All of lore.kernel.org
 help / color / mirror / Atom feed
From: "Artur Świgoń" <a.swigon@samsung.com>
To: Georgi Djakov <georgi.djakov@linaro.org>,
	devicetree@vger.kernel.org, linux-arm-kernel@lists.infradead.org,
	linux-samsung-soc@vger.kernel.org, linux-kernel@vger.kernel.org,
	linux-pm@vger.kernel.org, dri-devel@lists.freedesktop.org
Cc: cw00.choi@samsung.com, myungjoo.ham@samsung.com,
	inki.dae@samsung.com, sw0312.kim@samsung.com,
	leonard.crestez@nxp.com, m.szyprowski@samsung.com,
	b.zolnierkie@samsung.com, krzk@kernel.org
Subject: Re: [RFC PATCH v3 5/7] devfreq: exynos-bus: Add interconnect functionality to exynos-bus
Date: Fri, 24 Jan 2020 12:22:14 +0100	[thread overview]
Message-ID: <0de33b1ae7c52270d11c606ecccc9af5f44d0421.camel@samsung.com> (raw)
In-Reply-To: <15795421-db12-8466-bb7e-688e6625cf4a@linaro.org>

Hi Georgi,

On Wed, 2020-01-22 at 19:02 +0200, Georgi Djakov wrote:
> Hi Artur,
> 
> On 12/20/19 13:56, Artur Świgoń wrote:
> > This patch adds interconnect functionality to the exynos-bus devfreq
> > driver.
> > 
> > The SoC topology is a graph (or, more specifically, a tree) and its
> > edges are specified using the 'exynos,interconnect-parent-node' in the
> > DT. Due to unspecified relative probing order, -EPROBE_DEFER may be
> > propagated to ensure that the parent is probed before its children.
> > 
> > Each bus is now an interconnect provider and an interconnect node as well
> > (cf. Documentation/interconnect/interconnect.rst), i.e. every bus registers
> > itself as a node. Node IDs are not hardcoded but rather assigned at
> 
> Just to note that usually the provider consists of multiple nodes and each node
> represents a single master or slave port on the AXI bus for example. I am not
> sure whether this represents correctly the Exynos hardware, so it's up to
> you.
> 
> > runtime, in probing order (subject to the above-mentioned exception
> > regarding relative order). This approach allows for using this driver with
> > various Exynos SoCs.
> 
> This sounds good. I am wondering whether such dynamic probing would be useful
> for other platforms too. Then maybe it would make sense to even have a common DT
> property, but we will see.
> 
> Is this going to be used only together with devfreq?

Yes, this functions solely as an extension to devfreq, hence the slightly
unusual architecture (one icc_provider/icc_node per devfreq).

(Compared to a singleton icc_provider, this approach yields less code with
a very simple xlate()).

With exactly one icc_node for every devfreq device, I think I will actually
reuse the devfreq ID (as seen in the device name, e.g. the "3" in "devfreq3")
for the node ID. The devfreq framework already does the dynamic numbering
thing that I do in this patch using IDR.

> > Frequencies requested via the interconnect API for a given node are
> > propagated to devfreq using dev_pm_qos_update_request(). Please note that
> > it is not an error when CONFIG_INTERCONNECT is 'n', in which case all
> > interconnect API functions are no-op.
> 
> How about the case where CONFIG_INTERCONNECT=m. Looks like the build will fail
> if CONFIG_ARM_EXYNOS_BUS_DEVFREQ=y, so this dependency should be expressed in
> Kconfig.

I think adding:
	depends on INTERCONNECT || !INTERCONNECT

under ARM_EXYNOS_BUS_DEVFREQ does the trick.

