Re: [PATCH v7 7/7] cpufreq: dt: Add support for interconnect bandwidth scaling

[Sibi Sankar]

Hey Georgi,

On 2020-04-24 21:24, Georgi Djakov wrote:
> In addition to clocks and regulators, some devices can scale the 
> bandwidth
> of their on-chip interconnect - for example between CPU and DDR memory. 
> Add
> support for that, so that platforms which support it can make use of 
> it.
>
> Signed-off-by: Georgi Djakov <georgi.djakov@xxxxxxxxxx>
> ---
> v7:
> * Drop using dev_pm_opp_set_paths(), as it has been removed.
> * Add Kconfig dependency on INTERCONNECT, as it can be module.
>
>
> v2: 
> https://lore.kernel.org/r/20190423132823.7915-6-georgi.djakov@xxxxxxxxxx
>
>  drivers/cpufreq/Kconfig      |  1 +
>  drivers/cpufreq/cpufreq-dt.c | 15 +++++++++++++++
>  2 files changed, 16 insertions(+)
>
> diff --git a/drivers/cpufreq/Kconfig b/drivers/cpufreq/Kconfig
> index c3e6bd59e920..db2ad54ee67f 100644
> --- a/drivers/cpufreq/Kconfig
> +++ b/drivers/cpufreq/Kconfig
> @@ -217,6 +217,7 @@ config CPUFREQ_DT
>
>  config CPUFREQ_DT_PLATDEV
>  	bool
> +	depends on INTERCONNECT || !INTERCONNECT
>  	help
>  	  This adds a generic DT based cpufreq platdev driver for frequency
>  	  management.  This creates a 'cpufreq-dt' platform device, on the
> diff --git a/drivers/cpufreq/cpufreq-dt.c 
> b/drivers/cpufreq/cpufreq-dt.c
> index 26fe8dfb9ce6..4ecef3257532 100644
> --- a/drivers/cpufreq/cpufreq-dt.c
> +++ b/drivers/cpufreq/cpufreq-dt.c
> @@ -13,6 +13,7 @@
>  #include <linux/cpufreq.h>
>  #include <linux/cpumask.h>
>  #include <linux/err.h>
> +#include <linux/interconnect.h>
>  #include <linux/module.h>
>  #include <linux/of.h>
>  #include <linux/pm_opp.h>
> @@ -95,6 +96,7 @@ static int resources_available(void)
>  	struct device *cpu_dev;
>  	struct regulator *cpu_reg;
>  	struct clk *cpu_clk;
> +	struct icc_path *cpu_path;
>  	int ret = 0;
>  	const char *name;
>
> @@ -121,6 +123,19 @@ static int resources_available(void)
>
>  	clk_put(cpu_clk);
>
> +	cpu_path = of_icc_get(cpu_dev, NULL);
> +	ret = PTR_ERR_OR_ZERO(cpu_path);

Wouldn't we want to verify all
available paths instead of just
the first path?

> +	if (ret) {
> +		if (ret == -EPROBE_DEFER)
> +			dev_dbg(cpu_dev, "defer icc path: %d\n", ret);
> +		else
> +			dev_err(cpu_dev, "failed to get icc path: %d\n", ret);
> +
> +		return ret;
> +	}
> +
> +	icc_put(cpu_path);
> +
>  	name = find_supply_name(cpu_dev);
>  	/* Platform doesn't require regulator */
>  	if (!name)

--
Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum,
a Linux Foundation Collaborative Project.