[PATCH v5 00/10] platform/chrome: Introduce DT hardware prober

[PATCH v5 00/10] platform/chrome: Introduce DT hardware prober

[Chen-Yu Tsai]

Hi everyone,

This is v5 of my "of: Introduce hardware prober driver" [1] series.
v5 mainly addresses comments from Andy.

v2 continued Doug's "of: device: Support 2nd sources of probeable but
undiscoverable devices" [2] series, but follows the scheme suggested by
Rob, marking all second source component device nodes as "fail-needs-probe",
and having a hardware prober driver enable the one of them.


Changes since v4:
- Link to v4: 
  https://lore.kernel.org/all/20240808095931.2649657-1-wenst@chromium.org/
- Patch 1 "of: dynamic: Add of_changeset_update_prop_string"
  - Use modern designated initializer for |prop|
- Patch 2 "regulator: Move OF-specific regulator lookup code to of_regulator.c"
  - New patch only moving code
- Patch 3 "regulator: Split up _regulator_get()"
  - New patch splitting existing code into smaller functions
- Patch 4 "regulator: Do pure DT regulator lookup in of_regulator_bulk_get_all()"
  - Was "regulator: Add regulator_of_get_optional() for pure DT" in v4
  - Add OF-specific _of_regulator_get() function
  - Rename regulator_of_get_optional() to of_regulator_get_optional() for
    consistency
  - Make of_regulator_get_optional() static, as it is only used internally
  - Convert of_regulator_bulk_get_all()
- Patch 5 "gpiolib: Add gpio_property_name_length()"
  - New patch, split out from patch 7/8
- Patch 6 "i2c: Introduce OF component probe function"
  - Split code into helper functions
  - Use scoped helpers and __free() to reduce error path
- Patch 7 "i2c: of-prober: Add regulator support"
  - Split out GPIO handling as patch 8
  - Rewrote using of_regulator_bulk_get_all()
  - Replaced "regulators" with "regulator supplies" in debug messages
- Patch 8 "i2c: of-prober: Add GPIO support"
  - New patch split out from patch 7
  - Moved GPIO property name check to common function in gpiolib.c in new
    patch
  - Moved i2c_of_probe_free_gpios() into for_each_child_of_node_scoped()
  - Rewrote in gpiod_*_array-esque fashion
- Patch 9 "platform/chrome: Introduce device tree hardware prober"
  - Fix Kconfig dependency
  - Update copyright year
  - Drop "linux/of.h" header
  - Include "linux/errno.h"
  - Move |int ret| declaration to top of block
  - Return -ENODEV on no match instead of 0
  - Unregister platform driver and device unconditionally after previous
    change
- Patch 10 "arm64: dts: mediatek: mt8173-elm-hana: Mark touchscreens and
	    trackpads as fail"
  - Rebased

Changes since v3:
- Patch 1 "of: dynamic: Add of_changeset_update_prop_string"
  - Use new __of_prop_free() helper (helper was added by Rob in 2024/04)
  - Add new line before header declaration
- Patch 2 "regulator: Add regulator_of_get_optional() for pure DT"
  - New patch
- Patch 3 "i2c: Introduce OF component probe function"
  - Complete kernel-doc
  - Return different error if I2C controller is disabled
  - Expand comment to explain assumptions and constraints
  - Split for-loop finding target node and operations on target node
  - Add missing i2c_put_adapter()
  - Move prober code to separate file
- Patch 4 "i2c: of-prober: Add GPIO and regulator support"
  - New patch
- Patch 5 "platform/chrome: Introduce device tree hardware prober"
  - Include linux/init.h
  - Rewrite for loop in driver probe function as suggested by Andy
  - Make prober driver buildable as module
  - Ignore prober errors other than probe deferral
- Patch 6 "arm64: dts: mediatek: mt8173-elm-hana: Mark touchscreens and trackpads as fail"
  - Rebased

Patch "arm64: dts: mediatek: mt8173-elm-hana: Add G2touch G7500 touchscreen"
was merged separately and thus removed from this series.

Changes since v2:
- Added of_changeset_update_prop_string()
- Moved generic I2C code to the I2C core
- Moved remaining platform specific code to platform/chrome/
- Switched to of_node_is_available() to check if node is enabled.
- Switched to OF changeset API to update status property
- I2C probe helper function now accepts "struct device *dev" instead to
  reduce line length and dereferences
- Moved "ret = 0" to just before for_each_child_of_node(i2c_node, node)
- Depend on rather than select CONFIG_I2C
- Copied machine check to driver init function
- Explicitly mentioned "device tree" or OF in driver name, description
  and Kconfig symbol
- Dropped filename from inside the file
- Made loop variable size_t (instead of unsigned int as Andy asked)
- Switched to PLATFORM_DEVID_NONE instead of raw -1
- Switched to standard goto error path pattern in hw_prober_driver_init()
- Dropped device class from status property

Patches removed from v3 and saved for later:
- of: base: Add of_device_is_fail
- of: hw_prober: Support Chromebook SKU ID based component selection
- dt-bindings: arm: mediatek: Remove SKU specific compatibles for Google Krane
- arm64: dts: mediatek: mt8183-kukui: Merge Krane device trees

For the I2C component (touchscreens and trackpads) case from the
original series, the hardware prober driver finds the particular
class of device in the device tree, gets its parent I2C adapter,
and tries to initiate a simple I2C read for each device under that
I2C bus. When it finds one that responds, it considers that one
present, marks it as "okay", and returns, letting the driver core
actually probe the device.

This works fine in most cases since these components are connected
via a ribbon cable and always have the same resources. The prober
will also grab these resources and enable them.

The other case, selecting a display panel to use based on the SKU ID
from the firmware, hit a bit of an issue with fixing the OF graph.
It has been left out since v3.

Patch 1 adds of_changeset_update_prop_string(), as requested by Rob.

Patches 2 through 4 reorganize the OF-specific regulator core code and
reworks the existing of_regulator_bulk_get_all() function to look up
regulator supplies solely using device tree nodes.

Patch 5 adds a function to the GPIO library that checks whether a
given string (property name) matches the GPIO property pattern, and
if it does, returns the length of the GPIO name.

Patch 6 implements probing the I2C bus for presence of components as
a helper function in the I2C core.

Patch 7 implements regulator supply support for the I2C component
prober.

Patch 8 implements GPIO support for the I2C component prober.

Patch 9 adds a ChromeOS specific DT hardware prober. This initial
version targets the Hana Chromebooks, probing its I2C trackpads and
touchscreens.

Patch 10 modifies the Hana device tree and marks the touchscreens
and trackpads as "fail-needs-probe", ready for the driver to probe.


The patch and build time dependencies for this series is now quite
complicated:

  of_regulator.c cleanups [1] -> regulator patches here ----
							   |
							   v
  gpio patch in -next [2] -> gpiolib patch here  -----> i2c of-prober --
								       |
                      platform/chrome device tree hardware prober <----- 
 
The regulator patches in this series depend on other cleanup patches [1]
I sent earlier. The gpiolib patch depends on a commit in -next [2]
changing the GPIO suffixes array. Patches 6 through 8 introducting i2c
of-prober depend on the first 5 patches. Patch 9, The chrome prober,
depends on patches 6 through 8.

I think it might be easier if the respective maintainers take the first
five patches for -rc1, and patches 6 through 9 go through either the i2c
or chrome trees either very late in the merge window or right after it.
Patch 10 can go in only after everything else is in. This should be
better than having an immutable branch on top of some commit in -next
for three other trees to consume.


Thanks
ChenYu

[1] https://lore.kernel.org/all/20240822072047.3097740-1-wenst@chromium.org/
[2] commit 4b91188dced8 ("gpiolib: Replace gpio_suffix_count with NULL-terminated array")

Chen-Yu Tsai (10):
  of: dynamic: Add of_changeset_update_prop_string
  regulator: Move OF-specific regulator lookup code to of_regulator.c
  regulator: Split up _regulator_get()
  regulator: Do pure DT regulator lookup in of_regulator_bulk_get_all()
  gpiolib: Add gpio_property_name_length()
  i2c: Introduce OF component probe function
  i2c: of-prober: Add regulator support
  i2c: of-prober: Add GPIO support
  platform/chrome: Introduce device tree hardware prober
  arm64: dts: mediatek: mt8173-elm-hana: Mark touchscreens and trackpads
    as fail

 .../boot/dts/mediatek/mt8173-elm-hana.dtsi    |  13 +
 arch/arm64/boot/dts/mediatek/mt8173-elm.dtsi  |   4 +-
 drivers/gpio/gpiolib.c                        |  44 ++
 drivers/i2c/Makefile                          |   1 +
 drivers/i2c/i2c-core-of-prober.c              | 391 ++++++++++++++++++
 drivers/of/dynamic.c                          |  44 ++
 drivers/platform/chrome/Kconfig               |  11 +
 drivers/platform/chrome/Makefile              |   1 +
 .../platform/chrome/chromeos_of_hw_prober.c   | 104 +++++
 drivers/regulator/core.c                      | 139 +++----
 drivers/regulator/internal.h                  |  15 +-
 drivers/regulator/of_regulator.c              | 142 ++++++-
 include/linux/gpio/consumer.h                 |   2 +
 include/linux/i2c.h                           |   4 +
 include/linux/of.h                            |   4 +
 15 files changed, 820 insertions(+), 99 deletions(-)
 create mode 100644 drivers/i2c/i2c-core-of-prober.c
 create mode 100644 drivers/platform/chrome/chromeos_of_hw_prober.c

-- 
2.46.0.184.g6999bdac58-goog

[PATCH v5 01/10] of: dynamic: Add of_changeset_update_prop_string

[Chen-Yu Tsai]

Add a helper function to add string property updates to an OF changeset.
This is similar to of_changeset_add_prop_string(), but instead of adding
the property (and failing if it exists), it will update the property.

This shall be used later in the DT hardware prober.

Signed-off-by: Chen-Yu Tsai <wenst@chromium.org>
---
Changes since v4:
- Use modern designated initializer for |prop|

Changes since v3:
- Use new __of_prop_free() helper
- Add new line before header declaration

Changes since v2:
- New patch added in v3
---
 drivers/of/dynamic.c | 44 ++++++++++++++++++++++++++++++++++++++++++++
 include/linux/of.h   |  4 ++++
 2 files changed, 48 insertions(+)

diff --git a/drivers/of/dynamic.c b/drivers/of/dynamic.c
index 110104a936d9..daa69d160a78 100644
--- a/drivers/of/dynamic.c
+++ b/drivers/of/dynamic.c
@@ -1072,3 +1072,47 @@ int of_changeset_add_prop_bool(struct of_changeset *ocs, struct device_node *np,
 	return of_changeset_add_prop_helper(ocs, np, &prop);
 }
 EXPORT_SYMBOL_GPL(of_changeset_add_prop_bool);
+
+static int of_changeset_update_prop_helper(struct of_changeset *ocs,
+					   struct device_node *np,
+					   const struct property *pp)
+{
+	struct property *new_pp;
+	int ret;
+
+	new_pp = __of_prop_dup(pp, GFP_KERNEL);
+	if (!new_pp)
+		return -ENOMEM;
+
+	ret = of_changeset_update_property(ocs, np, new_pp);
+	if (ret)
+		__of_prop_free(new_pp);
+
+	return ret;
+}
+
+/**
+ * of_changeset_update_prop_string - Add a string property update to a changeset
+ *
+ * @ocs:	changeset pointer
+ * @np:		device node pointer
+ * @prop_name:	name of the property to be updated
+ * @str:	pointer to null terminated string
+ *
+ * Create a string property to be updated and add it to a changeset.
+ *
+ * Return: 0 on success, a negative error value in case of an error.
+ */
+int of_changeset_update_prop_string(struct of_changeset *ocs,
+				    struct device_node *np,
+				    const char *prop_name, const char *str)
+{
+	struct property prop = {
+		.name = (char *)prop_name,
+		.length = strlen(str) + 1,
+		.value = (void *)str,
+	};
+
+	return of_changeset_update_prop_helper(ocs, np, &prop);
+}
+EXPORT_SYMBOL_GPL(of_changeset_update_prop_string);
diff --git a/include/linux/of.h b/include/linux/of.h
index 85b60ac9eec5..046283be1cd3 100644
--- a/include/linux/of.h
+++ b/include/linux/of.h
@@ -1651,6 +1651,10 @@ static inline int of_changeset_add_prop_u32(struct of_changeset *ocs,
 	return of_changeset_add_prop_u32_array(ocs, np, prop_name, &val, 1);
 }
 
+int of_changeset_update_prop_string(struct of_changeset *ocs,
+				    struct device_node *np,
+				    const char *prop_name, const char *str);
+
 int of_changeset_add_prop_bool(struct of_changeset *ocs, struct device_node *np,
 			       const char *prop_name);
 
-- 
2.46.0.184.g6999bdac58-goog

Re: [PATCH v5 01/10] of: dynamic: Add of_changeset_update_prop_string

[Rob Herring]

On Thu, Aug 22, 2024 at 4:20 AM Chen-Yu Tsai <wenst@chromium.org> wrote:
>
> Add a helper function to add string property updates to an OF changeset.
> This is similar to of_changeset_add_prop_string(), but instead of adding
> the property (and failing if it exists), it will update the property.
>
> This shall be used later in the DT hardware prober.
>
> Signed-off-by: Chen-Yu Tsai <wenst@chromium.org>
> ---
> Changes since v4:
> - Use modern designated initializer for |prop|
>
> Changes since v3:
> - Use new __of_prop_free() helper
> - Add new line before header declaration
>
> Changes since v2:
> - New patch added in v3
> ---
>  drivers/of/dynamic.c | 44 ++++++++++++++++++++++++++++++++++++++++++++
>  include/linux/of.h   |  4 ++++
>  2 files changed, 48 insertions(+)

Reviewed-by: Rob Herring (Arm) <robh@kernel.org>

[PATCH v5 02/10] regulator: Move OF-specific regulator lookup code to of_regulator.c

[Chen-Yu Tsai]

There's still a bit of OF-specific code in the regulator device lookup
function.

Move those bits of code over to of_regulator.c, and create a new
function of_regulator_dev_lookup() to encapsulate the code moved out of
regulator_dev_lookup().

Also mark of_find_regulator_by_node() as static, since there are no
other users in other compile units.

There are no functional changes.

Signed-off-by: Chen-Yu Tsai <wenst@chromium.org>
---
Changes since v4:
- New patch
---
 drivers/regulator/core.c         |  85 ++------------------------
 drivers/regulator/internal.h     |   9 +--
 drivers/regulator/of_regulator.c | 102 ++++++++++++++++++++++++++++++-
 3 files changed, 111 insertions(+), 85 deletions(-)

diff --git a/drivers/regulator/core.c b/drivers/regulator/core.c
index 9029de5395ee..361309fcae57 100644
--- a/drivers/regulator/core.c
+++ b/drivers/regulator/core.c
@@ -419,72 +419,6 @@ static void regulator_lock_dependent(struct regulator_dev *rdev,
 	mutex_unlock(&regulator_list_mutex);
 }
 
-/**
- * of_get_child_regulator - get a child regulator device node
- * based on supply name
- * @parent: Parent device node
- * @prop_name: Combination regulator supply name and "-supply"
- *
- * Traverse all child nodes.
- * Extract the child regulator device node corresponding to the supply name.
- * returns the device node corresponding to the regulator if found, else
- * returns NULL.
- */
-static struct device_node *of_get_child_regulator(struct device_node *parent,
-						  const char *prop_name)
-{
-	struct device_node *regnode = NULL;
-	struct device_node *child = NULL;
-
-	for_each_child_of_node(parent, child) {
-		regnode = of_parse_phandle(child, prop_name, 0);
-
-		if (!regnode) {
-			regnode = of_get_child_regulator(child, prop_name);
-			if (regnode)
-				goto err_node_put;
-		} else {
-			goto err_node_put;
-		}
-	}
-	return NULL;
-
-err_node_put:
-	of_node_put(child);
-	return regnode;
-}
-
-/**
- * of_get_regulator - get a regulator device node based on supply name
- * @dev: Device pointer for the consumer (of regulator) device
- * @supply: regulator supply name
- *
- * Extract the regulator device node corresponding to the supply name.
- * returns the device node corresponding to the regulator if found, else
- * returns NULL.
- */
-static struct device_node *of_get_regulator(struct device *dev, const char *supply)
-{
-	struct device_node *regnode = NULL;
-	char prop_name[64]; /* 64 is max size of property name */
-
-	dev_dbg(dev, "Looking up %s-supply from device tree\n", supply);
-
-	snprintf(prop_name, 64, "%s-supply", supply);
-	regnode = of_parse_phandle(dev->of_node, prop_name, 0);
-
-	if (!regnode) {
-		regnode = of_get_child_regulator(dev->of_node, prop_name);
-		if (regnode)
-			return regnode;
-
-		dev_dbg(dev, "Looking up %s property in node %pOF failed\n",
-				prop_name, dev->of_node);
-		return NULL;
-	}
-	return regnode;
-}
-
 /* Platform voltage constraint check */
 int regulator_check_voltage(struct regulator_dev *rdev,
 			    int *min_uV, int *max_uV)
@@ -2009,7 +1943,6 @@ static struct regulator_dev *regulator_dev_lookup(struct device *dev,
 						  const char *supply)
 {
 	struct regulator_dev *r = NULL;
-	struct device_node *node;
 	struct regulator_map *map;
 	const char *devname = NULL;
 
@@ -2017,19 +1950,11 @@ static struct regulator_dev *regulator_dev_lookup(struct device *dev,
 
 	/* first do a dt based lookup */
 	if (dev && dev->of_node) {
-		node = of_get_regulator(dev, supply);
-		if (node) {
-			r = of_find_regulator_by_node(node);
-			of_node_put(node);
-			if (r)
-				return r;
-
-			/*
-			 * We have a node, but there is no device.
-			 * assume it has not registered yet.
-			 */
-			return ERR_PTR(-EPROBE_DEFER);
-		}
+		r = of_regulator_dev_lookup(dev, supply);
+		if (!IS_ERR(r))
+			return r;
+		if (PTR_ERR(r) == -EPROBE_DEFER)
+			return r;
 	}
 
 	/* if not found, try doing it non-dt way */
diff --git a/drivers/regulator/internal.h b/drivers/regulator/internal.h
index 77a502141089..ffeaef22169e 100644
--- a/drivers/regulator/internal.h
+++ b/drivers/regulator/internal.h
@@ -66,7 +66,8 @@ static inline struct regulator_dev *dev_to_rdev(struct device *dev)
 }
 
