On 11/11/2023 4:25 PM, Johan Hovold wrote:
On Sat, Nov 11, 2023 at 03:17:40PM +0530, Krishna Kurapati PSSNV wrote:
On 11/11/2023 2:00 PM, Krishna Kurapati PSSNV wrote:
On 11/10/2023 6:58 PM, Johan Hovold wrote:
phy-names:
minItems: 1
- maxItems: 2
- items:
- enum:
- - usb2-phy
- - usb3-phy
+ maxItems: 8
+ oneOf:
+ - items:
+ enum: [ usb2-phy, usb3-phy ]
+ - items:
+ pattern: "^usb[23]-port[0-3]$"
Shouldn't this just be
pattern: "^usb[23]-[0-3]$"
so that it matches the names that are used by the nvidia bindings?
We already have some inconsistency in that Amlogic uses a variant based
on the legacy names that needlessly includes "phy" in the names:
const: usb2-phy0
const: usb2-phy1
const: usb3-phy0
...
I don't think we should be introducing a third naming scheme here so I
suggest just following the nvidia bindings.
In that case, why don't we use "^usb[23]-phy[0-3]$". I think its close
to what we have on dwc3 core already today (usb2-phy/usb3-phy).
I mean, it isn't needless. It is a phy and shouldn't the binding suggest
that and include "-phy" in the name ?
No, adding a '-phy' suffix to each name is unnecessary since the
property is called 'phy-names'.
This is also documented:
For names used in {clock,dma,interrupt,reset}-names, do not add
any suffix, e.g.: "tx" instead of "txirq" (for interrupt).
https://docs.kernel.org/devicetree/bindings/writing-bindings.html
Thanks for the explanation.
and we've already discussed this when I asked you to drop the likewise
redundant '_irq' suffix from the interrupt names.
Yes, we did discuss this in irq context. But in this case I thought it
was fine because we already have usb(2/3)-"phy" already present.
When pushing v14, will make this change to usb(2/3)-(0-3) and skip port/phy.
Regards,
Krishna,