Date: Tue, 6 Nov 2018 14:21:13 -0800
From: Andres Freund <andres@anarazel.de>
To: Tom Lane <tgl@sss.pgh.pa.us>
Cc: =?utf-8?Q?Ond=C5=99ej?= Bouda <obouda@email.cz>,
	Alvaro Herrera <alvherre@2ndquadrant.com>,
	pgsql-hackers@lists.postgresql.org,
	pgsql-general@lists.postgresql.org
Subject: Re: backend crash on DELETE, reproducible locally
Message-ID: <20181106222113.lcxpuuo2ht2rtw2n@alap3.anarazel.de>
References: <20181106194553.dynhd2xuivo56ymd@alvherre.pgsql>
 <33d144b9-5208-8d32-7e95-94ed4d2931b8@email.cz>
 <450.1541535425@sss.pgh.pa.us>
 <134c09b5-3eb0-bafd-2aeb-f383e1e21b8a@email.cz>
 <2432.1541538215@sss.pgh.pa.us>
 <2877.1541538838@sss.pgh.pa.us>
 <fc1fde8f-2e1d-3f8d-a583-82ba383c7fd2@email.cz>
 <4375.1541540840@sss.pgh.pa.us>
 <20181106215457.h36ccb5wywatcqhz@alap3.anarazel.de>
 <5323.1541542300@sss.pgh.pa.us>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <5323.1541542300@sss.pgh.pa.us>
Precedence: bulk

Hi,

On 2018-11-06 17:11:40 -0500, Tom Lane wrote:
> Andres Freund <andres@anarazel.de> writes:
> > On 2018-11-06 16:47:20 -0500, Tom Lane wrote:
> >> Looks like somebody forgot to list
> >> RELOPT_KIND_GIST in RELOPT_KIND_INDEX, but is that the
> >> fault of commit c203d6cf8 or was it busted before?
> 
> > Looks new:
> > +   RELOPT_KIND_INDEX = RELOPT_KIND_BTREE|RELOPT_KIND_HASH|RELOPT_KIND_GIN|RELOPT_KIND_SPGIST,
> > there aren't any other "for all indexes" type options, so the whole
> > category didn't exist before.
> 
> > It also strikes me as a really bad idea, even if RELOPT_KIND_GIST
> > wouldn't have been omitted: It breaks index am extensibility.
> 
> Hm, well, that enum already knows about all index types, doesn't it?

Not quite sure what you mean.

Before c203d6cf8 there weren't reloptions across index types. But after
it, if one adds a new index AM via an extension, one can't set
recheck_on_update = false for indexes using it, even though the feature
affects such indexes. We support adding indexes AMs at runtime these
days, including WAL logging (even though that's a bit
voluminous). There's even a contrib index AM...

The list of AMs is supposed to be extensible at runtime, cf
add_reloption_kind().

Greetings,

Andres Freund