Re: [PATCH] socketcan: add a driver for FlexCAN controllers. - Marc Kleine-Budde

From: Marc Kleine-Budde <mkl-bIcnvbaLZ9MEGnE8C9+IrQ@public.gmane.org>
To: Wolfgang Grandegger <wg-5Yr1BZd7O62+XT7JhA+gdA@public.gmane.org>
Cc: socketcan-core-0fE9KPoRgkgATYTw5x5z8w@public.gmane.org,
	netdev-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
Subject: Re: [PATCH] socketcan: add a driver for FlexCAN controllers.
Date: Fri, 18 Jun 2010 13:21:32 +0200	[thread overview]
Message-ID: <4C1B56BC.1050303@pengutronix.de> (raw)
In-Reply-To: <4C1B52A7.3060607-5Yr1BZd7O62+XT7JhA+gdA@public.gmane.org>

[-- Attachment #1.1: Type: text/plain, Size: 2970 bytes --]

Wolfgang Grandegger wrote:
> On 06/18/2010 12:44 PM, Marc Kleine-Budde wrote:
>> Wolfgang Grandegger wrote:
>>> On 06/18/2010 12:16 PM, Marc Kleine-Budde wrote:
>>>> Wolfgang Grandegger wrote:
>>>>> Hi Hans-Jürgen,
>>>>>
>>>>> On 06/17/2010 12:52 PM, Hans J. Koch wrote:
>>>>>> This adds a driver for FlexCAN based CAN controllers,
>>>>>> e.g. found in Freescale i.MX35 SoCs.
>>>>>>
>>>>>> The original version of this driver was posted by Sascha Hauer in July 2009:
>>>>>> http://kerneltrap.org/mailarchive/linux-netdev/2009/7/29/6251621
>>>>>>
>>>>>> I took this version, added NAPI support, and fixed some problems found
>>>>>> during testing. Well, here is the result. Please review.
>>>>> I briefly browsed the patch and various bits and pieces are missing or
>>>>> not correctly implemented. Marc already pointed out a few of them:
>>>>>
>>>>> - I do not find can_put/get_echo_skb functions in the code. How is
>>>>>   IFF_ECHO supposed to work?
>>>> the driver uses hardware loopback
>>> OK, then
>>>
>>>   dev = alloc_candev(sizeof(struct flexcan_priv), 0);
>>>
>>> should be used (and TX_ECHO_SKB_MAX removed) in Hans-Jürgens driver.
>>>
>>>>> - Support for CAN_CTRLMODE_BERR_REPORTING and do_get_berr_counter()
>>>>>   seems to be missing.
>>>>>
>>>>> - Make use of alloc_can_skb() and alloc_can_err_skb().
>>>> the last two points are already addressed in my version of the driver.
>>> I do not see support for CAN_CTRLMODE_BERR_REPORTING in your driver
>>> (which has nothing to do with do_get_berr_counter).
>> oh yes...sorry, got confused.
>>
>> However I implemented CAN_CTRLMODE_BERR_REPORTING, i.e. turning of the
>> bit error interrupts by default. This has the effect that no bus warning
>> and bus passive interrupt was signalled.
>>
>> I should add a comment to my driver.
> 
> I'm confused, CAN_CTRLMODE_BERR_REPORTING means that the user can enable
> bus error reporting, which seems not be handled in the driver your sent.
> See:
> 
> http://lxr.linux.no/linux/drivers/net/can/sja1000/sja1000.c#L134
> http://lxr.linux.no/linux/drivers/net/can/sja1000/sja1000.c#L588

Which interrupts does "IRQ_BEI" include? What should
CAN_CTRLMODE_BERR_REPORTING do?

Looking at:
http://lxr.linux.no/linux+v2.6.34/drivers/net/can/sja1000/sja1000.c#L393
it seems that BEI on the SJA just effects bit, form and stuff errors.

If I disable the corresponding interrupt in the flexcan. This is
ERR_MSK, (1 << 14 in CTRL), I don't get error warning or error passive
interrupts either. I'm not sure about the bus off interrupt.

From my point of view this is not that what CAN_CTRLMODE_BERR_REPORTING
should do, is it?

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 #1.2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 260 bytes --]

[-- Attachment #2: Type: text/plain, Size: 188 bytes --]

_______________________________________________
Socketcan-core mailing list
Socketcan-core-0fE9KPoRgkgATYTw5x5z8w@public.gmane.org
https://lists.berlios.de/mailman/listinfo/socketcan-core