Re: pgsql: Have test_fsync honor -f filename argument.

From: Bruce Momjian <pgman(at)candle(dot)pha(dot)pa(dot)us>
To: Qingqing Zhou <zhouqq(at)cs(dot)toronto(dot)edu>
Cc: pgsql-committers(at)postgresql(dot)org
Subject: Re: pgsql: Have test_fsync honor -f filename argument.
Date: 2005-11-16 03:51:39
Message-ID: 200511160351.jAG3pdi14093@candle.pha.pa.us
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-committers

Qingqing Zhou wrote:
>
> "Bruce Momjian" <momjian(at)svr1(dot)postgresql(dot)org> wrote
> > -----------
> > Have test_fsync honor -f filename argument.
> >
>
> The patch may forget to change the error message (Also, it would be better
> to append a '\n' in die() function). Another possible problem of this test
> is that it pickups the "/var/tmp". OS may have IO optimization on temporary
> directories, which will affect the confidence of the results.
>
> Example:
> ---
> $#./test_fsync -f $HOME/test.out
> Simple write timing:
> write 4.265149
> $#./test_fsync -f /var/tmp/test.out
> Simple write timing:
> write 0.018404
> $#./test_fsync -f /tmp/test.out
> Simple write timing:
> write 0.018511
> ---

Thanks. Updated error message and added mention of memory file systems
to README.

--
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

In response to

Browse pgsql-committers by date

  From Date Subject
Next Message Bruce Momjian 2005-11-16 03:56:09 pgsql: Properly document return value of strpos().
Previous Message Bruce Momjian 2005-11-16 03:44:51 pgsql: Update error message and documentation for fsync test.