From: Tom Lane <tgl@sss.pgh.pa.us>
To: Andres Freund <andres@anarazel.de>
cc: Alexander Kuzmenkov <a.kuzmenkov@postgrespro.ru>,
        pgsql-hackers@lists.postgresql.org
Subject: Re: Performance improvements for src/port/snprintf.c
In-reply-to: <20180926221814.el5nwmkez6zfmzbz@alap3.anarazel.de>
References: <11787.1534530779@sss.pgh.pa.us> <2238.1534716465@sss.pgh.pa.us>
 <f5dd41da-911c-fd97-1b78-e0f8c0141897@postgrespro.ru>
 <15785.1536776055@sss.pgh.pa.us>
 <20180926220419.lorxpb564gakdjkz@alap3.anarazel.de>
 <20180926221814.el5nwmkez6zfmzbz@alap3.anarazel.de>
Comments: In-reply-to Andres Freund <andres@anarazel.de>
	message dated "Wed, 26 Sep 2018 15:18:14 -0700"
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-ID: <14301.1538005506.1@sss.pgh.pa.us>
Date: Wed, 26 Sep 2018 19:45:07 -0400
Message-ID: <14302.1538005507@sss.pgh.pa.us>
Precedence: bulk

Andres Freund <andres@anarazel.de> writes:
> On 2018-09-26 15:04:20 -0700, Andres Freund wrote:
>> I assume this partially is just the additional layers of function calls
>> (psprintf, pvsnprintf, pg_vsnprintf, dopr) that are now done, in
>> addition to pretty much the same work as before (i.e. sprintf("%.*f")).

No, there are no additional layers that weren't there before ---
snprintf.c's snprintf() slots in directly where the platform's did before.

Well, ok, dopr() wasn't there before, but I trust you're not claiming
that glibc's implementation of snprintf() is totally flat either.

I think it's just that snprintf.c is a bit slower in this case.  If you
look at glibc's implementation, they've expended a heck of a lot of code
and sweat on it.  The only reason we could hope to beat it is that we're
prepared to throw out some functionality, like LC_NUMERIC handling.

> I'm *NOT* proposing that as the actual solution, but as a datapoint, it
> might be interesting that hardcoding the precision and thus allowing use
> ofusing strfromd() instead of sprintf yields a *better* runtime than
> master.

Interesting.  strfromd() is a glibc-ism, and a fairly recent one at
that (my RHEL6 box doesn't seem to have it).  But we could use it where
available.  And it doesn't seem unreasonable to have a fast path for
the specific precision value(s) that float4/8out will actually use.

			regards, tom lane