Re: [PATCH v5 4/6] can: introduce the data bitrate configuration for CAN FD

[Marc Kleine-Budde <mkl@pengutronix.de>]

[-- Attachment #1: Type: text/plain, Size: 1572 bytes --]

On 02/28/2014 01:33 PM, Oliver Hartkopp wrote:
> 
> 
> On 28.02.2014 13:30, Marc Kleine-Budde wrote:
>> On 02/28/2014 01:20 PM, Oliver Hartkopp wrote:
>>>
>>>
>>> On 28.02.2014 11:56, Marc Kleine-Budde wrote:
>>>> On 02/27/2014 07:05 PM, Oliver Hartkopp wrote:
>>>>> On 27.02.2014 09:43, Marc Kleine-Budde wrote:
>>>>>
>>>>>>> +		if (priv->do_set_data_bittiming) {
>>>
>>>
>>>>
>>>>> The bitrate is always set via do_set[_data]_bittiming() when the interface is
>>>>> down - and *not* during the open process.
>>>>
>>>> No, not _always_. The flexcan and at91 drivers (maybe others, too) write
>>>> the values during open().
>>>
>>> flexcan does not assign priv->do_set_bittiming() and does the bitrate setting
>>> at open time.
>>>
>>> Thats ok too.
>>>
>>> This is up to the driver, whether do_set[_data]_bittiming() is assigned and
>>> when the stuff in the bitrate structures is set into the registers.
>>>
>>> I don't understand your problem.
>>
>> A new do_set_data_bittiming() callback is not needed, as it's better do
>> set the data and "normal" bit timing registers during open().
> 
> With this argumentation do_set_bittiming() has to be removed too.
> 
> Is this your intention?

Yes. But it's in many drivers so let's keep it, for now.

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: 242 bytes --]