Skip to contentSkip to conversation

media: ipu3-cio2: Don't dereference fwnode handle

PATCH Superseded Andy Shevchenko 3 participants 10 messages Nov 21, 2022 – Dec 8, 2022

Subsystems (5)

v2 of 3 Superseded Latest is v3 →

lore
Share & tools
lore

Andy Shevchenko

author reviewer
#0 lore raw

PATCH v2 1/4 media: ipu3-cio2: Don't dereference fwnode handle

Use acpi_fwnode_handle() instead of dereferencing an fwnode handle directly, which is a better coding practice.

---
v2: added tags (Heikki, Daniel)

 drivers/media/pci/intel/ipu3/cio2-bridge.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)
drivers/media/pci/intel/ipu3/cio2-bridge.c +4 -2
index df6c94da2f6a..18974a72e94a 100644
@@ -263,7 +263,7 @@ static int cio2_bridge_connect_sensor(const struct cio2_sensor_config *cfg,
 				      struct cio2_bridge *bridge,
 				      struct pci_dev *cio2)
 {
-	struct fwnode_handle *fwnode;
+	struct fwnode_handle *fwnode, *primary;
 	struct cio2_sensor *sensor;
 	struct acpi_device *adev;
 	acpi_status status;
@@ -322,7 +322,9 @@ static int cio2_bridge_connect_sensor(const struct cio2_sensor_config *cfg,
 		}
 
 		sensor->adev = acpi_dev_get(adev);
-		adev->fwnode.secondary = fwnode;
+
+		primary = acpi_fwnode_handle(adev);
+		primary->secondary = fwnode;
 
 		cio2_bridge_instantiate_vcm_i2c_client(sensor);
 
Signature · 1 line
2.35.1
author reviewer
#2 lore raw

PATCH v2 3/4 software node: Switch property entry test to a new API

Switch property entry test to use software_node_register_node_group() API. The current one is going to be removed soon.

---
v2: added tags (Heikki, Daniel)

 drivers/base/test/property-entry-test.c | 30 ++++++++++++-------------
 1 file changed, 14 insertions(+), 16 deletions(-)
drivers/base/test/property-entry-test.c +14 -16
index 6071d5bc128c..dd2b606d76a3 100644
@@ -405,20 +405,18 @@ static void pe_test_move_inline_str(struct kunit *test)
 /* Handling of reference properties */
 static void pe_test_reference(struct kunit *test)
 {
-	static const struct software_node nodes[] = {
-		{ .name = "1", },
-		{ .name = "2", },
-		{ }
-	};
+	static const struct software_node node1 = { .name = "1" };
+	static const struct software_node node2 = { .name = "2" };
+	static const struct software_node *group[] = { &node1, &node2, NULL };
 
 	static const struct software_node_ref_args refs[] = {
-		SOFTWARE_NODE_REFERENCE(&nodes[0]),
-		SOFTWARE_NODE_REFERENCE(&nodes[1], 3, 4),
+		SOFTWARE_NODE_REFERENCE(&node1),
+		SOFTWARE_NODE_REFERENCE(&node2, 3, 4),
 	};
 
 	const struct property_entry entries[] = {
-		PROPERTY_ENTRY_REF("ref-1", &nodes[0]),
-		PROPERTY_ENTRY_REF("ref-2", &nodes[1], 1, 2),
+		PROPERTY_ENTRY_REF("ref-1", &node1),
+		PROPERTY_ENTRY_REF("ref-2", &node2, 1, 2),
 		PROPERTY_ENTRY_REF_ARRAY("ref-3", refs),
 		{ }
 	};
@@ -427,7 +425,7 @@ static void pe_test_reference(struct kunit *test)
 	struct fwnode_reference_args ref;
 	int error;
 
-	error = software_node_register_nodes(nodes);
+	error = software_node_register_node_group(group);
 	KUNIT_ASSERT_EQ(test, error, 0);
 
 	node = fwnode_create_software_node(entries, NULL);
@@ -436,7 +434,7 @@ static void pe_test_reference(struct kunit *test)
 	error = fwnode_property_get_reference_args(node, "ref-1", NULL,
 						   0, 0, &ref);
 	KUNIT_ASSERT_EQ(test, error, 0);
-	KUNIT_EXPECT_PTR_EQ(test, to_software_node(ref.fwnode), &nodes[0]);
+	KUNIT_EXPECT_PTR_EQ(test, to_software_node(ref.fwnode), &node1);
 	KUNIT_EXPECT_EQ(test, ref.nargs, 0U);
 
