Horton

Re: Is RecoveryConflictInterrupt() entirely safe in a signal handler?

Andres Freund <andres@anarazel.de>

View thread View on postgr.es
From: Andres Freund <andres@anarazel.de>
To: Noah Misch <noah@leadboat.com>
Cc: Thomas Munro <thomas.munro@gmail.com>, Tom Lane <tgl@sss.pgh.pa.us>, Robert Haas <robertmhaas@gmail.com>, Michael Paquier <michael@paquier.xyz>, pgsql-hackers <pgsql-hackers@postgresql.org>
Date: 2023-01-04T22:16:50Z
Lists: pgsql-hackers 
Hi,

On 2022-12-29 00:40:52 -0800, Noah Misch wrote:
> Incidentally, the affected test contains comment "# DROP TABLE containing
> block which standby has in a pinned buffer".  The standby holds no pin at
> that moment; the LOCK TABLE pins system catalog pages, but it drops every
> pin it acquires.

I guess that comment survived from an earlier version of that test (or another
test where it was copied from).

I'm inclined to just delete it.

Greetings,

Andres Freund

Commits

  1. Fix recovery conflict SIGUSR1 handling.

  2. Redesign interrupt/cancel API for regex engine.

  3. Update contrib/trgm_regexp's memory management.

  4. Update tsearch regex memory management.

  5. Use MemoryContext API for regex memory management.