From: Dave Taht <dave.taht@gmail.com>
To: Vincent Prince <vincent.prince.fr@gmail.com>
Cc: "Marc Kleine-Budde" <mkl@pengutronix.de>,
"David S. Miller" <davem@davemloft.net>,
"Jamal Hadi Salim" <jhs@mojatatu.com>,
"Jiří Pírko" <jiri@resnulli.us>,
kernel@pengutronix.de, linux-can@vger.kernel.org,
"Linux Kernel Network Developers" <netdev@vger.kernel.org>,
"Cong Wang" <xiyou.wangcong@gmail.com>
Subject: Re: [PATCH v4] net: sch_generic: Use pfifo_fast as fallback scheduler for CAN hardware
Date: Wed, 23 Oct 2019 04:13:22 -0700 [thread overview]
Message-ID: <CAA93jw6koAuvuAXYBghJYcxwTz-uhvxPohFj9kBrpBDcQmmxMg@mail.gmail.com> (raw)
In-Reply-To: <1571827920-9024-1-git-send-email-vincent.prince.fr@gmail.com>
On Wed, Oct 23, 2019 at 3:52 AM Vincent Prince
<vincent.prince.fr@gmail.com> wrote:
>
> There is networking hardware that isn't based on Ethernet for layers 1 and 2.
>
> For example CAN.
>
> CAN is a multi-master serial bus standard for connecting Electronic Control
> Units [ECUs] also known as nodes. A frame on the CAN bus carries up to 8 bytes
> of payload. Frame corruption is detected by a CRC. However frame loss due to
> corruption is possible, but a quite unusual phenomenon.
>
> While fq_codel works great for TCP/IP, it doesn't for CAN. There are a lot of
> legacy protocols on top of CAN, which are not build with flow control or high
> CAN frame drop rates in mind.
>
> When using fq_codel, as soon as the queue reaches a certain delay based length,
> skbs from the head of the queue are silently dropped. Silently meaning that the
> user space using a send() or similar syscall doesn't get an error. However
> TCP's flow control algorithm will detect dropped packages and adjust the
> bandwidth accordingly.
>
> When using fq_codel and sending raw frames over CAN, which is the common use
> case, the user space thinks the package has been sent without problems, because
> send() returned without an error. pfifo_fast will drop skbs, if the queue
> length exceeds the maximum. But with this scheduler the skbs at the tail are
> dropped, an error (-ENOBUFS) is propagated to user space. So that the user
> space can slow down the package generation.
>
> On distributions, where fq_codel is made default via CONFIG_DEFAULT_NET_SCH
> during compile time, or set default during runtime with sysctl
> net.core.default_qdisc (see [1]), we get a bad user experience. In my test case
> with pfifo_fast, I can transfer thousands of million CAN frames without a frame
> drop. On the other hand with fq_codel there is more then one lost CAN frame per
> thousand frames.
>
> As pointed out fq_codel is not suited for CAN hardware, so this patch changes
> attach_one_default_qdisc() to use pfifo_fast for "ARPHRD_CAN" network devices.
>
> During transition of a netdev from down to up state the default queuing
> discipline is attached by attach_default_qdiscs() with the help of
> attach_one_default_qdisc(). This patch modifies attach_one_default_qdisc() to
> attach the pfifo_fast (pfifo_fast_ops) if the network device type is
> "ARPHRD_CAN".
