pgsql: bufmgr: Fix ordering of checks in PinBuffer()

From: Andres Freund <andres(at)anarazel(dot)de>
To: pgsql-committers(at)lists(dot)postgresql(dot)org
Subject: pgsql: bufmgr: Fix ordering of checks in PinBuffer()
Date: 2026-04-01 00:27:47
Message-ID: E1w7jQw-002NfP-0p@gemulon.postgresql.org
Views: Whole Thread | Raw Message | Download mbox | Resend email
Thread:
Lists: pgsql-committers

bufmgr: Fix ordering of checks in PinBuffer()

The check for skip_if_not_valid added in 819dc118c0f6 was put at the start of
the loop. A CAS loop in theory does allow to make that check in a race free
manner. However, just after the check, there's a
old_buf_state = WaitBufHdrUnlocked(buf);
which introduces a race, because it would allow BM_VALID to be cleared, after
the skip_if_not_valid check.

Fix by restarting the loop after WaitBufHdrUnlocked().

Reported-by: Yura Sokolov <y(dot)sokolov(at)postgrespro(dot)ru>
Discussion: https://postgr.es/m/5bf667f3-5270-4b19-a08f-0facbecdff68@postgrespro.ru

Branch
------
master

Details
-------
https://git.postgresql.org/pg/commitdiff/c0af4eb4e71ecd4826ae6c9985f1d10ace5e1be6

Modified Files
--------------
src/backend/storage/buffer/bufmgr.c | 7 ++++++-
1 file changed, 6 insertions(+), 1 deletion(-)

Browse pgsql-committers by date

  From Date Subject
Next Message Thomas Munro 2026-04-01 01:37:33 pgsql: Fix pg_waldump/t/001_basic.pl with BSD tar on ZFS.
Previous Message Tom Lane 2026-03-31 20:36:06 pgsql: Doc: warn that parallel pg_restore may fail if --no-schema was u