> > 
> > Signed-off-by: Artur Świgoń <a.swigon@samsung.com>
> > ---
> >  drivers/devfreq/exynos-bus.c | 144 +++++++++++++++++++++++++++++++++++
> >  1 file changed, 144 insertions(+)
> > 
> > diff --git a/drivers/devfreq/exynos-bus.c b/drivers/devfreq/exynos-bus.c
> > index 9fdb188915e8..694a9581dcdb 100644
> > --- a/drivers/devfreq/exynos-bus.c
> > +++ b/drivers/devfreq/exynos-bus.c
> > @@ -14,14 +14,19 @@
> >  #include <linux/devfreq-event.h>
> >  #include <linux/device.h>
> >  #include <linux/export.h>
> > +#include <linux/idr.h>
> > +#include <linux/interconnect-provider.h>
> >  #include <linux/module.h>
> >  #include <linux/of.h>
> >  #include <linux/pm_opp.h>
> > +#include <linux/pm_qos.h>
> >  #include <linux/platform_device.h>
> >  #include <linux/regulator/consumer.h>
> >  
> >  #define DEFAULT_SATURATION_RATIO	40
> >  
> > +#define kbps_to_khz(x) ((x) / 8)
> > +
> >  struct exynos_bus {
> >  	struct device *dev;
> >  
> > @@ -35,6 +40,12 @@ struct exynos_bus {
> >  	struct opp_table *opp_table;
> >  	struct clk *clk;
> >  	unsigned int ratio;
> > +
> > +	/* One provider per bus, one node per provider */
> > +	struct icc_provider provider;
> > +	struct icc_node *node;
> > +
> > +	struct dev_pm_qos_request qos_req;
> >  };
> >  
> >  /*
> > @@ -205,6 +216,39 @@ static void exynos_bus_passive_exit(struct device *dev)
> >  	clk_disable_unprepare(bus->clk);
> >  }
> >  
> > +static int exynos_bus_icc_set(struct icc_node *src, struct icc_node *dst)
> > +{
> > +	struct exynos_bus *src_bus = src->data, *dst_bus = dst->data;
> > +	s32 src_freq = kbps_to_khz(src->avg_bw);
> > +	s32 dst_freq = kbps_to_khz(dst->avg_bw);
> > +	int ret;
> > +
> > +	ret = dev_pm_qos_update_request(&src_bus->qos_req, src_freq);
> > +	if (ret < 0) {
> > +		dev_err(src_bus->dev, "failed to update PM QoS request");
> > +		return ret;
> > +	}
> > +
> > +	ret = dev_pm_qos_update_request(&dst_bus->qos_req, dst_freq);
> > +	if (ret < 0) {
> > +		dev_err(dst_bus->dev, "failed to update PM QoS request");
> > +		return ret;
> > +	}
> > +
> > +	return 0;
> > +}
> > +
> > +static struct icc_node *exynos_bus_icc_xlate(struct of_phandle_args *spec,
> > +					     void *data)
> > +{
> > +	struct exynos_bus *bus = data;
> > +
> > +	if (spec->np != bus->dev->of_node)
> > +		return ERR_PTR(-EINVAL);
> > +
> > +	return bus->node;
> > +}
> > +
> >  static int exynos_bus_parent_parse_of(struct device_node *np,
> >  					struct exynos_bus *bus)
> >  {
> > @@ -419,6 +463,96 @@ static int exynos_bus_profile_init_passive(struct exynos_bus *bus,
> >  	return 0;
> >  }
> >  
> > +static struct icc_node *exynos_bus_icc_get_parent(struct exynos_bus *bus)
> > +{
> > +	struct device_node *np = bus->dev->of_node;
> > +	struct of_phandle_args args;
> > +	int num, ret;
> > +
> > +	num = of_count_phandle_with_args(np, "exynos,interconnect-parent-node",
> > +					"#interconnect-cells");
> > +	if (num != 1)
> > +		return NULL; /* parent nodes are optional */
> > +
> > +	ret = of_parse_phandle_with_args(np, "exynos,interconnect-parent-node",
> > +					"#interconnect-cells", 0, &args);
> > +	if (ret < 0)
> > +		return ERR_PTR(ret);
> > +
> > +	of_node_put(args.np);
> > +
> > +	return of_icc_get_from_provider(&args);
> > +}
> > +
> > +static int exynos_bus_icc_init(struct exynos_bus *bus)
> > +{
> > +	static DEFINE_IDA(ida);
> > +
> > +	struct device *dev = bus->dev;
> > +	struct icc_provider *provider = &bus->provider;
> > +	struct icc_node *node, *parent_node;
> > +	int id, ret;
> > +
> > +	/* Initialize the interconnect provider */
> > +	provider->set = exynos_bus_icc_set;
> > +	provider->aggregate = icc_std_aggregate;
> > +	provider->xlate = exynos_bus_icc_xlate;
> > +	provider->dev = dev;
> > +	provider->inter_set = true;
> > +	provider->data = bus;
> > +
> > +	ret = icc_provider_add(provider);
> > +	if (ret < 0)
> > +		return ret;
> > +
> > +	ret = id = ida_alloc(&ida, GFP_KERNEL);
> > +	if (ret < 0)
> > +		goto err_id;
> > +
> > +	node = icc_node_create(id);
> > +	if (IS_ERR(node)) {
> > +		ret = PTR_ERR(node);
> > +		goto err_node;
> > +	}
> > +
> > +	bus->node = node;
> > +	node->name = dev->of_node->name;
> > +	node->data = bus;
> > +	icc_node_add(node, provider);
> > +
> > +	parent_node = exynos_bus_icc_get_parent(bus);
> > +	if (IS_ERR(parent_node)) {
> > +		ret = PTR_ERR(parent_node);
> > +		goto err_parent;
> > +	}
> > +
> > +	if (parent_node) {
> > +		ret = icc_link_create(node, parent_node->id);
> > +		if (ret < 0)
> > +			goto err_parent;
> > +	}
> > +
> > +	ret = dev_pm_qos_add_request(bus->devfreq->dev.parent, &bus->qos_req,
> > +					DEV_PM_QOS_MIN_FREQUENCY, 0);
> > +	if (ret < 0)
> > +		goto err_request;
> > +
> > +	return 0;
> > +
> > +err_request:
> > +	if (parent_node)
> > +		icc_link_destroy(node, parent_node);
> > +err_parent:
> > +	icc_node_del(node);
> > +	icc_node_destroy(id);
> > +err_node:
> > +	ida_free(&ida, id);
> > +err_id:
> > +	icc_provider_del(provider);
> > +
> > +	return ret;
> > +}
> > +
> >  static int exynos_bus_probe(struct platform_device *pdev)
> >  {
> >  	struct device *dev = &pdev->dev;
> > @@ -468,6 +602,16 @@ static int exynos_bus_probe(struct platform_device *pdev)
> >  	if (ret < 0)
> >  		goto err;
> >  
> > +	/*
> > +	 * Initialize interconnect provider. A return value of -ENOTSUPP means
> > +	 * that CONFIG_INTERCONNECT is disabled.
> > +	 */
> > +	ret = exynos_bus_icc_init(bus);
> > +	if (ret < 0 && ret != -ENOTSUPP) {
> > +		dev_err(dev, "failed to initialize the interconnect provider");
> > +		goto err;
> > +	}
> > +
> >  	max_state = bus->devfreq->profile->max_state;
> >  	min_freq = (bus->devfreq->profile->freq_table[0] / 1000);
> >  	max_freq = (bus->devfreq->profile->freq_table[max_state - 1] / 1000);
> > 
> 
> 
-- 
Artur Świgoń
Samsung R&D Institute Poland
Samsung Electronics



