Horton

Re: isTempNamespaceInUse() is incorrect with its handling of MyBackendId

Michael Paquier <michael@paquier.xyz>

View thread View on postgr.es
From: Michael Paquier <michael@paquier.xyz>
To: Julien Rouhaud <rjuju123@gmail.com>
Cc: Postgres hackers <pgsql-hackers@lists.postgresql.org>
Date: 2020-01-13T22:23:19Z
Lists: pgsql-hackers 
On Mon, Jan 13, 2020 at 02:56:13PM +0100, Julien Rouhaud wrote:
> Well, since isTempNamespaceInUse is for now only called by autovacuum, and
> shouldn't change soon, this really feels premature optimzation, so your
> original approach looks better.

Yes, I'd rather keep this routine in its simplest shape for now.  If
the optimization makes sense, though in most cases it won't because it
just helps sessions to detect faster their own temp schema, then let's
do it.  I'll let this patch aside for a couple of days to let others
comment on it, and if there are no objections, I'll commit the fix.
Thanks for the lookup!
--
Michael

Commits

  1. Fix buggy logic in isTempNamespaceInUse()