Re: [v2 1/2] dt-bindings: Add bindings for aspeed pwm-tach and pwm.

From: Rob Herring
Date: Thu Apr 15 2021 - 10:44:01 EST


On Wed, Apr 14, 2021 at 10:44 PM Billy Tsai <billy_tsai@xxxxxxxxxxxxxx> wrote:
>
> Hi Rob,
>
> On 2021/4/15, 6:16 AM,Rob Herringwrote:
>
> On Wed, Apr 14, 2021 at 06:49:38PM +0800, Billy Tsai wrote:
> >> This patch adds device bindings for aspeed pwm-tach device which is a
> >> multi-function device include pwn and tach function and pwm device which
> >> should be the sub-node of pwm-tach device.
> >>
> >> Signed-off-by: Billy Tsai <billy_tsai@xxxxxxxxxxxxxx>
> >> Change-Id: I18d9dea14c3a04e1b7e38ffecd49d45917b9b545
> >
> >Drop
> >
> >> ---
> >> .../bindings/mfd/aspeed,ast2600-pwm-tach.yaml | 60 +++++++++++++++++++
> >> .../bindings/pwm/aspeed,ast2600-pwm.yaml | 44 ++++++++++++++
> >> 2 files changed, 104 insertions(+)
> >> create mode 100644 Documentation/devicetree/bindings/mfd/aspeed,ast2600-pwm-tach.yaml
> >> create mode 100644 Documentation/devicetree/bindings/pwm/aspeed,ast2600-pwm.yaml
> >>
> >> diff --git a/Documentation/devicetree/bindings/mfd/aspeed,ast2600-pwm-tach.yaml b/Documentation/devicetree/bindings/mfd/aspeed,ast2600-pwm-tach.yaml
> >> new file mode 100644
> >> index 000000000000..eaf8bdf8d44e
> >> --- /dev/null
> >> +++ b/Documentation/devicetree/bindings/mfd/aspeed,ast2600-pwm-tach.yaml
> >> @@ -0,0 +1,60 @@
> >> +# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
> >> +# Copyright (C) 2021 ASPEED, Inc.
> >> +%YAML 1.2
> >> +---
> >> +$id: http://devicetree.org/schemas/mfd/aspeed,ast2600-pwm-tach.yaml#
> >> +$schema: http://devicetree.org/meta-schemas/core.yaml#
> >> +
> >> +title: PWM Tach controller Device Tree Bindings
> >> +
> >> +description: |
> >> + The PWM Tach controller is represented as a multi-function device which
> >> + includes:
> >> + PWM
> >> + Tach
>
> > But is it really? A PWM and tach sounds like a fan controller. Look at
>
> Our PWM is not only for fans but also used for the motor, led, buzzer, and so on.
> So I want to split the function into two devices with a multi-function device.
> One for PWM output and one for tach monitor.
>
> > other existing PWM+tach bindings we have for fans.
>
> I didn't see the PWM+tach bindings can you give some example for me, thanks.

Let me grep 'tach' for you:

Documentation/devicetree/bindings/hwmon/aspeed-pwm-tacho.txt
Documentation/devicetree/bindings/hwmon/npcm750-pwm-fan.txt

Hey, look at that, there's already one for ASpeed. So the question is
how is the newer h/w different?

>
> >> +
> >> +maintainers:
> >> + - Billy Tsai <billy_tsai@xxxxxxxxxxxxxx>
> >> +
> >> +properties:
> >> + compatible:
> >> + items:
> >> + - enum:
> >> + - aspeed,ast2600-pwm-tach
> >> + - const: syscon
> >> + - const: simple-mfd
> >> + reg:
> >> + maxItems: 1
> >> + "#address-cells":
> >> + const: 1
> >> + "#size-cells":
> >> + const: 1
> >> +
> >> +required:
> >> + - compatible
> >> + - reg
> >> + - "#address-cells"
> >> + - "#size-cells"
> >> +
> >> +additionalProperties:
> >> + type: object
>
> > As you know the 2 node names, they should be documented. However, see
> > below.
>
> >> +
> >> +examples:
> >> + - |
> >> + pwm_tach: pwm_tach@1e610000 {
> >> + compatible = "aspeed,ast2600-pwm-tach", "syscon", "simple-mfd";
> >> + #address-cells = <1>;
> >> + #size-cells = <1>;
> >> + reg = <0x1e610000 0x100>;
> >> +
> >> + pwm: pwm@0 {
> >> + compatible = "aspeed,ast2600-pwm";
> >> + #pwm-cells = <3>;
> >> + reg = <0x0 0x100>;
> >> + };
> >> +
> >> + tach: tach@1 {
> >> + compatible = "aspeed,ast2600-tach";
> >> + reg = <0x0 0x100>;
>
> > You have 2 nodes at the same address. Not valid.
>
> Our pwm and tach is used the same base address and the offset is like below:
>
> PWM0 used 0x0 0x4, Tach0 used 0x8 0xc
> PWM1 used 0x10 0x14, Tach1 used 0x18 0x1c
> ...
>
> I will remove the reg property from pwm and tach node and remove the "#address-cells" and
> "#size-cells" from the parent node.

That's not really the solution...

>
> >> + };
>
> > There's no real need for 2 child nodes. The parent node can be a PWM
> > provider.
>
> However, In our usage, the parent node is a mfd, not a simple PWM device only. I don't want to
> combine the different functions with the one device node.

Looks like a single h/w block to me. If you want to divide that up
into multiple drivers, then that's an OS problem. A single node can be
multiple providers. For example, on the existing aspeed binding, just
add '#pwm-cells' to the node if you want to also expose it as a PWM
provider.

Rob