WARNING: multiple messages have this Message-ID (diff)
From: "Artur Świgoń" <a.swigon@samsung.com>
To: Georgi Djakov <georgi.djakov@linaro.org>,
	devicetree@vger.kernel.org, linux-arm-kernel@lists.infradead.org,
	linux-samsung-soc@vger.kernel.org,  linux-kernel@vger.kernel.org,
	linux-pm@vger.kernel.org,  dri-devel@lists.freedesktop.org
Cc: b.zolnierkie@samsung.com, sw0312.kim@samsung.com,
	krzk@kernel.org, inki.dae@samsung.com, cw00.choi@samsung.com,
	myungjoo.ham@samsung.com, leonard.crestez@nxp.com,
	m.szyprowski@samsung.com
Subject: Re: [RFC PATCH v3 5/7] devfreq: exynos-bus: Add interconnect functionality to exynos-bus
Date: Fri, 24 Jan 2020 12:22:14 +0100	[thread overview]
Message-ID: <0de33b1ae7c52270d11c606ecccc9af5f44d0421.camel@samsung.com> (raw)
In-Reply-To: <15795421-db12-8466-bb7e-688e6625cf4a@linaro.org>

Hi Georgi,

On Wed, 2020-01-22 at 19:02 +0200, Georgi Djakov wrote:
> Hi Artur,
> 
> On 12/20/19 13:56, Artur Świgoń wrote:
> > This patch adds interconnect functionality to the exynos-bus devfreq
> > driver.
> > 
> > The SoC topology is a graph (or, more specifically, a tree) and its
> > edges are specified using the 'exynos,interconnect-parent-node' in the
> > DT. Due to unspecified relative probing order, -EPROBE_DEFER may be
> > propagated to ensure that the parent is probed before its children.
> > 
> > Each bus is now an interconnect provider and an interconnect node as well
> > (cf. Documentation/interconnect/interconnect.rst), i.e. every bus registers
> > itself as a node. Node IDs are not hardcoded but rather assigned at
> 
> Just to note that usually the provider consists of multiple nodes and each node
> represents a single master or slave port on the AXI bus for example. I am not
> sure whether this represents correctly the Exynos hardware, so it's up to
> you.
> 
> > runtime, in probing order (subject to the above-mentioned exception
> > regarding relative order). This approach allows for using this driver with
> > various Exynos SoCs.
> 
> This sounds good. I am wondering whether such dynamic probing would be useful
> for other platforms too. Then maybe it would make sense to even have a common DT
> property, but we will see.
> 
> Is this going to be used only together with devfreq?

Yes, this functions solely as an extension to devfreq, hence the slightly
unusual architecture (one icc_provider/icc_node per devfreq).

(Compared to a singleton icc_provider, this approach yields less code with
a very simple xlate()).

With exactly one icc_node for every devfreq device, I think I will actually
reuse the devfreq ID (as seen in the device name, e.g. the "3" in "devfreq3")
for the node ID. The devfreq framework already does the dynamic numbering
thing that I do in this patch using IDR.

> > Frequencies requested via the interconnect API for a given node are
> > propagated to devfreq using dev_pm_qos_update_request(). Please note that
> > it is not an error when CONFIG_INTERCONNECT is 'n', in which case all
> > interconnect API functions are no-op.
> 
> How about the case where CONFIG_INTERCONNECT=m. Looks like the build will fail
> if CONFIG_ARM_EXYNOS_BUS_DEVFREQ=y, so this dependency should be expressed in
> Kconfig.

I think adding:
	depends on INTERCONNECT || !INTERCONNECT

under ARM_EXYNOS_BUS_DEVFREQ does the trick.