 #ifdef CONFIG_OF
-struct regulator_dev *of_find_regulator_by_node(struct device_node *np);
+struct regulator_dev *of_regulator_dev_lookup(struct device *dev,
+					      const char *supply);
 struct regulator_init_data *regulator_of_get_init_data(struct device *dev,
 			         const struct regulator_desc *desc,
 				 struct regulator_config *config,
@@ -80,10 +81,10 @@ int of_get_n_coupled(struct regulator_dev *rdev);
 bool of_check_coupling_data(struct regulator_dev *rdev);
 
 #else
-static inline struct regulator_dev *
-of_find_regulator_by_node(struct device_node *np)
+static inline struct regulator_dev *of_regulator_dev_lookup(struct device *dev,
+							    const char *supply)
 {
-	return NULL;
+	return -ENODEV;
 }
 
 static inline struct regulator_init_data *
diff --git a/drivers/regulator/of_regulator.c b/drivers/regulator/of_regulator.c
index d557f7b1ec7c..d924f5c1de59 100644
--- a/drivers/regulator/of_regulator.c
+++ b/drivers/regulator/of_regulator.c
@@ -550,7 +550,73 @@ struct regulator_init_data *regulator_of_get_init_data(struct device *dev,
 	return NULL;
 }
 
-struct regulator_dev *of_find_regulator_by_node(struct device_node *np)
+/**
+ * of_get_child_regulator - get a child regulator device node
+ * based on supply name
+ * @parent: Parent device node
+ * @prop_name: Combination regulator supply name and "-supply"
+ *
+ * Traverse all child nodes.
+ * Extract the child regulator device node corresponding to the supply name.
+ * returns the device node corresponding to the regulator if found, else
+ * returns NULL.
+ */
+static struct device_node *of_get_child_regulator(struct device_node *parent,
+						  const char *prop_name)
+{
+	struct device_node *regnode = NULL;
+	struct device_node *child = NULL;
+
+	for_each_child_of_node(parent, child) {
+		regnode = of_parse_phandle(child, prop_name, 0);
+
+		if (!regnode) {
+			regnode = of_get_child_regulator(child, prop_name);
+			if (regnode)
+				goto err_node_put;
+		} else {
+			goto err_node_put;
+		}
+	}
+	return NULL;
+
+err_node_put:
+	of_node_put(child);
+	return regnode;
+}
+
+/**
+ * of_get_regulator - get a regulator device node based on supply name
+ * @dev: Device pointer for the consumer (of regulator) device
+ * @supply: regulator supply name
+ *
+ * Extract the regulator device node corresponding to the supply name.
+ * returns the device node corresponding to the regulator if found, else
+ * returns NULL.
+ */
+static struct device_node *of_get_regulator(struct device *dev, const char *supply)
+{
+	struct device_node *regnode = NULL;
+	char prop_name[64]; /* 64 is max size of property name */
+
+	dev_dbg(dev, "Looking up %s-supply from device tree\n", supply);
+
+	snprintf(prop_name, 64, "%s-supply", supply);
+	regnode = of_parse_phandle(dev->of_node, prop_name, 0);
+
+	if (!regnode) {
+		regnode = of_get_child_regulator(dev->of_node, prop_name);
+		if (regnode)
+			return regnode;
+
+		dev_dbg(dev, "Looking up %s property in node %pOF failed\n",
+				prop_name, dev->of_node);
+		return NULL;
+	}
+	return regnode;
+}
+
+static struct regulator_dev *of_find_regulator_by_node(struct device_node *np)
 {
 	struct device *dev;
 
@@ -559,6 +625,40 @@ struct regulator_dev *of_find_regulator_by_node(struct device_node *np)
 	return dev ? dev_to_rdev(dev) : NULL;
 }
 
+/** of_regulator_dev_lookup - lookup a regulator device with device tree only
+ * @dev: Device pointer for regulator supply lookup.
+ * @supply: Supply name or regulator ID.
+ *
+ * If successful, returns a struct regulator_dev that corresponds to the name
+ * @supply and with the embedded struct device refcount incremented by one.
+ * The refcount must be dropped by calling put_device().
+ * On failure one of the following ERR-PTR-encoded values is returned:
+ * -ENODEV if lookup fails permanently, -EPROBE_DEFER if lookup could succeed
+ * in the future.
+ */
+struct regulator_dev *of_regulator_dev_lookup(struct device *dev,
+					      const char *supply)
+{
+	struct regulator_dev *r;
+	struct device_node *node;
+
+	node = of_get_regulator(dev, supply);
+	if (node) {
+		r = of_find_regulator_by_node(node);
+		of_node_put(node);
+		if (r)
+			return r;
+
+		/*
+		 * We have a node, but there is no device.
+		 * assume it has not registered yet.
+		 */
+		return ERR_PTR(-EPROBE_DEFER);
+	}
+
+	return ERR_PTR(-ENODEV);
+}
+
 /*
  * Returns number of regulators coupled with rdev.
  */
-- 
2.46.0.184.g6999bdac58-goog

Re: [PATCH v5 02/10] regulator: Move OF-specific regulator lookup code to of_regulator.c

[Andy Shevchenko]

On Thu, Aug 22, 2024 at 05:19:55PM +0800, Chen-Yu Tsai wrote:
> There's still a bit of OF-specific code in the regulator device lookup
> function.
> 
> Move those bits of code over to of_regulator.c, and create a new
> function of_regulator_dev_lookup() to encapsulate the code moved out of
> regulator_dev_lookup().
> 
> Also mark of_find_regulator_by_node() as static, since there are no
> other users in other compile units.
> 
> There are no functional changes.

...

> +/**
> + * of_get_child_regulator - get a child regulator device node
> + * based on supply name
> + * @parent: Parent device node
> + * @prop_name: Combination regulator supply name and "-supply"
> + *
> + * Traverse all child nodes.
> + * Extract the child regulator device node corresponding to the supply name.
> + * returns the device node corresponding to the regulator if found, else
> + * returns NULL.

At the same time you may fix kernel-doc warnings (no "Return" section) in these
three (on your wish you may fix others in a separate change, but it's not
related to this series).

> + */

...

> +/** of_regulator_dev_lookup - lookup a regulator device with device tree only

Something went wrong with the indentation.

> + * @dev: Device pointer for regulator supply lookup.
> + * @supply: Supply name or regulator ID.
> + *
> + * If successful, returns a struct regulator_dev that corresponds to the name
> + * @supply and with the embedded struct device refcount incremented by one.
> + * The refcount must be dropped by calling put_device().
> + * On failure one of the following ERR-PTR-encoded values is returned:
> + * -ENODEV if lookup fails permanently, -EPROBE_DEFER if lookup could succeed
> + * in the future.
> + */

-- 
With Best Regards,
Andy Shevchenko

Re: [PATCH v5 02/10] regulator: Move OF-specific regulator lookup code to of_regulator.c

[Chen-Yu Tsai]

On Thu, Aug 22, 2024 at 9:47 PM Andy Shevchenko
<andriy.shevchenko@linux.intel.com> wrote:
>
> On Thu, Aug 22, 2024 at 05:19:55PM +0800, Chen-Yu Tsai wrote:
> > There's still a bit of OF-specific code in the regulator device lookup
> > function.
> >
> > Move those bits of code over to of_regulator.c, and create a new
> > function of_regulator_dev_lookup() to encapsulate the code moved out of
> > regulator_dev_lookup().
> >
> > Also mark of_find_regulator_by_node() as static, since there are no
> > other users in other compile units.
> >
> > There are no functional changes.
>
> ...
>
> > +/**
> > + * of_get_child_regulator - get a child regulator device node
> > + * based on supply name
> > + * @parent: Parent device node
> > + * @prop_name: Combination regulator supply name and "-supply"
> > + *
> > + * Traverse all child nodes.
> > + * Extract the child regulator device node corresponding to the supply name.
> > + * returns the device node corresponding to the regulator if found, else
> > + * returns NULL.
>
> At the same time you may fix kernel-doc warnings (no "Return" section) in these
> three (on your wish you may fix others in a separate change, but it's not
> related to this series).

As you said some other functions are missing it as well, so I'll do a
patch separate from this series to fix them all.

> > + */
>
> ...
>
> > +/** of_regulator_dev_lookup - lookup a regulator device with device tree only
>
> Something went wrong with the indentation.

Will fix, and also add a "Return" section.


Thanks
ChenYu

> > + * @dev: Device pointer for regulator supply lookup.
> > + * @supply: Supply name or regulator ID.
> > + *
> > + * If successful, returns a struct regulator_dev that corresponds to the name
> > + * @supply and with the embedded struct device refcount incremented by one.
> > + * The refcount must be dropped by calling put_device().
> > + * On failure one of the following ERR-PTR-encoded values is returned:
> > + * -ENODEV if lookup fails permanently, -EPROBE_DEFER if lookup could succeed
> > + * in the future.
> > + */
>
> --
> With Best Regards,
> Andy Shevchenko
>
>

Re: [PATCH v5 02/10] regulator: Move OF-specific regulator lookup code to of_regulator.c

[Andy Shevchenko]

On Fri, Aug 23, 2024 at 02:49:59PM +0800, Chen-Yu Tsai wrote:
> On Thu, Aug 22, 2024 at 9:47 PM Andy Shevchenko
> <andriy.shevchenko@linux.intel.com> wrote:
> > On Thu, Aug 22, 2024 at 05:19:55PM +0800, Chen-Yu Tsai wrote:

...

> > > +/**
> > > + * of_get_child_regulator - get a child regulator device node
> > > + * based on supply name
> > > + * @parent: Parent device node
> > > + * @prop_name: Combination regulator supply name and "-supply"
> > > + *
> > > + * Traverse all child nodes.
> > > + * Extract the child regulator device node corresponding to the supply name.
> > > + * returns the device node corresponding to the regulator if found, else
> > > + * returns NULL.
> >
> > At the same time you may fix kernel-doc warnings (no "Return" section) in these
> > three (on your wish you may fix others in a separate change, but it's not
> > related to this series).
> 
> As you said some other functions are missing it as well, so I'll do a
> patch separate from this series to fix them all.

But you need to fix them in this patch series. We do not add patches with known
issues, which are really easy to fix beforehand.

(And below seems you indirectly agrees on that)

> > > + */

...

> > > +/** of_regulator_dev_lookup - lookup a regulator device with device tree only
> >
> > Something went wrong with the indentation.
> 
> Will fix, and also add a "Return" section.

Thank you!

> > > + * @dev: Device pointer for regulator supply lookup.
> > > + * @supply: Supply name or regulator ID.
> > > + *
> > > + * If successful, returns a struct regulator_dev that corresponds to the name
> > > + * @supply and with the embedded struct device refcount incremented by one.
> > > + * The refcount must be dropped by calling put_device().
> > > + * On failure one of the following ERR-PTR-encoded values is returned:
> > > + * -ENODEV if lookup fails permanently, -EPROBE_DEFER if lookup could succeed
> > > + * in the future.
> > > + */

-- 
With Best Regards,
Andy Shevchenko

Re: [PATCH v5 02/10] regulator: Move OF-specific regulator lookup code to of_regulator.c

[Chen-Yu Tsai]

On Fri, Aug 23, 2024 at 9:43 PM Andy Shevchenko
<andriy.shevchenko@linux.intel.com> wrote:
>
> On Fri, Aug 23, 2024 at 02:49:59PM +0800, Chen-Yu Tsai wrote:
> > On Thu, Aug 22, 2024 at 9:47 PM Andy Shevchenko
> > <andriy.shevchenko@linux.intel.com> wrote:
> > > On Thu, Aug 22, 2024 at 05:19:55PM +0800, Chen-Yu Tsai wrote:
>
> ...
>
> > > > +/**
> > > > + * of_get_child_regulator - get a child regulator device node
> > > > + * based on supply name
> > > > + * @parent: Parent device node
> > > > + * @prop_name: Combination regulator supply name and "-supply"
> > > > + *
> > > > + * Traverse all child nodes.
> > > > + * Extract the child regulator device node corresponding to the supply name.
> > > > + * returns the device node corresponding to the regulator if found, else
> > > > + * returns NULL.
> > >
> > > At the same time you may fix kernel-doc warnings (no "Return" section) in these
> > > three (on your wish you may fix others in a separate change, but it's not
> > > related to this series).
> >
> > As you said some other functions are missing it as well, so I'll do a
> > patch separate from this series to fix them all.
>
> But you need to fix them in this patch series. We do not add patches with known
> issues, which are really easy to fix beforehand.

I'd say that's a preexisting thing, and code movement shouldn't change it.
But, yeah, I can send another patch fixing up all the kerneldoc issues
in both files.

ChenYu

> (And below seems you indirectly agrees on that)
>
> > > > + */
>
> ...
>
> > > > +/** of_regulator_dev_lookup - lookup a regulator device with device tree only
> > >
> > > Something went wrong with the indentation.
> >
> > Will fix, and also add a "Return" section.
>
> Thank you!
>
> > > > + * @dev: Device pointer for regulator supply lookup.
> > > > + * @supply: Supply name or regulator ID.
> > > > + *
> > > > + * If successful, returns a struct regulator_dev that corresponds to the name
> > > > + * @supply and with the embedded struct device refcount incremented by one.
> > > > + * The refcount must be dropped by calling put_device().
> > > > + * On failure one of the following ERR-PTR-encoded values is returned:
> > > > + * -ENODEV if lookup fails permanently, -EPROBE_DEFER if lookup could succeed
> > > > + * in the future.
> > > > + */
>
> --
> With Best Regards,
> Andy Shevchenko
>
>

Re: [PATCH v5 02/10] regulator: Move OF-specific regulator lookup code to of_regulator.c

[kernel test robot]

Hi Chen-Yu,

kernel test robot noticed the following build warnings:

[auto build test WARNING on broonie-regulator/for-next]
[also build test WARNING on robh/for-next wsa/i2c/for-next linus/master v6.11-rc5 next-20240823]
[If your patch is applied to the wrong git tree, kindly drop us a note.
And when submitting patch, we suggest to use '--base' as documented in
https://git-scm.com/docs/git-format-patch#_base_tree_information]

url:    https://github.com/intel-lab-lkp/linux/commits/Chen-Yu-Tsai/of-dynamic-Add-of_changeset_update_prop_string/20240826-105737
base:   https://git.kernel.org/pub/scm/linux/kernel/git/broonie/regulator.git for-next
patch link:    https://lore.kernel.org/r/20240822092006.3134096-3-wenst%40chromium.org
patch subject: [PATCH v5 02/10] regulator: Move OF-specific regulator lookup code to of_regulator.c
config: i386-randconfig-005-20240826 (https://download.01.org/0day-ci/archive/20240826/202408261940.R6VmGtGO-lkp@intel.com/config)
compiler: gcc-12 (Debian 12.2.0-14) 12.2.0
reproduce (this is a W=1 build): (https://download.01.org/0day-ci/archive/20240826/202408261940.R6VmGtGO-lkp@intel.com/reproduce)

If you fix the issue in a separate patch/commit (i.e. not just a new version of
the same patch/commit), kindly add following tags
| Reported-by: kernel test robot <lkp@intel.com>
| Closes: https://lore.kernel.org/oe-kbuild-all/202408261940.R6VmGtGO-lkp@intel.com/

All warnings (new ones prefixed by >>):

   In file included from drivers/regulator/helpers.c:17:
   drivers/regulator/internal.h: In function 'of_regulator_dev_lookup':
>> drivers/regulator/internal.h:87:16: warning: returning 'int' from a function with return type 'struct regulator_dev *' makes pointer from integer without a cast [-Wint-conversion]
      87 |         return -ENODEV;
         |                ^


vim +87 drivers/regulator/internal.h

    82	
    83	#else
    84	static inline struct regulator_dev *of_regulator_dev_lookup(struct device *dev,
    85								    const char *supply)
    86	{
  > 87		return -ENODEV;
    88	}
    89	

-- 
0-DAY CI Kernel Test Service
https://github.com/intel/lkp-tests/wiki

Re: [PATCH v5 02/10] regulator: Move OF-specific regulator lookup code to of_regulator.c

[kernel test robot]

Hi Chen-Yu,

kernel test robot noticed the following build errors:

[auto build test ERROR on broonie-regulator/for-next]
[also build test ERROR on robh/for-next wsa/i2c/for-next linus/master v6.11-rc5 next-20240823]
[If your patch is applied to the wrong git tree, kindly drop us a note.
And when submitting patch, we suggest to use '--base' as documented in
https://git-scm.com/docs/git-format-patch#_base_tree_information]

url:    https://github.com/intel-lab-lkp/linux/commits/Chen-Yu-Tsai/of-dynamic-Add-of_changeset_update_prop_string/20240826-105737
base:   https://git.kernel.org/pub/scm/linux/kernel/git/broonie/regulator.git for-next
patch link:    https://lore.kernel.org/r/20240822092006.3134096-3-wenst%40chromium.org
patch subject: [PATCH v5 02/10] regulator: Move OF-specific regulator lookup code to of_regulator.c
config: parisc-randconfig-002-20240826 (https://download.01.org/0day-ci/archive/20240826/202408261956.PHxQEruO-lkp@intel.com/config)
compiler: hppa-linux-gcc (GCC) 14.1.0
reproduce (this is a W=1 build): (https://download.01.org/0day-ci/archive/20240826/202408261956.PHxQEruO-lkp@intel.com/reproduce)

If you fix the issue in a separate patch/commit (i.e. not just a new version of
the same patch/commit), kindly add following tags
| Reported-by: kernel test robot <lkp@intel.com>
| Closes: https://lore.kernel.org/oe-kbuild-all/202408261956.PHxQEruO-lkp@intel.com/

All errors (new ones prefixed by >>):

   In file included from drivers/regulator/helpers.c:17:
   drivers/regulator/internal.h: In function 'of_regulator_dev_lookup':
>> drivers/regulator/internal.h:87:16: error: returning 'int' from a function with return type 'struct regulator_dev *' makes pointer from integer without a cast [-Wint-conversion]
      87 |         return -ENODEV;
         |                ^


vim +87 drivers/regulator/internal.h

    82	
    83	#else
    84	static inline struct regulator_dev *of_regulator_dev_lookup(struct device *dev,
    85								    const char *supply)
    86	{
  > 87		return -ENODEV;
    88	}
    89	

-- 
0-DAY CI Kernel Test Service
https://github.com/intel/lkp-tests/wiki

Re: [PATCH v5 02/10] regulator: Move OF-specific regulator lookup code to of_regulator.c

[kernel test robot]

Hi Chen-Yu,

kernel test robot noticed the following build errors:

[auto build test ERROR on broonie-regulator/for-next]
[also build test ERROR on robh/for-next wsa/i2c/for-next linus/master v6.11-rc5 next-20240826]
[If your patch is applied to the wrong git tree, kindly drop us a note.
And when submitting patch, we suggest to use '--base' as documented in
https://git-scm.com/docs/git-format-patch#_base_tree_information]

url:    https://github.com/intel-lab-lkp/linux/commits/Chen-Yu-Tsai/of-dynamic-Add-of_changeset_update_prop_string/20240826-105737
base:   https://git.kernel.org/pub/scm/linux/kernel/git/broonie/regulator.git for-next
patch link:    https://lore.kernel.org/r/20240822092006.3134096-3-wenst%40chromium.org
patch subject: [PATCH v5 02/10] regulator: Move OF-specific regulator lookup code to of_regulator.c
config: i386-buildonly-randconfig-003-20240826 (https://download.01.org/0day-ci/archive/20240826/202408262013.YYc4obi2-lkp@intel.com/config)
compiler: clang version 18.1.5 (https://github.com/llvm/llvm-project 617a15a9eac96088ae5e9134248d8236e34b91b1)
reproduce (this is a W=1 build): (https://download.01.org/0day-ci/archive/20240826/202408262013.YYc4obi2-lkp@intel.com/reproduce)

If you fix the issue in a separate patch/commit (i.e. not just a new version of
the same patch/commit), kindly add following tags
| Reported-by: kernel test robot <lkp@intel.com>
| Closes: https://lore.kernel.org/oe-kbuild-all/202408262013.YYc4obi2-lkp@intel.com/

All errors (new ones prefixed by >>):

   In file included from drivers/regulator/devres.c:15:
>> drivers/regulator/internal.h:87:9: error: incompatible integer to pointer conversion returning 'int' from a function with result type 'struct regulator_dev *' [-Wint-conversion]
      87 |         return -ENODEV;
         |                ^~~~~~~
   1 error generated.


vim +87 drivers/regulator/internal.h

    82	
    83	#else
    84	static inline struct regulator_dev *of_regulator_dev_lookup(struct device *dev,
    85								    const char *supply)
    86	{
  > 87		return -ENODEV;
    88	}
    89	

-- 
0-DAY CI Kernel Test Service
https://github.com/intel/lkp-tests/wiki

[PATCH v5 03/10] regulator: Split up _regulator_get()

[Chen-Yu Tsai]

_regulator_get() contains a lot of common code doing checks prior to the
regulator lookup and housekeeping work after the lookup. Almost all the
code could be shared with a OF-specific variant of _regulator_get().

