On Thu, 1 Aug 2024 at 18:12, Conor Dooley <conor@xxxxxxxxxx> wrote:
>
> On Mon, Jul 29, 2024 at 10:51:50AM +0200, Geert Uytterhoeven wrote:
> > Hi Ilya,
> >
> > On Sun, Jul 28, 2024 at 10:52 AM Ilya Orazov <ilordash02@xxxxxxxxx> wrote:
> > > On Tue, 23 Jul 2024 at 23:14, Conor Dooley <conor@xxxxxxxxxx> wrote:
> > > > On Tue, Jul 23, 2024 at 10:55:17PM +0300, Ilya Orazov wrote:
> > > > > On Tue, 23 Jul 2024 at 21:50, Conor Dooley <conor@xxxxxxxxxx> wrote:
> > > > > > On Tue, Jul 23, 2024 at 08:20:04PM +0300, IlorDash wrote:
> > > > > > > On Fri, 19 Jul 2024 at 18:07, Conor Dooley <conor@xxxxxxxxxx> wrote:
> > > > > > > >
> > > > > > > > On Fri, Jul 19, 2024 at 12:03:21AM +0300, Ilya Orazov wrote:
> > > > > > > > > Microchip ATA6561 is High-Speed CAN Transceiver with Standby Mode.
> > > > > > > > > It is pin-compatible with TI TCAN1042.
> > > > > > > > >
> > > > > > > > > Signed-off-by: Ilya Orazov <ilordash02@xxxxxxxxx>
> > > > > > > > > ---
> > > > > > > > > Documentation/devicetree/bindings/phy/ti,tcan104x-can.yaml | 1 +
> > > > > > > > > 1 file changed, 1 insertion(+)
> > > > > > > > >
> > > > > > > > > diff --git a/Documentation/devicetree/bindings/phy/ti,tcan104x-can.yaml b/Documentation/devicetree/bindings/phy/ti,tcan104x-can.yaml
> > > > > > > > > index 79dad3e89aa6..03de361849d2 100644
> > > > > > > > > --- a/Documentation/devicetree/bindings/phy/ti,tcan104x-can.yaml
> > > > > > > > > +++ b/Documentation/devicetree/bindings/phy/ti,tcan104x-can.yaml
> > > > > > > > > @@ -18,6 +18,7 @@ properties:
> > > > > > > > > - nxp,tjr1443
> > > > > > > > > - ti,tcan1042
> > > > > > > > > - ti,tcan1043
> > > > > > > > > + - microchip,ata6561
> > > > > > > >
> > > > > > > > Given that your driver patch has
> > > > > > > > | diff --git a/drivers/phy/phy-can-transceiver.c b/drivers/phy/phy-can-transceiver.c
> > > > > > > > | index ee4ce4249698..dbcd99213ba1 100644
> > > > > > > > | --- a/drivers/phy/phy-can-transceiver.c
> > > > > > > > | +++ b/drivers/phy/phy-can-transceiver.c
> > > > > > > > | @@ -89,6 +89,10 @@ static const struct of_device_id can_transceiver_phy_ids[] = {
> > > > > > > > | .compatible = "nxp,tjr1443",
> > > > > > > > | .data = &tcan1043_drvdata
> > > > > > > > | },
> > > > > > > > | + {
> > > > > > > > | + .compatible = "microchip,ata6561",
> > > > > > > > | + .data = &tcan1042_drvdata
> > > > > > > > | + },
> > > > > > > > | { }
> > > > > > > > | };
> > > > > > > >
> > > > > > > > the driver patch is actually not needed at all, and you just need to
> > > > > > > > allow ti,tcan1042 as fallback compatible in the binding, so something
> > > > > > > > like:
> > > > > > > >
> > > > > > > > compatible:
> > > > > > > > oneOf:
> > > > > > > > - enum:
> > > > > > > > - nxp,tjr1443
> > > > > > > > - ti,tcan1042
> > > > > > > > - ti,tcan1043
> > > > > > > > - items:
> > > > > > > > - const: microchip,ata6561
> > > > > > > > - const: ti,tcan1042
> > > > > > > >
> > > > > > > > '#phy-cells':
> > > > > > > > const: 0
> > > > > > >
> > > > > > > I tested the build with fallback compatible:
> > > > > > >
> > > > > > > compatible:
> > > > > > > oneOf:
> > > > > > > - items:
> > > > > > > - enum:
> > > > > > > - microchip,ata6561
> > > > > > > - const: ti,tcan1042
> > > > > > > - items:
> > > > > > > - enum:
> > > > > > > - nxp,tjr1443
> > > > > > > - const: ti,tcan1043
> > > > > > >
> > > > > > > and modified compatible property in DTS:
> > > > > > >
> > > > > > > compatible = "microchip,ata6561", "ti,tcan1042";
> > > > > > >
> > > > > > > Build succeeded, phy-can-transceiver driver was used. So I would like
> > > > > > > to add a fallback compatible for both "microchip,ata6561" and
> > > > > > > "nxp,tjr1443" in this binding and modify other DTS files with
> > > > > > > compatible = "nxp,tjr1443". What do you think?
