid" <Ehrenreich(at)consist(dot)de>
Cc: David Rowley <dgrowleyml(at)gmail(dot)com>, "pgsql-performance(at)lists(dot)postgresql(dot)org" <pgsql-performance(at)lists(dot)postgresql(dot)org>
Subject: Re: Postgres Optimizer ignores information about foreign key relationship, severly misestimating number of returned rows in join
Date: 2020-10-28 16:55:26
Message-ID: 1345771.1603904126@sss.pgh.pa.us
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-performance

"Ehrenreich, Sigrid" <Ehrenreich(at)consist(dot)de> writes:
> A patch would be very much appreciated.
> We are currently running on Version 12, but could upgrade to 13, if necessary.
> Could you send me a notification if you managed to program a patch for that?

I've pushed a patch for this to HEAD, but current thinking is that we
will not be back-patching it. Still, if you're desperate you could
consider running a custom build of v13 with that patch --- a quick
check suggests that it would back-patch easily. v12 would be a
slightly harder lift (I see one hunk doesn't apply) but probably
not by much.

https://git.postgresql.org/gitweb/?p=postgresql.git;a=patch;h=ad1c36b0709e47cdb3cc4abd6c939fe64279b63f

regards, tom lane

In response to

Responses

Browse pgsql-performance by date

  From Date Subject
Next Message Olivier Poquet 2020-10-28 22:46:12 query plan using partial index expects a much larger number of rows than is possible
Previous Message Mats Olsen 2020-10-28 06:45:10 Re: Query Performance / Planner estimate off