ARM: dts: meson: fixing USB support on Meson8b

[Martin Blumenstingl]

Hi Emiliano

On Fri, Sep 22, 2017 at 2:08 PM, Emiliano Ingrassia
<ingrassia at epigenesys.com> wrote:
> On Fri, Sep 22, 2017 at 08:59:42AM +0200, Martin Blumenstingl wrote:
>
> Hi Martin,
>
>> Hi Emiliano,
>>
>> On Wed, Sep 20, 2017 at 4:38 PM, Emiliano Ingrassia
>> <ingrassia at epigenesys.com> wrote:
>> > This patch fixes the Meson8b USB1 dts node, changing the IRQ type
>> > from edge rising to level high.
>> > This avoids errors like "usb 1-1-port1: cannot reset (err = -110)" on
>> > Odroid-C1+ board.
>> many thanks for investigating and even fixing this!
>>
>> > The patch also improves the same node adding the description of the digital
>> > and analog regulators found in DWC2 USB controller.
>> I could not find any other board that passes fixed regulators here.
>
> Actually there are some exynos boards which have those properties in their USB
> controllers dts nodes.
I see - those boards seem to use a real regulator for these supplies
(which needs to be turned on, which is why they are configuring it).
maybe we should get feedback from the dwc2 maintainers here

>> are you trying to fix a separate problem here?
>>
>
> Nothing in particular, just trying to avoid the messages:
> "Looking up vusb_d-supply property in node /soc/usb at c90c0000 failed" and
> "Looking up vusb_a-supply property in node /soc/usb at c90c0000 failed".
>
> The voltage values were taken from a comment regarding the dwc2_hstog_supply_names
> array, in drivers/usb/dwc2/core.h.
>
> Do you think that should be in a separate patch or require a better investigation?
yes, I think a separate patch would be good
if it is really needed then we need something similar for GXBB as well

>> can you add the following line before your Signed-off-by please:
>> Fixes: e29b1cf87473 ("ARM: dts: meson: add USB support on Meson8 and Meson8b")
>>
>
> Of course.
thank you!

>> > Signed-off-by: Emiliano Ingrassia <ingrassia at epigenesys.com>
>> > ---
>> >  arch/arm/boot/dts/meson8b.dtsi | 19 +++++++++++++++++++
>> >  1 file changed, 19 insertions(+)
>> >
>> > diff --git a/arch/arm/boot/dts/meson8b.dtsi b/arch/arm/boot/dts/meson8b.dtsi
>> > index bc278da7df0d..3306bff3d44d 100644
>> > --- a/arch/arm/boot/dts/meson8b.dtsi
>> > +++ b/arch/arm/boot/dts/meson8b.dtsi
>> > @@ -87,6 +87,22 @@
>> >                 compatible = "arm,cortex-a5-scu";
>> >                 reg = <0xc4300000 0x100>;
>> >         };
>> > +
>> > +       soc {
>> > +               vusb_d: regulator-vusb_d {
>> > +                       compatible = "regulator-fixed";
>> > +                       regulator-name = "vusb_d";
>> > +                       regulator-min-microvolt = <1200000>;
>> > +                       regulator-max-microvolt = <1200000>;
>> > +               };
>> > +
>> > +               vusb_a: regulator-vusb_a {
>> > +                       compatible = "regulator-fixed";
>> > +                       regulator-name = "vusb_a";
>> > +                       regulator-min-microvolt = <1100000>;
>> > +                       regulator-max-microvolt = <1100000>;
>> > +               };
>> > +       };
>> >  }; /* end of / */
>> >
>> >  &aobus {
>> > @@ -214,6 +230,9 @@
>> >
>> >  &usb1 {
>> >         compatible = "amlogic,meson8b-usb", "snps,dwc2";
>> > +       interrupts = <GIC_SPI 31 IRQ_TYPE_LEVEL_HIGH>;
>> I checked Amlogic's code: even Meson6 uses IRQ_TYPE_LEVEL_HIGH
>> (drivers/amlogic/usb/dwc_otg/310/dwc_otg_pcd_linux.c in Amlogic's 3.10
>> kernel sources)
>> can you please update meson.dtsi (instead of just updating
>> meson8b.dtsi) and do the same for the usb0 node?
>>
>
> Ok, I just submitted a new patch with the suggested corrections.
perfect, I will check that in a minute

>> > +       vusb_d-supply = <&vusb_d>;
>> > +       vusb_a-supply = <&vusb_a>;
>> >         clocks = <&clkc CLKID_USB1_DDR_BRIDGE>;
>> >         clock-names = "otg";
>> >  };
>> > --
>> > 2.14.1
>> >
>> >
>> > _______________________________________________
>> > linux-amlogic mailing list
>> > linux-amlogic at lists.infradead.org
>> > http://lists.infradead.org/mailman/listinfo/linux-amlogic
>>
>>
>> Regards,
>> Martin
>
> Regards,
>
> Emiliano


Regards,
Martin