From: Tom Lane <tgl@sss.pgh.pa.us>
To: Alexander Korotkov <aekorotkov@gmail.com>
cc: Tender Wang <tndrwang@gmail.com>, nicolas.maus@bertelsmann.de,
        pgsql-bugs@lists.postgresql.org
Subject: Re: BUG #18692: Segmentation fault when extending a varchar column
 with a gist index with custom signal length
In-reply-to: 
 <CAPpHfdsH6ACTzcBAs_YC+UzFgjCWTzhNpaAOxssEJ75jH5xVXw@mail.gmail.com>
References: <18692-72ea398df3ec6712@postgresql.org>
 <3261639.1730912933@sss.pgh.pa.us>
 <CAPpHfdsedeW4qiF_Ky88u=+0Ep7ZGH5yBd47a-rueQ2_jh3Dhw@mail.gmail.com>
 <CAHewXN=eBAk2wMn00CCYqcqC-1UtCHBMxr0nREUp_Ch8XYSQEg@mail.gmail.com>
 <CAPpHfdvc1C4Q9LQ-x9SarnM14UcaqxvursrZWrQbF8r2CT9ghg@mail.gmail.com>
 <3534898.1731031548@sss.pgh.pa.us>
 <CAPpHfdtZ2a=1FbZXUeEbx0i-avsQb=6UE9Kg8eTLoQzt3KaY0w@mail.gmail.com>
 <3557257.1731042654@sss.pgh.pa.us>
 <CAPpHfdsH6ACTzcBAs_YC+UzFgjCWTzhNpaAOxssEJ75jH5xVXw@mail.gmail.com>
Comments: In-reply-to Alexander Korotkov <aekorotkov@gmail.com>
	message dated "Fri, 08 Nov 2024 10:47:35 +0200"
MIME-Version: 1.0
Content-Type: text/plain; charset="UTF-8"
Content-ID: <3629172.1731077350.1@sss.pgh.pa.us>
Content-Transfer-Encoding: 8bit
Date: Fri, 08 Nov 2024 09:49:10 -0500
Message-ID: <3629173.1731077350@sss.pgh.pa.us>
Archived-At: 
 <https://www.postgresql.org/message-id/3629173.1731077350%40sss.pgh.pa.us>
Precedence: bulk

Alexander Korotkov <aekorotkov@gmail.com> writes:
> On Fri, Nov 8, 2024 at 7:10 AM Tom Lane <tgl@sss.pgh.pa.us> wrote:
>> WFM.  But I'm concerned that you couldn't build a test case
>> where the comparison fails.  Seems like we need one, in view
>> of this bug having escaped detection for so long.

> AFAICS, the only use case of CheckIndexCompatible() is ALTER TYPE.  In
> that case we only serialize and deserialize opclass options without
> any involvement of opclass or data type.  So, with the current usage
> scenario CompareOpclassOptions() is only a paranoid check, but
> CheckIndexCompatible() has to do it due to its general semantic.

OK.  Well, let's settle for having a test case that at least runs
the code, even if only the success path is taken.

With the release freeze bearing down on us, do you want to wait
till after the releases to fix this?  It seems non-urgent given
how long it took for anyone to notice.

			regards, tom lane