Re: [PATCH v2] net: sch_generic: Use pfifo_fast as fallback scheduler for CAN hardware - Marc Kleine-Budde

From: Marc Kleine-Budde <mkl@pengutronix.de>
To: Stephen Hemminger <stephen@networkplumber.org>
Cc: Vincent Prince <vincent.prince.fr@gmail.com>,
	jiri@resnulli.us, jhs@mojatatu.com, netdev@vger.kernel.org,
	dave.taht@gmail.com, linux-can@vger.kernel.org,
	kernel@pengutronix.de, xiyou.wangcong@gmail.com,
	davem@davemloft.net
Subject: Re: [PATCH v2] net: sch_generic: Use pfifo_fast as fallback scheduler for CAN hardware
Date: Tue, 22 Oct 2019 18:48:56 +0200	[thread overview]
Message-ID: <d0d865bd-7bc7-abb5-2b54-2c100c4b9a09@pengutronix.de> (raw)
In-Reply-To: <20191022094254.489fd6a4@hermes.lan>

[-- Attachment #1.1: Type: text/plain, Size: 2010 bytes --]

On 10/22/19 6:42 PM, Stephen Hemminger wrote:
> On Tue, 22 Oct 2019 16:53:44 +0200
> Marc Kleine-Budde <mkl@pengutronix.de> wrote:
> 
>> On 10/22/19 3:23 PM, Vincent Prince wrote:
>>> Signed-off-by: Vincent Prince <vincent.prince.fr@gmail.com>  
>>
>> please include a patch description. I.e. this one:
>>
>> -------->8-------->8-------->8-------->8-------->8-------->8-------->8--------  
>> 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.

> Why not fix fq_codel to return the same errors as other qdisc?

The head drop is the problem. After a send() system call returned to
user space, one would not expect that a later send() will knock an
earlier from the queue.

It's too late to throttle the package generation, as one frame is lost
already.

Marc

-- 
Pengutronix e.K.                  | Marc Kleine-Budde           |
Industrial Linux Solutions        | Phone: +49-231-2826-924     |
Vertretung West/Dortmund          | Fax:   +49-5121-206917-5555 |
Amtsgericht Hildesheim, HRA 2686  | http://www.pengutronix.de   |

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 488 bytes --]