 	/* wrong index */
@@ -447,7 +445,7 @@ static void pe_test_reference(struct kunit *test)
 	error = fwnode_property_get_reference_args(node, "ref-2", NULL,
 						   1, 0, &ref);
 	KUNIT_ASSERT_EQ(test, error, 0);
-	KUNIT_EXPECT_PTR_EQ(test, to_software_node(ref.fwnode), &nodes[1]);
+	KUNIT_EXPECT_PTR_EQ(test, to_software_node(ref.fwnode), &node2);
 	KUNIT_EXPECT_EQ(test, ref.nargs, 1U);
 	KUNIT_EXPECT_EQ(test, ref.args[0], 1LLU);
 
@@ -455,7 +453,7 @@ static void pe_test_reference(struct kunit *test)
 	error = fwnode_property_get_reference_args(node, "ref-2", NULL,
 						   3, 0, &ref);
 	KUNIT_ASSERT_EQ(test, error, 0);
-	KUNIT_EXPECT_PTR_EQ(test, to_software_node(ref.fwnode), &nodes[1]);
+	KUNIT_EXPECT_PTR_EQ(test, to_software_node(ref.fwnode), &node2);
 	KUNIT_EXPECT_EQ(test, ref.nargs, 3U);
 	KUNIT_EXPECT_EQ(test, ref.args[0], 1LLU);
 	KUNIT_EXPECT_EQ(test, ref.args[1], 2LLU);
@@ -470,14 +468,14 @@ static void pe_test_reference(struct kunit *test)
 	error = fwnode_property_get_reference_args(node, "ref-3", NULL,
 						   0, 0, &ref);
 	KUNIT_ASSERT_EQ(test, error, 0);
-	KUNIT_EXPECT_PTR_EQ(test, to_software_node(ref.fwnode), &nodes[0]);
+	KUNIT_EXPECT_PTR_EQ(test, to_software_node(ref.fwnode), &node1);
 	KUNIT_EXPECT_EQ(test, ref.nargs, 0U);
 
 	/* second reference in the array */
 	error = fwnode_property_get_reference_args(node, "ref-3", NULL,
 						   2, 1, &ref);
 	KUNIT_ASSERT_EQ(test, error, 0);
-	KUNIT_EXPECT_PTR_EQ(test, to_software_node(ref.fwnode), &nodes[1]);
+	KUNIT_EXPECT_PTR_EQ(test, to_software_node(ref.fwnode), &node2);
 	KUNIT_EXPECT_EQ(test, ref.nargs, 2U);
 	KUNIT_EXPECT_EQ(test, ref.args[0], 3LLU);
 	KUNIT_EXPECT_EQ(test, ref.args[1], 4LLU);
@@ -488,7 +486,7 @@ static void pe_test_reference(struct kunit *test)
 	KUNIT_EXPECT_NE(test, error, 0);
 
 	fwnode_remove_software_node(node);
-	software_node_unregister_nodes(nodes);
+	software_node_unregister_node_group(group);
 }
 
 static struct kunit_case property_entry_test_cases[] = {
Signature · 1 line
2.35.1
author reviewer
#1 lore raw

PATCH v2 2/4 media: ipu3-cio2: Convert to use software_node_register_node_group()

In reply to Andy Shevchenko

The currently used software_node_register_nodes() is going to be removed. Prepare driver by switching to a new API.

---
v2: added tags (Heikki, Daniel)

 drivers/media/pci/intel/ipu3/cio2-bridge.c | 21 ++++++++++++++++++---
 drivers/media/pci/intel/ipu3/cio2-bridge.h |  5 +++--
 2 files changed, 21 insertions(+), 5 deletions(-)
drivers/media/pci/intel/ipu3/cio2-bridge.c +18 -3
index 18974a72e94a..dfefe0d8aa95 100644
@@ -195,6 +195,19 @@ static void cio2_bridge_init_swnode_names(struct cio2_sensor *sensor)
 		 SWNODE_GRAPH_ENDPOINT_NAME_FMT, 0); /* And endpoint 0 */
 }
 
