public inbox for [email protected]  
help / color / mirror / Atom feed
From: Andrei Lepikhov <[email protected]>
To: Alexander Korotkov <[email protected]>
Cc: Alena Rybakina <[email protected]>
Cc: Nikolay Shaplov <[email protected]>
Cc: [email protected]
Cc: jian he <[email protected]>
Cc: Robert Haas <[email protected]>
Cc: [email protected]
Cc: Peter Geoghegan <[email protected]>
Cc: Marcos Pegoraro <[email protected]>
Cc: [email protected]
Cc: Peter Eisentraut <[email protected]>
Cc: Ranier Vilela <[email protected]>
Subject: Re: POC, WIP: OR-clause support for indexes
Date: Sun, 25 Aug 2024 16:16:07 +0200
Message-ID: <[email protected]> (raw)
In-Reply-To: <CAPpHfdvF864n=Lzmjd2XBi9TwboZvrhRtLSt2hCP+JVUv6XKzg@mail.gmail.com>
References: <[email protected]>
	<CAPpHfdsrtAB=0B-qYuQMPgyyy=-CycFJfzTWO_30nx27vXc3dA@mail.gmail.com>
	<[email protected]>
	<CAPpHfdvUEjkuFNUbg7GuLkXotwvpDgBYz4S8YcE=LRf9aBQh-Q@mail.gmail.com>
	<[email protected]>
	<CAPpHfdu5iQOjF93vGbjidsQkhHvY2NSm29duENYH_cbhC6x+Mg@mail.gmail.com>
	<CAPpHfdv9GrXfCOdPsMszBrORig+nW2+JBuMD-np_keVuKJ7bdQ@mail.gmail.com>
	<[email protected]>
	<CAPpHfduTM2W3kmiN1=5fvgcu6yA5=ghq82OkvLvZC8Mh9PpEuA@mail.gmail.com>
	<[email protected]>
	<[email protected]>
	<CAPpHfduFJHTNGCTFWt4GFs-E6Ts8xS5mCsKOYKBsZAmoOr8ZSA@mail.gmail.com>
	<[email protected]>
	<CAPpHfdtR3cBKCSj_7dd7EDF_pa-rUUQBx3-guK6kZZ=_hSeMfg@mail.gmail.com>
	<[email protected]>
	<CAPpHfdvF864n=Lzmjd2XBi9TwboZvrhRtLSt2hCP+JVUv6XKzg@mail.gmail.com>

On 23/8/2024 14:58, Alexander Korotkov wrote:
> On Fri, Aug 23, 2024 at 1:23 PM Andrei Lepikhov <[email protected]> wrote:
>> The last query doesn't group clauses into two indexes. The reason is in
>> match_index_to_operand which classifies all 'x=' to one class. I'm not
>> sure because of overhead, but it may be resolved by using
>> predicate_implied_by to partial indexes.
> 
> Yes, this is the conscious limitation of my patch: to consider similar
> OR arguments altogether and one-by-one, not in arbitrary groups.  The
> important thing here is that we still generating BitmapOR patch as we
> do without the patch.  So, there is no regression.  I would leave this
> as is to not make this feature too complicated. This could be improved
> in future though.
It looks reasonable for me, thanks for the explanation.

What's more, I suspicious about the line:
*subrinfo = *rinfo;

Here, you copy everything, including cached estimations like norm_selec 
or eval_cost. I see that the match_orclause_to_indexcol creates a new 
SAOP where all caches will be cleaned, but just to be sure, maybe we 
should reset any cached estimations to default values — in that case, 
anyone who tries to build a new path based on these grouped OR clauses 
will recalculate that data.
At least, incorrect eval_cost of iclause->rinfo can slightly change the 
cost of rechecking operation, can't it?

-- 
regards, Andrei Lepikhov







view thread (78+ messages)  latest in thread

reply

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Reply to all the recipients using the --to and --cc options:
  reply via email

  To: [email protected]
  Cc: [email protected], [email protected], [email protected], [email protected], [email protected], [email protected], [email protected], [email protected], [email protected], [email protected], [email protected], [email protected]
  Subject: Re: POC, WIP: OR-clause support for indexes
  In-Reply-To: <[email protected]>

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

This inbox is served by agora; see mirroring instructions
for how to clone and mirror all data and code used for this inbox