Re: Check volatile functions in ppi_clauses for memoize node - Mailing list pgsql-hackers

From Richard Guo
Subject Re: Check volatile functions in ppi_clauses for memoize node
Date
Msg-id CAMbWs49OsS+T4XepK9iV9YuQKYzgWF0QY6DA9iPb40fKr4n0gQ@mail.gmail.com
Whole thread Raw
In response to Re: Check volatile functions in ppi_clauses for memoize node  (David Rowley <dgrowleyml@gmail.com>)
List pgsql-hackers

On Mon, Aug 7, 2023 at 6:19 PM David Rowley <dgrowleyml@gmail.com> wrote:
On Fri, 4 Aug 2023 at 22:26, Richard Guo <guofenglinux@gmail.com> wrote:
> explain (costs off)
> select * from t t1 left join lateral
>     (select t1.a as t1a, t2.a as t2a from t t2) s
> on t1.a = s.t2a + random();
>                   QUERY PLAN
> -----------------------------------------------
>  Nested Loop Left Join
>    ->  Seq Scan on t t1
>    ->  Memoize
>          Cache Key: t1.a
>          Cache Mode: binary
>          ->  Seq Scan on t t2
>                Filter: (t1.a = (a + random()))
> (7 rows)
>
> According to the theory we should not use memoize node for this query
> because of the volatile function in the inner side.  So propose a patch
> to fix that.

Thanks for the patch.  I've pushed a variation of it.

I didn't really see the need to make a single list with all the
RestrictInfos. It seems fine just to write code and loop over the
ppi_clauses checking for volatility.

That looks good.  Thanks for pushing it!

Thanks
Richard

pgsql-hackers by date:

Previous
From: Yugo NAGATA
Date:
Subject: Re: pgbench: allow to exit immediately when any client is aborted
Next
From: Michael Paquier
Date:
Subject: Re: Support to define custom wait events for extensions