Split out the common parts so that they can be reused. The OF-specific
version of _regulator_get() will be added in a subsequent patch.
No functional changes were made.

Signed-off-by: Chen-Yu Tsai <wenst@chromium.org>
---
Changes since v4:
- New patch
---
 drivers/regulator/core.c     | 54 ++++++++++++++++++++++++++++--------
 drivers/regulator/internal.h |  4 +++
 2 files changed, 47 insertions(+), 11 deletions(-)

diff --git a/drivers/regulator/core.c b/drivers/regulator/core.c
index 361309fcae57..4d1c640cc030 100644
--- a/drivers/regulator/core.c
+++ b/drivers/regulator/core.c
@@ -2093,26 +2093,43 @@ static int regulator_resolve_supply(struct regulator_dev *rdev)
 	return ret;
 }
 
-/* Internal regulator request function */
-struct regulator *_regulator_get(struct device *dev, const char *id,
-				 enum regulator_get_type get_type)
+/* common pre-checks for regulator requests */
+int _regulator_get_common_check(struct device *dev, const char *id,
+				enum regulator_get_type get_type)
 {
-	struct regulator_dev *rdev;
-	struct regulator *regulator;
-	struct device_link *link;
-	int ret;
-
 	if (get_type >= MAX_GET_TYPE) {
 		dev_err(dev, "invalid type %d in %s\n", get_type, __func__);
-		return ERR_PTR(-EINVAL);
+		return -EINVAL;
 	}
 
 	if (id == NULL) {
 		dev_err(dev, "regulator request with no identifier\n");
-		return ERR_PTR(-EINVAL);
+		return -EINVAL;
 	}
 
-	rdev = regulator_dev_lookup(dev, id);
+	return 0;
+}
+
+/**
+ * _regulator_get_common - Common code for regulator requests
+ * @rdev: regulator device pointer as returned by *regulator_dev_lookup()
+ *       Its reference count is expected to have been incremented.
+ * @dev: device used for dev_printk messages
+ * @id: Supply name or regulator ID
+ * @get_type: enum regulator_get_type value corresponding to type of request
+ *
+ * Returns a struct regulator corresponding to @rdev,
+ * or IS_ERR() condition containing errno.
+ *
+ * This function should be chained with *regulator_dev_lookup() functions.
+ */
+struct regulator *_regulator_get_common(struct regulator_dev *rdev, struct device *dev,
+					const char *id, enum regulator_get_type get_type)
+{
+	struct regulator *regulator;
+	struct device_link *link;
+	int ret;
+
 	if (IS_ERR(rdev)) {
 		ret = PTR_ERR(rdev);
 
@@ -2228,6 +2245,21 @@ struct regulator *_regulator_get(struct device *dev, const char *id,
 	return regulator;
 }
 
+/* Internal regulator request function */
+struct regulator *_regulator_get(struct device *dev, const char *id,
+				 enum regulator_get_type get_type)
+{
+	struct regulator_dev *rdev;
+	int ret;
+
+	ret = _regulator_get_common_check(dev, id, get_type);
+	if (ret)
+		return ERR_PTR(ret);
+
+	rdev = regulator_dev_lookup(dev, id);
+	return _regulator_get_common(rdev, dev, id, get_type);
+}
+
 /**
  * regulator_get - lookup and obtain a reference to a regulator.
  * @dev: device for regulator "consumer"
diff --git a/drivers/regulator/internal.h b/drivers/regulator/internal.h
index ffeaef22169e..3718459fb0c5 100644
--- a/drivers/regulator/internal.h
+++ b/drivers/regulator/internal.h
@@ -121,6 +121,10 @@ enum regulator_get_type {
 	MAX_GET_TYPE
 };
 
+int _regulator_get_common_check(struct device *dev, const char *id,
+				enum regulator_get_type get_type);
+struct regulator *_regulator_get_common(struct regulator_dev *rdev, struct device *dev,
+					const char *id, enum regulator_get_type get_type);
 struct regulator *_regulator_get(struct device *dev, const char *id,
 				 enum regulator_get_type get_type);
 int _regulator_bulk_get(struct device *dev, int num_consumers,
-- 
2.46.0.184.g6999bdac58-goog

Re: [PATCH v5 03/10] regulator: Split up _regulator_get()

[Andy Shevchenko]

On Thu, Aug 22, 2024 at 05:19:56PM +0800, Chen-Yu Tsai wrote:
> _regulator_get() contains a lot of common code doing checks prior to the
> regulator lookup and housekeeping work after the lookup. Almost all the
> code could be shared with a OF-specific variant of _regulator_get().
> 
> Split out the common parts so that they can be reused. The OF-specific
> version of _regulator_get() will be added in a subsequent patch.
> No functional changes were made.

> +/**
> + * _regulator_get_common - Common code for regulator requests
> + * @rdev: regulator device pointer as returned by *regulator_dev_lookup()
> + *       Its reference count is expected to have been incremented.
> + * @dev: device used for dev_printk messages
> + * @id: Supply name or regulator ID
> + * @get_type: enum regulator_get_type value corresponding to type of request
> + *
> + * Returns a struct regulator corresponding to @rdev,
> + * or IS_ERR() condition containing errno.
> + *
> + * This function should be chained with *regulator_dev_lookup() functions.

kernel-doc will warn here: No "Return" section.

> + */

-- 
With Best Regards,
Andy Shevchenko

Re: [PATCH v5 03/10] regulator: Split up _regulator_get()

[Chen-Yu Tsai]

On Thu, Aug 22, 2024 at 9:49 PM Andy Shevchenko
<andriy.shevchenko@linux.intel.com> wrote:
>
> On Thu, Aug 22, 2024 at 05:19:56PM +0800, Chen-Yu Tsai wrote:
> > _regulator_get() contains a lot of common code doing checks prior to the
> > regulator lookup and housekeeping work after the lookup. Almost all the
> > code could be shared with a OF-specific variant of _regulator_get().
> >
> > Split out the common parts so that they can be reused. The OF-specific
> > version of _regulator_get() will be added in a subsequent patch.
> > No functional changes were made.
>
> > +/**
> > + * _regulator_get_common - Common code for regulator requests
> > + * @rdev: regulator device pointer as returned by *regulator_dev_lookup()
> > + *       Its reference count is expected to have been incremented.
> > + * @dev: device used for dev_printk messages
> > + * @id: Supply name or regulator ID
> > + * @get_type: enum regulator_get_type value corresponding to type of request
> > + *
> > + * Returns a struct regulator corresponding to @rdev,
> > + * or IS_ERR() condition containing errno.
> > + *
> > + * This function should be chained with *regulator_dev_lookup() functions.
>
> kernel-doc will warn here: No "Return" section.

Will fix.

ChenYu

> > + */
>
> --
> With Best Regards,
> Andy Shevchenko
>
>

[PATCH v5 04/10] regulator: Do pure DT regulator lookup in of_regulator_bulk_get_all()

[Chen-Yu Tsai]

The to-be-introduced I2C component prober needs to enable regulator
supplies (and toggle GPIO pins) for the various components it intends
to probe. To support this, a new "pure DT lookup" method for getting
regulator supplies is needed, since the device normally requesting
the supply won't get created until after the component is probed to
be available.

Convert the existing of_regulator_bulk_get_all() for this purpose.
This function has no in-tree users, as the original patch [1] that
used it was never landed. This patch changes the function ABI, but
it is straightforward to convert users.

The underlying code that supports the existing regulator_get*()
functions has been reworked in previous patches to support this
specific case. An internal OF-specific version of regulator_get(),
of_regulator_get_optional(), is added for this.

[1] https://lore.kernel.org/all/20231220203537.83479-2-jernej.skrabec@gmail.com/

Signed-off-by: Chen-Yu Tsai <wenst@chromium.org>
---
Changes since v4:
- Was "regulator: Add regulator_of_get_optional() for pure DT" in v4
- Restore platform-agnostic regulator consumer code to original state
- Move OF-specific regulator code to of_regulator.c (separate patch)
- Split _regulator_get() into three parts for reuse (separate patch)
- Add OF-specific _of_regulator_get() function
- Rename regulator_of_get_optional() to of_regulator_get_optional() for
  consistency
- Make of_regulator_get_optional() static, as it is only used internally
- Convert of_regulator_bulk_get_all()
Changes since v3:
- New patch
---
 drivers/regulator/core.c         |  2 +-
 drivers/regulator/internal.h     |  2 ++
 drivers/regulator/of_regulator.c | 54 +++++++++++++++++++++++++++-----
 3 files changed, 49 insertions(+), 9 deletions(-)

diff --git a/drivers/regulator/core.c b/drivers/regulator/core.c
index 4d1c640cc030..d70302e30731 100644
--- a/drivers/regulator/core.c
+++ b/drivers/regulator/core.c
@@ -1950,7 +1950,7 @@ static struct regulator_dev *regulator_dev_lookup(struct device *dev,
 
 	/* first do a dt based lookup */
 	if (dev && dev->of_node) {
-		r = of_regulator_dev_lookup(dev, supply);
+		r = of_regulator_dev_lookup(dev, dev->of_node, supply);
 		if (!IS_ERR(r))
 			return r;
 		if (PTR_ERR(r) == -EPROBE_DEFER)
diff --git a/drivers/regulator/internal.h b/drivers/regulator/internal.h
index 3718459fb0c5..44680ce3587b 100644
--- a/drivers/regulator/internal.h
+++ b/drivers/regulator/internal.h
@@ -67,6 +67,7 @@ static inline struct regulator_dev *dev_to_rdev(struct device *dev)
 
 #ifdef CONFIG_OF
 struct regulator_dev *of_regulator_dev_lookup(struct device *dev,
+					      struct device_node *np,
 					      const char *supply);
 struct regulator_init_data *regulator_of_get_init_data(struct device *dev,
 			         const struct regulator_desc *desc,
@@ -82,6 +83,7 @@ bool of_check_coupling_data(struct regulator_dev *rdev);
 
 #else
 static inline struct regulator_dev *of_regulator_dev_lookup(struct device *dev,
+							    struct device_node *np,
 							    const char *supply)
 {
 	return -ENODEV;
diff --git a/drivers/regulator/of_regulator.c b/drivers/regulator/of_regulator.c
index d924f5c1de59..67a5882a917e 100644
--- a/drivers/regulator/of_regulator.c
+++ b/drivers/regulator/of_regulator.c
@@ -587,22 +587,24 @@ static struct device_node *of_get_child_regulator(struct device_node *parent,
 
 /**
  * of_get_regulator - get a regulator device node based on supply name
- * @dev: Device pointer for the consumer (of regulator) device
+ * @dev: Device pointer for dev_printk messages
+ * @node: Device node pointer for supply property lookup
  * @supply: regulator supply name
  *
  * Extract the regulator device node corresponding to the supply name.
  * returns the device node corresponding to the regulator if found, else
  * returns NULL.
  */
-static struct device_node *of_get_regulator(struct device *dev, const char *supply)
+static struct device_node *of_get_regulator(struct device *dev, struct device_node *node,
+					    const char *supply)
 {
 	struct device_node *regnode = NULL;
 	char prop_name[64]; /* 64 is max size of property name */
 
-	dev_dbg(dev, "Looking up %s-supply from device tree\n", supply);
+	dev_dbg(dev, "Looking up %s-supply from device node %pOF\n", supply, node);
 
 	snprintf(prop_name, 64, "%s-supply", supply);
-	regnode = of_parse_phandle(dev->of_node, prop_name, 0);
+	regnode = of_parse_phandle(node, prop_name, 0);
 
 	if (!regnode) {
 		regnode = of_get_child_regulator(dev->of_node, prop_name);
@@ -626,7 +628,8 @@ static struct regulator_dev *of_find_regulator_by_node(struct device_node *np)
 }
 
 /** of_regulator_dev_lookup - lookup a regulator device with device tree only
- * @dev: Device pointer for regulator supply lookup.
+ * @dev: Device pointer for dev_printk messages.
+ * @node: Device node pointer for regulator supply lookup.
  * @supply: Supply name or regulator ID.
  *
  * If successful, returns a struct regulator_dev that corresponds to the name
@@ -636,13 +639,13 @@ static struct regulator_dev *of_find_regulator_by_node(struct device_node *np)
  * -ENODEV if lookup fails permanently, -EPROBE_DEFER if lookup could succeed
  * in the future.
  */
-struct regulator_dev *of_regulator_dev_lookup(struct device *dev,
+struct regulator_dev *of_regulator_dev_lookup(struct device *dev, struct device_node *np,
 					      const char *supply)
 {
 	struct regulator_dev *r;
 	struct device_node *node;
 
-	node = of_get_regulator(dev, supply);
+	node = of_get_regulator(dev, np, supply);
 	if (node) {
 		r = of_find_regulator_by_node(node);
 		of_node_put(node);
@@ -659,6 +662,41 @@ struct regulator_dev *of_regulator_dev_lookup(struct device *dev,
 	return ERR_PTR(-ENODEV);
 }
 
+static struct regulator *_of_regulator_get(struct device *dev, struct device_node *node,
+					   const char *id, enum regulator_get_type get_type)
+{
+	struct regulator_dev *r;
+	int ret;
+
+	ret = _regulator_get_common_check(dev, id, get_type);
+	if (ret)
+		return ERR_PTR(ret);
+
+	r = of_regulator_dev_lookup(dev, node, id);
+	return _regulator_get_common(r, dev, id, get_type);
+}
+
+/**
+ * of_regulator_get_optional - get optional regulator via device tree lookup
+ * @dev: device used for dev_printk messages
+ * @node: device node for regulator "consumer"
+ * @id: Supply name
+ *
+ * Returns a struct regulator corresponding to the regulator producer,
+ * or IS_ERR() condition containing errno.
+ *
+ * This is intended for use by consumers that want to get a regulator
+ * supply directly from a device node, and can and want to deal with
+ * absence of such supplies. This will _not_ consider supply aliases.
+ * See regulator_dev_lookup().
+ */
+static struct regulator *of_regulator_get_optional(struct device *dev,
+						   struct device_node *node,
+						   const char *id)
+{
+	return _of_regulator_get(NULL, node, id, OPTIONAL_GET);
+}
+
 /*
  * Returns number of regulators coupled with rdev.
  */
@@ -875,7 +913,7 @@ int of_regulator_bulk_get_all(struct device *dev, struct device_node *np,
 		} else {
 			memcpy(name, prop->name, i);
 			name[i] = '\0';
-			tmp = regulator_get(dev, name);
+			tmp = of_regulator_get_optional(dev, np, name);
 			if (IS_ERR(tmp)) {
 				ret = PTR_ERR(tmp);
 				goto error;
-- 
2.46.0.184.g6999bdac58-goog

Re: [PATCH v5 04/10] regulator: Do pure DT regulator lookup in of_regulator_bulk_get_all()

[Andy Shevchenko]

On Thu, Aug 22, 2024 at 05:19:57PM +0800, Chen-Yu Tsai wrote:
> The to-be-introduced I2C component prober needs to enable regulator
> supplies (and toggle GPIO pins) for the various components it intends
> to probe. To support this, a new "pure DT lookup" method for getting
> regulator supplies is needed, since the device normally requesting
> the supply won't get created until after the component is probed to
> be available.
> 
> Convert the existing of_regulator_bulk_get_all() for this purpose.
> This function has no in-tree users, as the original patch [1] that
> used it was never landed. This patch changes the function ABI, but
> it is straightforward to convert users.
> 
> The underlying code that supports the existing regulator_get*()
> functions has been reworked in previous patches to support this
> specific case. An internal OF-specific version of regulator_get(),
> of_regulator_get_optional(), is added for this.

> [1] https://lore.kernel.org/all/20231220203537.83479-2-jernej.skrabec@gmail.com/

Make it Link tag

Link: https://lore.kernel.org/all/20231220203537.83479-2-jernej.skrabec@gmail.com/ [1]
Signed-off-by: Chen-Yu Tsai <wenst@chromium.org>

...

>  	/* first do a dt based lookup */
>  	if (dev && dev->of_node) {

	if (dev_of_node())


> -		r = of_regulator_dev_lookup(dev, supply);
> +		r = of_regulator_dev_lookup(dev, dev->of_node, supply);

	dev_of_node()

>  		if (!IS_ERR(r))
>  			return r;
>  		if (PTR_ERR(r) == -EPROBE_DEFER)

...

>  /**
>   * of_get_regulator - get a regulator device node based on supply name
> - * @dev: Device pointer for the consumer (of regulator) device
> + * @dev: Device pointer for dev_printk messages

dev_printk()

> + * @node: Device node pointer for supply property lookup
>   * @supply: regulator supply name
>   *
>   * Extract the regulator device node corresponding to the supply name.
>   * returns the device node corresponding to the regulator if found, else
>   * returns NULL.
>   */

...

