[PATCH/RFC,V8,1/1] clk: Support for clock parents and rates assigned from device tree

This patch adds helper functions to configure clock parents and rates
as specified through 'assigned-clock-parents', 'assigned-clock-rates'
DT properties for a clock provider or clock consumer device.
The helpers are now being called by the bus code for the platform, I2C
and SPI busses, before the driver probing and also in the clock core
after registration of a clock provider.

Signed-off-by: Sylwester Nawrocki <s.nawrocki@samsung.com>
Acked-by: Kyungmin Park <kyungmin.park@samsung.com>
---
Changes since v6:
 - use a set of separate DT properties to specify the default parent
   clocks and rates;
 - the clock defaults setting extended to the I2C and SPI busses.

Changes since v5:
 - updated the DT binding description (dropped 'assigned-clocks' node);
 - fixed detecting of null phandles (ENOENT error handling);
 - modified of_clk_init() to account for that the clocks property may now
   contain a clock specifier with a phandle that points to our node;

Changes since v4:
 - added note explaining how to skip setting parent and rate
   of a clock,
 - moved of_clk_dev_init() calls to the platform bus,
 - added missing call to of_node_put(),
 - dropped debug traces.

Changes since v3:
 - added detailed description of the assigned-clocks subnode,
 - added missing 'static inline' to the function stub definition,
 - clk-conf.c is now excluded when CONFIG_OF is not set,
 - s/of_clk_device_setup/of_clk_device_init.

Changes since v2:
 - edited in clock-bindings.txt, added note about 'assigned-clocks'
   subnode which may be used to specify "global" clocks configuration
   at a clock provider node,
 - moved of_clk_device_setup() function declaration from clk-provider.h
   to clk-conf.h so required function stubs are available when
   CONFIG_COMMON_CLK is not enabled,

Changes since v1:
 - the helper function to parse and set assigned clock parents and
   rates made public so it is available to clock providers to call
   directly;
 - dropped the platform bus notification and call of_clk_device_setup()
   is is now called from the driver core, rather than from the
   notification callback;
 - s/of_clk_get_list_entry/of_clk_get_by_property.
---
 .../devicetree/bindings/clock/clock-bindings.txt   |   36 +++++
 drivers/base/platform.c                            |    5 +
 drivers/clk/Makefile                               |    3 +
 drivers/clk/clk-conf.c                             |  143 ++++++++++++++++++++
 drivers/clk/clk.c                                  |   12 +-
 drivers/i2c/i2c-core.c                             |    5 +
 drivers/spi/spi.c                                  |    5 +
 include/linux/clk/clk-conf.h                       |   20 +++
 8 files changed, 227 insertions(+), 2 deletions(-)
 create mode 100644 drivers/clk/clk-conf.c
 create mode 100644 include/linux/clk/clk-conf.h

--
1.7.9.5

--
To unsubscribe from this list: send the line "unsubscribe linux-i2c" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

On 18/06/14 17:29, Sylwester Nawrocki wrote:
> This patch adds helper functions to configure clock parents and rates
> as specified through 'assigned-clock-parents', 'assigned-clock-rates'
> DT properties for a clock provider or clock consumer device.
> The helpers are now being called by the bus code for the platform, I2C
> and SPI busses, before the driver probing and also in the clock core
> after registration of a clock provider.
> 
> Signed-off-by: Sylwester Nawrocki <s.nawrocki@samsung.com>
> Acked-by: Kyungmin Park <kyungmin.park@samsung.com>

Could someone please take a look and review that patch ?
Any further suggestions, ACKs/NAKs ?

I would appreciate a DT, SPI or the I2C maintainer opinions.

