Re: More speedups for tuple deformation

From: John Naylor <johncnaylorls(at)gmail(dot)com>
To: Andres Freund <andres(at)anarazel(dot)de>
Cc: David Rowley <dgrowleyml(at)gmail(dot)com>, Chao Li <li(dot)evan(dot)chao(at)gmail(dot)com>, PostgreSQL Developers <pgsql-hackers(at)lists(dot)postgresql(dot)org>
Subject: Re: More speedups for tuple deformation
Date: 2026-02-03 05:24:39
Message-ID: CANWCAZabO1oj+khF+YNVpmkTQwRRyNJesbsBhRFL5emZJh3tow@mail.gmail.com
Views: Whole Thread | Raw Message | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

On Tue, Feb 3, 2026 at 7:33 AM Andres Freund <andres(at)anarazel(dot)de> wrote:
> It's somewhat scary to see this level of code size increase in a case where
> the compiler really has no information to think vectorizing really is
> beneficial...

I tried building on gcc 15.2 with -fno-tree-loop-vectorize, and the
server .text segment was only 5kB smaller, but that may understate the
impact on the decisions that get made. Maybe it's better to opt in for
unrolling and vectorization? I think this option also controls whether
to turn loops into libc memset/memmove etc calls.

--
John Naylor
Amazon Web Services

In response to

Browse pgsql-hackers by date

  From Date Subject
Next Message John Naylor 2026-02-03 05:39:48 Re: [PATCH] ANALYZE: hash-accelerate MCV tracking for equality-only types
Previous Message Ashutosh Bapat 2026-02-03 05:18:30 Remove freelist reference in buf_init.c