MIME-Version: 1.0
References: <662101e2-5ca5-4bf5-9307-2519efde8690@ya.ru> <8687df7d49bc4b28b41c7cf4e9ad1e15d383b2fa.camel@cybertec.at>
 <Z/bk5E6pqIBWTF3j@ubby> <c857b827-3a17-4600-aa98-d0710216ccfe@aklaver.com>
 <Z/foWNmbdDOWH6/M@ubby> <CAKFQuwa-H6ZzST_dFNhviqQ8b4fPfLT+o3btfygrVOxQJhgWNQ@mail.gmail.com>
In-Reply-To: <CAKFQuwa-H6ZzST_dFNhviqQ8b4fPfLT+o3btfygrVOxQJhgWNQ@mail.gmail.com>
From: Merlin Moncure <mmoncure@gmail.com>
Date: Thu, 10 Apr 2025 22:18:42 -0500
Message-ID: <CAHyXU0ypfKJ6LmEWJHKF63m5UEcZwB5gny3DWGQygvVeePM8mw@mail.gmail.com>
Subject: Re: Interesting case of IMMUTABLE significantly hurting performance
To: "David G. Johnston" <david.g.johnston@gmail.com>
Cc: Nico Williams <nico@cryptonector.com>, Adrian Klaver <adrian.klaver@aklaver.com>, 
	Laurenz Albe <laurenz.albe@cybertec.at>, Olleg Samoylov <splarv@ya.ru>, 
	"pgsql-generallists.postgresql.org" <pgsql-general@lists.postgresql.org>
Content-Type: multipart/alternative; boundary="000000000000b24931063278297a"
Archived-At: <https://www.postgresql.org/message-id/CAHyXU0ypfKJ6LmEWJHKF63m5UEcZwB5gny3DWGQygvVeePM8mw%40mail.gmail.com>
Precedence: bulk

--000000000000b24931063278297a
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

On Thu, Apr 10, 2025 at 10:59=E2=80=AFAM David G. Johnston <
david.g.johnston@gmail.com> wrote:

> On Thu, Apr 10, 2025 at 8:49=E2=80=AFAM Nico Williams <nico@cryptonector.=
com>
> wrote:
>
>> On Wed, Apr 09, 2025 at 02:43:11PM -0700, Adrian Klaver wrote:
>> > On 4/9/25 14:21, Nico Williams wrote:
>> > > That to_char is not immutable is not documented though.  Though it's
>> > > clear when looking at the docs for the `jsonb_.*_tz()` functions.
>> >
>> > From here:
>> >
>> > https://www.postgresql.org/docs/current/catalog-pg-proc.html
>> >
>> > select proname, provolatile, prosrc  from pg_proc where
>> proname=3D'to_char';
>> > [...]
>>
>> I'm surprised to see that counted as docs, but good to know.
>>
>>
> Consulting pg_proc constitutes, IMO, going outside the documentation to
> retrieve information.  It is just not high up on anyone's annoyance list =
to
> try and get this piece of information incorporated into the documentation=
.
> Partly because \df+ does show this information as well, so at least one
> doesn't have to go write the catalog query themself.
>

Facts.  This is black magic.   This has come up over and over.

I guess the real problems here are lack of feedback on a number of fronts:
*) the server knows the function is not immutable but lets you create it
anyway, even though it can have negative downstream consequences
*) there is no way to discern inline vs non-inlined execution in explain
*) the planner is clearly not modelling function scan overhead give the
relative costing discrepancies

I think the first point is the most serious. A warning on function
creation, 'WARNING: this function is not provably immutable and therefore
is not eligible for inlining" with some appropriate hints might do the
trick, in the very specific situation where the function is otherwise
eligible.

merlin

--000000000000b24931063278297a
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr"><div dir=3D"ltr">On Thu, Apr 10, 2025 at 10:59=E2=80=AFAM =
David G. Johnston &lt;<a href=3D"mailto:david.g.johnston@gmail.com">david.g=
.johnston@gmail.com</a>&gt; wrote:</div><div class=3D"gmail_quote gmail_quo=
te_container"><blockquote class=3D"gmail_quote" style=3D"margin:0px 0px 0px=
 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div dir=3D=
"ltr"><div dir=3D"ltr"><div style=3D"font-family:arial,helvetica,sans-serif=
"><span style=3D"font-family:Arial,Helvetica,sans-serif">On Thu, Apr 10, 20=
25 at 8:49=E2=80=AFAM Nico Williams &lt;<a href=3D"mailto:nico@cryptonector=
.com" target=3D"_blank">nico@cryptonector.com</a>&gt; wrote:</span></div></=
div><div class=3D"gmail_quote"><blockquote class=3D"gmail_quote" style=3D"m=
argin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left=
:1ex">On Wed, Apr 09, 2025 at 02:43:11PM -0700, Adrian Klaver wrote:<br>
&gt; On 4/9/25 14:21, Nico Williams wrote:<br>
&gt; &gt; That to_char is not immutable is not documented though.=C2=A0 Tho=
ugh it&#39;s<br>
&gt; &gt; clear when looking at the docs for the `jsonb_.*_tz()` functions.=
<br>
&gt; <br>
&gt; From here:<br>
&gt; <br>
&gt; <a href=3D"https://www.postgresql.org/docs/current/catalog-pg-proc.htm=
l" rel=3D"noreferrer" target=3D"_blank">https://www.postgresql.org/docs/cur=
rent/catalog-pg-proc.html</a><br>
&gt; <br>
&gt; select proname, provolatile, prosrc=C2=A0 from pg_proc where proname=
=3D&#39;to_char&#39;;<br>
&gt; [...]<br>
<br>
I&#39;m surprised to see that counted as docs, but good to know.<br><br></b=
lockquote><div><br></div><div style=3D"font-family:arial,helvetica,sans-ser=
if">Consulting pg_proc constitutes, IMO, going outside the documentation=C2=
=A0to retrieve information.=C2=A0 It is just not high up on anyone&#39;s an=
noyance list to try and get this piece of information incorporated into the=
 documentation.=C2=A0 Partly because \df+ does show this information as wel=
l, so at least one doesn&#39;t have to go write the catalog query themself.=
</div></div></div></blockquote><div><br></div><div>Facts.=C2=A0 This is bla=
ck magic.=C2=A0 =C2=A0This has come up over and over.=C2=A0</div><div><br><=
/div><div>I guess the real problems here are lack of feedback on a number o=
f fronts:</div><div>*) the server knows the function is not immutable but l=
ets you create it anyway, even though it can have negative downstream conse=
quences</div><div>*) there is no way to discern inline vs non-inlined execu=
tion in explain</div><div>*) the planner is clearly not modelling function =
scan overhead give the relative costing discrepancies</div><div><br></div><=
div>I think the first point is the most serious. A warning on function crea=
tion, &#39;WARNING: this function is not provably immutable and therefore i=
s not eligible for inlining&quot; with some appropriate=C2=A0hints might do=
 the trick, in the very specific situation where the function is otherwise =
eligible.=C2=A0=C2=A0</div><div><br></div><div>merlin</div></div></div>

--000000000000b24931063278297a--