Horton

Re: [HACKERS] Should logtape.c blocks be of type long?

Heikki Linnakangas <hlinnaka@iki.fi>

View thread View on postgr.es
From: Heikki Linnakangas <hlinnaka@iki.fi>
To: Michael Paquier <michael@paquier.xyz>, Peter Geoghegan <pg@bowt.ie>
Cc: Tom Lane <tgl@sss.pgh.pa.us>, PostgreSQL Hackers <pgsql-hackers@postgresql.org>
Date: 2023-11-16T14:03:46Z
Lists: pgsql-hackers 
On 26/09/2023 07:15, Michael Paquier wrote:
> On Sun, Sep 24, 2023 at 10:42:49AM +0900, Michael Paquier wrote:
>> Indeed, or Windows decides that making long 8-byte is wiser, but I
>> doubt that's ever going to happen on backward-compatibility ground.
> 
> While looking more at that, I've noticed that I missed BufFileAppend()
> and BufFileSeekBlock(), that themselves rely on long.  The other code
> paths calling these two routines rely on BlockNumber (aka uint32), so
> that seems to be the bottom of it.

BufFileTellBlock should be adjusted too. Or removed altogether; it's 
been commented out since year 2000. Other than that, looks good to me.

-- 
Heikki Linnakangas
Neon (https://neon.tech)

Commits

  1. Change logtape/tuplestore code to use int64 for block numbers

  2. Remove NOT_USED BufFileTellBlock() from buffile.c