Thanks,
Sylwester
> ---
> Changes since v6:
>  - use a set of separate DT properties to specify the default parent
>    clocks and rates;
>  - the clock defaults setting extended to the I2C and SPI busses.
> 
> Changes since v5:
>  - updated the DT binding description (dropped 'assigned-clocks' node);
>  - fixed detecting of null phandles (ENOENT error handling);
>  - modified of_clk_init() to account for that the clocks property may now
>    contain a clock specifier with a phandle that points to our node;
> 
> Changes since v4:
>  - added note explaining how to skip setting parent and rate
>    of a clock,
>  - moved of_clk_dev_init() calls to the platform bus,
>  - added missing call to of_node_put(),
>  - dropped debug traces.
> 
> Changes since v3:
>  - added detailed description of the assigned-clocks subnode,
>  - added missing 'static inline' to the function stub definition,
>  - clk-conf.c is now excluded when CONFIG_OF is not set,
>  - s/of_clk_device_setup/of_clk_device_init.
> 
> Changes since v2:
>  - edited in clock-bindings.txt, added note about 'assigned-clocks'
>    subnode which may be used to specify "global" clocks configuration
>    at a clock provider node,
>  - moved of_clk_device_setup() function declaration from clk-provider.h
>    to clk-conf.h so required function stubs are available when
>    CONFIG_COMMON_CLK is not enabled,
> 
> Changes since v1:
>  - the helper function to parse and set assigned clock parents and
>    rates made public so it is available to clock providers to call
>    directly;
>  - dropped the platform bus notification and call of_clk_device_setup()
>    is is now called from the driver core, rather than from the
>    notification callback;
>  - s/of_clk_get_list_entry/of_clk_get_by_property.
> ---
>  .../devicetree/bindings/clock/clock-bindings.txt   |   36 +++++
>  drivers/base/platform.c                            |    5 +
>  drivers/clk/Makefile                               |    3 +
>  drivers/clk/clk-conf.c                             |  143 ++++++++++++++++++++
>  drivers/clk/clk.c                                  |   12 +-
>  drivers/i2c/i2c-core.c                             |    5 +
>  drivers/spi/spi.c                                  |    5 +
>  include/linux/clk/clk-conf.h                       |   20 +++
>  8 files changed, 227 insertions(+), 2 deletions(-)
>  create mode 100644 drivers/clk/clk-conf.c
>  create mode 100644 include/linux/clk/clk-conf.h
> 
> diff --git a/Documentation/devicetree/bindings/clock/clock-bindings.txt b/Documentation/devicetree/bindings/clock/clock-bindings.txt
> index f1578781..06fc6d5 100644
> --- a/Documentation/devicetree/bindings/clock/clock-bindings.txt
> +++ b/Documentation/devicetree/bindings/clock/clock-bindings.txt
> @@ -131,3 +131,39 @@ clock signal, and a UART.
>    ("pll" and "pll-switched").
>  * The UART has its baud clock connected the external oscillator and its
>    register clock connected to the PLL clock (the "pll-switched" signal)
> +
> +==Assigned clock parents and rates==
> +
> +Some platforms may require initial configuration of default parent clocks
> +and clock frequencies. Such a configuration can be specified in a device tree
> +node through assigned-clocks, assigned-clock-parents and assigned-clock-rates
> +properties. The assigned-clock-parents property should contain a list of parent
> +clocks in form of phandle and clock specifier pairs, the assigned-clock-parents
> +property the list of assigned clock frequency values - corresponding to clocks
> +listed in the assigned-clocks property.
> +
> +To skip setting parent or rate of a clock its corresponding entry should be
> +set to 0, or can be omitted if it is not followed by any non-zero entry.
> +
> +    uart@a000 {
> +        compatible = "fsl,imx-uart";
> +        reg = <0xa000 0x1000>;
> +        ...
> +        clocks = <&osc 0>, <&pll 1>;
> +        clock-names = "baud", "register";
> +
> +        assigned-clocks = <&clkcon 0>, <&pll 2>;
> +        assigned-clock-parents = <&pll 2>;
> +        assigned-clock-rates = <0>, <460800>;
> +    };
> +
> +In this example the <&pll 2> clock is set as parent of clock <&clkcon 0> and
> +the <&pll 2> clock is assigned a frequency value of 460800 Hz.
> +
> +Configuring a clock's parent and rate through the device node that consumes
> +the clock can be done only for clocks that have a single user. Specifying
> +conflicting parent or rate configuration in multiple consumer nodes for
> +a shared clock is forbidden.
> +
> +Configuration of common clocks, which affect multiple consumer devices can
> +be similarly specified in the clock provider node.
> diff --git a/drivers/base/platform.c b/drivers/base/platform.c
> index 5b47210..f622733 100644
> --- a/drivers/base/platform.c
> +++ b/drivers/base/platform.c
> @@ -23,6 +23,7 @@
>  #include <linux/pm_runtime.h>
>  #include <linux/idr.h>
>  #include <linux/acpi.h>
> +#include <linux/clk/clk-conf.h>
> 
>  #include "base.h"