> > > > > >
> > > > > > This is wrong on two counts. Firstly, were what you have correct, you
> > > > > > should
> > > > > > squash the two:
> > > > > > - items:
> > > > > > - enum:
> > > > > > - nxp,tjr1443
> > > > > > - microchip,ata6561
> > > > > > - const: ti,tcan1042
> > > > > >
> > > > > > However, that does not allow the TI compatibles in isolation, so you
> > > > > > still need to allow that for the actual TI devices, so you need:
> > > > > >
> > > > > > oneOf:
> > > > > > - items:
> > > > > > - enum:
> > > > > > - microchip,ata6561
> > > > > > - nxp,tjr1443
> > > > > > - ti,tcan1043
> > > > > > - const: ti,tcan1042
> > > > > > - const: ti,tcan1042
> > > > > >
> > > > > > There's probably some devicetrees that would need to be fixed up. I'm
> > > > > > just not convinced that this is worth retrofitting however.
> > > > >
> > > > > But nxp,tjr1443 is pin compatible with ti,tcan1043, so it should
> > > > > fallback only to ti,tcan1043 and not ti,tcan1042. That's why I decided
> > > > > to split them into different enums.
> > > >
> > > > Ah, sorry I missed that. I misread the match data. Then you need:
> > > > compatible:
> > > > oneOf:
> > > > - items:
> > > > - enum:
> > > > - microchip,ata6561
> > > > - const: ti,tcan1042
> > > > - items:
> > > > - enum:
> > > > - nxp,tjr1443
> > > > - const: ti,tcan1043
> > > > - enum:
> > > > const: ti,tcan1042
> > > > const: ti,tcan1043
> > > >
> > > > because the TI devices exist and we still need to be able to
> > > > differentiate the TI and NXP devices. If you have
> > > > compatible = "nxp,tjr1443", "ti,tcan1042";
> > > > that means the device is an nxp,tjr1443. If you have
> > > > compatible = "ti,tcan1042";
> > > > then that's a tcan1042.
> > > >
> > > > > I made my patch according to a similar one that adds support for
> > > > > nxp,tjr1443. You can find it's conversation on
> > > > > https://lore.kernel.org/all/6ee5e2ce00019bd3f77d6a702b38bab1a45f3bb0.1674037830.git.geert+renesas@xxxxxxxxx/t/#u.
> > > >
> > > > > I thought we want to hold all PHY chip names in one compatible enum
> > > > > and each in its own of_device_id struct in driver and extend them
> > > > > where appropriate.
> > > >
> > > > Nah, fallbacks are preferred when the programming model is either
> > > > identical or a "compatible superset" of an existing device. New
> > > > of_device_id structs should only be used where we need to account for
> > > > differences in the programming model.
> > >
> > > However, I am curious as to why the NXP CAN PHY transceiver was not
> > > included as fallback compatible. Geert, could you please share your
> > > thoughts on this matter?
> >
> > The TJR1443 looked sufficiently similar to the TCAN1043 to use the
> > same driver configuration (which is limited to having standby and/or
> > enable signals or not). However, I'm not sure it behaves exactly
> > the same, e.g. in case of reporting an error condition (which is not
> > yet supported by the driver). The part numbers are also different,
> > so this is not a simple case of SN74HCxx vs. CD74HCxx.
> >
> > Summary: I don't know if they are identical, or if TJR1443 is a
> > compatible superset of TCAN1043, or vice versa. Hence I went for the
> > safest way....
>
> If we don't know for sure what the craic is with compatibility, then we
> should leave the existing tjr1443 compatible as-is I think.
If I understood the kernel documentation correctly, we use fallback
compatibles when devices are similar or there is an iterative
relationship between them. In my case, the TCAN1042 and ATA6561 are
from different manufacturers, and I'm not sure about their fully
identical functionality.
Therefore, I'll go back to the original idea where I shouldn't use a
fallback compatible here and must leave it as another compatible
property with its own of_device_id struct.
What do you think about it? In my opinion, this is not a case for
fallback compatibility.
--
Best regards,
Ilya Orazov