>
> [1] https://github.com/systemd/systemd/issues/9194
>
> Suggested-by: Marc Kleine-Budde <mkl@pengutronix.de>
> Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
> Signed-off-by: Vincent Prince <vincent.prince.fr@gmail.com>
> ---
> Changes in v4:
> - add Marc credit to commit log
>
> Changes in v3:
> - add description
>
> Changes in v2:
> - reformat patch
>
> net/sched/sch_generic.c | 2 ++
> 1 file changed, 2 insertions(+)
>
> diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c
> index 77b289d..dfb2982 100644
> --- a/net/sched/sch_generic.c
> +++ b/net/sched/sch_generic.c
> @@ -1008,6 +1008,8 @@ static void attach_one_default_qdisc(struct net_device *dev,
>
> if (dev->priv_flags & IFF_NO_QUEUE)
> ops = &noqueue_qdisc_ops;
> + else if(dev->type == ARPHRD_CAN)
> + ops = &pfifo_fast_ops;
>
> qdisc = qdisc_create_dflt(dev_queue, ops, TC_H_ROOT, NULL);
> if (!qdisc) {
> --
> 2.7.4
>
While I'm delighted to see such a simple patch emerge, openwrt long
ago patched out pfifo_fast. pfifo_fast has
additional semantics not needed in the can use case either (I think)
and "pfifo" is fine, but sure, pfifo_fast if you must.
anyway, regardless, that's an easy fix and I hope this fix goes to
stable, as I've had nightmares about cars exploding due to out of
order can bus operations ever since I learned of this bug.
Acked-by: Dave Taht <dave.taht@gmail.com>
--
Dave Täht
CTO, TekLibre, LLC
http://www.teklibre.com
Tel: 1-831-205-9740
next prev parent reply other threads:[~2019-10-23 11:13 UTC|newest]
Thread overview: 25+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-03-27 16:56 [RFC] net: sch_generic: fq_codel vs pfifo_fast Marc Kleine-Budde
2019-03-27 16:56 ` [PATCH 1/2] net: sch_generic: add flag IFF_FIFO_QUEUE to use pfifo_fast as default scheduler Marc Kleine-Budde
2019-03-27 17:14 ` Cong Wang
2019-03-27 20:11 ` Marc Kleine-Budde
2019-03-27 20:53 ` Cong Wang
2019-04-02 17:22 ` Toke Høiland-Jørgensen
2019-03-27 18:53 ` Uwe Kleine-König
2019-03-27 19:27 ` Marc Kleine-Budde
2019-03-27 16:56 ` [PATCH 2/2] can: dev: let all CAN devices " Marc Kleine-Budde
2019-03-27 18:30 ` [RFC] net: sch_generic: fq_codel vs pfifo_fast Stephen Hemminger
2019-03-27 19:24 ` Marc Kleine-Budde
2019-10-22 12:47 ` [PATCH] net: sch_generic: Use pfifo_fast as fallback scheduler for CAN hardware Vincent Prince
2019-10-22 12:58 ` Marc Kleine-Budde
2019-10-22 13:23 ` [PATCH v2] " Vincent Prince
2019-10-22 14:53 ` Marc Kleine-Budde
2019-10-22 14:55 ` Marc Kleine-Budde
2019-10-22 16:42 ` Stephen Hemminger
2019-10-22 16:48 ` Marc Kleine-Budde
2019-10-22 17:28 ` Eric Dumazet
2019-10-22 18:21 ` Oliver Hartkopp
2019-10-22 15:09 ` [PATCH v3] " Vincent Prince
2019-10-23 10:52 ` [PATCH v4] " Vincent Prince
2019-10-23 11:13 ` Dave Taht [this message]
2019-10-23 13:44 ` [PATCH v5] " Vincent Prince
2019-10-26 2:20 ` David Miller
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=CAA93jw6koAuvuAXYBghJYcxwTz-uhvxPohFj9kBrpBDcQmmxMg@mail.gmail.com \
--to=dave.taht@gmail.com \
--cc=davem@davemloft.net \
--cc=jhs@mojatatu.com \
--cc=jiri@resnulli.us \
--cc=kernel@pengutronix.de \
--cc=linux-can@vger.kernel.org \
--cc=mkl@pengutronix.de \
--cc=netdev@vger.kernel.org \
--cc=vincent.prince.fr@gmail.com \
--cc=xiyou.wangcong@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 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).