MIME-Version: 1.0
References: <ddc34dbd-1efc-4710-824c-e101e7eb63e6@postgrespro.ru>
 <2542714.1774895703@sss.pgh.pa.us>
In-Reply-To: <2542714.1774895703@sss.pgh.pa.us>
From: Matthias van de Meent <boekewurm+postgres@gmail.com>
Date: Mon, 30 Mar 2026 20:50:37 +0200
Message-ID: 
 <CAEze2WhsfFwfjooci5gqpSvq_1ibJHnJZ29c58GbDJWH9nbYkA@mail.gmail.com>
Subject: Re: [BUG] Excessive memory usage with update on STORED generated
 columns.
To: Tom Lane <tgl@sss.pgh.pa.us>
Cc: "Anton A. Melnikov" <a.melnikov@postgrespro.ru>,
	PostgreSQL Hackers <pgsql-hackers@lists.postgresql.org>,
	Peter Eisentraut <peter@eisentraut.org>
Content-Type: text/plain; charset="UTF-8"
Archived-At: 
 <https://www.postgresql.org/message-id/CAEze2WhsfFwfjooci5gqpSvq_1ibJHnJZ29c58GbDJWH9nbYkA%40mail.gmail.com>
Precedence: bulk

On Mon, 30 Mar 2026 at 20:35, Tom Lane <tgl@sss.pgh.pa.us> wrote:
>
> So I think the correct fix is that there needs to be a separate
> boolean tracking whether this work has been done, akin to
> ExecGetExtraUpdatedCols's use of ri_extraUpdatedCols_valid.

Why would it need a new boolean? ri_extraUpdatedCols_valid tracks
exactly whether we've already gone through ExecInitGenerated(...,
CMD_UPDATE), and in doing so if both ri_extraUpdatedCols,
ri_GeneratedExprsU, and ri_NumGeneratedNeededU are valid or whether
they still need to be populated. Adding a new boolean would therefore
be rather duplicative.

See also the patch of a few hours ago at [0].

Kind regards,

Matthias van de Meent
Databricks (https://www.databricks.com)


[0] https://postgr.es/m/CAEze2Wh%2B_C8LtmiMRb58df%3DA1PrBVmMnYMOfbBUk9c%3Dm99CN%2Bw%40mail.gmail.com