Hi Chris,
On Tue, May 7, 2019 at 5:43 PM Chris Brandt <Chris.Brandt@xxxxxxxxxxx> wrote:
> On Tue, May 07, 2019, Geert Uytterhoeven wrote:
> > > + if (of_property_read_bool(dev->of_node, "renesas,uses_usb_x1"))
> > > + channel->uses_usb_x1 = true;
> > > +
> >
> > Perhaps this can be checked some other way (e.g. by checking for a non-
> > zero
> > clock rate of the USB_X1 clock referenced from DT), thus removing the need
> > for
> > adding a custom property?
>
> Currently, there is no USB_X1 in DT like there is for RZ/A1.
>
> For RZ/A2, those are dedicated pins that belong to the USB HW block
> itself. They do not feed into the system CPG or any dividers, so I
> never included it in the .dtsi.
Like pcie_bus_clk on R-Car?
We do have that in DT, with a "clock" link to it from the PCIe device node.
After all, it is provided by an external clock crystal, and consumed by the
PCIe device.
> So with that said, does a uses-usb-x1 property make more sense?
No ;-)
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
