Re: [PATCH v3 2/3] mmc: Add realtek USB sdmmc host driver

[Roger <rogerable@realtek.com>]

On 02/11/2014 05:50 PM, Ulf Hansson wrote:
> On 11 February 2014 10:27, Roger <rogerable@realtek.com> wrote:
>> On 02/10/2014 10:58 PM, Ulf Hansson wrote:
>>>
>>> On 6 February 2014 15:35,  <rogerable@realtek.com> wrote:
>>>>
>>>> From: Roger Tseng <rogerable@realtek.com>
>>>>
>>>> Realtek USB SD/MMC host driver provides mmc host support based on the
>>>> Realtek
>>>> USB card reader MFD driver.
>>>>
>>>> Signed-off-by: Roger Tseng <rogerable@realtek.com>
>>>> ---
>>>>    drivers/mmc/host/Kconfig          |    7 +
>>>>    drivers/mmc/host/Makefile         |    1 +
>>>>    drivers/mmc/host/rtsx_usb_sdmmc.c | 1500
>>>> +++++++++++++++++++++++++++++++++++++
>>>>    3 files changed, 1508 insertions(+)
>>>>    create mode 100644 drivers/mmc/host/rtsx_usb_sdmmc.c
>>
>> [snip]
>>
>>>> +#ifdef CONFIG_PM_RUNTIME
>>>
>>>
>>> There are stubs for pm_runtime* functions, thus the ifdefs can be removed.
>>> Please go though the complete patch and remove all instances.
>>>
>>>> +               pm_runtime_put(sdmmc_dev(host));
>>>
>>>
>>> I don't know so much about USB mmc hosts hardware, but I just wanted
>>> to find out if I have understood this correct.
>>>
>>> You can't do fine grained power management of the USB parent device,
>>> since it needs to be runtime resumed to be able keep the power the
>>> card? Once it becomes runtime suspended, the power to the card will
>>> thus also be dropped?
>>>
>> Yes, and to keep some internal state of the controller.
>
> Okay.
>
> But the internal state of the controller should be possible to restore
> at runtime_resume, so that should not be the reason, right?
>

I checked again. Internal registers will hold its state during suspend.
Card power is the exact reason.
>>> Kind regards
>>> Ulf Hansson
> .
>