Re: PG14: Avoid checking output-buffer-length for every encoded byte during pg_hex_encode

From: Michael Paquier <michael(at)paquier(dot)xyz>
To: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
Cc: Hans Buschmann <buschmann(at)nidsa(dot)net>, "pgsql-hackers(at)postgresql(dot)org" <pgsql-hackers(at)postgresql(dot)org>
Subject: Re: PG14: Avoid checking output-buffer-length for every encoded byte during pg_hex_encode
Date: 2021-08-18 00:14:14
Message-ID: YRxQ1sN5ceWPgwdM@paquier.xyz
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

On Tue, Aug 17, 2021 at 09:39:30AM -0400, Tom Lane wrote:
> OK, but the commit message should explain why they're getting reverted.

Sure. aef8948 gets down because of the performance impact. ccf4e27
was a cleanup following up aef8948, that loses its meaning. And
c3826f8 cannot be let alone because of the reasons why aef8948 was
introduced, as it has no safety net for out-of-bound handling in the
result buffer allocated.
--
Michael

In response to

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Peter Geoghegan 2021-08-18 00:31:08 Re: The Free Space Map: Problems and Opportunities
Previous Message Michael Paquier 2021-08-18 00:10:16 Re: PG14: Avoid checking output-buffer-length for every encoded byte during pg_hex_encode