Date: Wed, 24 Jul 2024 06:00:59 -0700
From: Noah Misch <noah@leadboat.com>
To: Michael Paquier <michael@paquier.xyz>
Cc: Aleksander Alekseev <aleksander@timescale.com>,
	Postgres hackers <pgsql-hackers@lists.postgresql.org>,
	Alexander Korotkov <aekorotkov@gmail.com>,
	Maxim Orlov <orlovmg@gmail.com>,
	Heikki Linnakangas <hlinnaka@iki.fi>,
	Japin Li <japinli@hotmail.com>, Andres Freund <andres@anarazel.de>,
	Pavel Borisov <pashkin.elfe@gmail.com>,
	Peter Eisentraut <peter.eisentraut@enterprisedb.com>
Subject: Re: XID formatting and SLRU refactorings (was: Add 64-bit XIDs into
 PostgreSQL 15)
Message-ID: <20240724130059.1f.nmisch@google.com>
References: <20240626002747.dc.nmisch@google.com>
 <CAJ7c6TOqhfD04kEUjDgLGaCpq_JSv0F_S=sWWjHm+zNw4pqh2w@mail.gmail.com>
 <20240626185844.00.nmisch@google.com>
 <CAJ7c6TMjJUBpmVHkA6VVv0GvUhp_WJQAUdLVjzp5niFF69OLBw@mail.gmail.com>
 <CAJ7c6TNRuQ5x1=i5voaSfa9K11Vv8XhSmeJscPDP19zUZqF0zQ@mail.gmail.com>
 <ZozTnFpuU3mcSuq9@paquier.xyz>
 <CAJ7c6TPLGJ=wPkfpO=W1EwGEPaSBj3f-PJXF0yecKEB-Et5H3g@mail.gmail.com>
 <ZpCDRtT0m04Qj591@paquier.xyz>
 <CAJ7c6TPGYD=neghnRyJn__YR53g40ZdU+xoHxp0QUH6wtN9ZvQ@mail.gmail.com>
 <Zp9xeKB9bKpnhlNs@paquier.xyz>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <Zp9xeKB9bKpnhlNs@paquier.xyz>
User-Agent: Mutt/2.2.12 (2023-09-09)
Archived-At: 
 <https://www.postgresql.org/message-id/20240724130059.1f.nmisch%40google.com>
Precedence: bulk

On Tue, Jul 23, 2024 at 06:01:44PM +0900, Michael Paquier wrote:
> Hearing nothing but cicadas as now is their season, I have taken the
> initiative here to address this open item.
> 
> 0001 felt a bit too complicated in slru.h, so I have simplified it and
> kept all the details in slru.c with SlruFileName().
> 
> I have reviewed all the code that uses SLRUs, and spotted three more
> problematic code paths in predicate.c that needed an update like the
> others for some pagenos.  I've added these, and applied 0002.  We
> should be good now.

I'm still seeing need for s/int/int64/ at:

- "pagesegno" variable
- return value of MultiXactIdToOffsetSegment()
- return value of MXOffsetToMemberSegment()
- callers of previous two

Only the first should be a live bug, since multixact isn't electing the higher
pageno ceiling.