[PATCH 2/6] ARM: OMAP3/4: iommu: adapt to runtime pm

[Felipe Contreras]

On Thu, Oct 18, 2012 at 1:51 AM, Omar Ramirez Luna <omar.luna at linaro.org> wrote:
> On 15 October 2012 22:23, Felipe Contreras <felipe.contreras at gmail.com> wrote:
>>> On probe this patch does pm_runtime_enable, however this doesn't mean
>>> the device is turned ON or resumed or kept ON all the time.
>>
>> In fact it's the other way around; pm_runtime_enable turns off the
>> power (if it's ON).
>
> pm_runtime_enable just decrements the disable_depth counter. Doesn't
> turn off anything by itself.

pm_runtime_enable turns on power management, without it the device
remains in whatever state it booted.

>> device_del
>>  device_pm_remove
>>   pm_runtime_remove
>>    __pm_runtime_disable <- HERE
>
> I'm not entirely convinced _iommu_ follows this path, it doesn't
> create any devices nor register them... whereas mailbox does create
> devices and mailbox does follow this path for the devices created
> which are child devices.
>
> So this path won't take care of the omap-iommu device pm_runtime_disable.

Are you sure? What code-path calls omap_iommu_remove then?

Anyway, it was just an observation. I've seen other code do this.

-- 
Felipe Contreras