Horton

Re: WAL Insertion Lock Improvements

Michael Paquier <michael@paquier.xyz>

View thread View on postgr.es
From: Michael Paquier <michael@paquier.xyz>
To: Bharath Rupireddy <bharath.rupireddyforpostgres@gmail.com>
Cc: Nathan Bossart <nathandbossart@gmail.com>, Andres Freund <andres@anarazel.de>, PostgreSQL Hackers <pgsql-hackers@lists.postgresql.org>
Date: 2023-05-09T06:25:29Z
Lists: pgsql-hackers 
On Tue, May 09, 2023 at 02:10:20PM +0900, Michael Paquier wrote:
> Should we split this patch into two parts, as they aim at tackling two
> different cases then?  One for LWLockConflictsWithVar() and
> LWLockReleaseClearVar() which are the straight-forward pieces
> (using one pg_atomic_write_u64() in LWLockUpdateVar instead), then
> a second for LWLockUpdateVar()?

I have been studying that a bit more, and I'd like to take this
suggestion back.  Apologies for the noise.
--
Michael

Commits

  1. Optimize pg_atomic_exchange_u32 and pg_atomic_exchange_u64.

  2. Document more assumptions of LWLock variable changes with WAL inserts

  3. Optimize WAL insertion lock acquisition and release with some atomics

  4. Avoid the use of a separate spinlock to protect a LWLock's wait queue.