Re: autovectorize page checksum code included elsewhere

From: Andres Freund <andres(at)anarazel(dot)de>
To: John Naylor <johncnaylorls(at)gmail(dot)com>
Cc: Nathan Bossart <nathandbossart(at)gmail(dot)com>, Ants Aasma <ants(at)cybertec(dot)at>, pgsql-hackers(at)postgresql(dot)org
Subject: Re: autovectorize page checksum code included elsewhere
Date: 2023-11-28 00:51:25
Message-ID: 20231128005125.hvq4jxz7gu6u5anc@awork3.anarazel.de
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

Hi,

On 2023-11-25 14:09:14 +0700, John Naylor wrote:
> * Note: I have seen the threads with the idea of compiling multiple
> entire binaries, and switching at postmaster start. I think it's a
> good idea, but I also suspect detecting flags from the packager is an
> easier intermediate step.

It's certainly an easier incremental step - but will it get us all that far?
Other architectures have similar issues, e.g. ARM with ARMv8.1-A having much
more scalable atomic instructions than plain ARMv8.0. And even on x86-64,
relying on distros to roll out new minimum x86_64-v2 doesn't get us that far -
we'd be up to a uarch from ~2010. And I doubt that they'll raise the bar to v4
in the next few years, so we'll leave years of improvements on the table. And
even PG specific repos can't just roll out a hard requirement for a new, more
modern, uarch that will crash some time after startup on some hardware.

Greetings,

Andres Freund

In response to

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Peter Smith 2023-11-28 00:54:33 Re: GUC names in messages
Previous Message Tristan Partin 2023-11-28 00:48:13 Re: SSL tests fail on OpenSSL v3.2.0