From: Thomas Gleixner <tglx@linutronix.de>
To: "Jason A. Donenfeld" <Jason@zx2c4.com>
Cc: LKML <linux-kernel@vger.kernel.org>, X86 ML <x86@kernel.org>,
Andy Lutomirski <luto@amacapital.net>,
Peter Zijlstra <peterz@infradead.org>
Subject: Re: Lazy FPU restoration / moving kernel_fpu_end() to context switch
Date: Fri, 15 Jun 2018 18:25:39 +0200 (CEST) [thread overview]
Message-ID: <alpine.DEB.2.21.1806151815080.1582@nanos.tec.linutronix.de> (raw)
In-Reply-To: <CAHmME9pBqGhCjdwx64GxYTKWiMkDNY3v2gnVL_Xm2q=3guOAsQ@mail.gmail.com>
On Fri, 15 Jun 2018, Jason A. Donenfeld wrote:
> In a loop this looks like:
>
> for (thing) {
> kernel_fpu_begin();
> encrypt(thing);
> kernel_fpu_end();
> }
>
> This is obviously very bad, because begin() and end() are slow, so
> WireGuard does the obvious:
>
> kernel_fpu_begin();
> for (thing)
> encrypt(thing);
> kernel_fpu_end();
>
> This is fine and well, and the crypto API I'm working on will enable
It might be fine crypto performance wise, but it's a total nightmare
latency wise because kernel_fpu_begin() disables preemption. We've seen
latencies in the larger millisecond range due to processing large data sets
with kernel FPU.
If you want to go there then we really need a better approach which allows
kernel FPU usage in preemptible context and in case of preemption a way to
stash the preempted FPU context and restore it when the task gets scheduled
in again. Just using the existing FPU stuff and moving the loops inside the
begin/end section and keeping preemption disabled for arbitrary time spans
is not going to fly.
Thanks,
tglx
next prev parent reply other threads:[~2018-06-15 16:25 UTC|newest]
Thread overview: 23+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-06-15 13:11 Lazy FPU restoration / moving kernel_fpu_end() to context switch Jason A. Donenfeld
2018-06-15 16:25 ` Thomas Gleixner [this message]
2018-06-15 18:33 ` Brian Gerst
2018-06-15 19:34 ` Peter Zijlstra
2018-06-15 20:30 ` Jason A. Donenfeld
2018-06-18 9:44 ` Peter Zijlstra
2018-06-18 15:25 ` Jason A. Donenfeld
2018-06-15 18:31 ` Andy Lutomirski
2018-06-15 18:40 ` Rik van Riel
2018-06-15 18:41 ` Dave Hansen
2018-06-15 18:49 ` Andy Lutomirski
2018-06-15 18:48 ` Dave Hansen
2018-06-15 18:53 ` Andy Lutomirski
2018-06-15 20:27 ` Jason A. Donenfeld
2018-06-15 20:48 ` Jason A. Donenfeld
2018-06-19 11:43 ` David Laight
2018-06-19 13:08 ` Thomas Gleixner
2018-06-15 20:33 ` Jason A. Donenfeld
2018-06-15 20:42 ` Dave Hansen
2018-06-15 20:52 ` Andy Lutomirski
2018-06-15 20:56 ` Rik van Riel
2018-07-11 16:28 ` Sebastian Andrzej Siewior
2018-07-11 20:10 ` Rik van Riel
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=alpine.DEB.2.21.1806151815080.1582@nanos.tec.linutronix.de \
--to=tglx@linutronix.de \
--cc=Jason@zx2c4.com \
--cc=linux-kernel@vger.kernel.org \
--cc=luto@amacapital.net \
--cc=peterz@infradead.org \
--cc=x86@kernel.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;
as well as URLs for NNTP newsgroup(s).