diff mbox

[v4] drivers/pinctrl: grab default handles from device core

Message ID 1358795855-21064-1-git-send-email-linus.walleij@stericsson.com (mailing list archive)
State New, archived
Headers show

Commit Message

Linus Walleij Jan. 21, 2013, 7:17 p.m. UTC
From: Linus Walleij <linus.walleij@linaro.org>

This makes the device core auto-grab the pinctrl handle and set
the "default" (PINCTRL_STATE_DEFAULT) state for every device
that is present in the device model right before probe. This will
account for the lion's share of embedded silicon devcies.

A modification of the semantics for pinctrl_get() is also done:
previously if the pinctrl handle for a certain device was already
taken, the pinctrl core would return an error. Now, since the
core may have already default-grabbed the handle and set its
state to "default", if the handle was already taken, this will
be disregarded and the located, previously instanitated handle
will be returned to the caller.

This way all code in drivers explicitly requesting their pinctrl
handlers will still be functional, and drivers that want to
explicitly retrieve and switch their handles can still do that.
But if the desired functionality is just boilerplate of this
type in the probe() function:

struct pinctrl  *p;

p = devm_pinctrl_get_select_default(&dev);
if (IS_ERR(p)) {
   if (PTR_ERR(p) == -EPROBE_DEFER)
        return -EPROBE_DEFER;
        dev_warn(&dev, "no pinctrl handle\n");
}

The discussion began with the addition of such boilerplate
to the omap4 keypad driver:
http://marc.info/?l=linux-input&m=135091157719300&w=2

A previous approach using notifiers was discussed:
http://marc.info/?l=linux-kernel&m=135263661110528&w=2
This failed because it could not handle deferred probes.

This patch alone does not solve the entire dilemma faced:
whether code should be distributed into the drivers or
if it should be centralized to e.g. a PM domain. But it
solves the immediate issue of the addition of boilerplate
to a lot of drivers that just want to grab the default
state. As mentioned, they can later explicitly retrieve
the handle and set different states, and this could as
well be done by e.g. PM domains as it is only related
to a certain struct device * pointer.

Cc: Felipe Balbi <balbi@ti.com>
Cc: Benoit Cousson <b-cousson@ti.com>
Cc: Dmitry Torokhov <dmitry.torokhov@gmail.com>
Cc: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Cc: Mitch Bradley <wmb@firmworks.com>
Cc: Ulf Hansson <ulf.hansson@linaro.org>
Cc: Rafael J. Wysocki <rjw@sisk.pl>
Cc: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
Cc: Rickard Andersson <rickard.andersson@stericsson.com>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Russell King <linux@arm.linux.org.uk>
Reviewed-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
---
ChangeLog v3->v4:
- Drop overzealous NULL checks.
- Move kref initialization to pinctrl_create().
- Seeking Tested-by from Stephen Warren so we do not disturb
  the Tegra platform.
- Seeking ACK on this from Greg (and others who like it) so I
  can merge it through the pinctrl subsystem.
ChangeLog v2->v3:
- Abstain from using IS_ERR_OR_NULL() in the driver core,
  Russell recently sent a patch to remove it. Handle the
  NULL case explicitly even though it's a bogus case.
- Make sure we handle probe deferral correctly in the device
  core file. devm_kfree() the container on error so we don't
  waste memory for devices without pinctrl handles.
- Introduce reference counting into the pinctrl core using
  <linux/kref.h> so that we don't release pinctrl handles
  that have been obtained for two or more places.
ChangeLog v1->v2:
- Only store a pointer in the device struct, and only allocate
  this if it's really used by the device.
---
 Documentation/pinctrl.txt       | 24 +++++++++--
 drivers/base/Makefile           |  1 +
 drivers/base/dd.c               |  7 +++
 drivers/base/pinctrl.c          | 95 +++++++++++++++++++++++++++++++++++++++++
 drivers/pinctrl/core.c          | 30 +++++++++++--
 drivers/pinctrl/core.h          |  3 ++
 include/linux/device.h          |  7 +++
 include/linux/pinctrl/devinfo.h | 45 +++++++++++++++++++
 8 files changed, 205 insertions(+), 7 deletions(-)
 create mode 100644 drivers/base/pinctrl.c
 create mode 100644 include/linux/pinctrl/devinfo.h

