Hi Geert-san,
> From: Geert Uytterhoeven, Sent: Wednesday, November 14, 2018 8:03 PM
>
> Hi Chris,
>
> On Wed, Nov 7, 2018 at 6:36 PM Chris Brandt <chris.brandt@xxxxxxxxxxx> wrote:
<snip>
> > @@ -16,8 +18,8 @@ Required properties:
> > R8A77990 SoC.
> > "renesas,usb2-phy-r8a77995" if the device is a part of an
> > R8A77995 SoC.
> > - "renesas,rcar-gen3-usb2-phy" for a generic R-Car Gen3 or RZ/G2
> > - compatible device.
> > + "renesas,rcar-gen3-usb2-phy" for a generic R-Car Gen3, RZ/G2 or
> > + RZ/A2 compatible device.
>
> Is it a good idea to declare RZ/A2 compatible to R-Car Gen3?
> Usually we don't do that for Renesas IP cores used in different families[*].
> Of course, I know you do have a good relationship with the actual RZ/A2
> hardware designers ;-)
>
> In light of Shimoda-san's comment w.r.t. R-Car H3 USB port1 in the driver
> code, perhaps "renesas,rcar-gen3-usb2-phy" should not be used?
Sorry for lack explanation.
- The default value of COMMCTRL on each port of R-Car Gen3 is the same (0x80000000).
- However, R-Car H3 USB port1 seems to ignore this value because
the port always acts as host. If I changed the register to 0 on R-Car H3 ES3.0,
the port1 can work as host.
So, I meant all "renesas,rcar-gen3-usb2-phy" devices without "is_otg_channel"
can set the COMMCTRL register to 0.
Of course, adding a new RZ/A2 family compatible is OK to me.
Best regards,
Yoshihiro Shimoda
> [*] Sole exceptions I'm aware of are:
> - "renesas,rmobile-iic", in addition to "renesas,rcar-gen2-iic" or
> "renesas,rcar-gen3-iic" on R-Car Gen2/3,
> - "renesas,rcar-gen2-cmt1" and "renesas,rcar-thermal" on R-Mobile APE6.
>
> Gr{oetje,eeting}s,
>
> Geert
>
> --
> Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@xxxxxxxxxxxxxx
>
> In personal conversations with technical people, I call myself a hacker. But
> when I'm talking to journalists I just say "programmer" or something like that.
> -- Linus Torvalds
