Message-ID: <23b8e1f917739be889e320a407316832cb1bf8f1.camel@j-davis.com>
Subject: Re: Rework of collation code, extensibility
From: Jeff Davis <pgsql@j-davis.com>
To: Peter Geoghegan <pg@bowt.ie>
Cc: pgsql-hackers@postgresql.org
Date: Fri, 20 Jan 2023 16:34:44 -0800
In-Reply-To: <45eb8e985d264bb7216b00d02eb28559d3d7b4bd.camel@j-davis.com>
References: <a581136455c940d7bd0ff482d3a2bd51af25a94f.camel@j-davis.com>
	 <a20de213c75fe471d067eecc1ac65e0edb9af377.camel@j-davis.com>
	 <34b42d5055eec98b7183f14e8db15e5f48c0d413.camel@j-davis.com>
	 <CAH2-WzmQ+qe_bemX=S6T_xvZ-cpfDOV06imo3kpmx4m7b6b7jg@mail.gmail.com>
	 <052a5ed874d110be2f3ae28752e363306b10966d.camel@j-davis.com>
	 <CAH2-WzmDEYEobEjWYpnHPx3F92pxaG1W4TVFdhCMjjHK+voq9g@mail.gmail.com>
	 <45eb8e985d264bb7216b00d02eb28559d3d7b4bd.camel@j-davis.com>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
User-Agent: Evolution 3.44.4-0ubuntu1 
MIME-Version: 1.0
Archived-At: <https://www.postgresql.org/message-id/23b8e1f917739be889e320a407316832cb1bf8f1.camel%40j-davis.com>
Precedence: bulk

On Fri, 2023-01-20 at 12:54 -0800, Jeff Davis wrote:
> Both of these are surprising, and I haven't investigated deeply yet.

It's just because autoconf defaults to -O2 and meson to -O3, at least
on my machine. It turns out that, at -O2, master and the refactoring
branch are even; but at -O3, both get faster, and the refactoring pulls
ahead by a few percentage points.

At least that's what's happening for en-US-x-icu on UTF-8 with my test
data set. I didn't see much of a difference in other situations, but I
didn't retest those other situations this time around.

We should still look into why disabling abbreviated keys improves
performance in some cases. Maybe we need a GUC for that?


--=20
Jeff Davis
PostgreSQL Contributor Team - AWS