Re: [PATCH V5 10/16] dt-bindings: PCI: tegra: Add device tree support for T194

[Vidya Sagar]

On 5/13/2019 8:50 PM, Rob Herring wrote:
> On Tue, May 7, 2019 at 4:20 AM Vidya Sagar <vidyas@xxxxxxxxxx> wrote:
> > On 4/26/2019 9:13 PM, Rob Herring wrote:
> > > On Wed, Apr 24, 2019 at 10:49:58AM +0530, Vidya Sagar wrote:
> > > > Add support for Tegra194 PCIe controllers. These controllers are based
> > > > on Synopsys DesignWare core IP.
> > > >
> > > > Signed-off-by: Vidya Sagar <vidyas@xxxxxxxxxx>
> > > > ---
>
> > > > +- nvidia,bpmp: Must contain a phandle to BPMP controller node.
> > > > +- nvidia,controller-id : Controller specific ID
> > > > +    0: C0
> > > > +    1: C1
> > > > +    2: C2
> > > > +    3: C3
> > > > +    4: C4
> > > > +    5: C5
> > >
> > > We don't normal put device indexes into DT. Why do you need this.
> > > Perhaps for accessing the BPMP? If so, make nvidia,bpmp a phandle+cell.
> > BPMP needs to know the controller number to enable it hence it needs to be
> > passed to BPMP. Just for accessing BPMP, I already added 'nvidia,bpmp' property.
>
> Then make nvidia,bpmp take the phandle and this number.
Ok. I'll take care of it in next patch series.

> > > > +- nvidia,disable-aspm-states: Controls advertisement of ASPM states
> > > > +    bit-0 to '1': Disables advertisement of ASPM-L0s
> > > > +    bit-1 to '1': Disables advertisement of ASPM-L1. This also disables
> > > > +                   advertisement of ASPM-L1.1 and ASPM-L1.2
> > > > +    bit-2 to '1': Disables advertisement of ASPM-L1.1
> > > > +    bit-3 to '1': Disables advertisement of ASPM-L1.2
> > >
> > > Can't this cover what 'supports-clkreq' does?
> > Well, they are related partially. i.e. if a platform doesn't have 'supports-clkreq' set,
> > then, by definition, it can't advertise support for ASPM L1.1 and L1.2 states. But, ASPM-L0s
> > and ASPM-L1 states don't depend on 'supports-clkreq' property.
> > Having this property gives more granularity as to support for which particular ASPM state
> > shouldn't be advertised by the root port.
>
> Okay, then it should be a common property then.
I'm planning to remove this given we have sysfs way (Heiner's patch series @
http://patchwork.ozlabs.org/project/linux-pci/list/?series=107392 to control which ASPM states
can be enabled run time. In case if that is not going to work for a given use case, I'll push
patches separately for controlling ASPM states advertisement/working.

> Rob