Comments

Greg Kroah-Hartman Jan. 21, 2013, 11:41 p.m. UTC | #1
On Mon, Jan 21, 2013 at 08:17:35PM +0100, Linus Walleij wrote:
> From: Linus Walleij <linus.walleij@linaro.org>
> 
> This makes the device core auto-grab the pinctrl handle and set
> the "default" (PINCTRL_STATE_DEFAULT) state for every device
> that is present in the device model right before probe. This will
> account for the lion's share of embedded silicon devcies.
> 
> A modification of the semantics for pinctrl_get() is also done:
> previously if the pinctrl handle for a certain device was already
> taken, the pinctrl core would return an error. Now, since the
> core may have already default-grabbed the handle and set its
> state to "default", if the handle was already taken, this will
> be disregarded and the located, previously instanitated handle
> will be returned to the caller.
> 
> This way all code in drivers explicitly requesting their pinctrl
> handlers will still be functional, and drivers that want to
> explicitly retrieve and switch their handles can still do that.
> But if the desired functionality is just boilerplate of this
> type in the probe() function:
> 
> struct pinctrl  *p;
> 
> p = devm_pinctrl_get_select_default(&dev);
> if (IS_ERR(p)) {
>    if (PTR_ERR(p) == -EPROBE_DEFER)
>         return -EPROBE_DEFER;
>         dev_warn(&dev, "no pinctrl handle\n");
> }
> 
> The discussion began with the addition of such boilerplate
> to the omap4 keypad driver:
> http://marc.info/?l=linux-input&m=135091157719300&w=2
> 
> A previous approach using notifiers was discussed:
> http://marc.info/?l=linux-kernel&m=135263661110528&w=2
> This failed because it could not handle deferred probes.
> 
> This patch alone does not solve the entire dilemma faced:
> whether code should be distributed into the drivers or
> if it should be centralized to e.g. a PM domain. But it
> solves the immediate issue of the addition of boilerplate
> to a lot of drivers that just want to grab the default
> state. As mentioned, they can later explicitly retrieve
> the handle and set different states, and this could as
> well be done by e.g. PM domains as it is only related
> to a certain struct device * pointer.
> 
> Cc: Felipe Balbi <balbi@ti.com>
> Cc: Benoit Cousson <b-cousson@ti.com>
> Cc: Dmitry Torokhov <dmitry.torokhov@gmail.com>
> Cc: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
> Cc: Mitch Bradley <wmb@firmworks.com>
> Cc: Ulf Hansson <ulf.hansson@linaro.org>
> Cc: Rafael J. Wysocki <rjw@sisk.pl>
> Cc: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
> Cc: Rickard Andersson <rickard.andersson@stericsson.com>
> Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
> Cc: Russell King <linux@arm.linux.org.uk>
> Reviewed-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
> Signed-off-by: Linus Walleij <linus.walleij@linaro.org>


Acked-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Feel free to take this through your tree.

thanks,

greg k-h
Linus Walleij Jan. 22, 2013, 12:42 p.m. UTC | #2
On Tue, Jan 22, 2013 at 12:41 AM, Greg Kroah-Hartman
<gregkh@linuxfoundation.org> wrote:
> On Mon, Jan 21, 2013 at 08:17:35PM +0100, Linus Walleij wrote:
>> From: Linus Walleij <linus.walleij@linaro.org>
>>
>> This makes the device core auto-grab the pinctrl handle and set
>> the "default" (PINCTRL_STATE_DEFAULT) state for every device
>> that is present in the device model right before probe. This will
>> account for the lion's share of embedded silicon devcies.
(...)
> Acked-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
>
> Feel free to take this through your tree.

Thanks Greg!

Stephen W, are you OK with me pushing this version of the patch
now?

I have your fix and hopefully the other required fixes in the patch
so as not to break Tegra, but prefer to have it solidly tested
if possible.

