Re: ECPG FETCH readahead

From: Boszormenyi Zoltan <zb(at)cybertec(dot)at>
To: Noah Misch <noah(at)leadboat(dot)com>, Michael Meskes <meskes(at)postgresql(dot)org>, Robert Haas <robertmhaas(at)gmail(dot)com>, PG Hackers <pgsql-hackers(at)postgresql(dot)org>, Heikki Linnakangas <heikki(dot)linnakangas(at)enterprisedb(dot)com>, Bruce Momjian <bruce(at)momjian(dot)us>
Subject: Re: ECPG FETCH readahead
Date: 2012-04-10 07:35:21
Message-ID: 4F83E2B9.3030302@cybertec.at
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

2012-04-08 19:38 keltezéssel, Michael Meskes írta:
> On Sun, Apr 08, 2012 at 06:35:33PM +0200, Boszormenyi Zoltan wrote:
>> Do you want me to change this or will you do it? I am on holiday
>> and will be back to work on wednesday.
> I don't think waiting till later this week is a real problem.

OK.

>
>> The possibility to test different readahead window sizes
>> without modifying the source and recompiling was useful.
> Sure, but you can still do that when not defining a fixed number in the
> statement.

OK.

>
>> The -R option simply provides a default without ornamenting
>> the DECLARE statement.
> Could you please incorporate these changes, too, when you're back from vacation?

Sure.

So, it's established that a specified READAHEAD N should not
be overridden. Even an explicit READAHEAD 1.

Only a non-decorated cursor can be overridden, even if
a different default readahead window size is specified with
e.g. "ecpg -R 8". If ECPGFETCHSZ is not present, 8 will be used,
if ECPGFETCHSZ is present, its value will be used. ECPGopen()
will need an extra bool argument to distinguish this.

Is this acceptable? Noah, Michael?

>
>>> I cannot find a test that tests the environment variable giving the fetch size.
>>> Could you please point me to that?
>> I didn't write such a test. The reason is that while variables are
>> exported by make from the Makefile to the binaries run by make
>> e.g. CFLAGS et.al. for $(CC), "make check" simply runs pg_regress
>> once which uses its own configuration file that doesn't have a
>> way to set or unset an environment variable. This could be a useful
>> extension to pg_regress though.
> How about calling setenv() from the test program itself?

Sure, I didn't think about it. It should be done before the
first EXEC SQL OPEN cursor.

Best regards,
Zoltán Böszörményi

--
----------------------------------
Zoltán Böszörményi
Cybertec Schönig& Schönig GmbH
Gröhrmühlgasse 26
A-2700 Wiener Neustadt, Austria
Web: http://www.postgresql-support.de
http://www.postgresql.at/

In response to

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message 乔志强 2012-04-10 08:09:17 答复: [GENERAL] [streaming replication] 9.1.3 streaming replication bug ?
Previous Message Boszormenyi Zoltan 2012-04-10 07:02:49 Re: [PATCH] lock_timeout and common SIGALRM framework