Re: [PATCH 3/3] Document geqo_seed variable.

Hi,

On Wednesday 15 July 2009 04:14:16 Robert Haas wrote:
> On Tue, Jul 14, 2009 at 6:34 PM, Andres Freund<andres(at)anarazel(dot)de> wrote:
> > ---
> > doc/src/sgml/config.sgml | 16 ++++++++++++++++
> > 1 files changed, 16 insertions(+), 0 deletions(-)
> >
> > diff --git a/doc/src/sgml/config.sgml b/doc/src/sgml/config.sgml
> > index 99d25d7..5d8eca9 100644
> > *** a/doc/src/sgml/config.sgml
> > --- b/doc/src/sgml/config.sgml
> > *************** archive_command = 'copy "%p" "C:\\server
> > *** 2150,2155 ****
> > --- 2150,2171 ----
> > </listitem>
> > </varlistentry>
> >
> > + <varlistentry id="guc-geqo-seed" xreflabel="geqo_seed">
> > + <term><varname>geqo_seed</varname>
> > + (<type>floating point</type>)</term>
> > + <indexterm>
> > + <primary><varname>geqo_seed</> configuration parameter</primary>
> > + </indexterm>
> > + <listitem>
> > + <para>
> > + Controls the initialization of the random number generator used
> > + by GEQO to select random paths through the join order search
> > space. + With the default setting of zero the join order planning
> > is not repeatable. + For repeatable planning set a value between
> > 0 (exclusive) and 1 (inclusive). + </para>
> > + </listitem>
> > + </varlistentry>
> > +
> > </variablelist>
> > </sect2>
> > <sect2 id="runtime-config-query-other">
> > --
> > 1.6.3.3.335.ge09a8
>
> I don't understand why people (including yourself, but you're not the
> only one) have begun submitting relatively trivial patches in multiple
> parts. This just creates multiple threads on the mailing list without
> adding any value. The doc changes are part of the patch; one email
> containing all the changes seems vastly preferable to me.
For one reason it got easy. For another doing so - except the doc patch
admittedly - makes it potentially easier to bisect changes (Spent to much time
bisecting kernel problems I guess).
I also find it easier to read a patch when I know what kind of issues to
expect.
For example the patch adding erand48 should definitely be looked at on by
somebody with windows whereas the repetitious changes to GEQO itself do not
necessarily be tested on windows.

It should not create multiple threads I think as all messages were a response
the first mail?

> IMHO, the only reason for submitting multiple patches if it there are
> pieces that are separately commitable.
PG has rather coarse grained commits. I guess some people are not used to that
from their own experience anymore....

Andres