On Tue, Sep 05, 2023 at 07:06:43PM +0530, Vignesh Raman wrote:
> Hi Maxime,
>
> On 05/09/23 17:27, Maxime Ripard wrote:
> > On Tue, Sep 05, 2023 at 05:11:43PM +0530, Vignesh Raman wrote:
> > > > > Also, that node actually has a label ("usb"), defined here:
> > > > > https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/arch/arm64/boot/dts/qcom/msm8916.dtsi#n2322
> > > > >
> > > > > So you can end up with
> > > > >
> > > > > &usb {
> > > > > dr_mode = "host";
> > > > > };
> > > >
> > > > ... which is the simplest and thus more robust one.
> > > >
> > >
> > > Should it be,
> > > &{/soc@0/usb} {
> > > dr_mode = "host";
> > > };
> >
> > No. The &{/...} syntax refers to a path. &... refers to a label. They
> > are not equivalent.
>
> Sorry I was not clear before.
>
> With,
> &usb {
> dr_mode = "host";
> };
>
> The target is <0xffffffff> and fdtoverlay fails to apply the dtbo.
You do have /plugin/ and have compiled the base device tree with overlay
support, right?
> With,
> &{/soc@0/usb} {
> dr_mode = "host";
> };
>
> The target-path is "/soc@0/usb" (usb: usb@78d9000)
Right, and that's not the path you want to modify. The path you want to
modify is /soc@0/usb@78d9000. usb is the label, it's absolute, and you
can't mix and match a path ("/soc@0/") and a label ("usb")
Maxime
Attachment:
signature.asc
Description: PGP signature
