Horton

Re: LogwrtResult contended spinlock

Alvaro Herrera <alvherre@alvh.no-ip.org>

View thread View on postgr.es
From: Alvaro Herrera <alvherre@alvh.no-ip.org>
To: Tom Lane <tgl@sss.pgh.pa.us>
Cc: Andres Freund <andres@anarazel.de>, pgsql-hackers@lists.postgresql.org, Jaime Casanova <jcasanov@systemguards.com.ec>
Date: 2022-04-07T18:29:44Z
Lists: pgsql-hackers 
On 2022-Apr-05, Alvaro Herrera wrote:

> Apologies -- I selected the wrong commit to extract the commit message
> from.  Here it is again.  I also removed an obsolete /* XXX */ comment.

I spent a lot of time staring at this to understand the needs for memory
barriers in the interactions.  In the end I decided not to get this out
for this cycle because I don't want to create subtle bugs in WAL.  I'll
come back with this for pg16.

-- 
Álvaro Herrera               48°01'N 7°57'E  —  https://www.EnterpriseDB.com/
"El que vive para el futuro es un iluso, y el que vive para el pasado,
un imbécil" (Luis Adler, "Los tripulantes de la noche")

Commits

  1. Remove bogus assertion in pg_atomic_monotonic_advance_u64

  2. Add XLogCtl->logInsertResult

  3. Operate XLogCtl->log{Write,Flush}Result with atomics

  4. Split XLogCtl->LogwrtResult into separate struct members

  5. Introduce atomic read/write functions with full barrier semantics.

  6. Reduce the number of GetFlushRecPtr() calls done by walsenders.

  7. Remove most volatile qualifiers from xlog.c