Re: [PATCH v2 4/5] dt-bindings: tpm: Add YAML schema for TPM TIS I2C options

[Rob Herring]

On Mon, Dec 16, 2019 at 7:53 AM Amir Mizinski <amirmizi6@xxxxxxxxx> wrote:
>
> On Sat, Dec 14, 2019 at 12:36 AM Rob Herring <robh@xxxxxxxxxx> wrote:
> >
> > On Mon, Dec 02, 2019 at 03:33:31PM +0200, amirmizi6@xxxxxxxxx wrote:
> > > From: Amir Mizinski <amirmizi6@xxxxxxxxx>
> > >
> > > Added a YAML schema to support tpm tis i2c realted dt-bindings for the I2c PTP based physical layer.
> >
> > Wrap your commmit message. And TPM, TIS?, and I2C should be capitalized.
>
> Thanks,  ill fix that.
>
> >
> > >
> > > Signed-off-by: Amir Mizinski <amirmizi6@xxxxxxxxx>
> > > ---
> > >  .../bindings/security/tpm/tpm-tis-i2c.yaml         | 38 ++++++++++++++++++++++
> > >  1 file changed, 38 insertions(+)
> > >  create mode 100644 Documentation/devicetree/bindings/security/tpm/tpm-tis-i2c.yaml
> >
> > Please read my comments on v1 (The first v1 from 11/10, not the 2nd v1
> > you sent).
>
> I sent a follow up comment regarding this:
> https://patchwork.kernel.org/patch/11236253/
> (2nd v1 was sent by mistake. sorry about that)

Sorry I missed your reply. However, you didn't address these comments:

> There's a bigger issue that the h/w here is more than just an I2C
> protocol. The chip may have multiple power supplies, clocks, reset
> lines, etc. HID over I2C seems like a similar case. Does the spec define
> *all* of that? If not, you need chip specific compatibles. You can keep
> this as a fallback though.

To rephrase this, a protocol does not fully describe the h/w and DT
should describe the h/w.

Also, you should include the interrupt whether you use it in the
driver currently or not. Again, it's about describing the h/w, not
what a driver happens to use ATM.

Rob