>  /** of_regulator_dev_lookup - lookup a regulator device with device tree only
> - * @dev: Device pointer for regulator supply lookup.
> + * @dev: Device pointer for dev_printk messages.

Ditto.

> + * @node: Device node pointer for regulator supply lookup.
>   * @supply: Supply name or regulator ID.
>   *
>   * If successful, returns a struct regulator_dev that corresponds to the name
> @@ -636,13 +639,13 @@ static struct regulator_dev *of_find_regulator_by_node(struct device_node *np)
>   * -ENODEV if lookup fails permanently, -EPROBE_DEFER if lookup could succeed
>   * in the future.
>   */

...

> +/**
> + * of_regulator_get_optional - get optional regulator via device tree lookup
> + * @dev: device used for dev_printk messages

Ditto.

> + * @node: device node for regulator "consumer"
> + * @id: Supply name
> + *
> + * Returns a struct regulator corresponding to the regulator producer,
> + * or IS_ERR() condition containing errno.
> + *
> + * This is intended for use by consumers that want to get a regulator
> + * supply directly from a device node, and can and want to deal with
> + * absence of such supplies. This will _not_ consider supply aliases.
> + * See regulator_dev_lookup().

Fix kernel-doc warning.

> + */

-- 
With Best Regards,
Andy Shevchenko

Re: [PATCH v5 04/10] regulator: Do pure DT regulator lookup in of_regulator_bulk_get_all()

[Chen-Yu Tsai]

On Thu, Aug 22, 2024 at 9:54 PM Andy Shevchenko
<andriy.shevchenko@linux.intel.com> wrote:
>
> On Thu, Aug 22, 2024 at 05:19:57PM +0800, Chen-Yu Tsai wrote:
> > The to-be-introduced I2C component prober needs to enable regulator
> > supplies (and toggle GPIO pins) for the various components it intends
> > to probe. To support this, a new "pure DT lookup" method for getting
> > regulator supplies is needed, since the device normally requesting
> > the supply won't get created until after the component is probed to
> > be available.
> >
> > Convert the existing of_regulator_bulk_get_all() for this purpose.
> > This function has no in-tree users, as the original patch [1] that
> > used it was never landed. This patch changes the function ABI, but
> > it is straightforward to convert users.
> >
> > The underlying code that supports the existing regulator_get*()
> > functions has been reworked in previous patches to support this
> > specific case. An internal OF-specific version of regulator_get(),
> > of_regulator_get_optional(), is added for this.
>
> > [1] https://lore.kernel.org/all/20231220203537.83479-2-jernej.skrabec@gmail.com/
>
> Make it Link tag
>
> Link: https://lore.kernel.org/all/20231220203537.83479-2-jernej.skrabec@gmail.com/ [1]
> Signed-off-by: Chen-Yu Tsai <wenst@chromium.org>
>
> ...
>
> >       /* first do a dt based lookup */
> >       if (dev && dev->of_node) {
>
>         if (dev_of_node())
>

Not part of the original change, but I don't see why not.

> > -             r = of_regulator_dev_lookup(dev, supply);
> > +             r = of_regulator_dev_lookup(dev, dev->of_node, supply);
>
>         dev_of_node()

Ack.

> >               if (!IS_ERR(r))
> >                       return r;
> >               if (PTR_ERR(r) == -EPROBE_DEFER)
>
> ...
>
> >  /**
> >   * of_get_regulator - get a regulator device node based on supply name
> > - * @dev: Device pointer for the consumer (of regulator) device
> > + * @dev: Device pointer for dev_printk messages
>
> dev_printk()

Ack.

> > + * @node: Device node pointer for supply property lookup
> >   * @supply: regulator supply name
> >   *
> >   * Extract the regulator device node corresponding to the supply name.
> >   * returns the device node corresponding to the regulator if found, else
> >   * returns NULL.
> >   */
>
> ...
>
> >  /** of_regulator_dev_lookup - lookup a regulator device with device tree only
> > - * @dev: Device pointer for regulator supply lookup.
> > + * @dev: Device pointer for dev_printk messages.
>
> Ditto.

Ack.

> > + * @node: Device node pointer for regulator supply lookup.
> >   * @supply: Supply name or regulator ID.
> >   *
> >   * If successful, returns a struct regulator_dev that corresponds to the name
> > @@ -636,13 +639,13 @@ static struct regulator_dev *of_find_regulator_by_node(struct device_node *np)
> >   * -ENODEV if lookup fails permanently, -EPROBE_DEFER if lookup could succeed
> >   * in the future.
> >   */
>
> ...
>
> > +/**
> > + * of_regulator_get_optional - get optional regulator via device tree lookup
> > + * @dev: device used for dev_printk messages
>
> Ditto.

Ack.

> > + * @node: device node for regulator "consumer"
> > + * @id: Supply name
> > + *
> > + * Returns a struct regulator corresponding to the regulator producer,
> > + * or IS_ERR() condition containing errno.
> > + *
> > + * This is intended for use by consumers that want to get a regulator
> > + * supply directly from a device node, and can and want to deal with
> > + * absence of such supplies. This will _not_ consider supply aliases.
> > + * See regulator_dev_lookup().
>
> Fix kernel-doc warning.

Ack.


Thanks
ChenYu

> > + */
>
> --
> With Best Regards,
> Andy Shevchenko
>
>

[PATCH v5 05/10] gpiolib: Add gpio_property_name_length()

[Chen-Yu Tsai]

The I2C device tree component prober needs to get and toggle GPIO lines
for the components it intends to probe. These components may not use the
same name for their GPIO lines, so the prober must go through the device
tree, check each property to see it is a GPIO property, and get the GPIO
line.

Instead of duplicating the GPIO suffixes, or exporting them to the
prober to do pattern matching, simply add and export a new function that
does the pattern matching and returns the length of the GPIO name. The
caller can then use that to copy out the name if it needs to.

Signed-off-by: Chen-Yu Tsai <wenst@chromium.org>

---
Changes since v4:
- new patch

Depends on commit 4b91188dced8 ("gpiolib: Replace gpio_suffix_count
with NULL-terminated array").
---
 drivers/gpio/gpiolib.c        | 44 +++++++++++++++++++++++++++++++++++
 include/linux/gpio/consumer.h |  2 ++
 2 files changed, 46 insertions(+)

diff --git a/drivers/gpio/gpiolib.c b/drivers/gpio/gpiolib.c
index 3903d0a75304..e4228ef6f131 100644
--- a/drivers/gpio/gpiolib.c
+++ b/drivers/gpio/gpiolib.c
@@ -4295,6 +4295,50 @@ struct gpio_desc *fwnode_gpiod_get_index(struct fwnode_handle *fwnode,
 }
 EXPORT_SYMBOL_GPL(fwnode_gpiod_get_index);
 
+/**
+ * gpio_property_name_length - Returns the GPIO name length from a property name
+ * @str:	string to check
+ *
+ * This function checks if the given name matches the GPIO property patterns, and
+ * returns the length of the name of the GPIO. The pattern is "*-<GPIO suffix>"
+ * or just "<GPIO suffix>".
+ *
+ * Returns:
+ * The length of the string before '-' if it matches "*-<GPIO suffix>", or
+ * 0 if no name part, just the suffix, or
+ * -EINVAL if the string doesn't match the pattern.
+ */
+int gpio_property_name_length(const char *str)
+{
+	size_t len;
+
+	len = strlen(str);
+
+	/* string need to be at minimum len(gpio) */
+	if (len < 4)
+		return -EINVAL;
+
+	/* Check for no-name case: "gpio" / "gpios" */
+	for (const char *const *p = gpio_suffixes; *p; p++)
+		if (!strcmp(str, *p))
+			return 0;
+
+	for (size_t i = len - 4; i > 0; i--) {
+		/* find right-most '-' and check if remainder matches suffix */
+		if (str[i] != '-')
+			continue;
+
+		for (const char *const *p = gpio_suffixes; *p; p++)
+			if (!strcmp(str + i + 1, *p))
+				return i;
+
+		return -EINVAL;
+	}
+
+	return -EINVAL;
+}
+EXPORT_SYMBOL_GPL(gpio_property_name_length);
+
 /**
  * gpiod_count - return the number of GPIOs associated with a device / function
  *		or -ENOENT if no GPIO has been assigned to the requested function
diff --git a/include/linux/gpio/consumer.h b/include/linux/gpio/consumer.h
index db2dfbae8edb..ce3a5f86a037 100644
--- a/include/linux/gpio/consumer.h
+++ b/include/linux/gpio/consumer.h
@@ -56,6 +56,8 @@ enum gpiod_flags {
 
 #ifdef CONFIG_GPIOLIB
 
+int gpio_property_name_length(const char *name);
+
 /* Return the number of GPIOs associated with a device / function */
 int gpiod_count(struct device *dev, const char *con_id);
 
-- 
2.46.0.184.g6999bdac58-goog

Re: [PATCH v5 05/10] gpiolib: Add gpio_property_name_length()

[Andy Shevchenko]

On Thu, Aug 22, 2024 at 05:19:58PM +0800, Chen-Yu Tsai wrote:
> The I2C device tree component prober needs to get and toggle GPIO lines
> for the components it intends to probe. These components may not use the
> same name for their GPIO lines, so the prober must go through the device
> tree, check each property to see it is a GPIO property, and get the GPIO
> line.
> 
> Instead of duplicating the GPIO suffixes, or exporting them to the
> prober to do pattern matching, simply add and export a new function that
> does the pattern matching and returns the length of the GPIO name. The
> caller can then use that to copy out the name if it needs to.

...

> +/**
> + * gpio_property_name_length - Returns the GPIO name length from a property name
> + * @str:	string to check

It's property name, so, I would name this 'propname'.

> + * This function checks if the given name matches the GPIO property patterns, and
> + * returns the length of the name of the GPIO. The pattern is "*-<GPIO suffix>"
> + * or just "<GPIO suffix>".
> + *
> + * Returns:
> + * The length of the string before '-' if it matches "*-<GPIO suffix>", or

What about "x-y-gpios"? It's unclear what will be the behaviour.

> + * 0 if no name part, just the suffix, or
> + * -EINVAL if the string doesn't match the pattern.
> + */
> +int gpio_property_name_length(const char *str)

gpio_get_... ?

> +{
> +	size_t len;
> +
> +	len = strlen(str);

If it has a thousands characters...?

> +	/* string need to be at minimum len(gpio) */
> +	if (len < 4)
> +		return -EINVAL;

Do we really need it here? See below as well.

> +	/* Check for no-name case: "gpio" / "gpios" */
> +	for (const char *const *p = gpio_suffixes; *p; p++)
> +		if (!strcmp(str, *p))
> +			return 0;

> +	for (size_t i = len - 4; i > 0; i--) {
> +		/* find right-most '-' and check if remainder matches suffix */
> +		if (str[i] != '-')
> +			continue;
> +
> +		for (const char *const *p = gpio_suffixes; *p; p++)
> +			if (!strcmp(str + i + 1, *p))
> +				return i;
> +
> +		return -EINVAL;
> +	}

This can be combined with the above

	for (const char *const *p = gpio_suffixes; *p; p++) {
		/*
		 * Find right-most '-' and check if remainder matches suffix.
		 * If no separator found, check for no-name cases.
		 */
		dash = strrchr(propname, '-');
		if (!strcmp(dash ? dash + 1 : propname, *p))
			return i;
	}

> +	return -EINVAL;
> +}

-- 
With Best Regards,
Andy Shevchenko

Re: [PATCH v5 05/10] gpiolib: Add gpio_property_name_length()

[Chen-Yu Tsai]

On Thu, Aug 22, 2024 at 10:37 PM Andy Shevchenko
<andriy.shevchenko@linux.intel.com> wrote:
>
> On Thu, Aug 22, 2024 at 05:19:58PM +0800, Chen-Yu Tsai wrote:
> > The I2C device tree component prober needs to get and toggle GPIO lines
> > for the components it intends to probe. These components may not use the
> > same name for their GPIO lines, so the prober must go through the device
> > tree, check each property to see it is a GPIO property, and get the GPIO
> > line.
> >
> > Instead of duplicating the GPIO suffixes, or exporting them to the
> > prober to do pattern matching, simply add and export a new function that
> > does the pattern matching and returns the length of the GPIO name. The
> > caller can then use that to copy out the name if it needs to.
>
> ...
>
> > +/**
> > + * gpio_property_name_length - Returns the GPIO name length from a property name
> > + * @str:     string to check
>
> It's property name, so, I would name this 'propname'.

Ack.

> > + * This function checks if the given name matches the GPIO property patterns, and
> > + * returns the length of the name of the GPIO. The pattern is "*-<GPIO suffix>"
> > + * or just "<GPIO suffix>".
> > + *
> > + * Returns:
> > + * The length of the string before '-' if it matches "*-<GPIO suffix>", or
>
> What about "x-y-gpios"? It's unclear what will be the behaviour.

I thought it was implied that the '-' mentioned here is the one before the
suffix. I made it more explicit.

> > + * 0 if no name part, just the suffix, or
> > + * -EINVAL if the string doesn't match the pattern.
> > + */
> > +int gpio_property_name_length(const char *str)
>
> gpio_get_... ?

Ack.

> > +{
> > +     size_t len;
> > +
> > +     len = strlen(str);
>
> If it has a thousands characters...?

Shouldn't matter much? I suppose using strrchr() as you suggested
requires one less pass.

> > +     /* string need to be at minimum len(gpio) */
> > +     if (len < 4)
> > +             return -EINVAL;
>
> Do we really need it here? See below as well.
>
> > +     /* Check for no-name case: "gpio" / "gpios" */
> > +     for (const char *const *p = gpio_suffixes; *p; p++)
> > +             if (!strcmp(str, *p))
> > +                     return 0;
>
> > +     for (size_t i = len - 4; i > 0; i--) {
> > +             /* find right-most '-' and check if remainder matches suffix */
> > +             if (str[i] != '-')
> > +                     continue;
> > +
> > +             for (const char *const *p = gpio_suffixes; *p; p++)
> > +                     if (!strcmp(str + i + 1, *p))
> > +                             return i;
> > +
> > +             return -EINVAL;
> > +     }
>
> This can be combined with the above
>
>         for (const char *const *p = gpio_suffixes; *p; p++) {
>                 /*
>                  * Find right-most '-' and check if remainder matches suffix.
>                  * If no separator found, check for no-name cases.
>                  */
>                 dash = strrchr(propname, '-');

I believe this line could be moved out of the for-loop. Otherwise it
looks much more concise compared to my version. I'll omit the comment
though, as it is just rehashing the kerneldoc description, and now
that the function is so short, it shouldn't be hard to read.

I'll add you as "Suggested-by".


Thanks
ChenYu

>                 if (!strcmp(dash ? dash + 1 : propname, *p))
>                         return i;
>         }
>
> > +     return -EINVAL;
> > +}
>
> --
> With Best Regards,
> Andy Shevchenko
>
>

Re: [PATCH v5 05/10] gpiolib: Add gpio_property_name_length()

[Andy Shevchenko]

On Fri, Aug 23, 2024 at 03:50:55PM +0800, Chen-Yu Tsai wrote:
> On Thu, Aug 22, 2024 at 10:37 PM Andy Shevchenko
> <andriy.shevchenko@linux.intel.com> wrote:
> > On Thu, Aug 22, 2024 at 05:19:58PM +0800, Chen-Yu Tsai wrote:

...

> > > +     len = strlen(str);
> >
> > If it has a thousands characters...?
> 
> Shouldn't matter much? I suppose using strrchr() as you suggested
> requires one less pass.

Yes, this is the point.

...

> > This can be combined with the above
> >
> >         for (const char *const *p = gpio_suffixes; *p; p++) {
> >                 /*
> >                  * Find right-most '-' and check if remainder matches suffix.
> >                  * If no separator found, check for no-name cases.
> >                  */
> >                 dash = strrchr(propname, '-');
> 
> I believe this line could be moved out of the for-loop. Otherwise it
> looks much more concise compared to my version. I'll omit the comment
> though, as it is just rehashing the kerneldoc description, and now
> that the function is so short, it shouldn't be hard to read.

Agree. And I put comment inside the loop, while it should be outside. But then,
as you said, the function becomes so little that kernel-doc above does the job,
hence no comment in the code needed.

> I'll add you as "Suggested-by".

Fine with me.

> >                 if (!strcmp(dash ? dash + 1 : propname, *p))
> >                         return i;
> >         }

-- 
With Best Regards,
Andy Shevchenko

[PATCH v5 06/10] i2c: Introduce OF component probe function

[Chen-Yu Tsai]

Some devices are designed and manufactured with some components having
multiple drop-in replacement options. These components are often
connected to the mainboard via ribbon cables, having the same signals
and pin assignments across all options. These may include the display
panel and touchscreen on laptops and tablets, and the trackpad on
laptops. Sometimes which component option is used in a particular device
can be detected by some firmware provided identifier, other times that
information is not available, and the kernel has to try to probe each
device.

This change attempts to make the "probe each device" case cleaner. The
current approach is to have all options added and enabled in the device
tree. The kernel would then bind each device and run each driver's probe
function. This works, but has been broken before due to the introduction
of asynchronous probing, causing multiple instances requesting "shared"
resources, such as pinmuxes, GPIO pins, interrupt lines, at the same
time, with only one instance succeeding. Work arounds for these include
moving the pinmux to the parent I2C controller, using GPIO hogs or
pinmux settings to keep the GPIO pins in some fixed configuration, and
requesting the interrupt line very late. Such configurations can be seen
on the MT8183 Krane Chromebook tablets, and the Qualcomm sc8280xp-based
Lenovo Thinkpad 13S.

Instead of this delicate dance between drivers and device tree quirks,
this change introduces a simple I2C component probe. function For a
given class of devices on the same I2C bus, it will go through all of
them, doing a simple I2C read transfer and see which one of them responds.
It will then enable the device that responds.

This requires some minor modifications in the existing device tree. The
status for all the device nodes for the component options must be set
to "failed-needs-probe". This makes it clear that some mechanism is
needed to enable one of them, and also prevents the prober and device
drivers running at the same time.

Signed-off-by: Chen-Yu Tsai <wenst@chromium.org>
---
Changes since v4:
- Split code into helper functions
- Use scoped helpers and __free() to reduce error path

Changes since v3:
- Complete kernel-doc
- Return different error if I2C controller is disabled
- Expand comment to explain assumptions and constraints
- Split for-loop finding target node and operations on target node
- Add missing i2c_put_adapter()
- Move prober code to separate file

Rob also asked why there was a limitation of "exactly one touchscreen
will be enabled across the whole tree".

The use case this prober currently targets is a component on consumer
electronics (tablet or laptop) being swapped out due to cost or supply
reasons. Designs with multiple components of the same type are pretty
rare. The way the next patch is written also assumes this for efficiency
reasons.