+static void cio2_bridge_init_swnode_group(struct cio2_sensor *sensor)
+{
+	struct software_node *nodes = sensor->swnodes;
+
+	sensor->group[SWNODE_SENSOR_HID] = &nodes[SWNODE_SENSOR_HID];
+	sensor->group[SWNODE_SENSOR_PORT] = &nodes[SWNODE_SENSOR_PORT];
+	sensor->group[SWNODE_SENSOR_ENDPOINT] = &nodes[SWNODE_SENSOR_ENDPOINT];
+	sensor->group[SWNODE_CIO2_PORT] = &nodes[SWNODE_CIO2_PORT];
+	sensor->group[SWNODE_CIO2_ENDPOINT] = &nodes[SWNODE_CIO2_ENDPOINT];
+	if (sensor->ssdb.vcmtype)
+		sensor->group[SWNODE_VCM] =  &nodes[SWNODE_VCM];
+}
+
 static void cio2_bridge_create_connection_swnodes(struct cio2_bridge *bridge,
 						  struct cio2_sensor *sensor)
 {
@@ -219,6 +232,8 @@ static void cio2_bridge_create_connection_swnodes(struct cio2_bridge *bridge,
 	if (sensor->ssdb.vcmtype)
 		nodes[SWNODE_VCM] =
 			NODE_VCM(cio2_vcm_types[sensor->ssdb.vcmtype - 1]);
+
+	cio2_bridge_init_swnode_group(sensor);
 }
 
 static void cio2_bridge_instantiate_vcm_i2c_client(struct cio2_sensor *sensor)
@@ -252,7 +267,7 @@ static void cio2_bridge_unregister_sensors(struct cio2_bridge *bridge)
 
 	for (i = 0; i < bridge->n_sensors; i++) {
 		sensor = &bridge->sensors[i];
-		software_node_unregister_nodes(sensor->swnodes);
+		software_node_unregister_node_group(sensor->group);
 		ACPI_FREE(sensor->pld);
 		acpi_dev_put(sensor->adev);
 		i2c_unregister_device(sensor->vcm_i2c_client);
@@ -310,7 +325,7 @@ static int cio2_bridge_connect_sensor(const struct cio2_sensor_config *cfg,
 		cio2_bridge_create_fwnode_properties(sensor, bridge, cfg);
 		cio2_bridge_create_connection_swnodes(bridge, sensor);
 
-		ret = software_node_register_nodes(sensor->swnodes);
+		ret = software_node_register_node_group(sensor->group);
 		if (ret)
 			goto err_free_pld;
 
@@ -337,7 +352,7 @@ static int cio2_bridge_connect_sensor(const struct cio2_sensor_config *cfg,
 	return 0;
 
 err_free_swnodes:
-	software_node_unregister_nodes(sensor->swnodes);
+	software_node_unregister_node_group(sensor->group);
 err_free_pld:
 	ACPI_FREE(sensor->pld);
 err_put_adev:
drivers/media/pci/intel/ipu3/cio2-bridge.h +3 -2
index 4418cbd08208..b93b749c65bd 100644
@@ -117,8 +117,9 @@ struct cio2_sensor {
 	struct acpi_device *adev;
 	struct i2c_client *vcm_i2c_client;
 
-	/* SWNODE_COUNT + 1 for terminating empty node */
-	struct software_node swnodes[SWNODE_COUNT + 1];
+	/* SWNODE_COUNT + 1 for terminating NULL */
+	const struct software_node *group[SWNODE_COUNT + 1];
+	struct software_node swnodes[SWNODE_COUNT];
 	struct cio2_node_names node_names;
 
 	struct cio2_sensor_ssdb ssdb;
Signature · 1 line
2.35.1
author reviewer
#3 lore raw

PATCH v2 4/4 software node: Remove unused APIs

In reply to Andy Shevchenko

There are no more users of software_node_register_nodes() and software_node_unregister_nodes(). Remove them.

---
v2: added tags (Heikki, Daniel)

 drivers/base/swnode.c    | 61 ----------------------------------------
 include/linux/property.h |  3 --
 2 files changed, 64 deletions(-)
drivers/base/swnode.c +0 -61
index 0a482212c7e8..da3c1c2cb9a8 100644
@@ -819,67 +819,6 @@ swnode_register(const struct software_node *node, struct swnode *parent,
 	return &swnode->fwnode;
 }
 
