On Friday 16 March 2018 08:22 AM, David Lechner wrote:
> +static int __init of_davinci_timer_init(struct device_node *np)
> +{
> + struct clk *clk;
> +
> + clk = of_clk_get(np, 0);
> + if (IS_ERR(clk)) {
> + struct of_phandle_args clkspec;
> +
> + /*
> + * Fall back to using ref_clk if the actual clock is not
> + * available. This currently always happens because platform
> + * clocks (i.e PLLs and PSCs) are registered as platform
> + * devices and therefore are not available at this point in
> + * the boot process.
It seems to me that this is not going to be a temporary problem (or at
least will be around for quite a while). So, I think we can as well just
look for ref_clk directly.
> + */
> + clkspec.np = of_find_node_by_name(NULL, "ref_clk");
> + if (IS_ERR(clkspec.np)) {
> + pr_err("%s: No clock available for timer!\n", __func__);
> + return PTR_ERR(clkspec.np);
> + }
> + clk = of_clk_get_from_provider(&clkspec);
> + of_node_put(clkspec.np);
> + }
> +
> + davinci_timer_init(clk);
> +
> + return 0;
> +}
> +TIMER_OF_DECLARE(davinci_timer, "ti,davinci-timer", of_davinci_timer_init);
Here, I think we should use "ti,da850-timer" so we can change the fixed
clock we are looking for based on the SoC. At a minimum, we should have
"ti,da850-timer" in the DT along with "ti,davinci-timer".
BTW, I noticed that "ti,davinci-timer" is not documented. I think we
need to add a binding documentation too.
Thanks,
Sekhar