> > 
> > Signed-off-by: Artur Świgoń <a.swigon@samsung.com>
> > ---
> >  drivers/devfreq/exynos-bus.c | 144 +++++++++++++++++++++++++++++++++++
> >  1 file changed, 144 insertions(+)
> > 
> > diff --git a/drivers/devfreq/exynos-bus.c b/drivers/devfreq/exynos-bus.c
> > index 9fdb188915e8..694a9581dcdb 100644
> > --- a/drivers/devfreq/exynos-bus.c
> > +++ b/drivers/devfreq/exynos-bus.c
> > @@ -14,14 +14,19 @@
> >  #include <linux/devfreq-event.h>
> >  #include <linux/device.h>
> >  #include <linux/export.h>
> > +#include <linux/idr.h>
> > +#include <linux/interconnect-provider.h>
> >  #include <linux/module.h>
> >  #include <linux/of.h>
> >  #include <linux/pm_opp.h>
> > +#include <linux/pm_qos.h>
> >  #include <linux/platform_device.h>
> >  #include <linux/regulator/consumer.h>
> >  
> >  #define DEFAULT_SATURATION_RATIO	40
> >  
> > +#define kbps_to_khz(x) ((x) / 8)
> > +
> >  struct exynos_bus {
> >  	struct device *dev;
> >  
> > @@ -35,6 +40,12 @@ struct exynos_bus {
> >  	struct opp_table *opp_table;
> >  	struct clk *clk;
> >  	unsigned int ratio;
> > +
> > +	/* One provider per bus, one node per provider */
> > +	struct icc_provider provider;
> > +	struct icc_node *node;
> > +
> > +	struct dev_pm_qos_request qos_req;
> >  };
> >  
> >  /*
> > @@ -205,6 +216,39 @@ static void exynos_bus_passive_exit(struct device *dev)
> >  	clk_disable_unprepare(bus->clk);
> >  }
> >  
> > +static int exynos_bus_icc_set(struct icc_node *src, struct icc_node *dst)
> > +{
> > +	struct exynos_bus *src_bus = src->data, *dst_bus = dst->data;
> > +	s32 src_freq = kbps_to_khz(src->avg_bw);
> > +	s32 dst_freq = kbps_to_khz(dst->avg_bw);
> > +	int ret;
> > +
> > +	ret = dev_pm_qos_update_request(&src_bus->qos_req, src_freq);
> > +	if (ret < 0) {
> > +		dev_err(src_bus->dev, "failed to update PM QoS request");
> > +		return ret;
> > +	}
> > +
> > +	ret = dev_pm_qos_update_request(&dst_bus->qos_req, dst_freq);
> > +	if (ret < 0) {
> > +		dev_err(dst_bus->dev, "failed to update PM QoS request");
> > +		return ret;
> > +	}
> > +
> > +	return 0;
> > +}
> > +
> > +static struct icc_node *exynos_bus_icc_xlate(struct of_phandle_args *spec,
> > +					     void *data)
> > +{
> > +	struct exynos_bus *bus = data;
> > +
> > +	if (spec->np != bus->dev->of_node)
> > +		return ERR_PTR(-EINVAL);
> > +
> > +	return bus->node;
> > +}
> > +
> >  static int exynos_bus_parent_parse_of(struct device_node *np,
> >  					struct exynos_bus *bus)
> >  {
> > @@ -419,6 +463,96 @@ static int exynos_bus_profile_init_passive(struct exynos_bus *bus,
> >  	return 0;
> >  }
> >  
> > +static struct icc_node *exynos_bus_icc_get_parent(struct exynos_bus *bus)
> > +{
> > +	struct device_node *np = bus->dev->of_node;
> > +	struct of_phandle_args args;
> > +	int num, ret;
> > +
> > +	num = of_count_phandle_with_args(np, "exynos,interconnect-parent-node",
> > +					"#interconnect-cells");
> > +	if (num != 1)
> > +		return NULL; /* parent nodes are optional */
> > +
> > +	ret = of_parse_phandle_with_args(np, "exynos,interconnect-parent-node",
> > +					"#interconnect-cells", 0, &args);
> > +	if (ret < 0)
> > +		return ERR_PTR(ret);
> > +
> > +	of_node_put(args.np);
> > +
> > +	return of_icc_get_from_provider(&args);
> > +}
> > +
> > +static int exynos_bus_icc_init(struct exynos_bus *bus)
> > +{
> > +	static DEFINE_IDA(ida);
> > +
> > +	struct device *dev = bus->dev;
> > +	struct icc_provider *provider = &bus->provider;
> > +	struct icc_node *node, *parent_node;
> > +	int id, ret;
> > +
> > +	/* Initialize the interconnect provider */
> > +	provider->set = exynos_bus_icc_set;
> > +	provider->aggregate = icc_std_aggregate;
> > +	provider->xlate = exynos_bus_icc_xlate;
> > +	provider->dev = dev;
> > +	provider->inter_set = true;
> > +	provider->data = bus;
> > +
> > +	ret = icc_provider_add(provider);
> > +	if (ret < 0)
> > +		return ret;
> > +
> > +	ret = id = ida_alloc(&ida, GFP_KERNEL);
> > +	if (ret < 0)
> > +		goto err_id;
> > +
> > +	node = icc_node_create(id);
> > +	if (IS_ERR(node)) {
> > +		ret = PTR_ERR(node);
> > +		goto err_node;
> > +	}
> > +
> > +	bus->node = node;
> > +	node->name = dev->of_node->name;
> > +	node->data = bus;
> > +	icc_node_add(node, provider);
> > +
> > +	parent_node = exynos_bus_icc_get_parent(bus);
> > +	if (IS_ERR(parent_node)) {
> > +		ret = PTR_ERR(parent_node);
> > +		goto err_parent;
> > +	}
> > +
> > +	if (parent_node) {
> > +		ret = icc_link_create(node, parent_node->id);
> > +		if (ret < 0)
> > +			goto err_parent;
> > +	}
> > +
> > +	ret = dev_pm_qos_add_request(bus->devfreq->dev.parent, &bus->qos_req,
> > +					DEV_PM_QOS_MIN_FREQUENCY, 0);
> > +	if (ret < 0)
> > +		goto err_request;
> > +
> > +	return 0;
> > +
> > +err_request:
> > +	if (parent_node)
> > +		icc_link_destroy(node, parent_node);
> > +err_parent:
> > +	icc_node_del(node);
> > +	icc_node_destroy(id);
> > +err_node:
> > +	ida_free(&ida, id);
> > +err_id:
> > +	icc_provider_del(provider);
> > +
> > +	return ret;
> > +}
> > +
> >  static int exynos_bus_probe(struct platform_device *pdev)
> >  {
> >  	struct device *dev = &pdev->dev;
> > @@ -468,6 +602,16 @@ static int exynos_bus_probe(struct platform_device *pdev)
> >  	if (ret < 0)
> >  		goto err;
> >  
> > +	/*
> > +	 * Initialize interconnect provider. A return value of -ENOTSUPP means
> > +	 * that CONFIG_INTERCONNECT is disabled.
> > +	 */
> > +	ret = exynos_bus_icc_init(bus);
> > +	if (ret < 0 && ret != -ENOTSUPP) {
> > +		dev_err(dev, "failed to initialize the interconnect provider");
> > +		goto err;
> > +	}
> > +
> >  	max_state = bus->devfreq->profile->max_state;
> >  	min_freq = (bus->devfreq->profile->freq_table[0] / 1000);
> >  	max_freq = (bus->devfreq->profile->freq_table[max_state - 1] / 1000);
> > 
> 
> 
-- 
Artur Świgoń
Samsung R&D Institute Poland
Samsung Electronics



