Re: Fix unsafe PlannedStmt access in pg_stat_statements

Michael Paquier <michael@paquier.xyz>

From: Michael Paquier <michael@paquier.xyz>

To: Andres Freund <andres@anarazel.de>

Cc: pgsql-hackers@lists.postgresql.org, Chao Li <li.evan.chao@gmail.com>

Date: 2026-05-13T04:26:36Z

Lists: pgsql-hackers

Commits

Same data as JSON: GET /api/v1/messages/:b64id/commits the thread's linked commits as JSON, with link sources. API reference →

pg_stat_statements: Set PlannedStmt to NULL after nested utility execution
- 66366217822e 19 (unreleased) landed
pg_stat_statements: Fix potential use-after-free of PlannedStmt
- 8268e41aca23 19 (unreleased) landed
pg_stat_statements: Add counters for generic and custom plans
- 3357471cf9f5 19 (unreleased) cited

Attachments

pgss-pstmt.patch (text/plain) patch

On Tue, May 12, 2026 at 11:00:16AM +0200, Andres Freund wrote:
> Seems like the code should make this clearer, by simply unsetting
> pstmt at the point it becomes unsafe to use? 

Sure, we could do that as well and crash hard if something decides to
do the same mistake.  Like the simple patch attached for example?  I
am not sure if we need to update the comment.  It's pretty clear what
the intention is, at least to me.
--
Michael