From: Tom Lane <tgl@sss.pgh.pa.us>
To: Thomas Munro <thomas.munro@gmail.com>
cc: Andres Freund <andres@anarazel.de>,
        Alvaro Herrera <alvherre@alvh.no-ip.org>,
        Fabien COELHO <coelho@cri.ensmp.fr>,
        Thomas Munro <tmunro@postgresql.org>,
        pgsql-committers <pgsql-committers@lists.postgresql.org>
Subject: Re: pgsql: Track LLVM 15 changes.
In-reply-to: 
 <CA+hUKGJwFNH2yTLeWPaUy9Le=AgvEripEAdmyem7naKGGCAFEQ@mail.gmail.com>
References: 
 <CA+hUKGLqengPszjoQeD_7nmHsjGZS380+ymke-Fm=nNGMEQuhw@mail.gmail.com>
 <202202142122.eqz4mu2ecfes@alvherre.pgsql>
 <CA+hUKG+LX+mjBTPV=u4RnZU_5sp=BgCHh_91a1f1NFoKphtJWQ@mail.gmail.com>
 <3155481.1644950069@sss.pgh.pa.us>
 <CA+hUKG+p=faBQR2PSAqWoWa+_tJdKPT0wjZPQe7XcDEttUCgdQ@mail.gmail.com>
 <20220215224123.fx2mxkvlbvhk7377@alap3.anarazel.de>
 <3224284.1644965080@sss.pgh.pa.us>
 <CA+hUKGKM6BWHy9oi9Oyt7wgVLgi3ht2hb=k0Tmf0RLDhkEEXkQ@mail.gmail.com>
 <1407185.1766682319@sss.pgh.pa.us>
 <CA+hUKGJwFNH2yTLeWPaUy9Le=AgvEripEAdmyem7naKGGCAFEQ@mail.gmail.com>
Comments: In-reply-to Thomas Munro <thomas.munro@gmail.com>
	message dated "Mon, 29 Dec 2025 22:15:57 +1300"
MIME-Version: 1.0
Content-Type: text/plain; charset="UTF-8"
Content-ID: <2864709.1767021157.1@sss.pgh.pa.us>
Content-Transfer-Encoding: 8bit
Date: Mon, 29 Dec 2025 10:12:37 -0500
Message-ID: <2864710.1767021157@sss.pgh.pa.us>
Archived-At: 
 <https://www.postgresql.org/message-id/2864710.1767021157%40sss.pgh.pa.us>
Precedence: bulk

Thomas Munro <thomas.munro@gmail.com> writes:
> On Fri, Dec 26, 2025 at 6:05 AM Tom Lane <tgl@sss.pgh.pa.us> wrote:
>> Surely we don't need this anymore?  I tried removing it locally
>> and didn't see any such warnings (with Red Hat's llvm 20.1.8).

> Yeah.  We should just delete these obsolete lines in 18 and master.
> For older branches, I think it should be enough to update the comment
> in 14-17, as attached.

LGTM.

> We could work harder and suppress the warnings
> only for LLVM 14 and 15 specifically in PostgreSQL 14-17, but it seems
> like the main point is surely to have *something* watching out for new
> warnings that are added in future LLVM versions, and we'll get that
> from the newer branches.

Agreed, not worth the trouble.

			regards, tom lane