Date: Mon, 1 Sep 2025 13:19:41 +0800
From: Julien Rouhaud <rjuju123@gmail.com>
To: Michael Paquier <michael@paquier.xyz>
Cc: pgsql-hackers@lists.postgresql.org
Subject: Re: Update outdated references to SLRU ControlLock
Message-ID: <aLUs7ap5paM8Vq9C@jrouhaud>
References: <aLUT2UO8RjJOzZNq@jrouhaud>
 <aLUptKkboud-OZAw@paquier.xyz>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <aLUptKkboud-OZAw@paquier.xyz>
Archived-At: 
 <https://www.postgresql.org/message-id/aLUs7ap5paM8Vq9C%40jrouhaud>
Precedence: bulk

On Mon, Sep 01, 2025 at 02:05:56PM +0900, Michael Paquier wrote:
> On Mon, Sep 01, 2025 at 11:32:41AM +0800, Julien Rouhaud wrote:
> > Note that the main comment of slru.c still has one paragraph that mentions
> > "bank control lock" consistently before switching to just "control lock" in the
> > next paragraph.  I'm assuming that it's ok in that context as it seems clear to
> > me that those are the same thing, just spelled with a less verbose name.
>
> Good catch, right.
>
> I am not seeing "control" used much as a term (3 times on HEAD).
> There is a lot of "bank lock" or "SLRU bank lock", both being mixed
> depending on the parts of the code using SLRUs (multixact, predicates,
> etc.).

Yes, and some other parts simply mentions "lock" (eg TransactionIdGetStatus)

> "SLRU bank lock" speaks a bit better to me, as the concept relates
> to..  SLRUs, but that's mostly a matter of taste between the three
> wordings, I guess.  Do you have a preference?

I don't really have a preference.  Bank lock is shorter but may be a bit more
obscure, especially outside slru.c, so using "SLRU bank lock" could be better
indeed.