public inbox for linux-can@vger.kernel.org
 help / color / mirror / Atom feed
From: Oliver Hartkopp <socketcan@hartkopp.net>
To: Marc Kleine-Budde <mkl@pengutronix.de>
Cc: puneet.sharma@moschip.com, linux-can@vger.kernel.org
Subject: Re: CAN device filterting
Date: Tue, 04 Sep 2012 18:53:25 +0200	[thread overview]
Message-ID: <50463205.8050705@hartkopp.net> (raw)
In-Reply-To: <504608CB.9080709@pengutronix.de>

On 04.09.2012 15:57, Marc Kleine-Budde wrote:

> On 09/04/2012 03:40 PM, Oliver Hartkopp wrote:
>> On 04.09.2012 15:31, Puneet Sharma wrote:
>>
>>> I need some guidance on implementing the CAN filtering of identifier in
>>> the driver code of CAN for atmel AT91SAM9263 board. Please guide me how
>>> to proceed with it..
>>
>>
>> Hm - i wrote that some days ago:
> 
> I talked to Puneet in private before, if I understand him correctly he
> wants to implement hardware filtering, although software filtering is
> available.


Ah, ok.

> 
> From my point of view it should become a per device callback, which you
> (admin, i.e. the root user) can set when the device is down. You have to
> extend the CAN netlink interface [1].


Good idea.

> Each device should identify what kind and how many hardware filters it
> supports:
> a) mask+id
> b) just id
> c) ...
> Analogue to ctrlmode [2][3].


Yes. Could be ambitious. E.g. a 2048 bit bitfield is an option too :-)

The question is, if the usual mailbox configurations in CAN controller
IP-cores can be stripped down to 2-3 standard types of filters, so that a good
interface can be specified.

> 
> The requested filers are transferred into the kernel via netlink, but
> you have to figure out how to encode a variable length list (netfilter
> has probably a solution for this). Then the code in can_changelink() [4]
> should check if the transferred filter is supported on that hardware.
> Then, during open the driver should code the filters into hardware, on
> at91 it should be done in at91_setup_mailboxes [5][6].


Yep. I'm looking forward to a patch :-)

Regards,
Oliver

> 
> Marc
> 
> [1] http://lxr.free-electrons.com/source/drivers/net/can/dev.c#L585
> [2] http://lxr.free-electrons.com/source/include/linux/can/netlink.h#L81
> [3] http://lxr.free-electrons.com/source/drivers/net/can/dev.c#L691
> [4] http://lxr.free-electrons.com/source/drivers/net/can/dev.c#L600
> [5] http://lxr.free-electrons.com/source/drivers/net/can/at91_can.c#L418
> [6] http://lxr.free-electrons.com/source/drivers/net/can/at91_can.c#L336



  reply	other threads:[~2012-09-04 16:53 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-09-04 13:31 CAN device filterting Puneet Sharma
2012-09-04 13:40 ` Oliver Hartkopp
2012-09-04 13:57   ` Marc Kleine-Budde
2012-09-04 16:53     ` Oliver Hartkopp [this message]
2012-09-04 17:13       ` Marc Kleine-Budde
2012-09-05  9:28         ` Puneet Sharma

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=50463205.8050705@hartkopp.net \
    --to=socketcan@hartkopp.net \
    --cc=linux-can@vger.kernel.org \
    --cc=mkl@pengutronix.de \
    --cc=puneet.sharma@moschip.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