Re: [PATCH 2/3] arm64: dts: qcom: qdu1000-idp: enable USB nodes

[Dmitry Baryshkov]

On Wed, 3 Apr 2024 at 10:50, Krishna Kurapati PSSNV
<quic_kriskura@xxxxxxxxxxx> wrote:
>
>
>
> On 4/2/2024 4:25 AM, Bjorn Andersson wrote:
> > On Tue, Mar 19, 2024 at 11:52:15AM +0200, Dmitry Baryshkov wrote:
> >> On Tue, 19 Mar 2024 at 11:11, Komal Bajaj <quic_kbajaj@xxxxxxxxxxx> wrote:
> >>>
> >>> Enable both USB controllers and associated hsphy and qmp phy
> >>> nodes on QDU1000 IDP. Add the usb type B port linked with the
> >>> DWC3 USB controller switched to OTG mode and tagged with
> >>> usb-role-switch.
> >>>
> >>> Co-developed-by: Amrit Anand <quic_amrianan@xxxxxxxxxxx>
> >>> Signed-off-by: Amrit Anand <quic_amrianan@xxxxxxxxxxx>
> >>> Signed-off-by: Komal Bajaj <quic_kbajaj@xxxxxxxxxxx>
> >>> ---
> >>>   arch/arm64/boot/dts/qcom/qdu1000-idp.dts | 65 ++++++++++++++++++++++++
> >>>   1 file changed, 65 insertions(+)
> >>>
> >>> diff --git a/arch/arm64/boot/dts/qcom/qdu1000-idp.dts b/arch/arm64/boot/dts/qcom/qdu1000-idp.dts
> >>> index 89b84fb0f70a..26442e707b5e 100644
> >>> --- a/arch/arm64/boot/dts/qcom/qdu1000-idp.dts
> >>> +++ b/arch/arm64/boot/dts/qcom/qdu1000-idp.dts
> >>> @@ -46,6 +46,33 @@ ppvar_sys: ppvar-sys-regulator {
> >>>                  regulator-boot-on;
> >>>          };
> >>>
> >>> +       usb_conn_gpio: usb-conn-gpio {
> >>> +               compatible = "gpio-usb-b-connector";
> >>
> >> If this board has only a USB-B connector, can it really handle USB 3.0?
> >>
> >
> > Here's a USB 3.0 Type-B cable, so no problem there:
> > https://en.wikipedia.org/wiki/USB_hardware#/media/File:USB_3.0_plug,_type_B_-_1709.jpg
> >
> >
> > @Komal, please confirm that this is the connector you have on the IDP?
> >
>
> Hi Bjorn,
>
>   Sorry for the confusion. The QDU1000 IDP has a Type-C connector. The
> type-c switch present between SoC and the connector is HD3SS3220 (from TI).
>
>   I think Dmitry's comment was that if it is 3.0, is it Type-C ? and if
> it is Type-C, then the compatible written in the being
> "gpio-usb-b-connector" would mean that there is a Type-B connector for
> someone who looks at the DT. (Dmitry, Please correct me if I understood
> the comment wrong).
>
>   I tried to push a series for adding a compatible to gpio conn driver
> [1] to resolve this and explained the connection specifics to Dmitry [2]
> and he suggested me to add a compatible for just the switch present on
> qdu1000 idp.
>
> Dmitry, Krzysztof,
>
> I was looking into the code again and it turns out there is a driver
> specific to HD3SS3220 switch [3] in linux already. I tried to check if
> it can be reused here but that driver relies on I2C communication
> between the SoC and the HD3SS3220 chip to get information on role
> switch. But in QDU1000 IDP board, there is no I2C communication present
> between SoC and the switch. Those lines have been cut off. The SoC only
> knows about VBUS/ID pins (other than DM/DP/SS Lanes) and no other I2C
> connections between the switch and the SoC. We still need to make use of
> vbus/id pins to decide which role we need to shift into. Can we still go
> ahead with using usb-conn-gpio driver by adding the compatible
> (qcom,qdu1000-hd3ss3220) and using it in DT ?

Is Qualcomm a manufacturer of the device? It is not.
Is qdu1000 a part of the device? It is not.

So the compatible string should be "ti,hd3ss3220". Which is fine to be
used in the platform driver. Just describe the differences in the
schema.


>
> Let me know your thoughts on this.
>
> [1]:
> https://lore.kernel.org/all/6f2df222-36d4-468e-99a7-9c48fae85aa9@xxxxxxxxxxx/
>
> [2]:
> https://lore.kernel.org/all/6f2df222-36d4-468e-99a7-9c48fae85aa9@xxxxxxxxxxx/
>
> [3]:
> https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/drivers/usb/typec/hd3ss3220.c?h=v6.9-rc2
>
> Regards,
> Krishna,
>
>
>


-- 
With best wishes
Dmitry