Re: [RFC PATCH 1/2] genirq: Add runtime resume/suspend support for IRQ chips

[Jon Hunter <jonathanh@nvidia.com>]

On 12/11/15 14:37, Lars-Peter Clausen wrote:
> On 11/12/2015 03:02 PM, Jon Hunter wrote:
> [...]
>>>>> One easy way out might be to always call pm_get/pm_but from
>>>>> bus_lock,/bus_unlock. This way the chip is guaranteed to be powered up when
>>>>> accessed happens. In addition pm_get is called when the IRQ is request and
>>>>> pm_put is called when the IRQ is release, this is to ensure the chip stays
>>>>> powered when it is actively monitoring the IRQ lines.
>>>>
>>>> Yes I had thought about that, but it is not quite that easy, because in
>>>> the case of request_irq() you don't want to pm_put() after the
>>>> bus_unlock(). However, the bus_lock/unlock() are good indicators of
>>>> different paths.
>>>
>>> You'd call pm_get() twice in request_irq() once from bus_lock() and once
>>> independently, that way you still have a reference even after the bus_unlock().
>>
>> Yes that is a possibility. However, there are places such as
>> show_interrupts() (kernel/irq/proc.c) and irq_gc_suspend() that do not
>> call bus_lock/unlock() which would need to be handled for PM. May be
>> these should also call bus_lock() as well?
> 
> show_interrupts() only accesses software state, not hardware state, or does it?

Good point. Today there only appears to be one user:

arch/powerpc/sysdev/fsl_msi.c:	.irq_print_chip = fsl_msi_print_chip,

This one is purely software. However, it would be easy to handle the
show_interrupts case if needed.

> suspend/resume is a bit tricky. It's kind of driver specific if it needs to
> actually access the hardware or whether the state is already shadowed in
> software. Maybe we can make this an exception for now and let drivers handle
> this on their own.

Yes I would agree with you on that.

Cheers
Jon