_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

WARNING: multiple messages have this Message-ID (diff)
From: "Artur Świgoń" <a.swigon@samsung.com>
To: Georgi Djakov <georgi.djakov@linaro.org>,
	devicetree@vger.kernel.org, linux-arm-kernel@lists.infradead.org,
	linux-samsung-soc@vger.kernel.org,  linux-kernel@vger.kernel.org,
	linux-pm@vger.kernel.org,  dri-devel@lists.freedesktop.org
Cc: b.zolnierkie@samsung.com, sw0312.kim@samsung.com,
	krzk@kernel.org, cw00.choi@samsung.com, myungjoo.ham@samsung.com,
	leonard.crestez@nxp.com, m.szyprowski@samsung.com
Subject: Re: [RFC PATCH v3 5/7] devfreq: exynos-bus: Add interconnect functionality to exynos-bus
Date: Fri, 24 Jan 2020 12:22:14 +0100	[thread overview]
Message-ID: <0de33b1ae7c52270d11c606ecccc9af5f44d0421.camel@samsung.com> (raw)
In-Reply-To: <15795421-db12-8466-bb7e-688e6625cf4a@linaro.org>

Hi Georgi,

On Wed, 2020-01-22 at 19:02 +0200, Georgi Djakov wrote:
> Hi Artur,
> 
> On 12/20/19 13:56, Artur Świgoń wrote:
> > This patch adds interconnect functionality to the exynos-bus devfreq
> > driver.
> > 
> > The SoC topology is a graph (or, more specifically, a tree) and its
> > edges are specified using the 'exynos,interconnect-parent-node' in the
> > DT. Due to unspecified relative probing order, -EPROBE_DEFER may be
> > propagated to ensure that the parent is probed before its children.
> > 
> > Each bus is now an interconnect provider and an interconnect node as well
> > (cf. Documentation/interconnect/interconnect.rst), i.e. every bus registers
> > itself as a node. Node IDs are not hardcoded but rather assigned at
> 
> Just to note that usually the provider consists of multiple nodes and each node
> represents a single master or slave port on the AXI bus for example. I am not
> sure whether this represents correctly the Exynos hardware, so it's up to
> you.
> 
> > runtime, in probing order (subject to the above-mentioned exception
> > regarding relative order). This approach allows for using this driver with
> > various Exynos SoCs.
> 
> This sounds good. I am wondering whether such dynamic probing would be useful
> for other platforms too. Then maybe it would make sense to even have a common DT
> property, but we will see.
> 
> Is this going to be used only together with devfreq?

Yes, this functions solely as an extension to devfreq, hence the slightly
unusual architecture (one icc_provider/icc_node per devfreq).

(Compared to a singleton icc_provider, this approach yields less code with
a very simple xlate()).

With exactly one icc_node for every devfreq device, I think I will actually
reuse the devfreq ID (as seen in the device name, e.g. the "3" in "devfreq3")
for the node ID. The devfreq framework already does the dynamic numbering
thing that I do in this patch using IDR.

> > Frequencies requested via the interconnect API for a given node are
> > propagated to devfreq using dev_pm_qos_update_request(). Please note that
> > it is not an error when CONFIG_INTERCONNECT is 'n', in which case all
> > interconnect API functions are no-op.
> 
> How about the case where CONFIG_INTERCONNECT=m. Looks like the build will fail
> if CONFIG_ARM_EXYNOS_BUS_DEVFREQ=y, so this dependency should be expressed in
> Kconfig.

I think adding:
	depends on INTERCONNECT || !INTERCONNECT

under ARM_EXYNOS_BUS_DEVFREQ does the trick.

