Re: Re: [PATCH v3] serial: mps2-uart: Check for error irq

Re: Re: [PATCH v3] serial: mps2-uart: Check for error irq

[Jiasheng Jiang]

On 2021-12-23 12:54, Robin Murphy wrote:
>> Because of the possible failure of the platform_get_irq(), it should be
>> better to check it to avoid the use of error irq.
>
> As far as I can see, the only "use" of error values is that they will be 
> passed to request_irq(), which will then return -EINVAL because they are 
> not valid IRQ numbers, and that error will be handled appropriately. I 
> think that's a relatively common pattern, so your commit message should 
> really describe why you think it's a problem and why this addition is a 
> meaningful improvement.

Thanks for your reminder, and I correct my commit message as follow.
If that's ok, I will correct my other patches like this.

For the possible failure of the platform_get_irq(), the returned irq
could be error number and will finally cause the failure of the request_irq().
So it might be better to check just now in order to avoid the waste of
later processes.

Sincerely thanks,
Jiasheng

Re: [PATCH v3] serial: mps2-uart: Check for error irq

[Robin Murphy]

On 2021-12-23 13:27, Jiasheng Jiang wrote:
> On 2021-12-23 12:54, Robin Murphy wrote:
>>> Because of the possible failure of the platform_get_irq(), it should be
>>> better to check it to avoid the use of error irq.
>>
>> As far as I can see, the only "use" of error values is that they will be
>> passed to request_irq(), which will then return -EINVAL because they are
>> not valid IRQ numbers, and that error will be handled appropriately. I
>> think that's a relatively common pattern, so your commit message should
>> really describe why you think it's a problem and why this addition is a
>> meaningful improvement.
> 
> Thanks for your reminder, and I correct my commit message as follow.
> If that's ok, I will correct my other patches like this.
> 
> For the possible failure of the platform_get_irq(), the returned irq
> could be error number and will finally cause the failure of the request_irq().
> So it might be better to check just now in order to avoid the waste of
> later processes.

Even better, consider that platform_get_irq() can now in certain cases 
return -EPROBE_DEFER, and the consequences of letting request_irq() 
effectively convert that into -EINVAL, even at probe time rather than 
later on ;)

Cheers,
Robin.

> 
> Sincerely thanks,
> Jiasheng
>