lkml.org 
[lkml]   [2020]   [Nov]   [12]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
From
Date
SubjectRe: [PATCH v2 1/7] dt-bindings: mfd: ti,j721e-system-controller.yaml: Document "syscon"
On Wed, Nov 11, 2020 at 11:25 PM Kishon Vijay Abraham I <kishon@ti.com> wrote:
>
> Hi Rob,
>
> On 12/11/20 2:58 am, Rob Herring wrote:
> > On Mon, Nov 09, 2020 at 10:34:03PM +0530, Kishon Vijay Abraham I wrote:
> >> Add binding documentation for "syscon" which should be a subnode of
> >> the system controller (scm-conf).
> >>
> >> Signed-off-by: Kishon Vijay Abraham I <kishon@ti.com>
> >> ---
> >> .../mfd/ti,j721e-system-controller.yaml | 40 +++++++++++++++++++
> >> 1 file changed, 40 insertions(+)
> >>
> >> diff --git a/Documentation/devicetree/bindings/mfd/ti,j721e-system-controller.yaml b/Documentation/devicetree/bindings/mfd/ti,j721e-system-controller.yaml
> >> index 19fcf59fd2fe..0b115b707ab2 100644
> >> --- a/Documentation/devicetree/bindings/mfd/ti,j721e-system-controller.yaml
> >> +++ b/Documentation/devicetree/bindings/mfd/ti,j721e-system-controller.yaml
> >> @@ -50,6 +50,38 @@ patternProperties:
> >> specified in
> >> Documentation/devicetree/bindings/mux/reg-mux.txt
> >>
> >> + "^syscon@[0-9a-f]+$":
> >> + type: object
> >> + description: |
> >
> > Don't need '|' if there's no formatting.
>
> Okay, will fix this.
> >
> >> + This is the system controller configuration required to configure PCIe
> >> + mode, lane width and speed.
> >> +
> >> + properties:
> >> + compatible:
> >> + items:
> >> + - enum:
> >> + - ti,j721e-system-controller
> >> + - const: syscon
> >> + - const: simple-mfd
> >
> > Humm, then what are this node's sub-nodes? And the same compatible as
> > the parent?
> >
>
> This node doesn't have sub-nodes.
>
> So one is the parent syscon node which has the entire system control
> region and then sub-nodes for each of the modules. In this case the PCIe
> in system control has only one 4 byte register that has to be configured.
>
> Both the parent node and sub-node are syscon, so given the same
> compatible for both.

'syscon' is just a hint. It doesn't define what any h/w is. IMO, we
never should have added it.

A compatible defines what the programming interface is for the node.
This one should only ever appear more than once if you have multiple
instances of the same block. So different registers, different
compatible. What you have here is just completely broken.

I don't think you even need a child node here. Just have PCIe node
point to the parent with an offset arg.

Rob

\
 
 \ /
  Last update: 2020-11-12 17:47    [W:3.089 / U:0.004 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site