> > 
> > Signed-off-by: Artur Świgoń <a.swigon@samsung.com>
> > ---
> >  drivers/devfreq/exynos-bus.c | 144 +++++++++++++++++++++++++++++++++++
> >  1 file changed, 144 insertions(+)
> > 
> > diff --git a/drivers/devfreq/exynos-bus.c b/drivers/devfreq/exynos-bus.c
> > index 9fdb188915e8..694a9581dcdb 100644
> > --- a/drivers/devfreq/exynos-bus.c
> > +++ b/drivers/devfreq/exynos-bus.c
> > @@ -14,14 +14,19 @@
> >  #include <linux/devfreq-event.h>
> >  #include <linux/device.h>
> >  #include <linux/export.h>
> > +#include <linux/idr.h>
> > +#include <linux/interconnect-provider.h>
> >  #include <linux/module.h>
> >  #include <linux/of.h>
> >  #include <linux/pm_opp.h>
> > +#include <linux/pm_qos.h>
> >  #include <linux/platform_device.h>
> >  #include <linux/regulator/consumer.h>
> >  
> >  #define DEFAULT_SATURATION_RATIO	40
> >  
> > +#define kbps_to_khz(x) ((x) / 8)
> > +
> >  struct exynos_bus {
> >  	struct device *dev;
> >  
> > @@ -35,6 +40,12 @@ struct exynos_bus {
> >  	struct opp_table *opp_table;
> >  	struct clk *clk;
> >  	unsigned int ratio;
> > +
> > +	/* One provider per bus, one node per provider */
> > +	struct icc_provider provider;
> > +	struct icc_node *node;
> > +
> > +	struct dev_pm_qos_request qos_req;
> >  };
> >  
> >  /*
> > @@ -205,6 +216,39 @@ static void exynos_bus_passive_exit(struct device *dev)
> >  	clk_disable_unprepare(bus->clk);
> >  }
> >  
> > +static int exynos_bus_icc_set(struct icc_node *src, struct icc_node *dst)
> > +{
> > +	struct exynos_bus *src_bus = src->data, *dst_bus = dst->data;
> > +	s32 src_freq = kbps_to_khz(src->avg_bw);
> > +	s32 dst_freq = kbps_to_khz(dst->avg_bw);
> > +	int ret;
> > +
> > +	ret = dev_pm_qos_update_request(&src_bus->qos_req, src_freq);
> > +	if (ret < 0) {
> > +		dev_err(src_bus->dev, "failed to update PM QoS request");
> > +		return ret;
> > +	}
> > +
> > +	ret = dev_pm_qos_update_request(&dst_bus->qos_req, dst_freq);
> > +	if (ret < 0) {
> > +		dev_err(dst_bus->dev, "failed to update PM QoS request");
> > +		return ret;
> > +	}
> > +
> > +	return 0;
> > +}
> > +
> > +static struct icc_node *exynos_bus_icc_xlate(struct of_phandle_args *spec,
> > +					     void *data)
> > +{
> > +	struct exynos_bus *bus = data;
> > +
> > +	if (spec->np != bus->dev->of_node)
> > +		return ERR_PTR(-EINVAL);
> > +
> > +	return bus->node;
> > +}
> > +
> >  static int exynos_bus_parent_parse_of(struct device_node *np,
> >  					struct exynos_bus *bus)
> >  {
> > @@ -419,6 +463,96 @@ static int exynos_bus_profile_init_passive(struct exynos_bus *bus,
> >  	return 0;
> >  }
> >  
> > +static struct icc_node *exynos_bus_icc_get_parent(struct exynos_bus *bus)
> > +{
> > +	struct device_node *np = bus->dev->of_node;
> > +	struct of_phandle_args args;
> > +	int num, ret;
> > +
> > +	num = of_count_phandle_with_args(np, "exynos,interconnect-parent-node",
> > +					"#interconnect-cells");
> > +	if (num != 1)
> > +		return NULL; /* parent nodes are optional */
> > +
> > +	ret = of_parse_phandle_with_args(np, "exynos,interconnect-parent-node",
> > +					"#interconnect-cells", 0, &args);
> > +	if (ret < 0)
> > +		return ERR_PTR(ret);
> > +
> > +	of_node_put(args.np);
> > +
> > +	return of_icc_get_from_provider(&args);
> > +}
> > +
> > +static int exynos_bus_icc_init(struct exynos_bus *bus)
> > +{
> > +	static DEFINE_IDA(ida);
> > +
> > +	struct device *dev = bus->dev;
> > +	struct icc_provider *provider = &bus->provider;
> > +	struct icc_node *node, *parent_node;
> > +	int id, ret;
> > +
> > +	/* Initialize the interconnect provider */
> > +	provider->set = exynos_bus_icc_set;
> > +	provider->aggregate = icc_std_aggregate;
> > +	provider->xlate = exynos_bus_icc_xlate;
> > +	provider->dev = dev;
> > +	provider->inter_set = true;
> > +	provider->data = bus;
> > +
> > +	ret = icc_provider_add(provider);
> > +	if (ret < 0)
> > +		return ret;
> > +
> > +	ret = id = ida_alloc(&ida, GFP_KERNEL);
> > +	if (ret < 0)
> > +		goto err_id;
> > +
> > +	node = icc_node_create(id);
> > +	if (IS_ERR(node)) {
> > +		ret = PTR_ERR(node);
> > +		goto err_node;
> > +	}
> > +
> > +	bus->node = node;
> > +	node->name = dev->of_node->name;
> > +	node->data = bus;
> > +	icc_node_add(node, provider);
> > +
> > +	parent_node = exynos_bus_icc_get_parent(bus);
> > +	if (IS_ERR(parent_node)) {
> > +		ret = PTR_ERR(parent_node);
> > +		goto err_parent;
> > +	}
> > +
> > +	if (parent_node) {
> > +		ret = icc_link_create(node, parent_node->id);
> > +		if (ret < 0)
> > +			goto err_parent;
> > +	}
> > +
> > +	ret = dev_pm_qos_add_request(bus->devfreq->dev.parent, &bus->qos_req,
> > +					DEV_PM_QOS_MIN_FREQUENCY, 0);
> > +	if (ret < 0)
> > +		goto err_request;
> > +
> > +	return 0;
> > +
> > +err_request:
> > +	if (parent_node)
> > +		icc_link_destroy(node, parent_node);
> > +err_parent:
> > +	icc_node_del(node);
> > +	icc_node_destroy(id);
> > +err_node:
> > +	ida_free(&ida, id);
> > +err_id:
> > +	icc_provider_del(provider);
> > +
> > +	return ret;
> > +}
> > +
> >  static int exynos_bus_probe(struct platform_device *pdev)
> >  {
> >  	struct device *dev = &pdev->dev;
> > @@ -468,6 +602,16 @@ static int exynos_bus_probe(struct platform_device *pdev)
> >  	if (ret < 0)
> >  		goto err;
> >  
> > +	/*
> > +	 * Initialize interconnect provider. A return value of -ENOTSUPP means
> > +	 * that CONFIG_INTERCONNECT is disabled.
> > +	 */
> > +	ret = exynos_bus_icc_init(bus);
> > +	if (ret < 0 && ret != -ENOTSUPP) {
> > +		dev_err(dev, "failed to initialize the interconnect provider");
> > +		goto err;
> > +	}
> > +
> >  	max_state = bus->devfreq->profile->max_state;
> >  	min_freq = (bus->devfreq->profile->freq_table[0] / 1000);
> >  	max_freq = (bus->devfreq->profile->freq_table[max_state - 1] / 1000);
> > 
> 
> 
-- 
Artur Świgoń
Samsung R&D Institute Poland
Samsung Electronics