>  #include "power/power.h"
> @@ -486,6 +487,10 @@ static int platform_drv_probe(struct device *_dev)
>  	struct platform_device *dev = to_platform_device(_dev);
>  	int ret;
> 
> +	ret = of_clk_set_defaults(_dev->of_node, false);
> +	if (ret < 0)
> +		return ret;
> +
>  	acpi_dev_pm_attach(_dev, true);
> 
>  	ret = drv->probe(dev);
> diff --git a/drivers/clk/Makefile b/drivers/clk/Makefile
> index b25a1fe..7fe358c 100644
> --- a/drivers/clk/Makefile
> +++ b/drivers/clk/Makefile
> @@ -8,6 +8,9 @@ obj-$(CONFIG_COMMON_CLK)	+= clk-fixed-rate.o
>  obj-$(CONFIG_COMMON_CLK)	+= clk-gate.o
>  obj-$(CONFIG_COMMON_CLK)	+= clk-mux.o
>  obj-$(CONFIG_COMMON_CLK)	+= clk-composite.o
> +ifeq ($(CONFIG_OF), y)
> +obj-$(CONFIG_COMMON_CLK)	+= clk-conf.o
> +endif
> 
>  # hardware specific clock types
>  # please keep this section sorted lexicographically by file/directory path name
> diff --git a/drivers/clk/clk-conf.c b/drivers/clk/clk-conf.c
> new file mode 100644
> index 0000000..1f73019
> --- /dev/null
> +++ b/drivers/clk/clk-conf.c
> @@ -0,0 +1,143 @@
> +/*
> + * Copyright (C) 2014 Samsung Electronics Co., Ltd.
> + * Sylwester Nawrocki <s.nawrocki@samsung.com>
> + *
> + * This program is free software; you can redistribute it and/or modify
> + * it under the terms of the GNU General Public License version 2 as
> + * published by the Free Software Foundation.
> + */
> +
> +#include <linux/clk.h>
> +#include <linux/clk-provider.h>
> +#include <linux/clk/clk-conf.h>
> +#include <linux/device.h>
> +#include <linux/of.h>
> +#include <linux/printk.h>
> +#include "clk.h"
> +
> +static int __set_clk_parents(struct device_node *node, bool clk_supplier)
> +{
> +	struct of_phandle_args clkspec;
> +	int index, rc, num_parents;
> +	struct clk *clk, *pclk;
> +
> +	num_parents = of_count_phandle_with_args(node, "assigned-clock-parents",
> +						 "#clock-cells");
> +	if (num_parents == -EINVAL)
> +		pr_err("clk: invalid value of clock-parents property at %s\n",
> +		       node->full_name);
> +
> +	for (index = 0; index < num_parents; index++) {
> +		rc = of_parse_phandle_with_args(node, "assigned-clock-parents",
> +					"#clock-cells",	index, &clkspec);
> +		if (rc < 0) {
> +			/* skip empty (null) phandles */
> +			if (rc == -ENOENT)
> +				continue;
> +			else
> +				return rc;
> +		}
> +		if (clkspec.np == node && !clk_supplier)
> +			return 0;
> +		pclk = of_clk_get_by_clkspec(&clkspec);
> +		if (IS_ERR(pclk)) {
> +			pr_warn("clk: couldn't get parent clock %d for %s\n",
> +				index, node->full_name);
> +			return PTR_ERR(pclk);
> +		}
> +
> +		rc = of_parse_phandle_with_args(node, "assigned-clocks",
> +					"#clock-cells", index, &clkspec);
> +		if (rc < 0)
> +			goto err;
> +		if (clkspec.np == node && !clk_supplier) {
> +			rc = 0;
> +			goto err;
> +		}
> +		clk = of_clk_get_by_clkspec(&clkspec);
> +		if (IS_ERR(pclk)) {
> +			pr_warn("clk: couldn't get parent clock %d for %s\n",
> +				index, node->full_name);
> +			rc = PTR_ERR(pclk);
> +			goto err;
> +		}
> +
> +		rc = clk_set_parent(clk, pclk);
> +		if (rc < 0)
> +			pr_err("clk: failed to reparent %s to %s: %d\n",
> +			       __clk_get_name(clk), __clk_get_name(pclk), rc);
> +		clk_put(clk);
> +		clk_put(pclk);
> +	}
> +	return 0;
> +err:
> +	clk_put(pclk);
> +	return rc;
> +}
> +
> +static int __set_clk_rates(struct device_node *node, bool clk_supplier)
> +{
> +	struct of_phandle_args clkspec;
> +	struct property	*prop;
> +	const __be32 *cur;
> +	int rc, index = 0;
> +	struct clk *clk;
> +	u32 rate;
> +
> +	of_property_for_each_u32(node, "assigned-clock-rates", prop, cur, rate) {
> +		if (rate) {
> +			rc = of_parse_phandle_with_args(node, "assigned-clocks",
> +					"#clock-cells",	index, &clkspec);
> +			if (rc < 0) {
> +				/* skip empty (null) phandles */
> +				if (rc == -ENOENT)
> +					continue;
> +				else
> +					return rc;
> +			}
> +			if (clkspec.np == node && !clk_supplier)
> +				return 0;
> +
> +			clk = of_clk_get_by_clkspec(&clkspec);
> +			if (IS_ERR(clk)) {
> +				pr_warn("clk: couldn't get clock %d for %s\n",
> +					index, node->full_name);
> +				return PTR_ERR(clk);
> +			}
> +
> +			rc = clk_set_rate(clk, rate);
> +			if (rc < 0)
> +				pr_err("clk: couldn't set %s clock rate: %d\n",
> +				       __clk_get_name(clk), rc);
> +			clk_put(clk);
> +		}
> +		index++;
> +	}
> +	return 0;
> +}
> +
> +/**
> + * of_clk_set_defaults() - parse and set assigned clocks configuration
> + * @node: device node to apply clock settings for
> + * @clk_supplier: true if clocks supplied by @node should also be considered
> + *
> + * This function parses 'assigned-{clocks/clock-parents/clock-rates}' properties
> + * and sets any specified clock parents and rates. The @clk_supplier argument
> + * should be set to true if @node may be also a clock supplier of any clock
> + * listed in its 'assigned-clocks' or 'assigned-clock-parents' properties.
> + * If @clk_supplier is false the function exits returnning 0 as soon as it