Changes since v2:
- New patch split out from "of: Introduce hardware prober driver"
- Addressed Rob's comments
  - Move i2c prober to i2c subsystem
  - Use of_node_is_available() to check if node is enabled.
  - Use OF changeset API to update status property
- Addressed Andy's comments
  - Probe function now accepts "struct device *dev" instead to reduce
    line length and dereferences
  - Move "ret = 0" to just before for_each_child_of_node(i2c_node, node)
---
 drivers/i2c/Makefile             |   1 +
 drivers/i2c/i2c-core-of-prober.c | 157 +++++++++++++++++++++++++++++++
 include/linux/i2c.h              |   4 +
 3 files changed, 162 insertions(+)
 create mode 100644 drivers/i2c/i2c-core-of-prober.c

diff --git a/drivers/i2c/Makefile b/drivers/i2c/Makefile
index 3f71ce4711e3..0c543c742895 100644
--- a/drivers/i2c/Makefile
+++ b/drivers/i2c/Makefile
@@ -9,6 +9,7 @@ i2c-core-objs 			:= i2c-core-base.o i2c-core-smbus.o
 i2c-core-$(CONFIG_ACPI)		+= i2c-core-acpi.o
 i2c-core-$(CONFIG_I2C_SLAVE) 	+= i2c-core-slave.o
 i2c-core-$(CONFIG_OF) 		+= i2c-core-of.o
+i2c-core-$(CONFIG_OF_DYNAMIC) 	+= i2c-core-of-prober.o
 
 obj-$(CONFIG_I2C_SMBUS)		+= i2c-smbus.o
 obj-$(CONFIG_I2C_CHARDEV)	+= i2c-dev.o
diff --git a/drivers/i2c/i2c-core-of-prober.c b/drivers/i2c/i2c-core-of-prober.c
new file mode 100644
index 000000000000..bb7b231201b0
--- /dev/null
+++ b/drivers/i2c/i2c-core-of-prober.c
@@ -0,0 +1,157 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * Linux I2C core OF component prober code
+ *
+ * Copyright (C) 2024 Google LLC
+ */
+
+#include <linux/cleanup.h>
+#include <linux/device.h>
+#include <linux/dev_printk.h>
+#include <linux/err.h>
+#include <linux/i2c.h>
+#include <linux/module.h>
+#include <linux/of.h>
+#include <linux/slab.h>
+
+/*
+ * Some devices, such as Google Hana Chromebooks, are produced by multiple
+ * vendors each using their preferred components. Such components are all
+ * in the device tree. Instead of having all of them enabled and having each
+ * driver separately try and probe its device while fighting over shared
+ * resources, they can be marked as "fail-needs-probe" and have a prober
+ * figure out which one is actually used beforehand.
+ *
+ * This prober assumes such drop-in parts are on the same I2C bus, have
+ * non-conflicting addresses, and can be directly probed by seeing which
+ * address responds.
+ *
+ * TODO:
+ * - Support handling common regulators and GPIOs.
+ * - Support I2C muxes
+ */
+
+static struct device_node *i2c_of_probe_get_i2c_node(struct device *dev, const char *type)
+{
+	struct device_node *node __free(device_node) = of_find_node_by_name(NULL, type);
+	if (!node)
+		return dev_err_ptr_probe(dev, -ENODEV, "Could not find %s device node\n", type);
+
+	struct device_node *i2c_node __free(device_node) = of_get_parent(node);
+	if (!of_node_name_eq(i2c_node, "i2c"))
+		return dev_err_ptr_probe(dev, -EINVAL, "%s device isn't on I2C bus\n", type);
+
+	if (!of_device_is_available(i2c_node))
+		return dev_err_ptr_probe(dev, -ENODEV, "I2C controller not available\n");
+
+	return no_free_ptr(i2c_node);
+}
+
+static int i2c_of_probe_enable_node(struct device *dev, struct device_node *node)
+{
+	int ret;
+
+	dev_info(dev, "Enabling %pOF\n", node);
+
+	struct of_changeset *ocs __free(kfree) = kzalloc(sizeof(*ocs), GFP_KERNEL);
+	if (!ocs)
+		return -ENOMEM;
+
+	of_changeset_init(ocs);
+	ret = of_changeset_update_prop_string(ocs, node, "status", "okay");
+	if (ret)
+		return ret;
+
+	ret = of_changeset_apply(ocs);
+	if (!ret) {
+		/*
+		 * ocs is intentionally kept around as it needs to
+		 * exist as long as the change is applied.
+		 */
+		void *ptr __always_unused = no_free_ptr(ocs);
+	} else {
+		/* ocs needs to be explicitly cleaned up before being freed. */
+		of_changeset_destroy(ocs);
+	}
+
+	return ret;
+}
+
+/**
+ * i2c_of_probe_component() - probe for devices of "type" on the same i2c bus
+ * @dev: &struct device of the caller, only used for dev_* printk messages
+ * @type: a string to match the device node name prefix to probe for
+ *
+ * Probe for possible I2C components of the same "type" on the same I2C bus
+ * that have their status marked as "fail".
+ *
+ * Assumes that across the entire device tree the only instances of nodes
+ * prefixed with "type" are the ones that need handling for second source
+ * components. In other words, if type is "touchscreen", then all device
+ * nodes named "touchscreen*" are the ones that need probing. There cannot
+ * be another "touchscreen" node that is already enabled.
+ *
+ * Assumes that for each "type" of component, only one actually exists. In
+ * other words, only one matching and existing device will be enabled.
+ *
+ * Context: Process context only. Does non-atomic I2C transfers.
+ *          Should only be used from a driver probe function, as the function
+ *          can return -EPROBE_DEFER if the I2C adapter is unavailable.
+ * Return: 0 on success or no-op, error code otherwise.
+ *         A no-op can happen when it seems like the device tree already
+ *         has components of the type to be probed already enabled. This
+ *         can happen when the device tree had not been updated to mark
+ *         the status of the to-be-probed components as "fail". Or this
+ *         function was already run with the same parameters and succeeded
+ *         in enabling a component. The latter could happen if the user
+ *         had multiple types of components to probe, and one of them down
+ *         the list caused a deferred probe. This is expected behavior.
+ */
+int i2c_of_probe_component(struct device *dev, const char *type)
+{
+	struct i2c_adapter *i2c;
+	int ret;
+
+	struct device_node *i2c_node __free(device_node) = i2c_of_probe_get_i2c_node(dev, type);
+	if (IS_ERR(i2c_node))
+		return PTR_ERR(i2c_node);
+
+	for_each_child_of_node_scoped(i2c_node, node) {
+		if (!of_node_name_prefix(node, type))
+			continue;
+		if (!of_device_is_available(node))
+			continue;
+
+		/*
+		 * Device tree has component already enabled. Either the
+		 * device tree isn't supported or we already probed once.
+		 */
+		return 0;
+	}
+
+	i2c = of_get_i2c_adapter_by_node(i2c_node);
+	if (!i2c)
+		return dev_err_probe(dev, -EPROBE_DEFER, "Couldn't get I2C adapter\n");
+
+	ret = 0;
+	for_each_child_of_node_scoped(i2c_node, node) {
+		union i2c_smbus_data data;
+		u32 addr;
+
+		if (!of_node_name_prefix(node, type))
+			continue;
+		if (of_property_read_u32(node, "reg", &addr))
+			continue;
+		if (i2c_smbus_xfer(i2c, addr, 0, I2C_SMBUS_READ, 0, I2C_SMBUS_BYTE, &data) < 0)
+			continue;
+
+		/* Found a device that is responding */
+		ret = i2c_of_probe_enable_node(dev, node);
+		break;
+	}
+
+	i2c_put_adapter(i2c);
+
+	return ret;
+}
+EXPORT_SYMBOL_GPL(i2c_of_probe_component);
diff --git a/include/linux/i2c.h b/include/linux/i2c.h
index 377def497298..4c48ea93ec49 100644
--- a/include/linux/i2c.h
+++ b/include/linux/i2c.h
@@ -1030,6 +1030,10 @@ const struct of_device_id
 int of_i2c_get_board_info(struct device *dev, struct device_node *node,
 			  struct i2c_board_info *info);
 
+#if IS_ENABLED(CONFIG_OF_DYNAMIC)
+int i2c_of_probe_component(struct device *dev, const char *type);
+#endif
+
 #else
 
 static inline struct i2c_client *of_find_i2c_device_by_node(struct device_node *node)
-- 
2.46.0.184.g6999bdac58-goog

Re: [PATCH v5 06/10] i2c: Introduce OF component probe function

[Andy Shevchenko]

On Thu, Aug 22, 2024 at 05:19:59PM +0800, Chen-Yu Tsai wrote:
> Some devices are designed and manufactured with some components having
> multiple drop-in replacement options. These components are often
> connected to the mainboard via ribbon cables, having the same signals
> and pin assignments across all options. These may include the display
> panel and touchscreen on laptops and tablets, and the trackpad on
> laptops. Sometimes which component option is used in a particular device
> can be detected by some firmware provided identifier, other times that
> information is not available, and the kernel has to try to probe each
> device.
> 
> This change attempts to make the "probe each device" case cleaner. The
> current approach is to have all options added and enabled in the device
> tree. The kernel would then bind each device and run each driver's probe
> function. This works, but has been broken before due to the introduction
> of asynchronous probing, causing multiple instances requesting "shared"
> resources, such as pinmuxes, GPIO pins, interrupt lines, at the same
> time, with only one instance succeeding. Work arounds for these include
> moving the pinmux to the parent I2C controller, using GPIO hogs or
> pinmux settings to keep the GPIO pins in some fixed configuration, and
> requesting the interrupt line very late. Such configurations can be seen
> on the MT8183 Krane Chromebook tablets, and the Qualcomm sc8280xp-based
> Lenovo Thinkpad 13S.
> 
> Instead of this delicate dance between drivers and device tree quirks,
> this change introduces a simple I2C component probe. function For a
> given class of devices on the same I2C bus, it will go through all of
> them, doing a simple I2C read transfer and see which one of them responds.
> It will then enable the device that responds.
> 
> This requires some minor modifications in the existing device tree. The
> status for all the device nodes for the component options must be set
> to "failed-needs-probe". This makes it clear that some mechanism is
> needed to enable one of them, and also prevents the prober and device
> drivers running at the same time.

...

> --- a/drivers/i2c/Makefile
> +++ b/drivers/i2c/Makefile
> @@ -9,6 +9,7 @@ i2c-core-objs 			:= i2c-core-base.o i2c-core-smbus.o
>  i2c-core-$(CONFIG_ACPI)		+= i2c-core-acpi.o
>  i2c-core-$(CONFIG_I2C_SLAVE) 	+= i2c-core-slave.o
>  i2c-core-$(CONFIG_OF) 		+= i2c-core-of.o
> +i2c-core-$(CONFIG_OF_DYNAMIC) 	+= i2c-core-of-prober.o

Seems like all the above (except ACPI) have the same issue, i.e. TABs/spaces
mixture.

...

> +	ret = of_changeset_apply(ocs);
> +	if (!ret) {

Why not positive conditional?

> +		/*
> +		 * ocs is intentionally kept around as it needs to
> +		 * exist as long as the change is applied.
> +		 */
> +		void *ptr __always_unused = no_free_ptr(ocs);
> +	} else {
> +		/* ocs needs to be explicitly cleaned up before being freed. */
> +		of_changeset_destroy(ocs);
> +	}

-- 
With Best Regards,
Andy Shevchenko

Re: [PATCH v5 06/10] i2c: Introduce OF component probe function

[Chen-Yu Tsai]

On Thu, Aug 22, 2024 at 10:02 PM Andy Shevchenko
<andriy.shevchenko@linux.intel.com> wrote:
>
> On Thu, Aug 22, 2024 at 05:19:59PM +0800, Chen-Yu Tsai wrote:
> > Some devices are designed and manufactured with some components having
> > multiple drop-in replacement options. These components are often
> > connected to the mainboard via ribbon cables, having the same signals
> > and pin assignments across all options. These may include the display
> > panel and touchscreen on laptops and tablets, and the trackpad on
> > laptops. Sometimes which component option is used in a particular device
> > can be detected by some firmware provided identifier, other times that
> > information is not available, and the kernel has to try to probe each
> > device.
> >
> > This change attempts to make the "probe each device" case cleaner. The
> > current approach is to have all options added and enabled in the device
> > tree. The kernel would then bind each device and run each driver's probe
> > function. This works, but has been broken before due to the introduction
> > of asynchronous probing, causing multiple instances requesting "shared"
> > resources, such as pinmuxes, GPIO pins, interrupt lines, at the same
> > time, with only one instance succeeding. Work arounds for these include
> > moving the pinmux to the parent I2C controller, using GPIO hogs or
> > pinmux settings to keep the GPIO pins in some fixed configuration, and
> > requesting the interrupt line very late. Such configurations can be seen
> > on the MT8183 Krane Chromebook tablets, and the Qualcomm sc8280xp-based
> > Lenovo Thinkpad 13S.
> >
> > Instead of this delicate dance between drivers and device tree quirks,
> > this change introduces a simple I2C component probe. function For a
> > given class of devices on the same I2C bus, it will go through all of
> > them, doing a simple I2C read transfer and see which one of them responds.
> > It will then enable the device that responds.
> >
> > This requires some minor modifications in the existing device tree. The
> > status for all the device nodes for the component options must be set
> > to "failed-needs-probe". This makes it clear that some mechanism is
> > needed to enable one of them, and also prevents the prober and device
> > drivers running at the same time.
>
> ...
>
> > --- a/drivers/i2c/Makefile
> > +++ b/drivers/i2c/Makefile
> > @@ -9,6 +9,7 @@ i2c-core-objs                         := i2c-core-base.o i2c-core-smbus.o
> >  i2c-core-$(CONFIG_ACPI)              += i2c-core-acpi.o
> >  i2c-core-$(CONFIG_I2C_SLAVE)         += i2c-core-slave.o
> >  i2c-core-$(CONFIG_OF)                += i2c-core-of.o
> > +i2c-core-$(CONFIG_OF_DYNAMIC)        += i2c-core-of-prober.o
>
> Seems like all the above (except ACPI) have the same issue, i.e. TABs/spaces
> mixture.

I was lazy and just copied the previous line. I'll fix the new line for now.

> ...
>
> > +     ret = of_changeset_apply(ocs);
> > +     if (!ret) {
>
> Why not positive conditional?

No real reason. I suppose having the error condition come first is more
common. Not sure if it makes any difference in this case though?


Thanks
ChenYu


> > +             /*
> > +              * ocs is intentionally kept around as it needs to
> > +              * exist as long as the change is applied.
> > +              */
> > +             void *ptr __always_unused = no_free_ptr(ocs);
> > +     } else {
> > +             /* ocs needs to be explicitly cleaned up before being freed. */
> > +             of_changeset_destroy(ocs);
> > +     }
>
> --
> With Best Regards,
> Andy Shevchenko
>
>

Re: [PATCH v5 06/10] i2c: Introduce OF component probe function

[Andy Shevchenko]

On Fri, Aug 23, 2024 at 04:40:36PM +0800, Chen-Yu Tsai wrote:
> On Thu, Aug 22, 2024 at 10:02 PM Andy Shevchenko
> <andriy.shevchenko@linux.intel.com> wrote:
> > On Thu, Aug 22, 2024 at 05:19:59PM +0800, Chen-Yu Tsai wrote:

...

> > > +     ret = of_changeset_apply(ocs);
> > > +     if (!ret) {
> >
> > Why not positive conditional?
> 
> No real reason. I suppose having the error condition come first is more
> common.

Yes, when you have something like

	if (err) {
		...
		return err;
	} else {
		...
	}


But you don't. That's why I commented on this.

> Not sure if it makes any difference in this case though?

! is hard to read by a human being, easy to make a mistake in the brain of
reader and with inverted logic the code reading becomes harder. So, it's pure
about cognitive function.

> > > +             /*
> > > +              * ocs is intentionally kept around as it needs to
> > > +              * exist as long as the change is applied.
> > > +              */
> > > +             void *ptr __always_unused = no_free_ptr(ocs);
> > > +     } else {
> > > +             /* ocs needs to be explicitly cleaned up before being freed. */
> > > +             of_changeset_destroy(ocs);
> > > +     }

-- 
With Best Regards,
Andy Shevchenko

[PATCH v5 07/10] i2c: of-prober: Add regulator support

[Chen-Yu Tsai]

This adds regulator management to the I2C OF component prober.
Components that the prober intends to probe likely require their
regulator supplies be enabled, and GPIOs be toggled to enable them or
bring them out of reset before they will respond to probe attempts.
GPIOs will be handled in the next patch.

Without specific knowledge of each component's resource names or
power sequencing requirements, the prober can only enable the
regulator supplies all at once, and toggle the GPIOs all at once.
Luckily, reset pins tend to be active low, while enable pins tend to
be active high, so setting the raw status of all GPIO pins to high
should work. The wait time before and after resources are enabled
are collected from existing drivers and device trees.

The prober collects resources from all possible components and enables
them together, instead of enabling resources and probing each component
one by one. The latter approach does not provide any boot time benefits
over simply enabling each component and letting each driver probe
sequentially.

The prober will also deduplicate the resources, since on a component
swap out or co-layout design, the resources are always the same.
While duplicate regulator supplies won't cause much issue, shared
GPIOs don't work reliably, especially with other drivers. For the
same reason, the prober will release the GPIOs before the successfully
probed component is actually enabled.

