On Wed, Aug 11, 2021 at 08:09:00PM +0800, tangbin wrote:
> On 2021/8/11 19:58, Mark Brown wrote:
> > On Wed, Aug 11, 2021 at 07:55:23PM +0800, Tang Bin wrote:

> > > The function stm32_spdifrx_parse_of() is only called by the function
> > > stm32_spdifrx_probe(), and the probe function is only called with
> > > an openfirmware platform device. Therefore there is no need to check
> > > the device_node in probe function.

> > What is the benefit of not doing the check?  It seems like reasonable
> > defensive programming.

> I think it's unnecessary, because we all know than the probe function is
> only trigger if

> the device and the driver matches, and the trigger mode is just Device Tree.
> So the device_node

> must be exist in the probe function if it works. That's the reason why I
> think it's redundant.

I see why it is redundant, I don't see what problem this redudnancy
causes.