From: John Ogness <john.ogness@linutronix.de>
To: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Cc: Paul Thomas <pthomas8589@gmail.com>,
Thomas Gleixner <tglx@linutronix.de>,
linux-rt-users <linux-rt-users@vger.kernel.org>,
dunc.mcphie@gmail.com, Anders Roxell <anders.roxell@linaro.org>
Subject: Re: poor cyclictest results with 5.0 series rt for arm64
Date: Thu, 15 Aug 2019 14:33:41 +0200 [thread overview]
Message-ID: <87a7ca4nh6.fsf@linutronix.de> (raw)
In-Reply-To: <20190815082955.olo3phbqj6disk3q@linutronix.de> (Sebastian Andrzej Siewior's message of "Thu, 15 Aug 2019 10:29:55 +0200")
On 2019-08-15, Sebastian Andrzej Siewior <bigeasy@linutronix.de> wrote:
>> The attachment is here:
>> https://bugzilla.kernel.org/attachment.cgi?id=284391
>
> both traces have `printk' in it. Could you please explain what it is
> doing? It should print something to the console. The printk code in
> v4.19 vs later changed and now (by default) ERR levels are printed
> immediately and will cause latency spikes.
>
> This would explain the spike but I don't understand how lazy preempt
> is related here.
Where in this trace is there a latency spike? I could not see one.
Emergency messages are not involved here. You can see at 84239098us that
systemd is writing to /dev/kmsg, but this is completed (haven only
written to the ring buffer) by 84239118us (20us).
printk schedules an irq_work to wake the printk thread. The irq_work
begins at 84241028us. All it does is call wake_up_interruptible_all() on
the queue.
John Ogness
next prev parent reply other threads:[~2019-08-15 12:33 UTC|newest]
Thread overview: 28+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-07-12 11:18 poor cyclictest results with 5.0 series rt for arm64 Paul Thomas
[not found] ` <CAD56B7c2=95r7r-6tr0Mby8jSeuQPDNSd4LELnUw1LVRn9Rqew@mail.gmail.com>
2019-07-12 11:32 ` Paul Thomas
2019-07-16 20:24 ` Thomas Gleixner
2019-07-16 21:50 ` Paul Thomas
2019-07-16 23:14 ` Thomas Gleixner
2019-07-17 0:44 ` Paul Thomas
2019-07-17 8:03 ` Thomas Gleixner
2019-07-17 8:10 ` Thomas Gleixner
2019-07-17 9:02 ` Thomas Gleixner
2019-07-17 10:21 ` Thomas Gleixner
2019-07-17 22:28 ` Paul Thomas
2019-08-13 9:53 ` Sebastian Andrzej Siewior
2019-08-13 18:02 ` Paul Thomas
2019-08-14 11:26 ` Sebastian Andrzej Siewior
2019-08-14 16:17 ` [PATCH RT] arm64: preempt: Check preemption level before looking at lazy-preempt Sebastian Andrzej Siewior
2019-08-14 16:17 ` [PATCH RT] x86: " Sebastian Andrzej Siewior
2019-08-14 17:04 ` poor cyclictest results with 5.0 series rt for arm64 Paul Thomas
2019-08-15 8:29 ` Sebastian Andrzej Siewior
2019-08-15 12:33 ` John Ogness [this message]
2019-08-15 13:30 ` Paul Thomas
2019-08-15 13:42 ` Sebastian Andrzej Siewior
2019-08-15 13:52 ` Paul Thomas
2019-08-15 14:19 ` Sebastian Andrzej Siewior
2019-08-15 15:12 ` Paul Thomas
2019-08-15 15:15 ` Sebastian Andrzej Siewior
2019-08-15 15:31 ` Paul Thomas
2019-08-15 18:00 ` Paul Thomas
2019-08-15 18:57 ` Sebastian Andrzej Siewior
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=87a7ca4nh6.fsf@linutronix.de \
--to=john.ogness@linutronix.de \
--cc=anders.roxell@linaro.org \
--cc=bigeasy@linutronix.de \
--cc=dunc.mcphie@gmail.com \
--cc=linux-rt-users@vger.kernel.org \
--cc=pthomas8589@gmail.com \
--cc=tglx@linutronix.de \
/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.