Signed-off-by: Chen-Yu Tsai <wenst@chromium.org>
---
Changes since v4:
- Split out GPIO handling to separate patch
- Rewrote using of_regulator_bulk_get_all()
- Replaced "regulators" with "regulator supplies" in debug messages

Changes since v3:
- New patch

This change is kept as a separate patch for now since the changes are
quite numerous.
---
 drivers/i2c/i2c-core-of-prober.c | 114 ++++++++++++++++++++++++++++++-
 1 file changed, 113 insertions(+), 1 deletion(-)

diff --git a/drivers/i2c/i2c-core-of-prober.c b/drivers/i2c/i2c-core-of-prober.c
index bb7b231201b0..32184cfd10f6 100644
--- a/drivers/i2c/i2c-core-of-prober.c
+++ b/drivers/i2c/i2c-core-of-prober.c
@@ -6,12 +6,14 @@
  */
 
 #include <linux/cleanup.h>
+#include <linux/delay.h>
 #include <linux/device.h>
 #include <linux/dev_printk.h>
 #include <linux/err.h>
 #include <linux/i2c.h>
 #include <linux/module.h>
 #include <linux/of.h>
+#include <linux/regulator/consumer.h>
 #include <linux/slab.h>
 
 /*
@@ -27,10 +29,93 @@
  * address responds.
  *
  * TODO:
- * - Support handling common regulators and GPIOs.
+ * - Support handling common GPIOs.
  * - Support I2C muxes
  */
 
+struct i2c_of_probe_data {
+	struct regulator_bulk_data *regulators;
+	unsigned int regulators_num;
+};
+
+/* Returns number of regulator supplies found for node, or error. */
+static int i2c_of_probe_get_regulator(struct device *dev, struct device_node *node,
+				      struct i2c_of_probe_data *data)
+{
+	struct regulator_bulk_data *tmp, *new_regulators;
+	int ret;
+
+	ret = of_regulator_bulk_get_all(dev, node, &tmp);
+	if (ret <= 0)
+		return ret;
+
+	if (!data->regulators) {
+		data->regulators = tmp;
+		data->regulators_num = ret;
+		return ret;
+	};
+
+	new_regulators = krealloc(data->regulators,
+				  sizeof(*tmp) * (data->regulators_num + ret),
+				  GFP_KERNEL);
+	if (!new_regulators) {
+		regulator_bulk_free(ret, tmp);
+		return -ENOMEM;
+	}
+
+	data->regulators = new_regulators;
+
+	for (unsigned int i = 0; i < ret; i++)
+		memcpy(&data->regulators[data->regulators_num++], &tmp[i], sizeof(*tmp));
+
+	return ret;
+}
+
+static void i2c_of_probe_free_res(struct i2c_of_probe_data *data)
+{
+	regulator_bulk_free(data->regulators_num, data->regulators);
+}
+
+static int i2c_of_probe_get_res(struct device *dev, struct device_node *node,
+				struct i2c_of_probe_data *data)
+{
+	struct property *prop;
+	int ret;
+
+	ret = i2c_of_probe_get_regulator(dev, node, data);
+	if (ret < 0) {
+		dev_err_probe(dev, ret, "Failed to get regulator supplies from %pOF\n", node);
+		goto err_cleanup;
+	}
+
+	return 0;
+
+err_cleanup:
+	i2c_of_probe_free_res(data);
+	return ret;
+}
+
+static int i2c_of_probe_enable_res(struct device *dev, struct i2c_of_probe_data *data)
+{
+	int ret = 0;
+
+	dev_dbg(dev, "Enabling regulator supplies\n");
+
+	ret = regulator_bulk_enable(data->regulators_num, data->regulators);
+	if (ret)
+		return ret;
+
+	/* largest post-power-on pre-reset-deassert delay seen among drivers */
+	msleep(500);
+
+	return 0;
+}
+
+static void i2c_of_probe_disable_regulators(struct i2c_of_probe_data *data)
+{
+	regulator_bulk_disable(data->regulators_num, data->regulators);
+}
+
 static struct device_node *i2c_of_probe_get_i2c_node(struct device *dev, const char *type)
 {
 	struct device_node *node __free(device_node) = of_find_node_by_name(NULL, type);
@@ -110,6 +195,7 @@ static int i2c_of_probe_enable_node(struct device *dev, struct device_node *node
 int i2c_of_probe_component(struct device *dev, const char *type)
 {
 	struct i2c_adapter *i2c;
+	struct i2c_of_probe_data probe_data = {0};
 	int ret;
 
 	struct device_node *i2c_node __free(device_node) = i2c_of_probe_get_i2c_node(dev, type);
@@ -133,6 +219,30 @@ int i2c_of_probe_component(struct device *dev, const char *type)
 	if (!i2c)
 		return dev_err_probe(dev, -EPROBE_DEFER, "Couldn't get I2C adapter\n");
 
+	/* Grab resources */
+	for_each_child_of_node_scoped(i2c_node, node) {
+		u32 addr;
+
+		if (!of_node_name_prefix(node, type))
+			continue;
+		if (of_property_read_u32(node, "reg", &addr))
+			continue;
+
+		dev_dbg(dev, "Requesting resources for %pOF\n", node);
+		ret = i2c_of_probe_get_res(dev, node, &probe_data);
+		if (ret)
+			return ret;
+	}
+
+	dev_dbg(dev, "Resources: # of regulator supplies = %d\n", probe_data.regulators_num);
+
+	/* Enable resources */
+	ret = i2c_of_probe_enable_res(dev, &probe_data);
+	if (ret) {
+		i2c_of_probe_free_res(&probe_data);
+		return dev_err_probe(dev, ret, "Failed to enable resources\n");
+	}
+
 	ret = 0;
 	for_each_child_of_node_scoped(i2c_node, node) {
 		union i2c_smbus_data data;
@@ -150,6 +260,8 @@ int i2c_of_probe_component(struct device *dev, const char *type)
 		break;
 	}
 
+	i2c_of_probe_disable_regulators(&probe_data);
+	i2c_of_probe_free_res(&probe_data);
 	i2c_put_adapter(i2c);
 
 	return ret;
-- 
2.46.0.184.g6999bdac58-goog

Re: [PATCH v5 07/10] i2c: of-prober: Add regulator support

[Andy Shevchenko]

On Thu, Aug 22, 2024 at 05:20:00PM +0800, Chen-Yu Tsai wrote:
> This adds regulator management to the I2C OF component prober.
> Components that the prober intends to probe likely require their
> regulator supplies be enabled, and GPIOs be toggled to enable them or
> bring them out of reset before they will respond to probe attempts.
> GPIOs will be handled in the next patch.
> 
> Without specific knowledge of each component's resource names or
> power sequencing requirements, the prober can only enable the
> regulator supplies all at once, and toggle the GPIOs all at once.
> Luckily, reset pins tend to be active low, while enable pins tend to
> be active high, so setting the raw status of all GPIO pins to high
> should work. The wait time before and after resources are enabled
> are collected from existing drivers and device trees.
> 
> The prober collects resources from all possible components and enables
> them together, instead of enabling resources and probing each component
> one by one. The latter approach does not provide any boot time benefits
> over simply enabling each component and letting each driver probe
> sequentially.
> 
> The prober will also deduplicate the resources, since on a component
> swap out or co-layout design, the resources are always the same.
> While duplicate regulator supplies won't cause much issue, shared
> GPIOs don't work reliably, especially with other drivers. For the
> same reason, the prober will release the GPIOs before the successfully
> probed component is actually enabled.

...

>  /*

>   * address responds.
>   *
>   * TODO:
> - * - Support handling common regulators and GPIOs.
> + * - Support handling common GPIOs.

You can split this to two lines in the first place and have less churn in this
patch and the other one.

>   * - Support I2C muxes
>   */

..

> +/* Returns number of regulator supplies found for node, or error. */
> +static int i2c_of_probe_get_regulator(struct device *dev, struct device_node *node,
> +				      struct i2c_of_probe_data *data)
> +{
> +	struct regulator_bulk_data *tmp, *new_regulators;
> +	int ret;
> +
> +	ret = of_regulator_bulk_get_all(dev, node, &tmp);
> +	if (ret <= 0)
> +		return ret;

I would split this and explain 0 case.


> +	if (!data->regulators) {
> +		data->regulators = tmp;
> +		data->regulators_num = ret;
> +		return ret;
> +	};
> +
> +	new_regulators = krealloc(data->regulators,
> +				  sizeof(*tmp) * (data->regulators_num + ret),

krealloc_array()

> +				  GFP_KERNEL);
> +	if (!new_regulators) {
> +		regulator_bulk_free(ret, tmp);
> +		return -ENOMEM;
> +	}
> +
> +	data->regulators = new_regulators;

> +	for (unsigned int i = 0; i < ret; i++)
> +		memcpy(&data->regulators[data->regulators_num++], &tmp[i], sizeof(*tmp));

Seems like copying array to array, no? If so, can't be done in a single memcpy() call?

> +	return ret;
> +}

...

> +static int i2c_of_probe_get_res(struct device *dev, struct device_node *node,
> +				struct i2c_of_probe_data *data)
> +{
> +	struct property *prop;
> +	int ret;
> +
> +	ret = i2c_of_probe_get_regulator(dev, node, data);
> +	if (ret < 0) {
> +		dev_err_probe(dev, ret, "Failed to get regulator supplies from %pOF\n", node);
> +		goto err_cleanup;
> +	}
> +
> +	return 0;
> +
> +err_cleanup:
> +	i2c_of_probe_free_res(data);
> +	return ret;
> +}

Hmm... why not

static int i2c_of_probe_get_res(struct device *dev, struct device_node *node,
				struct i2c_of_probe_data *data)
{
	struct property *prop;
	int ret;

	ret = i2c_of_probe_get_regulator(dev, node, data);
	if (ret < 0) {
		i2c_of_probe_free_res(data);
		return dev_err_probe(dev, ret, "Failed to get regulator supplies from %pOF\n", node);
	}

	return 0;
}

...

> +static int i2c_of_probe_enable_res(struct device *dev, struct i2c_of_probe_data *data)
> +{
> +	int ret = 0;

Redundant assignment.

> +	dev_dbg(dev, "Enabling regulator supplies\n");
> +
> +	ret = regulator_bulk_enable(data->regulators_num, data->regulators);
> +	if (ret)
> +		return ret;
> +
> +	/* largest post-power-on pre-reset-deassert delay seen among drivers */
> +	msleep(500);

How would we monitor if any [new] driver wants to use bigger timeout?

> +	return 0;
> +}

...

>  	struct i2c_adapter *i2c;
> +	struct i2c_of_probe_data probe_data = {0};

Reversed xmas tree order?

'0' is not needed.

...

> +	/* Grab resources */
> +	for_each_child_of_node_scoped(i2c_node, node) {
> +		u32 addr;
> +
> +		if (!of_node_name_prefix(node, type))
> +			continue;

Is it third or fourth copy of this code? At some point you probably want

#define for_each_child_of_node_with_prefix_scoped()
	for_each_if(...)

(or equivalent)

> +		if (of_property_read_u32(node, "reg", &addr))
> +			continue;
> +
> +		dev_dbg(dev, "Requesting resources for %pOF\n", node);
> +		ret = i2c_of_probe_get_res(dev, node, &probe_data);
> +		if (ret)
> +			return ret;
> +	}

-- 
With Best Regards,
Andy Shevchenko

Re: [PATCH v5 07/10] i2c: of-prober: Add regulator support

[Chen-Yu Tsai]

On Thu, Aug 22, 2024 at 10:09 PM Andy Shevchenko
<andriy.shevchenko@linux.intel.com> wrote:
>
> On Thu, Aug 22, 2024 at 05:20:00PM +0800, Chen-Yu Tsai wrote:
> > This adds regulator management to the I2C OF component prober.
> > Components that the prober intends to probe likely require their
> > regulator supplies be enabled, and GPIOs be toggled to enable them or
> > bring them out of reset before they will respond to probe attempts.
> > GPIOs will be handled in the next patch.
> >
> > Without specific knowledge of each component's resource names or
> > power sequencing requirements, the prober can only enable the
> > regulator supplies all at once, and toggle the GPIOs all at once.
> > Luckily, reset pins tend to be active low, while enable pins tend to
> > be active high, so setting the raw status of all GPIO pins to high
> > should work. The wait time before and after resources are enabled
> > are collected from existing drivers and device trees.
> >
> > The prober collects resources from all possible components and enables
> > them together, instead of enabling resources and probing each component
> > one by one. The latter approach does not provide any boot time benefits
> > over simply enabling each component and letting each driver probe
> > sequentially.
> >
> > The prober will also deduplicate the resources, since on a component
> > swap out or co-layout design, the resources are always the same.
> > While duplicate regulator supplies won't cause much issue, shared
> > GPIOs don't work reliably, especially with other drivers. For the
> > same reason, the prober will release the GPIOs before the successfully
> > probed component is actually enabled.
>
> ...
>
> >  /*
>
> >   * address responds.
> >   *
> >   * TODO:
> > - * - Support handling common regulators and GPIOs.
> > + * - Support handling common GPIOs.
>
> You can split this to two lines in the first place and have less churn in this
> patch and the other one.

Ack.

> >   * - Support I2C muxes
> >   */
>
> ..
>
> > +/* Returns number of regulator supplies found for node, or error. */
> > +static int i2c_of_probe_get_regulator(struct device *dev, struct device_node *node,
> > +                                   struct i2c_of_probe_data *data)
> > +{
> > +     struct regulator_bulk_data *tmp, *new_regulators;
> > +     int ret;
> > +
> > +     ret = of_regulator_bulk_get_all(dev, node, &tmp);
> > +     if (ret <= 0)
> > +             return ret;
>
> I would split this and explain 0 case.

Ack.

> > +     if (!data->regulators) {
> > +             data->regulators = tmp;
> > +             data->regulators_num = ret;
> > +             return ret;
> > +     };
> > +
> > +     new_regulators = krealloc(data->regulators,
> > +                               sizeof(*tmp) * (data->regulators_num + ret),
>
> krealloc_array()

Ack. Somehow I didn't find this function while I was rewriting the code.

> > +                               GFP_KERNEL);
> > +     if (!new_regulators) {
> > +             regulator_bulk_free(ret, tmp);
> > +             return -ENOMEM;
> > +     }
> > +
> > +     data->regulators = new_regulators;
>
> > +     for (unsigned int i = 0; i < ret; i++)
> > +             memcpy(&data->regulators[data->regulators_num++], &tmp[i], sizeof(*tmp));
>
> Seems like copying array to array, no? If so, can't be done in a single memcpy() call?

Ack.

> > +     return ret;
> > +}
>
> ...
>
> > +static int i2c_of_probe_get_res(struct device *dev, struct device_node *node,
> > +                             struct i2c_of_probe_data *data)
> > +{
> > +     struct property *prop;
> > +     int ret;
> > +
> > +     ret = i2c_of_probe_get_regulator(dev, node, data);
> > +     if (ret < 0) {
> > +             dev_err_probe(dev, ret, "Failed to get regulator supplies from %pOF\n", node);
> > +             goto err_cleanup;
> > +     }
> > +
> > +     return 0;
> > +
> > +err_cleanup:
> > +     i2c_of_probe_free_res(data);
> > +     return ret;
> > +}
>
> Hmm... why not
>
> static int i2c_of_probe_get_res(struct device *dev, struct device_node *node,
>                                 struct i2c_of_probe_data *data)
> {
>         struct property *prop;
>         int ret;
>
>         ret = i2c_of_probe_get_regulator(dev, node, data);
>         if (ret < 0) {
>                 i2c_of_probe_free_res(data);
>                 return dev_err_probe(dev, ret, "Failed to get regulator supplies from %pOF\n", node);
>         }
>
>         return 0;
> }
>
> ...

That would be more churn in the next patch, which introduces another
error condition requiring the same cleanup.

> > +static int i2c_of_probe_enable_res(struct device *dev, struct i2c_of_probe_data *data)
> > +{
> > +     int ret = 0;
>
> Redundant assignment.

Ack.

> > +     dev_dbg(dev, "Enabling regulator supplies\n");
> > +
> > +     ret = regulator_bulk_enable(data->regulators_num, data->regulators);
> > +     if (ret)
> > +             return ret;
> > +
> > +     /* largest post-power-on pre-reset-deassert delay seen among drivers */
> > +     msleep(500);
>
> How would we monitor if any [new] driver wants to use bigger timeout?

The assumption is that the person doing the integration should test for
this. This prober doesn't get called everywhere. It needs a driver to
call it, and that driver is written by someone for some specific platform.
Maybe I should explicitly spell that out in the function description?
Or even make it a parameter?

Also, having an arbitrarily large number here doesn't help platforms that
want to minimize boot time. On that front I'm also thinking about whether
it is possible to do a handover to the actual driver so that the latter
doesn't have to go through the whole power sequence again.

> > +     return 0;
> > +}
>
> ...
>
> >       struct i2c_adapter *i2c;
> > +     struct i2c_of_probe_data probe_data = {0};
>
> Reversed xmas tree order?

OK...

> '0' is not needed.

Ack.

> ...
>
> > +     /* Grab resources */
> > +     for_each_child_of_node_scoped(i2c_node, node) {
> > +             u32 addr;
> > +
> > +             if (!of_node_name_prefix(node, type))
> > +                     continue;
>
> Is it third or fourth copy of this code? At some point you probably want
>
> #define for_each_child_of_node_with_prefix_scoped()
>         for_each_if(...)
>
> (or equivalent)

Ack.


Thank you for the review.

ChenYu

> > +             if (of_property_read_u32(node, "reg", &addr))
> > +                     continue;
> > +
> > +             dev_dbg(dev, "Requesting resources for %pOF\n", node);
> > +             ret = i2c_of_probe_get_res(dev, node, &probe_data);
> > +             if (ret)
> > +                     return ret;
> > +     }
>
> --
> With Best Regards,
> Andy Shevchenko
>
>

Re: [PATCH v5 07/10] i2c: of-prober: Add regulator support

[Andy Shevchenko]

On Fri, Aug 23, 2024 at 05:35:59PM +0800, Chen-Yu Tsai wrote:
> On Thu, Aug 22, 2024 at 10:09 PM Andy Shevchenko
> <andriy.shevchenko@linux.intel.com> wrote:
> > On Thu, Aug 22, 2024 at 05:20:00PM +0800, Chen-Yu Tsai wrote:

...

> > Hmm... why not
> >
> > static int i2c_of_probe_get_res(struct device *dev, struct device_node *node,
> >                                 struct i2c_of_probe_data *data)
> > {
> >         struct property *prop;
> >         int ret;
> >
> >         ret = i2c_of_probe_get_regulator(dev, node, data);
> >         if (ret < 0) {
> >                 i2c_of_probe_free_res(data);
> >                 return dev_err_probe(dev, ret, "Failed to get regulator supplies from %pOF\n", node);
> >         }
> >
> >         return 0;
> > }
> 
> That would be more churn in the next patch, which introduces another
> error condition requiring the same cleanup.

OK!

...

> > > +     /* largest post-power-on pre-reset-deassert delay seen among drivers */
> > > +     msleep(500);
> >
> > How would we monitor if any [new] driver wants to use bigger timeout?
> 
> The assumption is that the person doing the integration should test for
> this. This prober doesn't get called everywhere. It needs a driver to
> call it, and that driver is written by someone for some specific platform.
> Maybe I should explicitly spell that out in the function description?
> Or even make it a parameter?
> 
> Also, having an arbitrarily large number here doesn't help platforms that
> want to minimize boot time. On that front I'm also thinking about whether
> it is possible to do a handover to the actual driver so that the latter
> doesn't have to go through the whole power sequence again.

Yeah, I think the best effort is to have a parameter.

-- 
With Best Regards,
Andy Shevchenko

[PATCH v5 08/10] i2c: of-prober: Add GPIO support

[Chen-Yu Tsai]

This adds GPIO management to the I2C OF component prober.
Components that the prober intends to probe likely require their
regulator supplies be enabled, and GPIOs be toggled to enable them or
bring them out of reset before they will respond to probe attempts.
regulator support was added in the previous patch.

