Re: Delay the variable initialization in get_rel_sync_entry
Amit Langote <amitlangote09@gmail.com>
From: Amit Langote <amitlangote09@gmail.com>
To: Michael Paquier <michael@paquier.xyz>
Cc: "houzj.fnst@fujitsu.com" <houzj.fnst@fujitsu.com>,
pgsql-hackers <pgsql-hackers@postgresql.org>, Euler Taveira <euler@eulerto.com>
Date: 2022-01-06T05:16:25Z
Lists: pgsql-hackers
On Wed, Jan 5, 2022 at 10:31 AM Michael Paquier <michael@paquier.xyz> wrote: > On Fri, Dec 24, 2021 at 01:27:26PM +0000, houzj.fnst@fujitsu.com wrote: > > Here is the perf result of pgoutput_change after applying the patch. > > I didn't notice something else that stand out. > > > > |--2.99%--pgoutput_change > > |--1.80%--get_rel_sync_entry > > |--1.56%--hash_search > > > > Also attach complete profiles. > > Thanks. I have also done my own set of measurements, and the > difference is noticeable in the profiles I looked at. So, applied > down to 13. Thanks. I agree the variables were being defined in the wrong place before this patch. -- Amit Langote EDB: http://www.enterprisedb.com
Commits
-
Reduce relcache access in WAL sender streaming logical changes
- 3f8062bcf7f9 13.6 landed
- 5ddfebded41f 14.2 landed
- 6ce16088bfed 15.0 landed