Re: [PATCH linux-next (v2) 1/2] clk: Add brcm, bcm6345-gate-clk device tree binding

[Michael Turquette]

Hi Simon,

Quoting Simon Arlott (2015-12-10 13:49:21)
> +periph_clk: periph_clk {
> +       compatible = "brcm,bcm63168-gate-clk", "brcm,bcm6345-gate-clk";
> +       regmap = <&periph_cntl>;
> +       offset = <0x4>;
> +
> +       #clock-cells = <1>;
> +       clock-indices =
> +               <1>,          <2>,        <3>,       <4>,     <5>,
> +               <6>,          <7>,        <8>,       <9>,     <10>,
> +               <11>,         <12>,       <13>,      <14>,    <15>,
> +               <16>,         <17>,       <18>,      <19>,    <20>,
> +               <27>,         <31>;
> +       clock-output-names =
> +               "vdsl_qproc", "vdsl_afe", "vdsl",    "mips",  "wlan_ocp",
> +               "dect",       "fap0",     "fap1",    "sar",   "robosw",
> +               "pcm",        "usbd",     "usbh",    "ipsec", "spi",
> +               "hsspi",      "pcie",     "phymips", "gmac",  "nand",
> +               "tbus",       "robosw250";

Why is clock-output-names required? Because you don't have any clock
data in your driver? Or is there another reason?

FYI, I'm not a fan of clock-output-names, and prefer for the clk
consumer devices to specify the clock-names property.

Another question, is it correct that this binding requires a DT node for
every register that contains clock control bits? If so, I'm skeptical of
that approach. What if you have a clock controller IP block on a future
soc that has several registers worth of clock controls?

Regards,
Mike

> +};
> +
> +timer_clk: timer_clk {
> +       compatible = "brcm,bcm63168-gate-clk", "brcm,bcm6345-gate-clk";
> +       regmap = <&timer_cntl>;
> +       offset = <0x4>;
> +
> +       #clock-cells = <1>;
> +       clock-indices = <17>, <18>;
> +       clock-output-names = "uto_extin", "usb_ref";
> +};
> +
> +ehci0: usb@10002500 {
> +       compatible = "brcm,bcm63168-ehci", "brcm,bcm6345-ehci", "generic-ehci";
> +       reg = <0x10002500 0x100>;
> +       big-endian;
> +       interrupt-parent = <&periph_intc>;
> +       interrupts = <10>;
> +       clocks = <&periph_clk 13>, <&timer_clk 18>;
> +       phys = <&usbh>;
> +};
> -- 
> 2.1.4
> 
> -- 
> Simon Arlott
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/