> + * determines the @node is also a supplier of any of the clocks.
> + */
> +int of_clk_set_defaults(struct device_node *node, bool clk_supplier)
> +{
> +	int rc;
> +
> +	if (!node)
> +		return 0;
> +
> +	rc = __set_clk_parents(node, clk_supplier);
> +	if (rc < 0)
> +		return rc;
> +
> +	return __set_clk_rates(node, clk_supplier);
> +}
> diff --git a/drivers/clk/clk.c b/drivers/clk/clk.c
> index 8b73ede..a77af45 100644
> --- a/drivers/clk/clk.c
> +++ b/drivers/clk/clk.c
> @@ -10,6 +10,7 @@
>   */
> 
>  #include <linux/clk-private.h>
> +#include <linux/clk/clk-conf.h>
>  #include <linux/module.h>
>  #include <linux/mutex.h>
>  #include <linux/spinlock.h>
> @@ -2414,6 +2415,7 @@ int of_clk_add_provider(struct device_node *np,
>  			void *data)
>  {
>  	struct of_clk_provider *cp;
> +	int ret;
> 
>  	cp = kzalloc(sizeof(struct of_clk_provider), GFP_KERNEL);
>  	if (!cp)
> @@ -2428,7 +2430,11 @@ int of_clk_add_provider(struct device_node *np,
>  	mutex_unlock(&of_clk_mutex);
>  	pr_debug("Added clock from %s\n", np->full_name);
> 
> -	return 0;
> +	ret = of_clk_set_defaults(np, true);
> +	if (ret < 0)
> +		of_clk_del_provider(np);
> +
> +	return ret;
>  }
>  EXPORT_SYMBOL_GPL(of_clk_add_provider);
> 
> @@ -2605,7 +2611,10 @@ void __init of_clk_init(const struct of_device_id *matches)
>  		list_for_each_entry_safe(clk_provider, next,
>  					&clk_provider_list, node) {
>  			if (force || parent_ready(clk_provider->np)) {
> +
>  				clk_provider->clk_init_cb(clk_provider->np);
> +				of_clk_set_defaults(clk_provider->np, true);
> +
>  				list_del(&clk_provider->node);
>  				kfree(clk_provider);
>  				is_init_done = true;
> @@ -2620,7 +2629,6 @@ void __init of_clk_init(const struct of_device_id *matches)
>  		 */
>  		if (!is_init_done)
>  			force = true;
> -
>  	}
>  }
>  #endif
> diff --git a/drivers/i2c/i2c-core.c b/drivers/i2c/i2c-core.c
> index 7c7f4b8..66aa83b 100644
> --- a/drivers/i2c/i2c-core.c
> +++ b/drivers/i2c/i2c-core.c
> @@ -42,6 +42,7 @@
>  #include <linux/of.h>
>  #include <linux/of_device.h>
>  #include <linux/of_irq.h>
> +#include <linux/clk/clk-conf.h>
>  #include <linux/completion.h>
>  #include <linux/hardirq.h>
>  #include <linux/irqflags.h>
> @@ -274,6 +275,10 @@ static int i2c_device_probe(struct device *dev)
>  					client->flags & I2C_CLIENT_WAKE);
>  	dev_dbg(dev, "probe\n");
> 
> +	status = of_clk_set_defaults(dev->of_node, false);
> +	if (status < 0)
> +		return status;
> +
>  	acpi_dev_pm_attach(&client->dev, true);
>  	status = driver->probe(client, i2c_match_id(driver->id_table, client));
>  	if (status)
> diff --git a/drivers/spi/spi.c b/drivers/spi/spi.c
> index 939edf4..8374620 100644
> --- a/drivers/spi/spi.c
> +++ b/drivers/spi/spi.c
> @@ -29,6 +29,7 @@
>  #include <linux/mutex.h>
>  #include <linux/of_device.h>
>  #include <linux/of_irq.h>
> +#include <linux/clk/clk-conf.h>
>  #include <linux/slab.h>
>  #include <linux/mod_devicetable.h>
>  #include <linux/spi/spi.h>
> @@ -259,6 +260,10 @@ static int spi_drv_probe(struct device *dev)
>  	const struct spi_driver		*sdrv = to_spi_driver(dev->driver);
>  	int ret;
> 
> +	ret = of_clk_set_defaults(dev->of_node, false);
> +	if (ret)
> +		return ret;
> +
>  	acpi_dev_pm_attach(dev, true);
>  	ret = sdrv->probe(to_spi_device(dev));
>  	if (ret)
> diff --git a/include/linux/clk/clk-conf.h b/include/linux/clk/clk-conf.h
> new file mode 100644
> index 0000000..f3050e1
> --- /dev/null
> +++ b/include/linux/clk/clk-conf.h
> @@ -0,0 +1,20 @@
> +/*
> + * Copyright (C) 2014 Samsung Electronics Co., Ltd.
> + * Sylwester Nawrocki <s.nawrocki@samsung.com>
> + *
> + * This program is free software; you can redistribute it and/or modify
> + * it under the terms of the GNU General Public License version 2 as
> + * published by the Free Software Foundation.
> + */
> +
> +struct device_node;
> +
> +#if defined(CONFIG_OF) && defined(CONFIG_COMMON_CLK)
> +int of_clk_set_defaults(struct device_node *node, bool clk_supplier);
> +#else
> +static inline int of_clk_set_defaults(struct device_node *node,
> +				      bool clk_supplier)
> +{
> +	return 0;
> +}
> +#endif
> --
> 1.7.9.5
--
To unsubscribe from this list: send the line "unsubscribe linux-i2c" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Quoting Sylwester Nawrocki (2014-07-03 10:25:53)
> On 18/06/14 17:29, Sylwester Nawrocki wrote:
> > This patch adds helper functions to configure clock parents and rates
> > as specified through 'assigned-clock-parents', 'assigned-clock-rates'
> > DT properties for a clock provider or clock consumer device.
> > The helpers are now being called by the bus code for the platform, I2C
> > and SPI busses, before the driver probing and also in the clock core
> > after registration of a clock provider.
> > 
> > Signed-off-by: Sylwester Nawrocki <s.nawrocki@samsung.com>
> > Acked-by: Kyungmin Park <kyungmin.park@samsung.com>
> 
> Could someone please take a look and review that patch ?
> Any further suggestions, ACKs/NAKs ?

