Re: [PATCH v4 1/6] dt-bindings: phy: ti,tcan104x-can: Document NXP TJA105X/1048

[Conor Dooley <conor@kernel.org>]

[-- Attachment #1: Type: text/plain, Size: 4336 bytes --]

On Tue, Sep 02, 2025 at 02:24:49AM +0000, Peng Fan wrote:
> Hi Conor,
> 
> > Subject: Re: [PATCH v4 1/6] dt-bindings: phy: ti,tcan104x-can:
> > Document NXP TJA105X/1048
> > 
> > On Mon, Sep 01, 2025 at 07:54:01PM +0100, Conor Dooley wrote:
> > > On Mon, Sep 01, 2025 at 11:18:11AM +0800, Peng Fan wrote:
> > > > The TJA105[1,7] is a high-speed CAN transceiver which is a
> > > > pin-compatible alternative for TI TCAN1043 with sleep mode
> > > > supported, and has a compatible programming model, therefore
> > use ti,tcan1043 as fallback compatible.
> > > >
> > > > The TJA1048 is a dual high-speed CAN transceiver with sleep mode
> > supported.
> > > >
> > > > Reviewed-by: Frank Li <Frank.Li@nxp.com>
> > > > Signed-off-by: Peng Fan <peng.fan@nxp.com>
> > > > ---
> > > >  .../devicetree/bindings/phy/ti,tcan104x-can.yaml   | 30
> > ++++++++++++++++++++--
> > > >  1 file changed, 28 insertions(+), 2 deletions(-)
> > > >
> > > > diff --git
> > > > a/Documentation/devicetree/bindings/phy/ti,tcan104x-can.yaml
> > > > b/Documentation/devicetree/bindings/phy/ti,tcan104x-can.yaml
> > > > index
> > > >
> > 4a8c3829d85d3c4a4963750d03567c1c345beb91..f8e0c24856a2ba83
> > b5c988b246
> > > > 464f47e11a032f 100644
> > > > --- a/Documentation/devicetree/bindings/phy/ti,tcan104x-can.yaml
> > > > +++ b/Documentation/devicetree/bindings/phy/ti,tcan104x-
> > can.yaml
> > > > @@ -19,18 +19,25 @@ properties:
> > > >            - enum:
> > > >                - microchip,ata6561
> > > >            - const: ti,tcan1042
> > > > +      - items:
> > > > +          - enum:
> > > > +              - nxp,tja1051
> > > > +              - nxp,tja1057
> > > > +          - const: ti,tcan1043
> > > >        - enum:
> > > >            - ti,tcan1042
> > > >            - ti,tcan1043
> > > > +          - nxp,tja1048
> > > >            - nxp,tjr1443
> > > >
> > > >    '#phy-cells':
> > > > -    const: 0
> > > > +    enum: [0, 1]
> > > >
> > > >    standby-gpios:
> > > >      description:
> > > >        gpio node to toggle standby signal on transceiver
> > > > -    maxItems: 1
> > > > +    minItems: 1
> > > > +    maxItems: 2
> > >
> > > You're adding a second standby gpio, which one is which?
> > > I assume you mean that item 1 is stbn1 and item 2 is stbn 2 for
> > tja1048.
> 
> Yes. There are two standby pins.
> 
> > > Might be kinda obvious, but I think it should be mentioned.
> 
> I could update description as below.
> "
> description:
>   gpio node to toggle standby signal on transceiver. For two Items,
>   item 1 is for stbn1, item 2 is for stbn2.

Sure.

> 
> > >
> > > tja105{1,7} don't have a standby gpio, but they do have a silent
> > mode.
> 
> Right. Tja105{1,7} has a pin S for silent mode.
> 
> > > silent mode seems fundamentally different to standby, since the
> > > receiver still works. Seems like that should be handled differently, no?
> > 
> > The docs for standby mode for the tcan1043 don't match with the
> > tja1051, "Standby mode is a low power mode where the driver and
> > receiver are disabled," so does the fallback compatible even make
> > sense? Seems like a combination of enable and standby gpios are used
> > to put the tcan device into silent mode but the tja1051 has a pin for
> > that alone and seemingly does not support standby at all?
> 
> Thanks for looking into the details. I also read more into the datasheet.
> 
> Tcan1043:
> EN   nSTB
> H       H    ---> normal mode
> L        H   ---> slient mode
> H       L    ---> standby mode
> L        X    --->off mode
> 
> TJA1051
> EN   S
> H     L   --> normal mode
> H     H  --> slient mode
> L      X  --> off mode
> 
> Your analysis is correct. silent is different with standby,
> but we only wanna to use normal and off mode

In addition, it looks like there are some tja1051 devices that don't
even have the enable pin? Of the 4 SKUs, there are 2 that use pin5 as
Vio and 1 that has it n/c. Only the T/E device has an enable there.

> If we need to handle differently, I need
> to add a new optional property
> 
> silent-gpios:
>   description:
>     gpio node to toggle silent signal on transceiver
>   minItems: 1

Ye, I think so. And probably add some if/then to use it instead of
standby on the relevant chips.

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 228 bytes --]