Date: Wed, 12 Feb 2025 09:38:04 -0600
From: Nathan Bossart <nathandbossart@gmail.com>
To: Bertrand Drouvot <bertranddrouvot.pg@gmail.com>
Cc: Julien Rouhaud <rjuju123@gmail.com>, pgsql-hackers@lists.postgresql.org
Subject: Re: Unneeded volatile qualifier in fmgr.c
Message-ID: <Z6zAXAKSqQHOwqyE@nathan>
References: <Z6xbAgXKY2L-3d5Q@jrouhaud>
 <Z6xevtT9vDJA2tcX@ip-10-97-1-34.eu-west-3.compute.internal>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <Z6xevtT9vDJA2tcX@ip-10-97-1-34.eu-west-3.compute.internal>
Archived-At: <https://www.postgresql.org/message-id/Z6zAXAKSqQHOwqyE%40nathan>
Precedence: bulk

On Wed, Feb 12, 2025 at 08:41:34AM +0000, Bertrand Drouvot wrote:
> On Wed, Feb 12, 2025 at 04:25:38PM +0800, Julien Rouhaud wrote:
>> While reading some code in fmgr.c I noticed that the save_nestlevel variable is
>> declared as volatile.  I'm assuming that's because a long time ago it was
>> modified in the PG_TRY / PG_CATCH block but it doesn't look needed anymore.
>> 
>> Trivial patch attached.
> 
> Yeah, makes sense to me.

It was modified in the PG_CATCH section when it was first added in commit
2abae34, but that part was removed a week later in commit 82a4798.  But I'm
not even sure it needed the volatile marker in the first commit, based on
the comment in elog.h:

 * Note: if a local variable of the function containing PG_TRY is modified
 * in the PG_TRY section and used in the PG_CATCH section, that variable
 * must be declared "volatile" for POSIX compliance.

AFAICT it was never modified in the PG_TRY section.

In any case, I don't see a problem with removing "volatile" now, so I'll
plan on committing this later today if there are no objections.

-- 
nathan