Patch looks good to me. I'm happy to take it.

> 
> I would appreciate a DT, SPI or the I2C maintainer opinions.

Yes, Acks from SPI and I2C maintainers would be good. I might need to
drop those parts of this patch if they don't come through :-(

Regards,
Mike

> 
> Thanks,
> Sylwester
> > ---
> > Changes since v6:
> >  - use a set of separate DT properties to specify the default parent
> >    clocks and rates;
> >  - the clock defaults setting extended to the I2C and SPI busses.
> > 
> > Changes since v5:
> >  - updated the DT binding description (dropped 'assigned-clocks' node);
> >  - fixed detecting of null phandles (ENOENT error handling);
> >  - modified of_clk_init() to account for that the clocks property may now
> >    contain a clock specifier with a phandle that points to our node;
> > 
> > Changes since v4:
> >  - added note explaining how to skip setting parent and rate
> >    of a clock,
> >  - moved of_clk_dev_init() calls to the platform bus,
> >  - added missing call to of_node_put(),
> >  - dropped debug traces.
> > 
> > Changes since v3:
> >  - added detailed description of the assigned-clocks subnode,
> >  - added missing 'static inline' to the function stub definition,
> >  - clk-conf.c is now excluded when CONFIG_OF is not set,
> >  - s/of_clk_device_setup/of_clk_device_init.
> > 
> > Changes since v2:
> >  - edited in clock-bindings.txt, added note about 'assigned-clocks'
> >    subnode which may be used to specify "global" clocks configuration
> >    at a clock provider node,
> >  - moved of_clk_device_setup() function declaration from clk-provider.h
> >    to clk-conf.h so required function stubs are available when
> >    CONFIG_COMMON_CLK is not enabled,
> > 
> > Changes since v1:
> >  - the helper function to parse and set assigned clock parents and
> >    rates made public so it is available to clock providers to call
> >    directly;
> >  - dropped the platform bus notification and call of_clk_device_setup()
> >    is is now called from the driver core, rather than from the
> >    notification callback;
> >  - s/of_clk_get_list_entry/of_clk_get_by_property.
> > ---
> >  .../devicetree/bindings/clock/clock-bindings.txt   |   36 +++++
> >  drivers/base/platform.c                            |    5 +
> >  drivers/clk/Makefile                               |    3 +
> >  drivers/clk/clk-conf.c                             |  143 ++++++++++++++++++++
> >  drivers/clk/clk.c                                  |   12 +-
> >  drivers/i2c/i2c-core.c                             |    5 +
> >  drivers/spi/spi.c                                  |    5 +
> >  include/linux/clk/clk-conf.h                       |   20 +++
> >  8 files changed, 227 insertions(+), 2 deletions(-)
> >  create mode 100644 drivers/clk/clk-conf.c
> >  create mode 100644 include/linux/clk/clk-conf.h
> > 
> > diff --git a/Documentation/devicetree/bindings/clock/clock-bindings.txt b/Documentation/devicetree/bindings/clock/clock-bindings.txt
> > index f1578781..06fc6d5 100644
> > --- a/Documentation/devicetree/bindings/clock/clock-bindings.txt
> > +++ b/Documentation/devicetree/bindings/clock/clock-bindings.txt
> > @@ -131,3 +131,39 @@ clock signal, and a UART.
> >    ("pll" and "pll-switched").
> >  * The UART has its baud clock connected the external oscillator and its
> >    register clock connected to the PLL clock (the "pll-switched" signal)
> > +
> > +==Assigned clock parents and rates==
> > +
> > +Some platforms may require initial configuration of default parent clocks
> > +and clock frequencies. Such a configuration can be specified in a device tree
> > +node through assigned-clocks, assigned-clock-parents and assigned-clock-rates
> > +properties. The assigned-clock-parents property should contain a list of parent
> > +clocks in form of phandle and clock specifier pairs, the assigned-clock-parents
> > +property the list of assigned clock frequency values - corresponding to clocks
> > +listed in the assigned-clocks property.
> > +
> > +To skip setting parent or rate of a clock its corresponding entry should be
> > +set to 0, or can be omitted if it is not followed by any non-zero entry.
> > +
> > +    uart@a000 {
> > +        compatible = "fsl,imx-uart";
> > +        reg = <0xa000 0x1000>;
> > +        ...
> > +        clocks = <&osc 0>, <&pll 1>;
> > +        clock-names = "baud", "register";
> > +
> > +        assigned-clocks = <&clkcon 0>, <&pll 2>;
> > +        assigned-clock-parents = <&pll 2>;
> > +        assigned-clock-rates = <0>, <460800>;
> > +    };
> > +
> > +In this example the <&pll 2> clock is set as parent of clock <&clkcon 0> and
> > +the <&pll 2> clock is assigned a frequency value of 460800 Hz.
> > +
> > +Configuring a clock's parent and rate through the device node that consumes
> > +the clock can be done only for clocks that have a single user. Specifying
> > +conflicting parent or rate configuration in multiple consumer nodes for
> > +a shared clock is forbidden.
> > +
> > +Configuration of common clocks, which affect multiple consumer devices can
> > +be similarly specified in the clock provider node.
> > diff --git a/drivers/base/platform.c b/drivers/base/platform.c
> > index 5b47210..f622733 100644
> > --- a/drivers/base/platform.c
> > +++ b/drivers/base/platform.c
> > @@ -23,6 +23,7 @@
> >  #include <linux/pm_runtime.h>
> >  #include <linux/idr.h>
> >  #include <linux/acpi.h>
> > +#include <linux/clk/clk-conf.h>
> > 
> >  #include "base.h"
> >  #include "power/power.h"
> > @@ -486,6 +487,10 @@ static int platform_drv_probe(struct device *_dev)
> >       struct platform_device *dev = to_platform_device(_dev);
> >       int ret;
> > 
> > +     ret = of_clk_set_defaults(_dev->of_node, false);