Yours,
Linus Walleij
Stephen Warren Jan. 22, 2013, 5:37 p.m. UTC | #3
On 01/21/2013 12:17 PM, Linus Walleij wrote:
> This makes the device core auto-grab the pinctrl handle and set
> the "default" (PINCTRL_STATE_DEFAULT) state for every device
> that is present in the device model right before probe. This will
> account for the lion's share of embedded silicon devcies.

> diff --git a/Documentation/pinctrl.txt b/Documentation/pinctrl.txt

> @@ -1144,6 +1156,12 @@ PIN_MAP_MUX_GROUP_HOG_DEFAULT("pinctrl-foo", NULL /* group */, "power_func")
>  
>  This gives the exact same result as the above construction.
>  
> +This should not be used for any kind of device which is represented in
> +the device model, as the pinctrl core will attempt to do the equal of
> +pinctrl_get_select_default() for these devices right before their device
> +drivers are probed, so hogging these will just make the model look
> +strange. Instead put in proper map entries.

This is a policy change unrelated to this change. There are good reasons
to use pinctrl hogs for everything except where dynamic changes are
required by drivers. As such, I think the wording above is overly
strong. Preferably, that paragraph should simply not be added.

> diff --git a/drivers/base/pinctrl.c b/drivers/base/pinctrl.c

> +int pinctrl_bind_pins(struct device *dev)
> +{
> +	struct dev_pin_info *dpi;
> +	int ret;
> +
> +	/* Allocate a pin state container on-the-fly */
> +	if (!dev->pins) {

This path is guaranteed to be take unless there is a bug. The first time
through, this field will be NULL. If probe fails, the code below
attempts to ensure this field is set back to NULL. As such, we should
simply remove this if condition and always allocate dpi. This would also
remove the requirement to set dev->pins back to NULL below, this
simplifying the code all around, although perhaps it'd be a good idea to
clear out any invalid pointers for clarity either way.

> +		dpi = devm_kzalloc(dev, sizeof(*dpi), GFP_KERNEL);
> +		if (!dpi)
> +			return -ENOMEM;
> +	} else
> +		dpi = dev->pins;

> +	/*
> +	 * Check if we already have a pinctrl handle, as we may arrive here
> +	 * after a deferral in the state selection below
> +	 */
> +	if (!dpi->p) {
...
> +	/*
> +	 * We may have looked up the state earlier as well.
> +	 */
> +	if (!dpi->default_state) {

The same argument applies to both those if conditions too.

Following on from this, I thought that the code looked OK; all the error
paths set dev->pins=NULL when expected, although I think that the
pinctrl_lookup_state() short-exit path should have cleared dpi->p and
dpi->default_state, so I thought this patch would test out OK. However,
pinctrl_bind_pins() is not the only source of errors during probe(); the
driver itself could fail to probe() due to -EPROBE_DEFER, and that would
then require clearing dev->pins. So in fact, this patch still doesn't
work. Again, this can all be solved simply by removing all the
conditionals in the code that I mention above.

I'll send an updated/tested patch in a minute.
diff mbox

Patch

diff --git a/Documentation/pinctrl.txt b/Documentation/pinctrl.txt
index da40efb..68836e5 100644
--- a/Documentation/pinctrl.txt
+++ b/Documentation/pinctrl.txt
@@ -972,6 +972,18 @@  pinmux core.
 Pin control requests from drivers
 =================================
 
+When a device driver is about to probe the device core will automatically
+attempt to issue pinctrl_get_select_default() on these devices.
+This way driver writers do not need to add any of the boilerplate code
+of the type found below. However when doing fine-grained state selection
+and not using the "default" state, you may have to do some device driver
+handling of the pinctrl handles and states.
+
+So if you just want to put the pins for a certain device into the default
+state and be done with it, there is nothing you need to do besides
+providing the proper mapping table. The device core will take care of
+the rest.
+
 Generally it is discouraged to let individual drivers get and enable pin
 control. So if possible, handle the pin control in platform code or some other
 place where you have access to all the affected struct device * pointers. In
@@ -1097,9 +1109,9 @@  situations that can be electrically unpleasant, you will certainly want to
 mux in and bias pins in a certain way before the GPIO subsystems starts to
 deal with them.
 