_______________________________________________
dri-devel mailing list
dri-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/dri-devel

  reply	other threads:[~2020-01-24 11:22 UTC|newest]

Thread overview: 93+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <CGME20191220120140eucas1p14ad33c20882f8f48e02337ea16754d91@eucas1p1.samsung.com>
2019-12-20 11:56 ` [RFC PATCH v3 0/7] PM / devfreq: Simple QoS for exynos-bus using interconnect Artur Świgoń
2019-12-20 11:56   ` Artur Świgoń
2019-12-20 11:56   ` Artur Świgoń
     [not found]   ` <CGME20191220120141eucas1p11f5fa9d76d17e80e06199cb7a911c482@eucas1p1.samsung.com>
2019-12-20 11:56     ` [RFC PATCH v3 1/7] interconnect: Export of_icc_get_from_provider() Artur Świgoń
2019-12-20 11:56       ` Artur Świgoń
2019-12-20 11:56       ` Artur Świgoń
     [not found]   ` <CGME20191220120142eucas1p1f43c7a862d9c0faa72e14b21d7d697e9@eucas1p1.samsung.com>
2019-12-20 11:56     ` [RFC PATCH v3 2/7] interconnect: Relax requirement in of_icc_get_from_provider() Artur Świgoń
2019-12-20 11:56       ` Artur Świgoń
2019-12-20 11:56       ` Artur Świgoń
2019-12-21 17:20       ` Chanwoo Choi
2019-12-21 17:20         ` Chanwoo Choi
2019-12-21 17:20         ` Chanwoo Choi
     [not found]   ` <CGME20191220120143eucas1p1c9b01ae8c2e4ecd70423ef9d8001536f@eucas1p1.samsung.com>
2019-12-20 11:56     ` [RFC PATCH v3 3/7] interconnect: Allow inter-provider pairs to be configured Artur Świgoń
2019-12-20 11:56       ` Artur Świgoń
2019-12-20 11:56       ` Artur Świgoń
2019-12-22 17:08       ` Chanwoo Choi
2019-12-22 17:08         ` Chanwoo Choi
2019-12-22 17:08         ` Chanwoo Choi
     [not found]   ` <CGME20191220120144eucas1p119ececf161a6d45a6a194e432bbbd1f9@eucas1p1.samsung.com>
2019-12-20 11:56     ` [RFC PATCH v3 4/7] arm: dts: exynos: Add interconnect bindings for Exynos4412 Artur Świgoń
2019-12-20 11:56       ` Artur Świgoń
2019-12-20 11:56       ` Artur Świgoń
2019-12-21 20:00       ` Chanwoo Choi
2019-12-21 20:00         ` Chanwoo Choi
2019-12-21 20:00         ` Chanwoo Choi
2019-12-30 15:44       ` Krzysztof Kozlowski
2019-12-30 15:44         ` Krzysztof Kozlowski
2019-12-30 15:44         ` Krzysztof Kozlowski
2019-12-31  7:18         ` Artur Świgoń
2019-12-31  7:18           ` Artur Świgoń
2019-12-31  7:18           ` Artur Świgoń
2019-12-31  9:22           ` Krzysztof Kozlowski
2019-12-31  9:22             ` Krzysztof Kozlowski
2019-12-31  9:22             ` Krzysztof Kozlowski
2019-12-31  9:41             ` Artur Świgoń
2019-12-31  9:41               ` Artur Świgoń
2019-12-31  9:41               ` Artur Świgoń
2019-12-31 10:02               ` Krzysztof Kozlowski
2019-12-31 10:02                 ` Krzysztof Kozlowski
2019-12-31 10:02                 ` Krzysztof Kozlowski
2019-12-31 10:23                 ` Artur Świgoń
2019-12-31 10:23                   ` Artur Świgoń
2019-12-31 10:23                   ` Artur Świgoń
2019-12-31 10:38                   ` Krzysztof Kozlowski
2019-12-31 10:38                     ` Krzysztof Kozlowski
2019-12-31 10:38                     ` Krzysztof Kozlowski
2019-12-31 11:03                     ` Artur Świgoń
2019-12-31 11:03                       ` Artur Świgoń
