Re: [patch 1/5] pinctrl: samsung: Remove bogus irq_[un]mask from resource management

[Tomasz Figa]

2017-06-30 15:02 GMT+09:00 Krzysztof Kozlowski <krzk@xxxxxxxxxx>:
> On Fri, Jun 30, 2017 at 4:47 AM, Tomasz Figa <tomasz.figa@xxxxxxxxx> wrote:
>> Hi Thomas,
>>
>> 2017-06-30 6:33 GMT+09:00 Thomas Gleixner <tglx@xxxxxxxxxxxxx>:
>>> The irq chip callbacks irq_request/release_resources() have absolutely no
>>> business with masking and unmasking the irq.
>>>
>>> The core code unmasks the interrupt after complete setup and masks it
>>> before invoking irq_release_resources().
>>>
>>> The unmask is actually harmful as it happens before the interrupt is
>>> completely initialized in __setup_irq().
>>>
>>> Remove it.
>>
>> Good catch, thanks! (Note that the original patch of mine [1] did that
>> in .irq_startup()/.irq_shutdown(), which was for some reason changed
>> later, but I don't remember the exact story.)
>>
>> [1] https://patchwork.kernel.org/patch/4466431/
>>
>> Acked-by: Tomasz Figa <tomasz.figa@xxxxxxxxx>
>>
>> Sylwester, Krzysztof, would you be able to do some basic test?
>
> I suppose this was not tested so yes - I have platforms do this. I
> understand that checking any non-shared GPIO interrupt should be
> sufficient to test, right?

I think any interrupt from the Exynos pin controller should work, even
shared one. I'd expect irq_request_resources() to be invoked for
shared interrupts as well, otherwise we have a problem... (I quickly
looked through kernel/irq/manage.c and it seems to be invoked for the
first __setup_irq() call even for shared interrupts.)

Thanks.

Best regards,
Tomasz