Without specific knowledge of each component's resource names or
power sequencing requirements, the prober can only enable the
regulator supplies all at once, and toggle the GPIOs all at once.
Luckily, reset pins tend to be active low, while enable pins tend to
be active high, so setting the raw status of all GPIO pins to high
should work. The wait time before and after resources are enabled
are collected from existing drivers and device trees.

The prober collects resources from all possible components and enables
them together, instead of enabling resources and probing each component
one by one. The latter approach does not provide any boot time benefits
over simply enabling each component and letting each driver probe
sequentially.

The prober will also deduplicate the resources, since on a component
swap out or co-layout design, the resources are always the same.
While duplicate regulator supplies won't cause much issue, shared
GPIOs don't work reliably, especially with other drivers. For the
same reason, the prober will release the GPIOs before the successfully
probed component is actually enabled.

Signed-off-by: Chen-Yu Tsai <wenst@chromium.org>

---
Changes since v4:
- Split out from previous patch
- Moved GPIO property name check to common function in gpiolib.c in new
  patch
- Moved i2c_of_probe_free_gpios() into for_each_child_of_node_scoped()
- Rewrote in gpiod_*_array-esque fashion
---
 drivers/i2c/i2c-core-of-prober.c | 126 ++++++++++++++++++++++++++++++-
 1 file changed, 124 insertions(+), 2 deletions(-)

diff --git a/drivers/i2c/i2c-core-of-prober.c b/drivers/i2c/i2c-core-of-prober.c
index 32184cfd10f6..046e6605053c 100644
--- a/drivers/i2c/i2c-core-of-prober.c
+++ b/drivers/i2c/i2c-core-of-prober.c
@@ -10,6 +10,7 @@
 #include <linux/device.h>
 #include <linux/dev_printk.h>
 #include <linux/err.h>
+#include <linux/gpio/consumer.h>
 #include <linux/i2c.h>
 #include <linux/module.h>
 #include <linux/of.h>
@@ -29,11 +30,11 @@
  * address responds.
  *
  * TODO:
- * - Support handling common GPIOs.
  * - Support I2C muxes
  */
 
 struct i2c_of_probe_data {
+	struct gpio_descs *gpiods;
 	struct regulator_bulk_data *regulators;
 	unsigned int regulators_num;
 };
@@ -71,8 +72,88 @@ static int i2c_of_probe_get_regulator(struct device *dev, struct device_node *no
 	return ret;
 }
 
+/*
+ * Returns 1 if property is GPIO and GPIO successfully requested,
+ * 0 if not a GPIO property, or error if request for GPIO failed.
+ */
+static int i2c_of_probe_get_gpiod(struct device_node *node, struct property *prop,
+				  struct i2c_of_probe_data *data)
+{
+	struct fwnode_handle *fwnode = of_fwnode_handle(node);
+	struct gpio_descs *gpiods;
+	struct gpio_desc *gpiod;
+	char con[32]; /* 32 is max size of property name */
+	char *con_id = NULL;
+	size_t new_size;
+	int len;
+
+	len = gpio_property_name_length(prop->name);
+	if (len < 0)
+		return 0;
+
+	if (len >= sizeof(con) - 1) {
+		pr_err("%pOF: length of GPIO name \"%s\" exceeds current limit\n",
+		       node, prop->name);
+		return -EINVAL;
+	}
+
+	if (len > 0) {
+		strscpy(con, prop->name, len + 1);
+		con_id = con;
+	}
+
+	/*
+	 * GPIO descriptors are not reference counted. GPIOD_FLAGS_BIT_NONEXCLUSIVE
+	 * can't differentiate between GPIOs shared between devices to be probed and
+	 * other devices (which is incorrect). If the initial request fails with
+	 * -EBUSY, retry with GPIOD_FLAGS_BIT_NONEXCLUSIVE and see if it matches
+	 * any existing ones.
+	 */
+	gpiod = fwnode_gpiod_get_index(fwnode, con_id, 0, GPIOD_ASIS, "i2c-of-prober");
+	if (IS_ERR(gpiod)) {
+		int i;
+
+		if (PTR_ERR(gpiod) != -EBUSY || !data->gpiods)
+			return PTR_ERR(gpiod);
+
+		gpiod = fwnode_gpiod_get_index(fwnode, con_id, 0,
+					       GPIOD_ASIS | GPIOD_FLAGS_BIT_NONEXCLUSIVE,
+					       "i2c-of-prober");
+		for (i = 0; i < data->gpiods->ndescs; i++)
+			if (gpiod == data->gpiods->desc[i])
+				return 1;
+
+		return -EBUSY;
+	}
+
+	new_size = struct_size(gpiods, desc, data->gpiods ? data->gpiods->ndescs + 1 : 1);
+	gpiods = krealloc(data->gpiods, new_size, GFP_KERNEL);
+	if (!gpiods) {
+		gpiod_put(gpiod);
+		return -ENOMEM;
+	}
+
+	data->gpiods = gpiods;
+	data->gpiods->desc[data->gpiods->ndescs++] = gpiod;
+
+	return 1;
+}
+
+/*
+ * This is split into two functions because in the normal flow the GPIOs
+ * have to be released before the actual driver probes so that the latter
+ * can acquire them.
+ */
+static void i2c_of_probe_free_gpios(struct i2c_of_probe_data *data)
+{
+	if (data->gpiods)
+		gpiod_put_array(data->gpiods);
+	data->gpiods = NULL;
+}
+
 static void i2c_of_probe_free_res(struct i2c_of_probe_data *data)
 {
+	i2c_of_probe_free_gpios(data);
 	regulator_bulk_free(data->regulators_num, data->regulators);
 }
 
@@ -88,6 +169,18 @@ static int i2c_of_probe_get_res(struct device *dev, struct device_node *node,
 		goto err_cleanup;
 	}
 
+	for_each_property_of_node(node, prop) {
+		dev_dbg(dev, "Trying property %pOF/%s\n", node, prop->name);
+
+		/* GPIOs */
+		ret = i2c_of_probe_get_gpiod(node, prop, data);
+		if (ret < 0) {
+			dev_err_probe(dev, ret, "Failed to get GPIO from %pOF/%s\n",
+				      node, prop->name);
+			goto err_cleanup;
+		}
+	}
+
 	return 0;
 
 err_cleanup:
@@ -98,6 +191,7 @@ static int i2c_of_probe_get_res(struct device *dev, struct device_node *node,
 static int i2c_of_probe_enable_res(struct device *dev, struct i2c_of_probe_data *data)
 {
 	int ret = 0;
+	int gpio_i;
 
 	dev_dbg(dev, "Enabling regulator supplies\n");
 
@@ -108,7 +202,32 @@ static int i2c_of_probe_enable_res(struct device *dev, struct i2c_of_probe_data
 	/* largest post-power-on pre-reset-deassert delay seen among drivers */
 	msleep(500);
 
+	if (!data->gpiods)
+		return 0;
+
+	for (gpio_i = 0; gpio_i < data->gpiods->ndescs; gpio_i++) {
+		/*
+		 * reset GPIOs normally have opposite polarity compared to
+		 * enable GPIOs. Instead of parsing the flags again, simply
+		 * set the raw value to high.
+		 */
+		dev_dbg(dev, "Setting GPIO %d\n", gpio_i);
+		ret = gpiod_direction_output_raw(data->gpiods->desc[gpio_i], 1);
+		if (ret)
+			goto disable_gpios;
+	}
+
+	/* largest post-reset-deassert delay seen in tree for Elan I2C HID */
+	msleep(300);
+
 	return 0;
+
+disable_gpios:
+	for (gpio_i--; gpio_i >= 0; gpio_i--)
+		gpiod_set_raw_value_cansleep(data->gpiods->desc[gpio_i], 0);
+	regulator_bulk_disable(data->regulators_num, data->regulators);
+
+	return ret;
 }
 
 static void i2c_of_probe_disable_regulators(struct i2c_of_probe_data *data)
@@ -234,7 +353,9 @@ int i2c_of_probe_component(struct device *dev, const char *type)
 			return ret;
 	}
 
-	dev_dbg(dev, "Resources: # of regulator supplies = %d\n", probe_data.regulators_num);
+	dev_dbg(dev, "Resources: # of GPIOs = %d, # of regulator supplies = %d\n",
+		probe_data.gpiods ? probe_data.gpiods->ndescs : 0,
+		probe_data.regulators_num);
 
 	/* Enable resources */
 	ret = i2c_of_probe_enable_res(dev, &probe_data);
@@ -256,6 +377,7 @@ int i2c_of_probe_component(struct device *dev, const char *type)
 			continue;
 
 		/* Found a device that is responding */
+		i2c_of_probe_free_gpios(&probe_data);
 		ret = i2c_of_probe_enable_node(dev, node);
 		break;
 	}
-- 
2.46.0.184.g6999bdac58-goog

Re: [PATCH v5 08/10] i2c: of-prober: Add GPIO support

[Andy Shevchenko]

On Thu, Aug 22, 2024 at 05:20:01PM +0800, Chen-Yu Tsai wrote:
> This adds GPIO management to the I2C OF component prober.
> Components that the prober intends to probe likely require their
> regulator supplies be enabled, and GPIOs be toggled to enable them or
> bring them out of reset before they will respond to probe attempts.
> regulator support was added in the previous patch.
> 
> Without specific knowledge of each component's resource names or
> power sequencing requirements, the prober can only enable the
> regulator supplies all at once, and toggle the GPIOs all at once.
> Luckily, reset pins tend to be active low, while enable pins tend to
> be active high, so setting the raw status of all GPIO pins to high
> should work. The wait time before and after resources are enabled
> are collected from existing drivers and device trees.
> 
> The prober collects resources from all possible components and enables
> them together, instead of enabling resources and probing each component
> one by one. The latter approach does not provide any boot time benefits
> over simply enabling each component and letting each driver probe
> sequentially.
> 
> The prober will also deduplicate the resources, since on a component
> swap out or co-layout design, the resources are always the same.
> While duplicate regulator supplies won't cause much issue, shared
> GPIOs don't work reliably, especially with other drivers. For the
> same reason, the prober will release the GPIOs before the successfully
> probed component is actually enabled.

...

> +	struct fwnode_handle *fwnode = of_fwnode_handle(node);
> +	struct gpio_descs *gpiods;
> +	struct gpio_desc *gpiod;
> +	char con[32]; /* 32 is max size of property name */

Use 'propname' to be aligned with GPIO library usages.

> +	char *con_id = NULL;
> +	size_t new_size;
> +	int len;

...

> +	if (len >= sizeof(con) - 1) {

This can be transformed to check the returned value from strscpy().

> +		pr_err("%pOF: length of GPIO name \"%s\" exceeds current limit\n",
> +		       node, prop->name);
> +		return -EINVAL;
> +	}
> +
> +	if (len > 0) {
> +		strscpy(con, prop->name, len + 1);

The correct (robust) call is with destination size. Which means here that you
may use 2-argument strscpy().

> +		con_id = con;
> +	}

...

> +	if (!data->gpiods)
> +		return 0;

If it comes a new code (something else besides GPIOs and regulators) this will be a (small) impediment. Better to have a helper for each case and do

	ret = ..._gpiods();
	if (ret)
		...

Same for regulators and anything else in the future, if any.

> +		/*
> +		 * reset GPIOs normally have opposite polarity compared to

"reset"

> +		 * enable GPIOs. Instead of parsing the flags again, simply

"enable"

> +		 * set the raw value to high.

This is quite a fragile assumption. Yes, it would work in 98% cases, but will
break if it's not true somewhere else.

> +		 */

...

> +	/* largest post-reset-deassert delay seen in tree for Elan I2C HID */
> +	msleep(300);

Same Q, how do you monitor _all_ the drivers?

...

> +disable_gpios:
> +	for (gpio_i--; gpio_i >= 0; gpio_i--)
> +		gpiod_set_raw_value_cansleep(data->gpiods->desc[gpio_i], 0);

Can't you call the _array() variant here?

...

> -	dev_dbg(dev, "Resources: # of regulator supplies = %d\n", probe_data.regulators_num);
> +	dev_dbg(dev, "Resources: # of GPIOs = %d, # of regulator supplies = %d\n",
> +		probe_data.gpiods ? probe_data.gpiods->ndescs : 0,
> +		probe_data.regulators_num);

I would issue one message per class of the devices (GPIOs, regulators, ...)

-- 
With Best Regards,
Andy Shevchenko

Re: [PATCH v5 08/10] i2c: of-prober: Add GPIO support

[Chen-Yu Tsai]

On Thu, Aug 22, 2024 at 10:20 PM Andy Shevchenko
<andriy.shevchenko@linux.intel.com> wrote:
>
> On Thu, Aug 22, 2024 at 05:20:01PM +0800, Chen-Yu Tsai wrote:
> > This adds GPIO management to the I2C OF component prober.
> > Components that the prober intends to probe likely require their
> > regulator supplies be enabled, and GPIOs be toggled to enable them or
> > bring them out of reset before they will respond to probe attempts.
> > regulator support was added in the previous patch.
> >
> > Without specific knowledge of each component's resource names or
> > power sequencing requirements, the prober can only enable the
> > regulator supplies all at once, and toggle the GPIOs all at once.
> > Luckily, reset pins tend to be active low, while enable pins tend to
> > be active high, so setting the raw status of all GPIO pins to high
> > should work. The wait time before and after resources are enabled
> > are collected from existing drivers and device trees.
> >
> > The prober collects resources from all possible components and enables
> > them together, instead of enabling resources and probing each component
> > one by one. The latter approach does not provide any boot time benefits
> > over simply enabling each component and letting each driver probe
> > sequentially.
> >
> > The prober will also deduplicate the resources, since on a component
> > swap out or co-layout design, the resources are always the same.
> > While duplicate regulator supplies won't cause much issue, shared
> > GPIOs don't work reliably, especially with other drivers. For the
> > same reason, the prober will release the GPIOs before the successfully
> > probed component is actually enabled.
>
> ...
>
> > +     struct fwnode_handle *fwnode = of_fwnode_handle(node);
> > +     struct gpio_descs *gpiods;
> > +     struct gpio_desc *gpiod;
> > +     char con[32]; /* 32 is max size of property name */
>
> Use 'propname' to be aligned with GPIO library usages.

Ack.

> > +     char *con_id = NULL;
> > +     size_t new_size;
> > +     int len;
>
> ...
>
> > +     if (len >= sizeof(con) - 1) {
>
> This can be transformed to check the returned value from strscpy().

Ack.

> > +             pr_err("%pOF: length of GPIO name \"%s\" exceeds current limit\n",
> > +                    node, prop->name);
> > +             return -EINVAL;
> > +     }
> > +
> > +     if (len > 0) {
> > +             strscpy(con, prop->name, len + 1);
>
> The correct (robust) call is with destination size. Which means here that you
> may use 2-argument strscpy().

Ack.

> > +             con_id = con;
> > +     }
>
> ...
>
> > +     if (!data->gpiods)
> > +             return 0;
>
> If it comes a new code (something else besides GPIOs and regulators) this will be a (small) impediment. Better to have a helper for each case and do
>
>         ret = ..._gpiods();
>         if (ret)
>                 ...
>
> Same for regulators and anything else in the future, if any.

I'm not sure I follow. Do you mean wrap each individual type in a wrapper
and call those here, like the following?

    i2c_of_probe_enable_res(...)
    {
        ret = i2c_of_probe_enable_regulators(...)
        if (ret)
              return ret;

        ret = i2c_of_probe_enable_gpios(...)
        if (ret)
              goto error_disable_regulators;

        ...
    }

> > +             /*
> > +              * reset GPIOs normally have opposite polarity compared to
>
> "reset"
>
> > +              * enable GPIOs. Instead of parsing the flags again, simply
>
> "enable"
>
> > +              * set the raw value to high.
>
> This is quite a fragile assumption. Yes, it would work in 98% cases, but will
> break if it's not true somewhere else.

Well, this seems to be the de facto standard. Or it would have to remember
what each GPIO descriptor's name is, and try to classify those into either
"enable" or "reset", and set their respective logical values to 1 or 0.
And then you run into a peripheral with a broken binding that has its
"reset" GPIO inverted, i.e. it's driver behavior needs to follow the
"enable" GPIO style. The class of devices this prober targets are
consumer electronics (laptops, tablets, phones) that at least have gone
through some component selection where the options won't have conflicting
requirements.

And if the polarities of the possible components don't line up, then this
probe structure can't really do anything. One would need something that
power sequences each component separately and probes it. I would really
like to avoid that if possible, as it makes the boot time (to peripheral
available) dependent on which component you have and how far down the
list it is. We have Chromebooks that have 4 touchscreen components
introduced over the years. In that case something more like Doug's
original proposal would work better: something that forces mutual
exclusivity among a class of devices.

> > +              */
>
> ...
>
> > +     /* largest post-reset-deassert delay seen in tree for Elan I2C HID */
> > +     msleep(300);
>
> Same Q, how do you monitor _all_ the drivers?

Discussion in the previous patch.

> ...
>
> > +disable_gpios:
> > +     for (gpio_i--; gpio_i >= 0; gpio_i--)
> > +             gpiod_set_raw_value_cansleep(data->gpiods->desc[gpio_i], 0);
>
> Can't you call the _array() variant here?

I thought that without |struct gpio_array| the _array() variant wouldn't
help, i.e. it would still be a loop internally. Looks like I was wrong.

> ...
>
> > -     dev_dbg(dev, "Resources: # of regulator supplies = %d\n", probe_data.regulators_num);
> > +     dev_dbg(dev, "Resources: # of GPIOs = %d, # of regulator supplies = %d\n",
> > +             probe_data.gpiods ? probe_data.gpiods->ndescs : 0,
> > +             probe_data.regulators_num);
>
> I would issue one message per class of the devices (GPIOs, regulators, ...)

Ack.


Thank you for the review.
ChenYu

Re: [PATCH v5 08/10] i2c: of-prober: Add GPIO support

[Andy Shevchenko]

On Fri, Aug 23, 2024 at 06:32:16PM +0800, Chen-Yu Tsai wrote:
> On Thu, Aug 22, 2024 at 10:20 PM Andy Shevchenko
> <andriy.shevchenko@linux.intel.com> wrote:
> > On Thu, Aug 22, 2024 at 05:20:01PM +0800, Chen-Yu Tsai wrote:

...

> > > +     if (!data->gpiods)
> > > +             return 0;
> >
> > If it comes a new code (something else besides GPIOs and regulators) this
> > will be a (small) impediment. Better to have a helper for each case and do
> >
> >         ret = ..._gpiods();
> >         if (ret)
> >                 ...
> >
> > Same for regulators and anything else in the future, if any.
> 
> I'm not sure I follow. Do you mean wrap each individual type in a wrapper
> and call those here, like the following?
> 
>     i2c_of_probe_enable_res(...)
>     {
>         ret = i2c_of_probe_enable_regulators(...)
>         if (ret)
>               return ret;
> 
>         ret = i2c_of_probe_enable_gpios(...)
>         if (ret)
>               goto error_disable_regulators;
> 
>         ...
>     }

Yes.