-The above can be hidden: using pinctrl hogs, the pin control driver may be
-setting up the config and muxing for the pins when it is probing,
-nevertheless orthogonal to the GPIO subsystem.
+The above can be hidden: using the device core, the pinctrl core may be
+setting up the config and muxing for the pins right before the device is
+probing, nevertheless orthogonal to the GPIO subsystem.
 
 But there are also situations where it makes sense for the GPIO subsystem
 to communicate directly with with the pinctrl subsystem, using the latter
@@ -1144,6 +1156,12 @@  PIN_MAP_MUX_GROUP_HOG_DEFAULT("pinctrl-foo", NULL /* group */, "power_func")
 
 This gives the exact same result as the above construction.
 
+This should not be used for any kind of device which is represented in
+the device model, as the pinctrl core will attempt to do the equal of
+pinctrl_get_select_default() for these devices right before their device
+drivers are probed, so hogging these will just make the model look
+strange. Instead put in proper map entries.
+
 
 Runtime pinmuxing
 =================
diff --git a/drivers/base/Makefile b/drivers/base/Makefile
index 5aa2d70..4e22ce3 100644
--- a/drivers/base/Makefile
+++ b/drivers/base/Makefile
@@ -21,6 +21,7 @@  endif
 obj-$(CONFIG_SYS_HYPERVISOR) += hypervisor.o
 obj-$(CONFIG_REGMAP)	+= regmap/
 obj-$(CONFIG_SOC_BUS) += soc.o
+obj-$(CONFIG_PINCTRL) += pinctrl.o
 
 ccflags-$(CONFIG_DEBUG_DRIVER) := -DDEBUG
 
diff --git a/drivers/base/dd.c b/drivers/base/dd.c
index e3bbed8..65631015 100644
--- a/drivers/base/dd.c
+++ b/drivers/base/dd.c
@@ -24,6 +24,7 @@ 
 #include <linux/wait.h>
 #include <linux/async.h>
 #include <linux/pm_runtime.h>
+#include <linux/pinctrl/devinfo.h>
 
 #include "base.h"
 #include "power/power.h"
@@ -269,6 +270,12 @@  static int really_probe(struct device *dev, struct device_driver *drv)
 	WARN_ON(!list_empty(&dev->devres_head));
 
