From: "Paul E. McKenney" <paulmck-23VcF4HTsmIX0ybBhKVfKdBPR1lH4CV8@public.gmane.org>
To: GeHao Kang <kanghao0928-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
Cc: "Peter Zijlstra" <peterz-wEGCiKHe2LqWVfeAwA7xHQ@public.gmane.org>,
"Chris Metcalf"
<cmetcalf-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>,
"Frédéric Weisbecker"
<fweisbec-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>,
linux-api-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
tglx-hfZtesqFncYOwBW4kG4KsQ@public.gmane.org,
mingo-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org
Subject: Re: Context switch latency in tickless isolated CPU
Date: Sun, 21 Aug 2016 07:53:02 -0700 [thread overview]
Message-ID: <20160821145302.GP3482@linux.vnet.ibm.com> (raw)
In-Reply-To: <CANv7uPRh4RiSXwpRsOooNAJumMj2TmiRc3Qq5_kwzf9SetE_FA-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
On Sun, Aug 21, 2016 at 07:26:04PM +0800, GeHao Kang wrote:
> On Fri, Aug 19, 2016 at 8:34 PM, Peter Zijlstra <peterz-wEGCiKHe2LqWVfeAwA7xHQ@public.gmane.org> wrote:
>
> > Why are you wanting to use nohz_full if you do syscalls?
>
> We hope to reduce the overhead of the tick while the real time
> applications run,
> and these applications might do some syscalls to operate the I/O devices like
> EtherCAT.
If latency is all you care about, one approach is to map the device
registers into userspace and do the I/O without assistance from the
kernel.
Alternatively, use in-memory mailbox/queuing techniques to hand the
I/O off to some other thread.
Thanx, Paul
WARNING: multiple messages have this Message-ID (diff)
From: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
To: GeHao Kang <kanghao0928@gmail.com>
Cc: "Peter Zijlstra" <peterz@infradead.org>,
"Chris Metcalf" <cmetcalf@mellanox.com>,
"Frédéric Weisbecker" <fweisbec@gmail.com>,
linux-api@vger.kernel.org, linux-kernel@vger.kernel.org,
tglx@linutronix.de, mingo@kernel.org
Subject: Re: Context switch latency in tickless isolated CPU
Date: Sun, 21 Aug 2016 07:53:02 -0700 [thread overview]
Message-ID: <20160821145302.GP3482@linux.vnet.ibm.com> (raw)
In-Reply-To: <CANv7uPRh4RiSXwpRsOooNAJumMj2TmiRc3Qq5_kwzf9SetE_FA@mail.gmail.com>
On Sun, Aug 21, 2016 at 07:26:04PM +0800, GeHao Kang wrote:
> On Fri, Aug 19, 2016 at 8:34 PM, Peter Zijlstra <peterz@infradead.org> wrote:
>
> > Why are you wanting to use nohz_full if you do syscalls?
>
> We hope to reduce the overhead of the tick while the real time
> applications run,
> and these applications might do some syscalls to operate the I/O devices like
> EtherCAT.
If latency is all you care about, one approach is to map the device
registers into userspace and do the I/O without assistance from the
kernel.
Alternatively, use in-memory mailbox/queuing techniques to hand the
I/O off to some other thread.
Thanx, Paul
next prev parent reply other threads:[~2016-08-21 14:53 UTC|newest]
Thread overview: 17+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-08-17 6:26 Context switch latency in tickless isolated CPU GeHao Kang
[not found] ` <CANv7uPR+RidX41Jyzy-xpYqYHSPiw3TrBgAm11PzOfNEeP7wJA-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2016-08-17 12:18 ` Chris Metcalf
2016-08-17 12:18 ` Chris Metcalf
2016-08-18 3:25 ` GeHao Kang
[not found] ` <CANv7uPSwu=nxYYjWQFABcyj6ZpAemHyXbn4JnZ6WaVJQd3RX4A-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2016-08-19 12:34 ` Peter Zijlstra
2016-08-19 12:34 ` Peter Zijlstra
2016-08-21 11:26 ` GeHao Kang
[not found] ` <CANv7uPRh4RiSXwpRsOooNAJumMj2TmiRc3Qq5_kwzf9SetE_FA-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2016-08-21 14:53 ` Paul E. McKenney [this message]
2016-08-21 14:53 ` Paul E. McKenney
[not found] ` <20160821145302.GP3482-23VcF4HTsmIX0ybBhKVfKdBPR1lH4CV8@public.gmane.org>
2016-08-22 9:40 ` GeHao Kang
2016-08-22 9:40 ` GeHao Kang
[not found] ` <CANv7uPQ8rSU8foZv1GiLStiNX0Z6e+7tnRLJ_v2fspY5NZ8nwg-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2016-08-22 14:48 ` Paul E. McKenney
2016-08-22 14:48 ` Paul E. McKenney
2016-08-22 15:12 ` Mark Hounschell
2016-08-22 15:37 ` Paul E. McKenney
[not found] ` <20160822153714.GC3482-23VcF4HTsmIX0ybBhKVfKdBPR1lH4CV8@public.gmane.org>
2016-08-22 16:35 ` Mark Hounschell
2016-08-22 16:35 ` Mark Hounschell
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=20160821145302.GP3482@linux.vnet.ibm.com \
--to=paulmck-23vcf4htsmix0ybbhkvfkdbpr1lh4cv8@public.gmane.org \
--cc=cmetcalf-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org \
--cc=fweisbec-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org \
--cc=kanghao0928-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org \
--cc=linux-api-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
--cc=linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
--cc=mingo-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org \
--cc=peterz-wEGCiKHe2LqWVfeAwA7xHQ@public.gmane.org \
--cc=tglx-hfZtesqFncYOwBW4kG4KsQ@public.gmane.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.