Horton

Re: valgrind versus pg_atomic_init()

Tom Lane <tgl@sss.pgh.pa.us>

View thread View on postgr.es
From: Tom Lane <tgl@sss.pgh.pa.us>
To: Andres Freund <andres@anarazel.de>
Cc: Noah Misch <noah@leadboat.com>, pgsql-hackers@lists.postgresql.org
Date: 2020-06-15T04:26:02Z
Lists: pgsql-hackers 
Andres Freund <andres@anarazel.de> writes:
> On 2020-06-14 22:30:25 -0400, Tom Lane wrote:
>> Perhaps it'd be worth putting a memory barrier at the end of the _init
>> function(s)?  As you say, this is hypothetical right now, but that'd be
>> a cheap improvement.

> I don't think it'd be that cheap for some cases. There's an atomic for
> every shared buffer, making their initialization full memory barriers
> would likely be noticable for larger shared_buffers values.

Fair point --- if we did need to do something to make this safer, doing it
at the level of individual atomic values would be the wrong thing anyway.

			regards, tom lane

Commits

  1. Avoid need for valgrind suppressions for pg_atomic_init_u64 on some platforms.