Re: [RFT][PATCH v2 2/2] PM / i2c: designware: Clean up system sleep handling without ACPI

["Rafael J. Wysocki" <rafael@kernel.org>]

On Tue, Sep 5, 2017 at 5:24 PM, Mika Westerberg
<mika.westerberg@linux.intel.com> wrote:
> On Tue, Sep 05, 2017 at 05:04:21PM +0200, Rafael J. Wysocki wrote:
>> On Tuesday, September 5, 2017 5:07:44 PM CEST Mika Westerberg wrote:
>> > On Tue, Sep 05, 2017 at 04:55:44PM +0200, Rafael J. Wysocki wrote:
>> > > On Tuesday, September 5, 2017 4:58:35 PM CEST Mika Westerberg wrote:
>> > > > On Tue, Sep 05, 2017 at 04:46:11PM +0200, Rafael J. Wysocki wrote:
>> > > > > On Tuesday, September 5, 2017 4:45:11 PM CEST Mika Westerberg wrote:
>> > > > > > On Mon, Sep 04, 2017 at 12:01:54PM +0200, Rafael J. Wysocki wrote:
>> > > > > > >  static const struct dev_pm_ops dw_i2c_dev_pm_ops = {
>> > > > > > > - .prepare = dw_i2c_plat_prepare,
>> > > > > > > - .complete = dw_i2c_plat_complete,
>> > > > > > > - SET_SYSTEM_SLEEP_PM_OPS(dw_i2c_plat_suspend, dw_i2c_plat_resume)
>> > > > > > > - SET_RUNTIME_PM_OPS(dw_i2c_plat_runtime_suspend,
>> > > > > > > -                    dw_i2c_plat_resume,
>> > > > > > > -                    NULL)
>> > > > > > > + SET_LATE_SYSTEM_SLEEP_PM_OPS(dw_i2c_plat_suspend, dw_i2c_plat_resume)
>> > > > > >
>> > > > > > This seems to cause problem with intel-lpss MFD driver because it uses
>> > > > > > .suspend() and .resume() instead of .suspend_late() and .resume_early().
>> > > > >
>> > > > > OK, so there is one more dependency here.
>> > > > >
>> > > > > Can you please point me to this code?
>> > > >
>> > > > It is in drivers/mfd/intel-lpss.c. See intel_lpss_resume().
>> > > >
>> > >
>> > > Looking at it, but I don't quite see how this is related to the
>> > > i2c-designware-platedv suspend/resume ...
>> >
>> > intel-lpss is the parent device for i2c-designware-platdrv. It is
>> > supposed to handle all LPSS specific stuff, like bringing the PCI device
>> > out of reset before the i2c-designware-platdrv does its own resume
>> > things.
>>
>> Yes, I see.
>>
>> OK, so what about moving its suspend/resume to the late/early stages?
>>
>> Would the parent of it be confused?
>
> It seems to work. I did following change and now suspend/resume works
> fine with your patch series.

OK, thanks!

> diff --git a/drivers/mfd/intel-lpss.h b/drivers/mfd/intel-lpss.h
> index 694116630ffa..c987f7fe6c74 100644
> --- a/drivers/mfd/intel-lpss.h
> +++ b/drivers/mfd/intel-lpss.h
> @@ -38,8 +38,8 @@ int intel_lpss_resume(struct device *dev);
>  #ifdef CONFIG_PM_SLEEP
>  #define INTEL_LPSS_SLEEP_PM_OPS                        \
>         .prepare = intel_lpss_prepare,          \
> -       .suspend = intel_lpss_suspend,          \
> -       .resume = intel_lpss_resume,            \
> +       .suspend_late = intel_lpss_suspend,     \
> +       .resume_early = intel_lpss_resume,      \
>         .freeze = intel_lpss_suspend,           \
>         .thaw = intel_lpss_resume,              \
>         .poweroff = intel_lpss_suspend,         \

Of course, freeze/thaw, poweroff/restore need to be moved to the
late/early stages too.

I'll add this patch to the series and resend, then.

BTW, is the parent of intel-lpss in this case a PCI device or something else?