 	dev->driver = drv;
+
+	/* If using pinctrl, bind pins now before probing */
+	ret = pinctrl_bind_pins(dev);
+	if (ret)
+		goto probe_failed;
+
 	if (driver_sysfs_add(dev)) {
 		printk(KERN_ERR "%s: driver_sysfs_add(%s) failed\n",
 			__func__, dev_name(dev));
diff --git a/drivers/base/pinctrl.c b/drivers/base/pinctrl.c
new file mode 100644
index 0000000..a1f6b49
--- /dev/null
+++ b/drivers/base/pinctrl.c
@@ -0,0 +1,95 @@ 
+/*
+ * Driver core interface to the pinctrl subsystem.
+ *
+ * Copyright (C) 2012 ST-Ericsson SA
+ * Written on behalf of Linaro for ST-Ericsson
+ * Based on bits of regulator core, gpio core and clk core
+ *
+ * Author: Linus Walleij <linus.walleij@linaro.org>
+ *
+ * License terms: GNU General Public License (GPL) version 2
+ */
+
+#include <linux/device.h>
+#include <linux/pinctrl/devinfo.h>
+#include <linux/pinctrl/consumer.h>
+#include <linux/slab.h>
+
+/**
+ * pinctrl_bind_pins() - called by the device core before probe
+ * @dev: the device that is just about to probe
+ */
+int pinctrl_bind_pins(struct device *dev)
+{
+	struct dev_pin_info *dpi;
+	int ret;
+
+	/* Allocate a pin state container on-the-fly */
+	if (!dev->pins) {
+		dpi = devm_kzalloc(dev, sizeof(*dpi), GFP_KERNEL);
+		if (!dpi)
+			return -ENOMEM;
+	} else
+		dpi = dev->pins;
+
+	/*
+	 * Check if we already have a pinctrl handle, as we may arrive here
+	 * after a deferral in the state selection below
+	 */
+	if (!dpi->p) {
+		dpi->p = devm_pinctrl_get(dev);
+		if (IS_ERR(dpi->p)) {
+			int ret = PTR_ERR(dpi->p);
+
+			dev_dbg(dev, "no pinctrl handle\n");
+			/*
+			 * If no pinctrl handle was found for this device,
+			 * let's explicitly free the pin container in the
+			 * device, there is no point in keeping it around.
+			 */
+			devm_kfree(dev, dpi);
+			dev->pins = NULL;
+			/* Only return deferrals */
+			if (ret == -EPROBE_DEFER)
+				return ret;
+			return 0;
+		}
+	}
+
+	/*
+	 * For a newly allocated info struct, here is where we keep it,
+	 * since at this point it actually contains something.
+	 */
+	dev->pins = dpi;
+
+	/*
+	 * We may have looked up the state earlier as well.
+	 */
+	if (!dpi->default_state) {
+		dpi->default_state = pinctrl_lookup_state(dpi->p,
+						PINCTRL_STATE_DEFAULT);
+		if (IS_ERR(dpi->default_state)) {
+			dev_dbg(dev, "no default pinctrl state\n");
+			return 0;
+		}
+	}
+
+	ret = pinctrl_select_state(dpi->p, dpi->default_state);
+	if (ret) {
+		dev_dbg(dev, "failed to activate default pinctrl state\n");
+
+		/* Only return deferrals */
+		if (ret == -EPROBE_DEFER) {
+			/*
+			 * The devm_* allocated container will be free():ed
+			 * during deferral, re-obtain handle and all up above
+			 * next time we try to probe.
+			 */
+			dev->pins = NULL;
+			return ret;
+		}
+		return 0;
+	}
+
+	return 0;
+}
diff --git a/drivers/pinctrl/core.c b/drivers/pinctrl/core.c
index 5b4885c..895b4b7 100644
--- a/drivers/pinctrl/core.c
+++ b/drivers/pinctrl/core.c
@@ -14,6 +14,7 @@ 
 #define pr_fmt(fmt) "pinctrl core: " fmt
 
 #include <linux/kernel.h>
+#include <linux/kref.h>
 #include <linux/export.h>
 #include <linux/init.h>
 #include <linux/device.h>
@@ -721,6 +722,8 @@  static struct pinctrl *create_pinctrl(struct device *dev)
 		return ERR_PTR(ret);
 	}
 
+	kref_init(&p->users);
+
 	/* Add the pinctrl handle to the global list */
 	list_add_tail(&p->node, &pinctrl_list);
 
@@ -734,9 +737,17 @@  static struct pinctrl *pinctrl_get_locked(struct device *dev)
 	if (WARN_ON(!dev))
 		return ERR_PTR(-EINVAL);
 
+	/*
+	 * See if somebody else (such as the device core) has already
+	 * obtained a handle to the pinctrl for this device. In that case,
+	 * return another pointer to it.
+	 */
 	p = find_pinctrl(dev);
-	if (p != NULL)
-		return ERR_PTR(-EBUSY);
+	if (p != NULL) {
+		dev_dbg(dev, "obtain a copy of previously claimed pinctrl\n");
+		kref_get(&p->users);
+		return p;
+	}
 
 	return create_pinctrl(dev);
 }
@@ -792,13 +803,24 @@  static void pinctrl_put_locked(struct pinctrl *p, bool inlist)
 }
 
 /**
- * pinctrl_put() - release a previously claimed pinctrl handle
+ * pinctrl_release() - release the pinctrl handle
+ * @kref: the kref in the pinctrl being released
+ */
+void pinctrl_release(struct kref *kref)
+{
+	struct pinctrl *p = container_of(kref, struct pinctrl, users);
+
+	pinctrl_put_locked(p, true);
+}
+
+/**
+ * pinctrl_put() - decrease use count on a previously claimed pinctrl handle
  * @p: the pinctrl handle to release
  */
 void pinctrl_put(struct pinctrl *p)
 {
 	mutex_lock(&pinctrl_mutex);
-	pinctrl_put_locked(p, true);
+	kref_put(&p->users, pinctrl_release);
 	mutex_unlock(&pinctrl_mutex);
 }
 EXPORT_SYMBOL_GPL(pinctrl_put);
diff --git a/drivers/pinctrl/core.h b/drivers/pinctrl/core.h
index 12f5694..bd483d6 100644
--- a/drivers/pinctrl/core.h
+++ b/drivers/pinctrl/core.h
@@ -9,6 +9,7 @@ 
  * License terms: GNU General Public License (GPL) version 2
  */
 
