Generic Linux architectural discussions
 help / color / mirror / Atom feed
From: Mateusz Guzik <mjguzik@gmail.com>
To: Linus Torvalds <torvalds@linux-foundation.org>
Cc: linux-kernel@vger.kernel.org, linux-arch@vger.kernel.org, bp@alien8.de
Subject: Re: [PATCH v2] x86: bring back rep movsq for user access on CPUs without ERMS
Date: Sun, 3 Sep 2023 22:48:58 +0200	[thread overview]
Message-ID: <20230903204858.lv7i3kqvw6eamhgz@f> (raw)
In-Reply-To: <CAHk-=wg6bzTdQHSsswHPYFUbb1DfszyWTZ97hZv7bYxaNHVkHw@mail.gmail.com>

On Sun, Sep 03, 2023 at 01:08:03PM -0700, Linus Torvalds wrote:
> On Sun, 3 Sept 2023 at 11:49, Linus Torvalds
> <torvalds@linux-foundation.org> wrote:
> >
> > So I have no idea why you claim that "currently they have no choice".
> > glibc is simply being incredibly stupid, and using newfstatat() for no
> > good reason.
> 
> Do you have any good benchmark that shows the effects of this?
> 
> And if you do, does the attached patch (ENTIRELY UNTESTED!) fix the
> silly glibc mis-feature?
> 

"real fstat" is syscall(5, fd, &sb).

Sapphire Rapids, will-it-scale, ops/s

stock fstat	5088199
patched fstat	7625244	(+49%)
real fstat	8540383	(+67% / +12%)

It dodges lockref et al, but it does not dodge SMAP which accounts for
the difference.

  reply	other threads:[~2023-09-03 20:49 UTC|newest]

Thread overview: 35+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-08-30 14:03 [PATCH v2] x86: bring back rep movsq for user access on CPUs without ERMS Mateusz Guzik
2023-08-30 16:50 ` Linus Torvalds
2023-08-30 20:00 ` Linus Torvalds
2023-09-01 15:20   ` Mateusz Guzik
2023-09-01 15:29     ` Linus Torvalds
2023-09-03 18:49     ` Linus Torvalds
2023-09-03 19:14       ` Linus Torvalds
2023-09-03 20:08       ` Linus Torvalds
2023-09-03 20:48         ` Mateusz Guzik [this message]
2023-09-03 20:57           ` Linus Torvalds
2023-09-03 21:06             ` Mateusz Guzik
2023-09-03 21:08               ` Linus Torvalds
2023-09-03 21:18                 ` Mateusz Guzik
2023-09-03 23:28                   ` Al Viro
2023-09-03 20:58           ` Mateusz Guzik
2023-09-03 21:05           ` Linus Torvalds
2023-09-03 21:48             ` Ingo Molnar
2023-09-03 22:34               ` Linus Torvalds
2023-09-03 23:15                 ` Mateusz Guzik
2023-09-04  3:07                   ` Linus Torvalds
2023-09-04  3:17                     ` Linus Torvalds
2023-09-04  6:03                       ` Mateusz Guzik
2023-09-04 17:28                         ` Linus Torvalds
2023-09-05 20:41                           ` Mateusz Guzik
2023-09-06  0:16                             ` Linus Torvalds
2023-09-06  4:11                               ` Mateusz Guzik
2023-09-01 13:33 ` David Laight
2023-09-01 15:28   ` Mateusz Guzik
2023-09-03 20:42     ` David Laight
2023-09-10 10:53       ` Mateusz Guzik
2023-09-11 10:37         ` David Laight
2023-09-12 18:48           ` Linus Torvalds
2023-09-12 19:41             ` David Laight
2023-09-12 20:48               ` Linus Torvalds
2023-09-13  8:25                 ` David Laight

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20230903204858.lv7i3kqvw6eamhgz@f \
    --to=mjguzik@gmail.com \
    --cc=bp@alien8.de \
    --cc=linux-arch@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=torvalds@linux-foundation.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox