[PATCH v7 05/11] ARM: dts: enable hi4511 with device tree

[Haojian Zhuang]

On 23 August 2013 02:50, Stephen Warren <swarren at wwwdotorg.org> wrote:
> On 08/22/2013 12:07 PM, Kevin Hilman wrote:
>> [+ DT maintainers]
>>
>> Haojian Zhuang <haojian.zhuang at linaro.org> writes:
>>
>>> Enable Hisilicon Hi4511 development platform with device tree support.
>>>
>>> Signed-off-by: Haojian Zhuang <haojian.zhuang at linaro.org>
> ...
>>> +/include/ "skeleton.dtsi"
>>> +
>>> +/ {
>>> +    aliases {
>>> +            serial0 = &uart0;
>>> +            serial1 = &uart1;
>>> +            serial2 = &uart2;
>>> +            serial3 = &uart3;
>>> +            serial4 = &uart4;
>>> +    };
>>> +
>>> +    cpus {
>>> +            #address-cells = <1>;
>>> +            #size-cells = <0>;
>>> +
>>> +            cpu0: cpu at 0 {
>>> +                    device_type = "cpu";
>>> +                    compatible = "arm,cortex-a9";
>>> +                    reg = <0x0>;
>>> +                    next-level-cache = <&L2>;
>>> +            };
>>> +    };
>>> +
>>> +    osc32k: osc32k {
>>> +            compatible = "fixed-clock";
>>> +            #clock-cells = <0>;
>>> +            clock-frequency = <32768>;
>>> +            clock-output-names = "osc32khz";
>>> +    };
>>
>> ...seems many of the recent users of clocks have grouped them into a
>> clocks {} grouping on a "simple-bus".
>>
>> DT folks: is there a rule of thumb on how whether these fixed clocks
>> should be grouped on a simple bus?
>
> I would expect all the clock node names to be just "clock", since the
> node names should describe the type of device not their identity (i.e.
> clock name).
>
> In turn, this means that each clock node name needs to use a unit
> address ("@nnn") to make them unique. In turn, this means they must have
> a reg property since the unit address must match the first entry in the
> reg property.

No, it's really bad on using a unit address. The register always contains
multiple mux or gate or divider. It would cause duplicated unit address.

I tried to use index number also. And it's also bad to append new clock nodes.
So I use the label name instead.

>
> Now I assume these clocks don't have any memory-mapped IO registers, so
> they would need an arbitrary reg value rather than a real one. So it
> doesn't make sense to place them directly under the root DT node, since
> their reg values would make no sense within the context of the
> CPU-visible MMIO space that the root node describes.
>
> In this case, it's typical to put all the clock nodes into e.g. a
> /clocks node, since that node can introduce a separate numbering-space
> for clocks. For example, I'd expect something like:
>
>         clocks {
>                 #address-cells = <1>;
>                 #size-cells = <0>;
>
>                 osc32k: clock at 0 {
>                         compatible = "fixed-clock";
>                         reg = <0>;
>                         #clock-cells = <0>;
>                         clock-frequency = <32768>;
>                         clock-output-names = "osc32khz";
>                 };
>
>                 osc26m: clock at 1 {
>                         compatible = "fixed-clock";
>                         reg = <1>;
>                         #clock-cells = <0>;
>                         clock-frequency = <26000000>;
>                         clock-output-names = "osc26mhz";
>                 };
>                 ...
>         };

Those fixed-clock doesn't contain reg property. Since it needs not to access
any clock register. It only provides the clock rate those child clock node.

>
> However, it also depends on what is actually providing those clocks. If
> every one of them is some standalone device on the board (e.g. a
> crystal), then just dumping them all in /clocks makes sense. However, if
> the clocks are provided by some on-SoC clock module, then I'd likely
> expect the clocks to be contained within the DT node that represents
> that clock module, which presumably does have some registers, and hence
> could be a direct child of the root node. For example, I wonder if the
> following is more accurate:
>
>         sctrl: sctrl at fc802000 {
>                 compatible = "hisilicon,sctrl";
>                 reg = <0xfc802000 0x1000>;
>                 #address-cells = <1>;
>                 #size-cells = <0>;
>
>                 osc32k: clock at 0 {
>                         compatible = "fixed-clock";
>                         reg = <0>;
>                         #clock-cells = <0>;
>                         clock-frequency = <32768>;
>                         clock-output-names = "osc32khz";
>                 };
>
>                 osc26m: clock at 1 {
>                         compatible = "fixed-clock";
>                         reg = <1>;
>                         #clock-cells = <0>;
>                         clock-frequency = <26000000>;
>                         clock-output-names = "osc26mhz";
>                 };
>                 ...
>         };
>
> ... since I see there are already quite a few clocks inside the sctrl node.

I can move all others clock nodes into clocks node, likes osc32k, osc26m.
Since they're not belong to sctrl register bank. And I also move all clock nodes
into a new dtsi file to make it more clearly.