2019-12-31 11:03                       ` Artur Świgoń
2020-01-22 16:54       ` Georgi Djakov
2020-01-22 16:54         ` Georgi Djakov
2020-01-22 16:54         ` Georgi Djakov
2020-01-24 11:22         ` Artur Świgoń
2020-01-24 11:22           ` Artur Świgoń
2020-01-24 11:22           ` Artur Świgoń
     [not found]   ` <CGME20191220120145eucas1p295af63eed7b23982d8c49fcf875cec8c@eucas1p2.samsung.com>
2019-12-20 11:56     ` [RFC PATCH v3 5/7] devfreq: exynos-bus: Add interconnect functionality to exynos-bus Artur Świgoń
2019-12-20 11:56       ` Artur Świgoń
2019-12-20 11:56       ` Artur Świgoń
2019-12-21 19:53       ` Chanwoo Choi
2019-12-21 19:53         ` Chanwoo Choi
2019-12-21 19:53         ` Chanwoo Choi
2019-12-21 19:55         ` Chanwoo Choi
2019-12-21 19:55           ` Chanwoo Choi
2019-12-21 19:55           ` Chanwoo Choi
2020-01-22 17:02       ` Georgi Djakov
2020-01-22 17:02         ` Georgi Djakov
2020-01-22 17:02         ` Georgi Djakov
2020-01-24 11:22         ` Artur Świgoń [this message]
2020-01-24 11:22           ` Artur Świgoń
2020-01-24 11:22           ` Artur Świgoń
2020-01-24 12:32           ` Georgi Djakov
2020-01-24 12:32             ` Georgi Djakov
2020-01-24 12:32             ` Georgi Djakov
     [not found]   ` <CGME20191220120146eucas1p25dada01c315215d18bb8a15e3173b52c@eucas1p2.samsung.com>
2019-12-20 11:56     ` [RFC PATCH v3 6/7] arm: dts: exynos: Add interconnects to Exynos4412 mixer Artur Świgoń
2019-12-20 11:56       ` Artur Świgoń
2019-12-20 11:56       ` Artur Świgoń
2019-12-21 20:08       ` Chanwoo Choi
2019-12-21 20:08         ` Chanwoo Choi
2019-12-21 20:08         ` Chanwoo Choi
     [not found]   ` <CGME20191220120146eucas1p22a7b0457be4f378b113f67dc25f2eba7@eucas1p2.samsung.com>
2019-12-20 11:56     ` [RFC PATCH v3 7/7] drm: exynos: mixer: Add interconnect support Artur Świgoń
2019-12-20 11:56       ` Artur Świgoń
2019-12-20 11:56       ` Artur Świgoń
2019-12-21 20:15       ` Chanwoo Choi
2019-12-21 20:15         ` Chanwoo Choi
2019-12-21 20:15         ` Chanwoo Choi
2019-12-24  4:56       ` Inki Dae
2019-12-24  4:56         ` Inki Dae
2019-12-24  4:56         ` Inki Dae
2019-12-30  9:35         ` Artur Świgoń
2019-12-30  9:35           ` Artur Świgoń
2019-12-30  9:35           ` Artur Świgoń
2019-12-21 19:58   ` [RFC PATCH v3 0/7] PM / devfreq: Simple QoS for exynos-bus using interconnect Chanwoo Choi
2019-12-21 19:58     ` Chanwoo Choi
2019-12-21 19:58     ` Chanwoo Choi

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=0de33b1ae7c52270d11c606ecccc9af5f44d0421.camel@samsung.com \
    --to=a.swigon@samsung.com \
    --cc=b.zolnierkie@samsung.com \
    --cc=cw00.choi@samsung.com \
    --cc=devicetree@vger.kernel.org \
    --cc=dri-devel@lists.freedesktop.org \
    --cc=georgi.djakov@linaro.org \
    --cc=inki.dae@samsung.com \
    --cc=krzk@kernel.org \
    --cc=leonard.crestez@nxp.com \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-pm@vger.kernel.org \
    --cc=linux-samsung-soc@vger.kernel.org \
    --cc=m.szyprowski@samsung.com \
    --cc=myungjoo.ham@samsung.com \
    --cc=sw0312.kim@samsung.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.