Re: dropping datumSort field

From: Robert Haas <robertmhaas(at)gmail(dot)com>
To: Zhihong Yu <zyu(at)yugabyte(dot)com>
Cc: PostgreSQL Developers <pgsql-hackers(at)lists(dot)postgresql(dot)org>
Subject: Re: dropping datumSort field
Date: 2022-08-09 17:43:43
Message-ID: CA+TgmoacE2pqntkads=cgwxLmxH+RsCK5JxYa_XgMroatCWi_g@mail.gmail.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

On Tue, Aug 9, 2022 at 11:42 AM Zhihong Yu <zyu(at)yugabyte(dot)com> wrote:
> Though the datumSort field may be harmless for now, in the future, the (auto) alignment padding may not work if more fields are added to the struct.
> I think we should always leave some room in struct's for future expansion.

I doubt the size of this struct is particularly important, unlike
ExprEvalStep which needs to be small. But if it turns out in the
future that we need to try to squeeze this struct into fewer bytes, we
can always do something like this then. Right now there's no obvious
point to it.

Sure, it might be valuable *if* we add more fields to the struct and
*if* that means that the byte taken up by this flag actually makes the
struct bigger and *if* the size of the struct is demonstrated to be a
performance problem. But right now none of that has happened, and
maybe none of it will ever happen.

--
Robert Haas
EDB: http://www.enterprisedb.com

In response to

Browse pgsql-hackers by date

  From Date Subject
Next Message David Steele 2022-08-09 17:49:34 Re: moving basebackup code to its own directory
Previous Message Robert Haas 2022-08-09 17:32:49 Re: moving basebackup code to its own directory