Re: [PATCH] Use optimized single-datum tuplesort in ExecSort - Mailing list pgsql-hackers

From John Naylor
Subject Re: [PATCH] Use optimized single-datum tuplesort in ExecSort
Date
Msg-id CAFBsxsEegvJsvUUL-rr7=-gE92=ku5w7FQHK6694VRNFt=SSwQ@mail.gmail.com
Whole thread Raw
In response to Re: [PATCH] Use optimized single-datum tuplesort in ExecSort  (David Rowley <dgrowleyml@gmail.com>)
Responses Re: [PATCH] Use optimized single-datum tuplesort in ExecSort  (Ronan Dunklau <ronan.dunklau@aiven.io>)
List pgsql-hackers

On Wed, Jul 14, 2021 at 6:14 AM David Rowley <dgrowleyml@gmail.com> wrote:

> It would be good to get a 2nd opinion about this idea.  Also, more
> benchmark results with v6 and v8 would be good too.

I tested this on an older Xeon, gcc 8.4 (here median of each test, full results attached):

test HEAD v6 v8

Test1 588 1007 998
Test2 198 202 197
Test3 374 516 512
Test4 172 165 166
Test5 255 279 283
Test6 227 251 251
Test7 145 147 146
Test8 474 783 770

Test4 could be a regression, but 2 and 7 look fine here.

--
John Naylor
EDB: http://www.enterprisedb.com

Attachment

pgsql-hackers by date:

Previous
From: Tom Lane
Date:
Subject: Re: Add missing function abs (interval)
Next
From: David Rowley
Date:
Subject: Re: [PATCH] Use optimized single-datum tuplesort in ExecSort