MIME-Version: 1.0
References: 
 <CA+K2Runi_H2CBL0yMm3De2KqcR9RMA0HK5cLJjEhoNszC7myeg@mail.gmail.com>
 <aY5C6Xa5im72NF_Y@alap3.anarazel.de>
 <CA+K2Rum_QTZqTUrdMOL5hr-OOpCwGR_9Nj1z15BFObjktMOY6A@mail.gmail.com>
 <abBuKalOno33MQFw@nathan>
 <CA+K2Rum7+Jm2rm65K5msxaiAM8QTkhSNAYarPBP9O7nBXYo12Q@mail.gmail.com>
 <abmiNPQOqBrRlf_m@nathan>
 <CA+K2Rum-TB_iNzDWoXOJspf=jq0gd-wees8+9tBTJNyhy9cK5g@mail.gmail.com>
 <CA+K2Ru=PdZuXQbcfvqKysTkebTyXNd9j7dp+mTFQEYpLdGw1eA@mail.gmail.com>
 <acWj5FntidHJ9nVP@nathan>
 <CA+K2Runq+1gy8p6a-DsxpT2OkEkEu3cUGsZ9tdiGNrg_=P39gg@mail.gmail.com>
 <acv2vu8miagnHG1B@nathan>
In-Reply-To: <acv2vu8miagnHG1B@nathan>
From: KAZAR Ayoub <ma_kazar@esi.dz>
Date: Thu, 2 Apr 2026 20:07:38 +0200
Message-ID: 
 <CA+K2Ru=JK5NUEaxA77pCEer40QnV1TMxeg68Et9RL0zMZw_Jyw@mail.gmail.com>
Subject: Re: Speed up COPY TO text/CSV parsing using SIMD
To: Nathan Bossart <nathandbossart@gmail.com>
Cc: Andres Freund <andres@anarazel.de>,
 Pg Hackers <pgsql-hackers@postgresql.org>,
	Neil Conway <neil.conway@gmail.com>,
 Manni Wood <manni.wood@enterprisedb.com>,
	Andrew Dunstan <andrew@dunslane.net>, Shinya Kato <shinya11.kato@gmail.com>,
	Mark Wong <markwkm@gmail.com>, Nazir Bilal Yavuz <byavuz81@gmail.com>
Content-Type: multipart/alternative; boundary="00000000000018bee7064e7e1482"
Archived-At: 
 <https://www.postgresql.org/message-id/CA%2BK2Ru%3DJK5NUEaxA77pCEer40QnV1TMxeg68Et9RL0zMZw_Jyw%40mail.gmail.com>
Precedence: bulk

--00000000000018bee7064e7e1482
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

On Tue, Mar 31, 2026 at 6:30=E2=80=AFPM Nathan Bossart <nathandbossart@gmai=
l.com>
wrote:

> On Fri, Mar 27, 2026 at 07:48:38PM +0100, KAZAR Ayoub wrote:
> > I added a prescan loop inside the simd helpers trying to catch special
> > chars in sizeof(Vector8) characters, i measured how good is this at
> > reducing the overhead of starting simd and exiting at first vector:
> > the scalar loop is better than SIMD for one vector if it finds a specia=
l
> > character before 6th character, worst case is not a clean vector, where
> the
> > scalar loop needs 20 more cycles compared to SIMD.
> > This helps mitigate the case of JSON(B) in CSV format, this is why I on=
ly
> > added this for CSV case only.
>
> Interesting.
>
> > In a benchmark with 10M early SIMD exit like the JSONB case, the previo=
us
> > 3% regression is gone.
>
> While these are nice results, I think it's best that we target v20 for th=
is
> patch so that we have more time to benchmark and explore edge cases.
>
Thanks for the review.
Fair enough, I'll try many more cases in the upcoming weeks to make sure
we're not missing anything.

>
> --
> nathan

Regards,
Ayoub

--00000000000018bee7064e7e1482
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr"><div dir=3D"ltr">On Tue, Mar 31, 2026 at 6:30=E2=80=AFPM N=
athan Bossart &lt;<a href=3D"mailto:nathandbossart@gmail.com">nathandbossar=
t@gmail.com</a>&gt; wrote:<br></div><div class=3D"gmail_quote gmail_quote_c=
ontainer"><blockquote class=3D"gmail_quote" style=3D"margin:0px 0px 0px 0.8=
ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">On Fri, Mar 27,=
 2026 at 07:48:38PM +0100, KAZAR Ayoub wrote:<br>
&gt; I added a prescan loop inside the simd helpers trying to catch special=
<br>
&gt; chars in sizeof(Vector8) characters, i measured how good is this at<br=
>
&gt; reducing the overhead of starting simd and exiting at first vector:<br=
>
&gt; the scalar loop is better than SIMD for one vector if it finds a speci=
al<br>
&gt; character before 6th character, worst case is not a clean vector, wher=
e the<br>
&gt; scalar loop needs 20 more cycles compared to SIMD.<br>
&gt; This helps mitigate the case of JSON(B) in CSV format, this is why I o=
nly<br>
&gt; added this for CSV case only.<br>
<br>
Interesting.<br>
<br>
&gt; In a benchmark with 10M early SIMD exit like the JSONB case, the previ=
ous<br>
&gt; 3% regression is gone.<br>
<br>
While these are nice results, I think it&#39;s best that we target v20 for =
this<br>
patch so that we have more time to benchmark and explore edge cases.<br></b=
lockquote><div>Thanks for the review.</div><div>Fair enough, I&#39;ll try m=
any more cases in the upcoming weeks=C2=A0to make sure we&#39;re not missin=
g anything.</div><blockquote class=3D"gmail_quote" style=3D"margin:0px 0px =
0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">
<br>
-- <br>
nathan</blockquote><div>Regards,</div><div>Ayoub</div></div></div>

--00000000000018bee7064e7e1482--