> > +     if (ret < 0)
> > +             return ret;
> > +
> >       acpi_dev_pm_attach(_dev, true);
> > 
> >       ret = drv->probe(dev);
> > diff --git a/drivers/clk/Makefile b/drivers/clk/Makefile
> > index b25a1fe..7fe358c 100644
> > --- a/drivers/clk/Makefile
> > +++ b/drivers/clk/Makefile
> > @@ -8,6 +8,9 @@ obj-$(CONFIG_COMMON_CLK)      += clk-fixed-rate.o
> >  obj-$(CONFIG_COMMON_CLK)     += clk-gate.o
> >  obj-$(CONFIG_COMMON_CLK)     += clk-mux.o
> >  obj-$(CONFIG_COMMON_CLK)     += clk-composite.o
> > +ifeq ($(CONFIG_OF), y)
> > +obj-$(CONFIG_COMMON_CLK)     += clk-conf.o
> > +endif
> > 
> >  # hardware specific clock types
> >  # please keep this section sorted lexicographically by file/directory path name
> > diff --git a/drivers/clk/clk-conf.c b/drivers/clk/clk-conf.c
> > new file mode 100644
> > index 0000000..1f73019
> > --- /dev/null
> > +++ b/drivers/clk/clk-conf.c
> > @@ -0,0 +1,143 @@
> > +/*
> > + * Copyright (C) 2014 Samsung Electronics Co., Ltd.
> > + * Sylwester Nawrocki <s.nawrocki@samsung.com>
> > + *
> > + * This program is free software; you can redistribute it and/or modify
> > + * it under the terms of the GNU General Public License version 2 as
> > + * published by the Free Software Foundation.
> > + */
> > +
> > +#include <linux/clk.h>
> > +#include <linux/clk-provider.h>
> > +#include <linux/clk/clk-conf.h>
> > +#include <linux/device.h>
> > +#include <linux/of.h>
> > +#include <linux/printk.h>
> > +#include "clk.h"
> > +
> > +static int __set_clk_parents(struct device_node *node, bool clk_supplier)
> > +{
> > +     struct of_phandle_args clkspec;
> > +     int index, rc, num_parents;
> > +     struct clk *clk, *pclk;
> > +
> > +     num_parents = of_count_phandle_with_args(node, "assigned-clock-parents",
> > +                                              "#clock-cells");
> > +     if (num_parents == -EINVAL)
> > +             pr_err("clk: invalid value of clock-parents property at %s\n",
> > +                    node->full_name);
> > +
> > +     for (index = 0; index < num_parents; index++) {
> > +             rc = of_parse_phandle_with_args(node, "assigned-clock-parents",
> > +                                     "#clock-cells", index, &clkspec);
> > +             if (rc < 0) {
> > +                     /* skip empty (null) phandles */
> > +                     if (rc == -ENOENT)
> > +                             continue;
> > +                     else
> > +                             return rc;
> > +             }
> > +             if (clkspec.np == node && !clk_supplier)
> > +                     return 0;
> > +             pclk = of_clk_get_by_clkspec(&clkspec);
> > +             if (IS_ERR(pclk)) {
> > +                     pr_warn("clk: couldn't get parent clock %d for %s\n",
> > +                             index, node->full_name);
> > +                     return PTR_ERR(pclk);
> > +             }
> > +
> > +             rc = of_parse_phandle_with_args(node, "assigned-clocks",
> > +                                     "#clock-cells", index, &clkspec);
> > +             if (rc < 0)
> > +                     goto err;
> > +             if (clkspec.np == node && !clk_supplier) {
> > +                     rc = 0;
> > +                     goto err;
> > +             }
> > +             clk = of_clk_get_by_clkspec(&clkspec);
> > +             if (IS_ERR(pclk)) {
> > +                     pr_warn("clk: couldn't get parent clock %d for %s\n",
> > +                             index, node->full_name);
> > +                     rc = PTR_ERR(pclk);
> > +                     goto err;
> > +             }
> > +
> > +             rc = clk_set_parent(clk, pclk);
> > +             if (rc < 0)
> > +                     pr_err("clk: failed to reparent %s to %s: %d\n",
> > +                            __clk_get_name(clk), __clk_get_name(pclk), rc);
> > +             clk_put(clk);
> > +             clk_put(pclk);
> > +     }
> > +     return 0;
> > +err:
> > +     clk_put(pclk);
> > +     return rc;
> > +}
> > +
> > +static int __set_clk_rates(struct device_node *node, bool clk_supplier)
> > +{
> > +     struct of_phandle_args clkspec;
> > +     struct property *prop;
> > +     const __be32 *cur;
> > +     int rc, index = 0;
> > +     struct clk *clk;
> > +     u32 rate;
> > +
> > +     of_property_for_each_u32(node, "assigned-clock-rates", prop, cur, rate) {
> > +             if (rate) {
> > +                     rc = of_parse_phandle_with_args(node, "assigned-clocks",
> > +                                     "#clock-cells", index, &clkspec);
> > +                     if (rc < 0) {
> > +                             /* skip empty (null) phandles */
> > +                             if (rc == -ENOENT)
> > +                                     continue;
> > +                             else
> > +                                     return rc;
> > +                     }
> > +                     if (clkspec.np == node && !clk_supplier)
> > +                             return 0;
> > +
> > +                     clk = of_clk_get_by_clkspec(&clkspec);
> > +                     if (IS_ERR(clk)) {
> > +                             pr_warn("clk: couldn't get clock %d for %s\n",
> > +                                     index, node->full_name);
> > +                             return PTR_ERR(clk);
> > +                     }
> > +
> > +                     rc = clk_set_rate(clk, rate);
> > +                     if (rc < 0)
> > +                             pr_err("clk: couldn't set %s clock rate: %d\n",
> > +                                    __clk_get_name(clk), rc);
> > +                     clk_put(clk);
> > +             }
> > +             index++;