+#include <linux/kref.h>
 #include <linux/mutex.h>
 #include <linux/radix-tree.h>
 #include <linux/pinctrl/pinconf.h>
@@ -54,6 +55,7 @@  struct pinctrl_dev {
  * @state: the current state
  * @dt_maps: the mapping table chunks dynamically parsed from device tree for
  *	this device, if any
+ * @users: reference count
  */
 struct pinctrl {
 	struct list_head node;
@@ -61,6 +63,7 @@  struct pinctrl {
 	struct list_head states;
 	struct pinctrl_state *state;
 	struct list_head dt_maps;
+	struct kref users;
 };
 
 /**
diff --git a/include/linux/device.h b/include/linux/device.h
index 43dcda9..001f663 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -21,6 +21,7 @@ 
 #include <linux/compiler.h>
 #include <linux/types.h>
 #include <linux/mutex.h>
+#include <linux/pinctrl/devinfo.h>
 #include <linux/pm.h>
 #include <linux/atomic.h>
 #include <linux/ratelimit.h>
@@ -620,6 +621,8 @@  struct acpi_dev_node {
  * @pm_domain:	Provide callbacks that are executed during system suspend,
  * 		hibernation, system resume and during runtime PM transitions
  * 		along with subsystem-level and driver-level callbacks.
+ * @pins:	For device pin management.
+ *		See Documentation/pinctrl.txt for details.
  * @numa_node:	NUMA node this device is close to.
  * @dma_mask:	Dma mask (if dma'ble device).
  * @coherent_dma_mask: Like dma_mask, but for alloc_coherent mapping as not all
@@ -672,6 +675,10 @@  struct device {
 	struct dev_pm_info	power;
 	struct dev_pm_domain	*pm_domain;
 
+#ifdef CONFIG_PINCTRL
+	struct dev_pin_info	*pins;
+#endif
+
 #ifdef CONFIG_NUMA
 	int		numa_node;	/* NUMA node this device is close to */
 #endif
diff --git a/include/linux/pinctrl/devinfo.h b/include/linux/pinctrl/devinfo.h
new file mode 100644
index 0000000..6e5f8a9
--- /dev/null
+++ b/include/linux/pinctrl/devinfo.h
@@ -0,0 +1,45 @@ 
+/*
+ * Per-device information from the pin control system.
+ * This is the stuff that get included into the device
+ * core.
+ *
+ * Copyright (C) 2012 ST-Ericsson SA
+ * Written on behalf of Linaro for ST-Ericsson
+ * This interface is used in the core to keep track of pins.
+ *
+ * Author: Linus Walleij <linus.walleij@linaro.org>
+ *
+ * License terms: GNU General Public License (GPL) version 2
+ */
+
+#ifndef PINCTRL_DEVINFO_H
+#define PINCTRL_DEVINFO_H
+
+#ifdef CONFIG_PINCTRL
+
+/* The device core acts as a consumer toward pinctrl */
+#include <linux/pinctrl/consumer.h>
+
+/**
+ * struct dev_pin_info - pin state container for devices
+ * @p: pinctrl handle for the containing device
+ * @default_state: the default state for the handle, if found
+ */
+struct dev_pin_info {
+	struct pinctrl *p;
+	struct pinctrl_state *default_state;
+};
+
+extern int pinctrl_bind_pins(struct device *dev);
+
+#else
+
+/* Stubs if we're not using pinctrl */
+
+static inline int pinctrl_bind_pins(struct device *dev)
+{
+	return 0;
+}
+
+#endif /* CONFIG_PINCTRL */
+#endif /* PINCTRL_DEVINFO_H */