Re: [PATCH v3 2/2] dt-bindings: net: wireless: mt76: clarify backoff limit format

[Ryder Lee <Ryder.Lee@mediatek.com>]

On Wed, 2026-02-11 at 17:52 +0800, Chen-Yu Tsai wrote:
> On Wed, Feb 11, 2026 at 5:19 PM Ryder Lee <Ryder.Lee@mediatek.com>
> wrote:
> > 
> > On Wed, 2026-02-11 at 10:08 +0100, Krzysztof Kozlowski wrote:
> > > On 11/02/2026 09:59, Ryder Lee wrote:
> > > > On Wed, 2026-02-11 at 09:41 +0100, Krzysztof Kozlowski wrote:
> > > > > On 11/02/2026 09:33, Ryder Lee wrote:
> > > > > > > > > Why this cannot be a schema?
> > > > > > > > > 
> > > > > > > > > 
> > > > > > > > Well, actually, it's already a schema. This is just an
> > > > > > > > expanded
> > > > > > > 
> > > > > > > Where exactly?
> > > > > > > 
> > > > > > 
> > > > > > How 1T1ss is used across different generations is what my
> > > > > > example
> > > > > > above
> > > > > > was talking about.
> > > > > 
> > > > > Where exactly it is already a schema? Please point me line
> > > > > encoding
> > > > > this.
> > > > > 
> > > > > 
> > > > line 243 paths-ru
> > > > line 261 paths-ru-bf
> > > 
> > > I do not see there anything like you wrote here. You just list
> > > all of
> > > them, no device constraints.
> > > 
> > > Best regards,
> > > Krzysztof
> > > 
> > 
> > The original schema is a broad description. Now a reviewer want me
> > to
> > describe the differences for various connected devices, but I don’t
> > know how to add a compatible string for PCIe, USB, or even SDIO
> > devices
> > for their constraints. So I used the driver’s generation name...
> > can I
> > just write “mt7996”? Or do I need a complete and meaningful
> > compatible
> > string?
> 
> You can fill in the PCI or USB IDs as the compatible string.
> 
> See for example
> - Documentation/devicetree/bindings/net/wireless/brcm,bcm4329-
> fmac.yaml
> - Documentation/devicetree/bindings/net/wireless/brcm,bcm4329-
> fmac.yaml
> - Documentation/devicetree/bindings/net/wireless/qca,ath9k.yaml

Oh, I see. These examples are really helpful. I’ll add the PCI IDs and
replace the original connac2/3 patch with these PCI IDs.

> 
> There's no equivalent for SDIO, so they just have separate
> compatibles.
> The bcm4329-fmac binding also has examples of these.
> 
> If the hardware has something like a chip ID register, then you can
> have a common fallback string. At the extreme end of this is the ARM
> Mali bindings, which just have one compatible string for the core GPU
> as the fallback, and per platform/SoC compatibles to cover the glue
> layer:
> 
> - Documentation/devicetree/bindings/gpu/arm,mali-bifrost.yaml
> 
> 
> ChenYu
> 

Ryder