> > +     }
> > +     return 0;
> > +}
> > +
> > +/**
> > + * of_clk_set_defaults() - parse and set assigned clocks configuration
> > + * @node: device node to apply clock settings for
> > + * @clk_supplier: true if clocks supplied by @node should also be considered
> > + *
> > + * This function parses 'assigned-{clocks/clock-parents/clock-rates}' properties
> > + * and sets any specified clock parents and rates. The @clk_supplier argument
> > + * should be set to true if @node may be also a clock supplier of any clock
> > + * listed in its 'assigned-clocks' or 'assigned-clock-parents' properties.
> > + * If @clk_supplier is false the function exits returnning 0 as soon as it
> > + * determines the @node is also a supplier of any of the clocks.
> > + */
> > +int of_clk_set_defaults(struct device_node *node, bool clk_supplier)
> > +{
> > +     int rc;
> > +
> > +     if (!node)
> > +             return 0;
> > +
> > +     rc = __set_clk_parents(node, clk_supplier);
> > +     if (rc < 0)
> > +             return rc;
> > +
> > +     return __set_clk_rates(node, clk_supplier);
> > +}
> > diff --git a/drivers/clk/clk.c b/drivers/clk/clk.c
> > index 8b73ede..a77af45 100644
> > --- a/drivers/clk/clk.c
> > +++ b/drivers/clk/clk.c
> > @@ -10,6 +10,7 @@
> >   */
> > 
> >  #include <linux/clk-private.h>
> > +#include <linux/clk/clk-conf.h>
> >  #include <linux/module.h>
> >  #include <linux/mutex.h>
> >  #include <linux/spinlock.h>
> > @@ -2414,6 +2415,7 @@ int of_clk_add_provider(struct device_node *np,
> >                       void *data)
> >  {
> >       struct of_clk_provider *cp;
> > +     int ret;
> > 
> >       cp = kzalloc(sizeof(struct of_clk_provider), GFP_KERNEL);
> >       if (!cp)
> > @@ -2428,7 +2430,11 @@ int of_clk_add_provider(struct device_node *np,
> >       mutex_unlock(&of_clk_mutex);
> >       pr_debug("Added clock from %s\n", np->full_name);
> > 
> > -     return 0;
> > +     ret = of_clk_set_defaults(np, true);
> > +     if (ret < 0)
> > +             of_clk_del_provider(np);
> > +
> > +     return ret;
> >  }
> >  EXPORT_SYMBOL_GPL(of_clk_add_provider);
> > 
> > @@ -2605,7 +2611,10 @@ void __init of_clk_init(const struct of_device_id *matches)
> >               list_for_each_entry_safe(clk_provider, next,
> >                                       &clk_provider_list, node) {
> >                       if (force || parent_ready(clk_provider->np)) {
> > +
> >                               clk_provider->clk_init_cb(clk_provider->np);
> > +                             of_clk_set_defaults(clk_provider->np, true);
> > +
> >                               list_del(&clk_provider->node);
> >                               kfree(clk_provider);
> >                               is_init_done = true;
> > @@ -2620,7 +2629,6 @@ void __init of_clk_init(const struct of_device_id *matches)
> >                */
> >               if (!is_init_done)
> >                       force = true;
> > -
> >       }
> >  }
> >  #endif
> > diff --git a/drivers/i2c/i2c-core.c b/drivers/i2c/i2c-core.c
> > index 7c7f4b8..66aa83b 100644
> > --- a/drivers/i2c/i2c-core.c
> > +++ b/drivers/i2c/i2c-core.c
> > @@ -42,6 +42,7 @@
> >  #include <linux/of.h>
> >  #include <linux/of_device.h>
> >  #include <linux/of_irq.h>
> > +#include <linux/clk/clk-conf.h>
> >  #include <linux/completion.h>
> >  #include <linux/hardirq.h>
> >  #include <linux/irqflags.h>
> > @@ -274,6 +275,10 @@ static int i2c_device_probe(struct device *dev)
> >                                       client->flags & I2C_CLIENT_WAKE);
> >       dev_dbg(dev, "probe\n");
> > 
> > +     status = of_clk_set_defaults(dev->of_node, false);
> > +     if (status < 0)
> > +             return status;
> > +
> >       acpi_dev_pm_attach(&client->dev, true);
> >       status = driver->probe(client, i2c_match_id(driver->id_table, client));
> >       if (status)
> > diff --git a/drivers/spi/spi.c b/drivers/spi/spi.c
> > index 939edf4..8374620 100644
> > --- a/drivers/spi/spi.c
> > +++ b/drivers/spi/spi.c
> > @@ -29,6 +29,7 @@
> >  #include <linux/mutex.h>
> >  #include <linux/of_device.h>
> >  #include <linux/of_irq.h>
> > +#include <linux/clk/clk-conf.h>
> >  #include <linux/slab.h>
> >  #include <linux/mod_devicetable.h>
> >  #include <linux/spi/spi.h>
> > @@ -259,6 +260,10 @@ static int spi_drv_probe(struct device *dev)
> >       const struct spi_driver         *sdrv = to_spi_driver(dev->driver);
> >       int ret;
> > 
> > +     ret = of_clk_set_defaults(dev->of_node, false);
> > +     if (ret)
> > +             return ret;
> > +
> >       acpi_dev_pm_attach(dev, true);
> >       ret = sdrv->probe(to_spi_device(dev));
> >       if (ret)
> > diff --git a/include/linux/clk/clk-conf.h b/include/linux/clk/clk-conf.h
> > new file mode 100644
> > index 0000000..f3050e1
> > --- /dev/null
> > +++ b/include/linux/clk/clk-conf.h
> > @@ -0,0 +1,20 @@
> > +/*
> > + * Copyright (C) 2014 Samsung Electronics Co., Ltd.
> > + * Sylwester Nawrocki <s.nawrocki@samsung.com>
> > + *
> > + * This program is free software; you can redistribute it and/or modify
> > + * it under the terms of the GNU General Public License version 2 as
> > + * published by the Free Software Foundation.
> > + */
> > +
> > +struct device_node;
> > +
> > +#if defined(CONFIG_OF) && defined(CONFIG_COMMON_CLK)
> > +int of_clk_set_defaults(struct device_node *node, bool clk_supplier);
> > +#else
> > +static inline int of_clk_set_defaults(struct device_node *node,
> > +                                   bool clk_supplier)
> > +{
> > +     return 0;
> > +}
> > +#endif
> > --
> > 1.7.9.5
--
To unsubscribe from this list: send the line "unsubscribe linux-i2c" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

