All of lore.kernel.org
 help / color / mirror / Atom feed
From: Lorenzo Colitti <lorenzo@google.com>
To: Greg KH <gregkh@linuxfoundation.org>
Cc: "Maciej Żenczykowski" <zenczykowski@gmail.com>,
	"Ingo Molnar" <mingo@kernel.org>,
	"Anna-Maria Behnsen" <anna-maria@linutronix.de>,
	lkml <linux-kernel@vger.kernel.org>,
	mikael.beckius@windriver.com,
	"Thomas Gleixner" <tglx@linutronix.de>,
	"Maciej Żenczykowski" <maze@google.com>
Subject: Re: [PATCH] hrtimer: Update softirq_expires_next correctly after __hrtimer_get_next_event()
Date: Wed, 14 Apr 2021 11:49:01 +0900	[thread overview]
Message-ID: <CAKD1Yr1DnDTELUX2DQtPDtAoDMqCz6dV+TZbBuC1CFm32O8MrA@mail.gmail.com> (raw)
In-Reply-To: <YHXRWoVIYLL4rYG9@kroah.com>

On Wed, Apr 14, 2021 at 2:14 AM Greg KH <gregkh@linuxfoundation.org> wrote:
> To give context, the commit is now 46eb1701c046 ("hrtimer: Update
> softirq_expires_next correctly after __hrtimer_get_next_event()") and is
> attached below.
>
> The f_ncm.c driver is doing a lot of calls to hrtimer_start() with mode
> HRTIMER_MODE_REL_SOFT for I think every packet it gets.  If that should
> not be happening, we can easily fix it but I guess no one has actually
> had fast USB devices that noticed this until now :)

AIUI the purpose of this timer is to support packet aggregation. USB
transfers are relatively expensive/high latency. 6 Gbps is 500k
1500-byte packets per second, or one every 2us. So f_ncm buffers as
many packets as will fit into 16k (usually, 10 1500-byte packets), and
only initiates a USB transfer when those packets have arrived. That
ends up doing only one transfer every 20us. It sets a 300us timer to
ensure that if the 10 packets haven't arrived, it still sends out
whatever it has when the timer fires. The timer is set/updated on
every packet buffered by ncm.

Is this somehow much more expensive in 5.10.24 than it was before?
Even if this driver is somehow "holding it wrong", might there not be
other workloads that have a similar problem? What about regressions on
those workloads?

  reply	other threads:[~2021-04-14  2:49 UTC|newest]

Thread overview: 23+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-04-13 16:55 [PATCH] hrtimer: Update softirq_expires_next correctly after __hrtimer_get_next_event() Maciej Żenczykowski
2021-04-13 17:14 ` Greg KH
2021-04-14  2:49   ` Lorenzo Colitti [this message]
2021-04-15 16:47     ` Thomas Gleixner
2021-04-20  3:12       ` Maciej Żenczykowski
2021-04-20  6:44         ` Thomas Gleixner
2021-04-20  8:15       ` Lorenzo Colitti
2021-04-20 14:19         ` Thomas Gleixner
2021-04-21 14:08           ` Lorenzo Colitti
2021-04-21 14:40             ` Lorenzo Colitti
2021-04-21 15:22               ` Greg KH
2021-04-22  0:08             ` Thomas Gleixner
2021-04-22 10:07               ` Thomas Gleixner
2021-04-22 14:20               ` Lorenzo Colitti
2021-04-22 15:35                 ` Thomas Gleixner
2021-04-26  8:49           ` [PATCH] hrtimer: Avoid double reprogramming in __hrtimer_start_range_ns() Thomas Gleixner
2021-04-26  9:40             ` Peter Zijlstra
2021-04-26 12:25               ` Peter Zijlstra
2021-05-14 19:29                 ` Thomas Gleixner
2021-04-26 12:33               ` Thomas Gleixner
2021-04-26 12:40                 ` Peter Zijlstra
2021-04-26 14:27                   ` Thomas Gleixner
  -- strict thread matches above, loose matches on Subject: below --
2021-02-12 13:38 Sv: [PATCH] hrtimer: Interrupt storm on clock_settime Beckius, Mikael
2021-02-23 16:02 ` [PATCH] hrtimer: Update softirq_expires_next correctly after __hrtimer_get_next_event() Anna-Maria Behnsen

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=CAKD1Yr1DnDTELUX2DQtPDtAoDMqCz6dV+TZbBuC1CFm32O8MrA@mail.gmail.com \
    --to=lorenzo@google.com \
    --cc=anna-maria@linutronix.de \
    --cc=gregkh@linuxfoundation.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=maze@google.com \
    --cc=mikael.beckius@windriver.com \
    --cc=mingo@kernel.org \
    --cc=tglx@linutronix.de \
    --cc=zenczykowski@gmail.com \
    /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.