Date: Sat, 02 May 2026 13:38:28 +0900 (JST)
Message-Id: <20260502.133828.1601537265248588543.ishii@postgresql.org>
To: assam258@gmail.com
Cc: zsolt.parragi@percona.com, sjjang112233@gmail.com,
 vik@postgresfriends.org, er@xs4all.nl, jacob.champion@enterprisedb.com,
 david.g.johnston@gmail.com, peter@eisentraut.org, li.evan.chao@gmail.com,
 pgsql-hackers@postgresql.org
Subject: Re: Row pattern recognition
From: Tatsuo Ishii <ishii@postgresql.org>
In-Reply-To: 
 <CAAAe_zCe1gj_UkapC4e_GVZmud+dxru8YONpKRhtY6c2RmXcZg@mail.gmail.com>
References: 
 <CAAAe_zBdwAUDNs_WFdLkFF=ewhkDkv-AqizVEVzhsfremGFb4w@mail.gmail.com>
	<20260502.131718.1177156404282376744.ishii@postgresql.org>
	<CAAAe_zCe1gj_UkapC4e_GVZmud+dxru8YONpKRhtY6c2RmXcZg@mail.gmail.com>
Mime-Version: 1.0
Content-Type: Text/Plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Archived-At: 
 <https://www.postgresql.org/message-id/20260502.133828.1601537265248588543.ishii%40postgresql.org>
Precedence: bulk

Hi Henson,

> I know in v47 JIT compilation supports PREV/NEXT. What about
>> FIRST/LAST and compound forms?
>>
> 
> Yes, those are JIT-compiled too in v47.
> 
> The JIT side only deals with two opcodes -- EEOP_RPR_NAV_SET
> and EEOP_RPR_NAV_RESTORE.
> 
> FIRST/LAST and the compound forms differ only in how
> ExecEvalRPRNavSet() computes target_pos for the slot swap;
> the opcode interface and the call into ExecEvalRPRNavSet /
> ExecEvalRPRNavRestore are unchanged.
> 
> So 0018's JIT support carries over to FIRST/LAST and
> compounds without any JIT-side change.

Ok, from user's point of view, PREV/NEXT/FIRST/LAST and compound forms
are all JIT-compiled. I will add it to the "major changes" section in
the v47 patches posting.

Regards,
--
Tatsuo Ishii
SRA OSS K.K.
English: http://www.sraoss.co.jp/index_en/
Japanese:http://www.sraoss.co.jp