Re: WIP: extensible enums - Mailing list pgsql-hackers

From Robert Haas
Subject Re: WIP: extensible enums
Date
Msg-id AANLkTinxuFwcGBcPk9CUctTRQ2v4tFHjxgYeRD3vLw5E@mail.gmail.com
Whole thread Raw
In response to Re: WIP: extensible enums  (Merlin Moncure <mmoncure@gmail.com>)
List pgsql-hackers
On Wed, Oct 20, 2010 at 6:54 PM, Merlin Moncure <mmoncure@gmail.com> wrote:
> On Tue, Oct 19, 2010 at 9:15 PM, Andrew Dunstan <andrew@dunslane.net> wrote:
>> Efficiency has  always been one of the major reasons for using enums, so
>> it's important that we make them extensible without badly affecting
>> performance.
>
> on that note is it worthwhile backpatching recent versions to allocate
> enums with even numbered oids? That way people binary upgrading can
> get the benefit of the optimization they should qualify for...

Uh, -1 from me.  This is not a bug fix, and it will only help people
who create new enums between the time they upgrade to the relevant
minor release and the time they upgrade to 9.1.  We are not into the
business of back-patching marginal peformance enhancements.  If we
want to have a 9.0R2 release, or whatever, then so be it, but let's
not be modifying behavior in stable branches unless there's a *bug*.

--
Robert Haas
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company


pgsql-hackers by date:

Previous
From: Robert Haas
Date:
Subject: Re: default_statistics_target WAS: max_wal_senders must die
Next
From: Greg Stark
Date:
Subject: Re: default_statistics_target WAS: max_wal_senders must die