Re: [PATCH v4 5/9] usb: dwc3: Rework clock initialization to be more flexible

[Rob Herring]

On Tue, Oct 29, 2019 at 4:14 AM Felipe Balbi <balbi@xxxxxxxxxx> wrote:
>
>
> Hi,
>
> John Stultz <john.stultz@xxxxxxxxxx> writes:
>
> > The dwc3 core binding specifies three clocks:
> >   ref, bus_early, and suspend
> >
> > which are all controlled in the driver together.
> >
> > However some variants of the hardware my not have all three clks
>                                         ^^
>                                         may
>
> In fact *all* platforms have all three clocks. It's just that in some
> cases clock pins are shorted together (or take input from same clock).
>
> > So this patch reworks the reading of the clks from the dts to
> > use devm_clk_bulk_get_all() will will fetch all the clocks
>                               ^^^^
>                               which?
>
> > specified in the dts together.
> >
> > This patch was reccomended by Rob Herring <robh@xxxxxxxxxx>
> > as an alternative to creating multiple bindings for each variant
> > of hardware when the only unique bits were clocks and resets.
> >
> > Cc: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx>
> > Cc: Rob Herring <robh+dt@xxxxxxxxxx>
> > Cc: Mark Rutland <mark.rutland@xxxxxxx>
> > CC: ShuFan Lee <shufan_lee@xxxxxxxxxxx>
> > Cc: Heikki Krogerus <heikki.krogerus@xxxxxxxxxxxxxxx>
> > Cc: Suzuki K Poulose <suzuki.poulose@xxxxxxx>
> > Cc: Chunfeng Yun <chunfeng.yun@xxxxxxxxxxxx>
> > Cc: Yu Chen <chenyu56@xxxxxxxxxx>
> > Cc: Felipe Balbi <balbi@xxxxxxxxxx>
> > Cc: Hans de Goede <hdegoede@xxxxxxxxxx>
> > Cc: Andy Shevchenko <andy.shevchenko@xxxxxxxxx>
> > Cc: Jun Li <lijun.kernel@xxxxxxxxx>
> > Cc: Valentin Schneider <valentin.schneider@xxxxxxx>
> > Cc: Jack Pham <jackp@xxxxxxxxxxxxxx>
> > Cc: linux-usb@xxxxxxxxxxxxxxx
> > Cc: devicetree@xxxxxxxxxxxxxxx
> > Suggested-by: Rob Herring <robh@xxxxxxxxxx>
> > Signed-off-by: John Stultz <john.stultz@xxxxxxxxxx>
> > ---
> > v3: Rework dwc3 core rather then adding another dwc-of-simple
> >     binding.
> > ---
> >  drivers/usb/dwc3/core.c | 20 +++++---------------
> >  1 file changed, 5 insertions(+), 15 deletions(-)
> >
> > diff --git a/drivers/usb/dwc3/core.c b/drivers/usb/dwc3/core.c
> > index a039e35ec7ad..4d4f1836b62c 100644
> > --- a/drivers/usb/dwc3/core.c
> > +++ b/drivers/usb/dwc3/core.c
> > @@ -305,12 +305,6 @@ static int dwc3_core_soft_reset(struct dwc3 *dwc)
> >       return 0;
> >  }
> >
> > -static const struct clk_bulk_data dwc3_core_clks[] = {
> > -     { .id = "ref" },
> > -     { .id = "bus_early" },
> > -     { .id = "suspend" },
> > -};
>
> another option would be to pass three clocks with the same phandle. That
> would even make sure that clock usage counts are correct, no?

If you have the datasheet for the block, then perhaps some suggestion
of which clocks code be the same. My guess would be ref and suspend.

Maybe suspend is a fixed clock which is unmanaged on HiSilicon
platforms. If we allow for no clocks on some platforms, then why does
it have to be all or none?

Rob