From: Tom Lane <tgl@sss.pgh.pa.us>
To: shihao zhong <zhong950419@gmail.com>
cc: Bruce Momjian <bruce@momjian.us>,
        PostgreSQL-development <pgsql-hackers@lists.postgresql.org>
Subject: Re: First draft of PG 19 release notes
In-reply-to: 
 <CAGRkXqSsSL=5qXYymzN+D8hPZcBZGuEcT+5zF+MXE0KL_7Kxtg@mail.gmail.com>
References: <ad7ngXC125AvNqC2@momjian.us> <5973.1791788107@localhost>
 <as1_bGM_HKFA6HOs@momjian.us>
 <CAGRkXqSsSL=5qXYymzN+D8hPZcBZGuEcT+5zF+MXE0KL_7Kxtg@mail.gmail.com>
Comments: In-reply-to shihao zhong <zhong950419@gmail.com>
	message dated "Sat, 17 Oct 2026 00:03:07 -0400"
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-ID: <3382928.1792210380.1@sss.pgh.pa.us>
Date: Sat, 17 Oct 2026 00:13:00 -0400
Message-ID: <3382929.1792210380@sss.pgh.pa.us>
Archived-At: 
 <https://www.postgresql.org/message-id/3382929.1792210380%40sss.pgh.pa.us>
Precedence: bulk

shihao zhong <zhong950419@gmail.com> writes:
> 4. The hash join change (1811f1af98f) can write outer rows with NULL
> keys to a temp file, even with one batch. A LEFT JOIN with 297000 NULL
> keys wrote 33 MB of temp files, 18.6 wrote none. With temp_file_limit at
> 1MB it fails.

I'm a bit skeptical about this one.  The older code would have
put those rows into the primary hash table rather than a dedicated
tuplestore, but either way you risk having to dump them to disk.

			regards, tom lane