Re: [HACKERS] data on devel code perf dip

From: Bruce Momjian <pgman(at)candle(dot)pha(dot)pa(dot)us>
To: Mark Wong <markw(at)osdl(dot)org>
Cc: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>, Alvaro Herrera <alvherre(at)alvh(dot)no-ip(dot)org>, maryedie(at)osdl(dot)org, PostgreSQL-patches <pgsql-patches(at)postgresql(dot)org>, Josh Berkus <josh(at)agliodbs(dot)com>
Subject: Re: [HACKERS] data on devel code perf dip
Date: 2005-08-12 16:12:17
Message-ID: 200508121612.j7CGCHT03271@candle.pha.pa.us
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers pgsql-patches

Mark Wong wrote:
> On Thu, 11 Aug 2005 22:11:42 -0400 (EDT)
> Bruce Momjian <pgman(at)candle(dot)pha(dot)pa(dot)us> wrote:
>
> > Tom Lane wrote:
> > > Bruce Momjian <pgman(at)candle(dot)pha(dot)pa(dot)us> writes:
> > > >> O_DIRECT is only being used for WAL page writes (or I sure hope so
> > > >> anyway), so shared_buffers should be irrelevant.
> > >
> > > > Uh, O_DIRECT really just enables when open_sync is used, and I assume
> > > > that is not used for writing dirty buffers during a checkpoint.
> > >
> > > I double-checked that O_DIRECT is really just used for WAL, and only
> > > when the sync mode is open_sync or open_datasync. So it seems
> > > impossible that it affected a run with mode fdatasync. What seems the
> > > best theory at the moment is that the grouped-WAL-write part of the
> > > patch doesn't work so well as we thought.
> >
> > Yes, that's my only guess. Let us know if you want the patch to test,
> > rather than pulling CVS before and after the patch was applied.
>
> Yeah, a patch would be a little easier. :)

OK, patch attached. The code has been cleaned up a little since then but
this is the basic change that should be tested. It is based on CVS of
2005/07/29 03:22:33 GMT.

--
Bruce Momjian | http://candle.pha.pa.us
pgman(at)candle(dot)pha(dot)pa(dot)us | (610) 359-1001
+ If your life is a hard drive, | 13 Roberts Road
+ Christ can be your backup. | Newtown Square, Pennsylvania 19073

Attachment Content-Type Size
unknown_filename text/plain 13.3 KB

In response to

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Alvaro Herrera 2005-08-12 16:12:20 CREATE TABLE synopsis
Previous Message Mark Wong 2005-08-12 15:43:10 Re: data on devel code perf dip

Browse pgsql-patches by date

  From Date Subject
Next Message Alvaro Herrera 2005-08-12 16:37:22 createuser patch
Previous Message Bruce Momjian 2005-08-12 15:59:06 Re: [HACKERS] For review: Server instrumentation patch