Feedback-ID: id4a34324:Fastmail
Date: Wed, 5 Mar 2025 09:19:11 -0500
From: Andres Freund <andres@anarazel.de>
To: Tom Lane <tgl@sss.pgh.pa.us>
Cc: Peter Geoghegan <pg@bowt.ie>, 
	Matthias van de Meent <boekewurm+postgres@gmail.com>, Andrew Dunstan <andrew@dunslane.net>, 
	Melanie Plageman <melanieplageman@gmail.com>, Thomas Munro <thomas.munro@gmail.com>, 
	Konstantin Knizhnik <knizhnik@garret.ru>, PostgreSQL Hackers <pgsql-hackers@postgresql.org>, 
	Alexander Kuzmenkov <akuzmenkov@timescale.com>
Subject: Re: Incorrect result of bitmap heap scan.
Message-ID: <tf5pp2o2a5x5qjcseq354bd26ya4o7p2vjzm5z4w57ca3vy6bc@ep7enrljvdkr>
References: <873c33c5-ef9e-41f6-80b2-2f5e11869f1c@garret.ru>
 <CAEze2Wg1Q4gWzm9RZ0yXydm23Mj3iScu8LA__Zz3JJEgpnoGPQ@mail.gmail.com>
 <vdoqkxlh3apghactvvfw6ftgeog4hqnc5sinudx2qj4kbxyab4@sk6nalz5xxti>
 <CAEze2WiHrnpXU33p7fcbrUyRByfK9jStVHqjxJNTzAhObyCooA@mail.gmail.com>
 <CAEze2WibNxb__4gM3UoFug=vunqvc_DLm=E9Qsgq=tCsHX1p2Q@mail.gmail.com>
 <CAEze2Wg8sbYAtK9FDO-MV9tbdfWA-8AB-feq2qTMib0xSpfnwQ@mail.gmail.com>
 <CAH2-Wz=Prv5buOehZnVrkk24nVn5xDQFXJ9=L8y0PhKzPCnSJg@mail.gmail.com>
 <3091523.1740875715@sss.pgh.pa.us>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <3091523.1740875715@sss.pgh.pa.us>
Archived-At: <https://www.postgresql.org/message-id/tf5pp2o2a5x5qjcseq354bd26ya4o7p2vjzm5z4w57ca3vy6bc%40ep7enrljvdkr>
Precedence: bulk

Hi,

On 2025-03-01 19:35:15 -0500, Tom Lane wrote:
> Peter Geoghegan <pg@bowt.ie> writes:
> > Is everybody in agreement about committing and back patching this fix,
> > which simply disables the optimization altogether?
> > I myself don't see a better way, but thought I'd ask before proceeding
> > with review and commit.
> 
> If you don't see a clear path forward, then "disable" is the only
> reasonable choice for the back branches.  Maybe we'll find a fix
> in future, but it seems unlikely that it'd be back-patchable.

I don't think there's a realistic way to fix it in the backbranches. And even
on master, I doubt that much of the current code would survive.

Greetings,

Andres Freund