On Fri, Jul 25, 2014 at 02:42:31PM -0700, Mike Turquette wrote:
> Quoting Sylwester Nawrocki (2014-07-03 10:25:53)
> > On 18/06/14 17:29, Sylwester Nawrocki wrote:
> > > This patch adds helper functions to configure clock parents and rates
> > > as specified through 'assigned-clock-parents', 'assigned-clock-rates'
> > > DT properties for a clock provider or clock consumer device.
> > > The helpers are now being called by the bus code for the platform, I2C
> > > and SPI busses, before the driver probing and also in the clock core
> > > after registration of a clock provider.
> > > 
> > > Signed-off-by: Sylwester Nawrocki <s.nawrocki@samsung.com>
> > > Acked-by: Kyungmin Park <kyungmin.park@samsung.com>
> > 
> > Could someone please take a look and review that patch ?
> > Any further suggestions, ACKs/NAKs ?
> 
> Patch looks good to me. I'm happy to take it.
> 
> > 
> > I would appreciate a DT, SPI or the I2C maintainer opinions.
> 
> Yes, Acks from SPI and I2C maintainers would be good. I might need to
> drop those parts of this patch if they don't come through :-(

For the I2C part:

Acked-by: Wolfram Sang <wsa@the-dreams.de>

On Fri, Jul 25, 2014 at 02:42:31PM -0700, Mike Turquette wrote:
> Quoting Sylwester Nawrocki (2014-07-03 10:25:53)

> > I would appreciate a DT, SPI or the I2C maintainer opinions.

> Yes, Acks from SPI and I2C maintainers would be good. I might need to
> drop those parts of this patch if they don't come through :-(

So, I just noticed this.  The reason I didn't see this earlier is that
it was buried at the end of a large clock API patch rather than a split
out patch for the SPI subsystem so it fell towards the bottom of
my review queue.  There seems to be no dependency on adding the feature
as part of one commit so it should really have been split out.

Please don't assume that people are going to look in detail at patches
that aren't obviously for them - I know I end up ignoring a lot of what
I get sent since I get CCed on a lot of random things for no apparent
reason (or get tediously large numbers of resends of things that are
tangentially related), I imagine others do the same.  If you're doing
that please draw attention to it, splitting out per subsystem is the
best way of doing that.

This is particularly unfortunate since I am actually a bit confused as
to why we need to open code this for every bus rather than doing it in
the driver core, there doesn't seem to be anything at all bus specific
going on here.  Why can't we just call this from the driver core?
Having the feature work for some buses and not others is going to get
old fast.

It's also not clear to me why we're passing false to of_clk_set_defaults()
when we do call it - it's quite common for I2C and SPI devices to be
clock providers and have clock trees.  As far as I can tell the
reasoning is that what's actually going on here is that this is actually
a mechanism to defer the initialisation to adding of the clock providers
in which case contrary to the documentation for the function it's
actually about device registration.  Is that right?

On 13/08/14 22:34, Mark Brown wrote:
> On Fri, Jul 25, 2014 at 02:42:31PM -0700, Mike Turquette wrote:
>> Quoting Sylwester Nawrocki (2014-07-03 10:25:53)
> 
>>> I would appreciate a DT, SPI or the I2C maintainer opinions.
> 
>> Yes, Acks from SPI and I2C maintainers would be good. I might need to
>> drop those parts of this patch if they don't come through :-(
> 
> So, I just noticed this.  The reason I didn't see this earlier is that
> it was buried at the end of a large clock API patch rather than a split
> out patch for the SPI subsystem so it fell towards the bottom of
> my review queue.  There seems to be no dependency on adding the feature
> as part of one commit so it should really have been split out.
> 
> Please don't assume that people are going to look in detail at patches
> that aren't obviously for them - I know I end up ignoring a lot of what
> I get sent since I get CCed on a lot of random things for no apparent
> reason (or get tediously large numbers of resends of things that are
> tangentially related), I imagine others do the same.  If you're doing
> that please draw attention to it, splitting out per subsystem is the
> best way of doing that.

OK, will try to remember and improve things in future.

> This is particularly unfortunate since I am actually a bit confused as
> to why we need to open code this for every bus rather than doing it in
> the driver core, there doesn't seem to be anything at all bus specific
> going on here.  Why can't we just call this from the driver core?

I had it done in the driver core in first versions of this patch, but
then I learnt it is not something Greg would be fond of, so I went for
this method. In fact, clocks configuration is not something which
applies to each and every device, thus putting it in the bus code might
not be that bad after all.

> Having the feature work for some buses and not others is going to get
> old fast.
> 
> It's also not clear to me why we're passing false to of_clk_set_defaults()
> when we do call it - it's quite common for I2C and SPI devices to be
> clock providers and have clock trees.  As far as I can tell the
> reasoning is that what's actually going on here is that this is actually
> a mechanism to defer the initialisation to adding of the clock providers
> in which case contrary to the documentation for the function it's
> actually about device registration.  Is that right?

The flag is there to tell if clocks supplied by a particular device should
also be configured. The function is being called with the flag set right
after a clock provider registration in the clock core, this is where the
take care of things when an (I2C, SPI, etc.) device is a clock provider.
Obviously attempting to configure clocks which were not yet registered
before a device's driver probe() call would fail. The flag is there to
defer configuration of clocks after they are actually registered with
the clk core, not only parsable from the device tree.

Documentation of the function might be indeed a bit confusing :/
I'll consider changing the sentence:

"The @clk_supplier argument
 should be set to true if @node may be also a clock supplier of any clock
 listed in its 'assigned-clocks' or 'assigned-clock-parents' properties."

to

 "The @clk_supplier argument
  should be set to true if @node is or may be a clock supplier of any
  clock listed in its 'assigned-clocks' or 'assigned-clock-parents'
  properties and such clocks are also to be configured."

--
Regards,
Sylwester
--
To unsubscribe from this list: send the line "unsubscribe linux-i2c" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html