...

> > > +             /*
> > > +              * reset GPIOs normally have opposite polarity compared to
> >
> > "reset"
> >
> > > +              * enable GPIOs. Instead of parsing the flags again, simply
> >
> > "enable"
> >
> > > +              * set the raw value to high.
> >
> > This is quite a fragile assumption. Yes, it would work in 98% cases, but will
> > break if it's not true somewhere else.
> 
> Well, this seems to be the de facto standard. Or it would have to remember
> what each GPIO descriptor's name is, and try to classify those into either
> "enable" or "reset", and set their respective logical values to 1 or 0.
> And then you run into a peripheral with a broken binding that has its
> "reset" GPIO inverted, i.e. it's driver behavior needs to follow the
> "enable" GPIO style. The class of devices this prober targets are
> consumer electronics (laptops, tablets, phones) that at least have gone
> through some component selection where the options won't have conflicting
> requirements.

I'm talking from real life example(s) :-)

Recently I looked at the OV7251 sensor driver that expects "enable" GPIO while
all users supply "reset"-as-"enable" with the exact trouble I described.
Yet it's pure software / ABI issue in that case, but who knows what PCB
engineers may come up with.

> And if the polarities of the possible components don't line up, then this
> probe structure can't really do anything. One would need something that
> power sequences each component separately and probes it. I would really
> like to avoid that if possible, as it makes the boot time (to peripheral
> available) dependent on which component you have and how far down the
> list it is. We have Chromebooks that have 4 touchscreen components
> introduced over the years. In that case something more like Doug's
> original proposal would work better: something that forces mutual
> exclusivity among a class of devices.

Maybe. I just pointed out the potential problem.

> > > +              */

-- 
With Best Regards,
Andy Shevchenko

Re: [PATCH v5 08/10] i2c: of-prober: Add GPIO support

[Chen-Yu Tsai]

On Fri, Aug 23, 2024 at 10:01 PM Andy Shevchenko
<andriy.shevchenko@linux.intel.com> wrote:
>
> On Fri, Aug 23, 2024 at 06:32:16PM +0800, Chen-Yu Tsai wrote:
> > On Thu, Aug 22, 2024 at 10:20 PM Andy Shevchenko
> > <andriy.shevchenko@linux.intel.com> wrote:
> > > On Thu, Aug 22, 2024 at 05:20:01PM +0800, Chen-Yu Tsai wrote:
>
> ...
>
> > > > +     if (!data->gpiods)
> > > > +             return 0;
> > >
> > > If it comes a new code (something else besides GPIOs and regulators) this
> > > will be a (small) impediment. Better to have a helper for each case and do
> > >
> > >         ret = ..._gpiods();
> > >         if (ret)
> > >                 ...
> > >
> > > Same for regulators and anything else in the future, if any.
> >
> > I'm not sure I follow. Do you mean wrap each individual type in a wrapper
> > and call those here, like the following?
> >
> >     i2c_of_probe_enable_res(...)
> >     {
> >         ret = i2c_of_probe_enable_regulators(...)
> >         if (ret)
> >               return ret;
> >
> >         ret = i2c_of_probe_enable_gpios(...)
> >         if (ret)
> >               goto error_disable_regulators;
> >
> >         ...
> >     }
>
> Yes.
>
> ...
>
> > > > +             /*
> > > > +              * reset GPIOs normally have opposite polarity compared to
> > >
> > > "reset"
> > >
> > > > +              * enable GPIOs. Instead of parsing the flags again, simply
> > >
> > > "enable"
> > >
> > > > +              * set the raw value to high.
> > >
> > > This is quite a fragile assumption. Yes, it would work in 98% cases, but will
> > > break if it's not true somewhere else.
> >
> > Well, this seems to be the de facto standard. Or it would have to remember
> > what each GPIO descriptor's name is, and try to classify those into either
> > "enable" or "reset", and set their respective logical values to 1 or 0.
> > And then you run into a peripheral with a broken binding that has its
> > "reset" GPIO inverted, i.e. it's driver behavior needs to follow the
> > "enable" GPIO style. The class of devices this prober targets are
> > consumer electronics (laptops, tablets, phones) that at least have gone
> > through some component selection where the options won't have conflicting
> > requirements.
>
> I'm talking from real life example(s) :-)
>
> Recently I looked at the OV7251 sensor driver that expects "enable" GPIO while
> all users supply "reset"-as-"enable" with the exact trouble I described.
> Yet it's pure software / ABI issue in that case, but who knows what PCB
> engineers may come up with.

For the OV7251 case specifically, the current approach works fine, as the
polarity is the same: high electric level for a working chip.

But now that you mention it, camera sensors are exactly the case I had
in mind, though not the same chip. Some of the OmniVision and GalaxyCore
sensors have a "shutdown" pin that is active high, i.e. high electric
level shuts down the chip.

So I guess the alternative would be to remember each GPIO's name, and
do the appropriate thing based on the name, and also set the logical
value, not the raw value. If it says "shutdown" or "reset", set the
logical value to 0; if it says "enable", set it to 1. And hopefully
we don't run into a binding which has "shutdown" but is actually
"enable" ...

I don't have this case and I kind of want to leave it as a TODO though.
I feel like the scope of the series is expanding ever so slightly.


ChenYu

> > And if the polarities of the possible components don't line up, then this
> > probe structure can't really do anything. One would need something that
> > power sequences each component separately and probes it. I would really
> > like to avoid that if possible, as it makes the boot time (to peripheral
> > available) dependent on which component you have and how far down the
> > list it is. We have Chromebooks that have 4 touchscreen components
> > introduced over the years. In that case something more like Doug's
> > original proposal would work better: something that forces mutual
> > exclusivity among a class of devices.
>
> Maybe. I just pointed out the potential problem.
>
> > > > +              */
>
> --
> With Best Regards,
> Andy Shevchenko
>
>

[PATCH v5 09/10] platform/chrome: Introduce device tree hardware prober

[Chen-Yu Tsai]

Some devices are designed and manufactured with some components having
multiple drop-in replacement options. These components are often
connected to the mainboard via ribbon cables, having the same signals
and pin assignments across all options. These may include the display
panel and touchscreen on laptops and tablets, and the trackpad on
laptops. Sometimes which component option is used in a particular device
can be detected by some firmware provided identifier, other times that
information is not available, and the kernel has to try to probe each
device.

This change attempts to make the "probe each device" case cleaner. The
current approach is to have all options added and enabled in the device
tree. The kernel would then bind each device and run each driver's probe
function. This works, but has been broken before due to the introduction
of asynchronous probing, causing multiple instances requesting "shared"
resources, such as pinmuxes, GPIO pins, interrupt lines, at the same
time, with only one instance succeeding. Work arounds for these include
moving the pinmux to the parent I2C controller, using GPIO hogs or
pinmux settings to keep the GPIO pins in some fixed configuration, and
requesting the interrupt line very late. Such configurations can be seen
on the MT8183 Krane Chromebook tablets, and the Qualcomm sc8280xp-based
Lenovo Thinkpad 13S.

Instead of this delicate dance between drivers and device tree quirks,
this change introduces a simple I2C component prober. For any given
class of devices on the same I2C bus, it will go through all of them,
doing a simple I2C read transfer and see which one of them responds.
It will then enable the device that responds.

This requires some minor modifications in the existing device tree.
The status for all the device nodes for the component options must be
set to "failed-needs-probe". This makes it clear that some mechanism is
needed to enable one of them, and also prevents the prober and device
drivers running at the same time.

Signed-off-by: Chen-Yu Tsai <wenst@chromium.org>
---
Changes since v4:
- Fix Kconfig dependency
- Update copyright year
- Drop "linux/of.h" header
- Include "linux/errno.h"
- Move |int ret| declaration to top of block
- Return -ENODEV on no match instead of 0
- Unregister platform driver and device unconditionally after previous
  change

Changes since v3:
- Include linux/init.h
- Rewrite for loop in driver probe function as suggested by Andy
- Make prober driver buildable as module
- Ignore prober errors other than probe deferral

Changes since v2:
- Addressed Rob's comments
  - Move remaining driver code to drivers/platform/chrome/
  - Depend on rather than select CONFIG_I2C
  - Copy machine check to driver init function
- Addressed Andy's comments
  - Explicitly mention "device tree" or OF in driver name, description
    and Kconfig symbol
  - Drop filename from inside the file
  - Switch to passing "struct device *" to shorten lines
  - Move "ret = 0" to just before for_each_child_of_node(i2c_node, node)
  - Make loop variable size_t (instead of unsigned int as Andy asked)
  - Use PLATFORM_DEVID_NONE instead of raw -1
  - Use standard goto error path pattern in hw_prober_driver_init()

- Changes since v1:
  - New patch
---
 drivers/platform/chrome/Kconfig               |  11 ++
 drivers/platform/chrome/Makefile              |   1 +
 .../platform/chrome/chromeos_of_hw_prober.c   | 104 ++++++++++++++++++
 3 files changed, 116 insertions(+)
 create mode 100644 drivers/platform/chrome/chromeos_of_hw_prober.c

diff --git a/drivers/platform/chrome/Kconfig b/drivers/platform/chrome/Kconfig
index 7dbeb786352a..b7dbaf77b6db 100644
--- a/drivers/platform/chrome/Kconfig
+++ b/drivers/platform/chrome/Kconfig
@@ -61,6 +61,17 @@ config CHROMEOS_TBMC
 	  To compile this driver as a module, choose M here: the
 	  module will be called chromeos_tbmc.
 
+config CHROMEOS_OF_HW_PROBER
+	tristate "ChromeOS Device Tree Hardware Prober"
+	depends on OF
+	depends on I2C
+	select OF_DYNAMIC
+	default OF
+	help
+	  This option enables the device tree hardware prober for ChromeOS
+	  devices. The driver will probe the correct component variant in
+	  devices that have multiple drop-in options for one component.
+
 config CROS_EC
 	tristate "ChromeOS Embedded Controller"
 	select CROS_EC_PROTO
diff --git a/drivers/platform/chrome/Makefile b/drivers/platform/chrome/Makefile
index 2dcc6ccc2302..21a9d5047053 100644
--- a/drivers/platform/chrome/Makefile
+++ b/drivers/platform/chrome/Makefile
@@ -8,6 +8,7 @@ obj-$(CONFIG_CHROMEOS_ACPI)		+= chromeos_acpi.o
 obj-$(CONFIG_CHROMEOS_LAPTOP)		+= chromeos_laptop.o
 obj-$(CONFIG_CHROMEOS_PRIVACY_SCREEN)	+= chromeos_privacy_screen.o
 obj-$(CONFIG_CHROMEOS_PSTORE)		+= chromeos_pstore.o
+obj-$(CONFIG_CHROMEOS_OF_HW_PROBER)	+= chromeos_of_hw_prober.o
 obj-$(CONFIG_CHROMEOS_TBMC)		+= chromeos_tbmc.o
 obj-$(CONFIG_CROS_EC)			+= cros_ec.o
 obj-$(CONFIG_CROS_EC_I2C)		+= cros_ec_i2c.o
diff --git a/drivers/platform/chrome/chromeos_of_hw_prober.c b/drivers/platform/chrome/chromeos_of_hw_prober.c
new file mode 100644
index 000000000000..acf643a9eebe
--- /dev/null
+++ b/drivers/platform/chrome/chromeos_of_hw_prober.c
@@ -0,0 +1,104 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * ChromeOS Device Tree Hardware Prober
+ *
+ * Copyright (c) 2024 Google LLC
+ */
+
+#include <linux/array_size.h>
+#include <linux/errno.h>
+#include <linux/i2c.h>
+#include <linux/module.h>
+#include <linux/platform_device.h>
+
+#define DRV_NAME	"chromeos_of_hw_prober"
+
+/**
+ * struct hw_prober_entry - Holds an entry for the hardware prober
+ *
+ * @compatible:	compatible string to match against the machine
+ * @prober:	prober function to call when machine matches
+ * @data:	extra data for the prober function
+ */
+struct hw_prober_entry {
+	const char *compatible;
+	int (*prober)(struct device *dev, const void *data);
+	const void *data;
+};
+
+static int chromeos_i2c_component_prober(struct device *dev, const void *data)
+{
+	const char *type = data;
+
+	return i2c_of_probe_component(dev, type);
+}
+
+static const struct hw_prober_entry hw_prober_platforms[] = {
+	{ .compatible = "google,hana", .prober = chromeos_i2c_component_prober, .data = "touchscreen" },
+	{ .compatible = "google,hana", .prober = chromeos_i2c_component_prober, .data = "trackpad" },
+};
+
+static int chromeos_of_hw_prober_probe(struct platform_device *pdev)
+{
+	for (size_t i = 0; i < ARRAY_SIZE(hw_prober_platforms); i++) {
+		int ret;
+
+		if (!of_machine_is_compatible(hw_prober_platforms[i].compatible))
+			continue;
+
+		ret = hw_prober_platforms[i].prober(&pdev->dev, hw_prober_platforms[i].data);
+		/* Ignore unrecoverable errors and keep going through other probers */
+		if (ret == -EPROBE_DEFER)
+			return ret;
+	}
+
+	return 0;
+}
+
+static struct platform_driver chromeos_of_hw_prober_driver = {
+	.probe	= chromeos_of_hw_prober_probe,
+	.driver	= {
+		.name = DRV_NAME,
+	},
+};
+
+static struct platform_device *chromeos_of_hw_prober_pdev;
+
+static int chromeos_of_hw_prober_driver_init(void)
+{
+	size_t i;
+	int ret;
+
+	for (i = 0; i < ARRAY_SIZE(hw_prober_platforms); i++)
+		if (of_machine_is_compatible(hw_prober_platforms[i].compatible))
+			break;
+	if (i == ARRAY_SIZE(hw_prober_platforms))
+		return -ENODEV;
+
+	ret = platform_driver_register(&chromeos_of_hw_prober_driver);
+	if (ret)
+		return ret;
+
+	chromeos_of_hw_prober_pdev =
+			platform_device_register_simple(DRV_NAME, PLATFORM_DEVID_NONE, NULL, 0);
+	if (IS_ERR(chromeos_of_hw_prober_pdev))
+		goto err;
+
+	return 0;
+
+err:
+	platform_driver_unregister(&chromeos_of_hw_prober_driver);
+
+	return PTR_ERR(chromeos_of_hw_prober_pdev);
+}
+module_init(chromeos_of_hw_prober_driver_init);
+
+static void chromeos_of_hw_prober_driver_exit(void)
+{
+	platform_device_unregister(chromeos_of_hw_prober_pdev);
+	platform_driver_unregister(&chromeos_of_hw_prober_driver);
+}
+module_exit(chromeos_of_hw_prober_driver_exit);
+
+MODULE_LICENSE("GPL");
+MODULE_DESCRIPTION("ChromeOS device tree hardware prober");
-- 
2.46.0.184.g6999bdac58-goog

[PATCH v5 10/10] arm64: dts: mediatek: mt8173-elm-hana: Mark touchscreens and trackpads as fail

[Chen-Yu Tsai]

Instead of having them all available, mark them all as "fail-needs-probe"
and have the implementation try to probe which one is present.

Also remove the shared resource workaround by moving the pinctrl entry
for the trackpad interrupt line back into the individual trackpad nodes.

Signed-off-by: Chen-Yu Tsai <wenst@chromium.org>
---
Changes since v4:
- Rebased

Changes since v3:
- Also remove second source workaround, i.e. move the interrupt line
  pinctrl entry from the i2c node back to the components.

Changes since v2:
- Drop class from status
---
 arch/arm64/boot/dts/mediatek/mt8173-elm-hana.dtsi | 13 +++++++++++++
 arch/arm64/boot/dts/mediatek/mt8173-elm.dtsi      |  4 ++--
 2 files changed, 15 insertions(+), 2 deletions(-)

diff --git a/arch/arm64/boot/dts/mediatek/mt8173-elm-hana.dtsi b/arch/arm64/boot/dts/mediatek/mt8173-elm-hana.dtsi
index 8d1cbc92bce3..251e084bf7de 100644
--- a/arch/arm64/boot/dts/mediatek/mt8173-elm-hana.dtsi
+++ b/arch/arm64/boot/dts/mediatek/mt8173-elm-hana.dtsi
@@ -14,6 +14,7 @@ touchscreen2: touchscreen@34 {
 		compatible = "melfas,mip4_ts";
 		reg = <0x34>;
 		interrupts-extended = <&pio 88 IRQ_TYPE_LEVEL_LOW>;
+		status = "fail-needs-probe";
 	};
 
 	/*
@@ -26,6 +27,7 @@ touchscreen3: touchscreen@20 {
 		reg = <0x20>;
 		hid-descr-addr = <0x0020>;
 		interrupts-extended = <&pio 88 IRQ_TYPE_LEVEL_LOW>;
+		status = "fail-needs-probe";
 	};
 
 	/* Lenovo Ideapad C330 uses G2Touch touchscreen as a 2nd source touchscreen */
@@ -47,9 +49,12 @@ &i2c4 {
 	trackpad2: trackpad@2c {
 		compatible = "hid-over-i2c";
 		interrupts-extended = <&pio 117 IRQ_TYPE_LEVEL_LOW>;
+		pinctrl-names = "default";
+		pinctrl-0 = <&trackpad_irq>;
 		reg = <0x2c>;
 		hid-descr-addr = <0x0020>;
 		wakeup-source;
+		status = "fail-needs-probe";
 	};
 };
 
@@ -74,3 +79,11 @@ pins_wp {
 		};
 	};
 };
+
+&touchscreen {
+	status = "fail-needs-probe";
+};
+
+&trackpad {
+	status = "fail-needs-probe";
+};
diff --git a/arch/arm64/boot/dts/mediatek/mt8173-elm.dtsi b/arch/arm64/boot/dts/mediatek/mt8173-elm.dtsi
index b4d85147b77b..eee64461421f 100644
--- a/arch/arm64/boot/dts/mediatek/mt8173-elm.dtsi
+++ b/arch/arm64/boot/dts/mediatek/mt8173-elm.dtsi
@@ -358,12 +358,12 @@ touchscreen: touchscreen@10 {
 &i2c4 {
 	clock-frequency = <400000>;
 	status = "okay";
-	pinctrl-names = "default";
-	pinctrl-0 = <&trackpad_irq>;
 
 	trackpad: trackpad@15 {
 		compatible = "elan,ekth3000";
 		interrupts-extended = <&pio 117 IRQ_TYPE_LEVEL_LOW>;
+		pinctrl-names = "default";
+		pinctrl-0 = <&trackpad_irq>;
 		reg = <0x15>;
 		vcc-supply = <&mt6397_vgp6_reg>;
 		wakeup-source;
-- 
2.46.0.184.g6999bdac58-goog