Horton

Re: [HACKERS] AdvanceXLInsertBuffer vs. WAL segment compressibility

Michael Paquier <michael@paquier.xyz>

View thread View on postgr.es
From: Michael Paquier <michael@paquier.xyz>
To: Stephen Frost <sfrost@snowman.net>
Cc: Chapman Flack <chap@anastigmatix.net>, Daniel Gustafsson <daniel@yesql.se>, Tom Lane <tgl@sss.pgh.pa.us>, Michael Paquier <michael.paquier@gmail.com>, Heikki Linnakangas <hlinnaka@iki.fi>, Robert Haas <robertmhaas@gmail.com>, Bruce Momjian <bruce@momjian.us>, PostgreSQL Hackers <pgsql-hackers@postgresql.org>
Date: 2018-03-26T05:43:54Z
Lists: pgsql-hackers 
On Sun, Mar 25, 2018 at 11:27:31PM -0400, Stephen Frost wrote:
> AdvanceXLInsertBuffer() does quite a bit, so I'm a bit surprised to see
> this simply removing that call, you're confident there's nothing done
> which still needs doing..?

Have a look at BKP_REMOVABLE then.  This was moved to page headers in
2dd9322, still it seems to me that the small benefits outlined on this
thread don't justify breaking tools relying on this flag set, especially
if there is no replacement for it.

So I would vote to not commit this patch.
--
Michael

Commits

  1. Ensure that WAL pages skipped by a forced WAL switch are zero-filled.

  2. Improve scalability of WAL insertions.

  3. Move BKP_REMOVABLE bit from individual WAL records to WAL page headers.