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: 2023-05-12T22:56:56Z
Lists: pgsql-hackers 
On Fri, May 12, 2023 at 07:35:20AM +0530, Bharath Rupireddy wrote:
> --enable-atomics=no, -T60:
> --enable-spinlocks=no, -T60:
> --enable-atomics=no --enable-spinlocks=no, -T60:

Thanks for these extra tests, I have not done these specific cases but
the profiles look similar to what I've seen myself.  If I recall
correctly the fallback implementation of atomics just uses spinlocks
internally to force the barriers required.
--
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.