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: 2023-05-22T00:26:25Z
Lists: pgsql-hackers 
On Fri, May 19, 2023 at 08:34:16PM +0530, Bharath Rupireddy wrote:
> I get it. How about the following similar to what
> ProcessProcSignalBarrier() has?
> 
> +     * Note that pg_atomic_exchange_u64 is a full barrier, so we're guaranteed
> +     * that the variable is updated before waking up waiters.
> +     */
> 
> +     * Note that pg_atomic_exchange_u64 is a full barrier, so we're guaranteed
> +     * that the variable is updated before releasing the lock.
>       */
> 
> Please find the attached v8 patch with the above change.

Simpler and consistent, nice.  I don't have much more to add, so I
have switched the patch as RfC.
--
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.