On 10/8/19 12:24 PM, Kurt Van Dijck wrote:
> On di, 08 okt 2019 12:02:15 +0200, Marc Kleine-Budde wrote:
>> On 10/8/19 11:57 AM, Marc Kleine-Budde wrote:
>>> From: Kurt Van Dijck <dev.kurt@xxxxxxxxxxxxxxxxxxxxxx>
>>>
>>> When the status register is read without status IRQ pending, it may get
>>> into a state that it goes into busoff state without having its irq
>>> activated, so the driver will never know.
>>
>> Can you help rephrasing the patch description.
>>
>> I understand the problem is that we should only read the status register
>> if there is a status IRQ. But what are the consequences regarding the
>> busoff state?
>
> busoff is the last irq you'll get :-)
> If you miss that one, you'll never catch in again.
>
> The observation is that you have some bus errors, but the driver doesn't
> know that the chip went in busoff, so it will never wake up again.
>
> Only solution is to set link down & up again.
>
> What's your opinion about this one:
Perfect, now it's easier to understand the problem.
> When the status register is read without the status IRQ pending, the
> chip may not raise the interrupt line for an upcoming status interrupt
> and the driver may miss a status interrupt.
> It is critical that the BUSOFF status interrupt is informed to the
^^^^^^^^
replaced by forwarded
> higher layers, since no more interrupts will follow without
> intervention.
added to linux-can.
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:
signature.asc
Description: OpenPGP digital signature