-/**
- * software_node_register_nodes - Register an array of software nodes
- * @nodes: Zero terminated array of software nodes to be registered
- *
- * Register multiple software nodes at once. If any node in the array
- * has its .parent pointer set (which can only be to another software_node),
- * then its parent **must** have been registered before it is; either outside
- * of this function or by ordering the array such that parent comes before
- * child.
- */
-int software_node_register_nodes(const struct software_node *nodes)
-{
-	int ret;
-	int i;
-
-	for (i = 0; nodes[i].name; i++) {
-		const struct software_node *parent = nodes[i].parent;
-
-		if (parent && !software_node_to_swnode(parent)) {
-			ret = -EINVAL;
-			goto err_unregister_nodes;
-		}
-
-		ret = software_node_register(&nodes[i]);
-		if (ret)
-			goto err_unregister_nodes;
-	}
-
-	return 0;
-
-err_unregister_nodes:
-	software_node_unregister_nodes(nodes);
-	return ret;
-}
-EXPORT_SYMBOL_GPL(software_node_register_nodes);
-
-/**
- * software_node_unregister_nodes - Unregister an array of software nodes
- * @nodes: Zero terminated array of software nodes to be unregistered
- *
- * Unregister multiple software nodes at once. If parent pointers are set up
- * in any of the software nodes then the array **must** be ordered such that
- * parents come before their children.
- *
- * NOTE: If you are uncertain whether the array is ordered such that
- * parents will be unregistered before their children, it is wiser to
- * remove the nodes individually, in the correct order (child before
- * parent).
- */
-void software_node_unregister_nodes(const struct software_node *nodes)
-{
-	unsigned int i = 0;
-
-	while (nodes[i].name)
-		i++;
-
-	while (i--)
-		software_node_unregister(&nodes[i]);
-}
-EXPORT_SYMBOL_GPL(software_node_unregister_nodes);
-
 /**
  * software_node_register_node_group - Register a group of software nodes
  * @node_group: NULL terminated array of software node pointers to be registered
include/linux/property.h +0 -3
index 5d840299146d..6870abe12dc5 100644
@@ -492,9 +492,6 @@ const struct software_node *
 software_node_find_by_name(const struct software_node *parent,
 			   const char *name);
 
-int software_node_register_nodes(const struct software_node *nodes);
-void software_node_unregister_nodes(const struct software_node *nodes);
-
 int software_node_register_node_group(const struct software_node **node_group);
 void software_node_unregister_node_group(const struct software_node **node_group);
 
Signature · 1 line
2.35.1
author reviewer
#4 lore raw
In reply to Andy Shevchenko

+ Cc: Petr, Sergey

Quoted 2 lines · Andy Shevchenko

Use acpi_fwnode_handle() instead of dereferencing an fwnode handle directly, which is a better coding practice.

It appears that this series depends on fd070e8ceb90 ("test_printf: Refactor fwnode_pointer() to make it more readable") which is in PRINTK tree.

Sakari, Mauro, if you are okay to route this via that tree, can we get your tags for that? Otherwise we need to postpone this till v6.2-rc1 (but I would like to decrease the chances to appear a new user of the to be removed API).

Signature · 2 lines
With Best Regards,
Andy Shevchenko
author reviewer
#5 lore raw
In reply to Andy Shevchenko

Dunno what happened to my previous reply to this. Okay, trying again...

+ Cc: Petr, Sergey

Quoted 2 lines · Andy Shevchenko

Use acpi_fwnode_handle() instead of dereferencing an fwnode handle directly, which is a better coding practice.

It appears that this series depends on fd070e8ceb90 ("test_printf: Refactor fwnode_pointer() to make it more readable") which is in PRINTK tree.

Sakari, Mauro, if you are okay to route this via that tree, can we get your tags for that? Otherwise we need to postpone this till v6.2-rc1 (but I would like to decrease the chances to appear a new user of the to be removed API).

Note, that Greg Acked v1 of the swnode patches (which are the same in v2).

Signature · 2 lines
With Best Regards,
Andy Shevchenko
maintainer Acked-by
#6 lore raw

Hi Andy,

Quoted 8 lines · Andy Shevchenko

Dunno what happened to my previous reply to this. Okay, trying again...

+ Cc: Petr, Sergey

Quoted 2 lines · Andy Shevchenko

Use acpi_fwnode_handle() instead of dereferencing an fwnode handle directly, which is a better coding practice.

Quoted 8 lines

It appears that this series depends on fd070e8ceb90 ("test_printf: Refactor fwnode_pointer() to make it more readable") which is in PRINTK tree.

Sakari, Mauro, if you are okay to route this via that tree, can we get your tags for that? Otherwise we need to postpone this till v6.2-rc1 (but I would like to decrease the chances to appear a new user of the to be removed API).

Note, that Greg Acked v1 of the swnode patches (which are the same in v2).

Sorry for the late reply. Feel free to do that if it's not too late, with:

I don't think the linkelihood for having a new user for this API is high.

Signature · 3 lines
Kind regards,

Sakari Ailus
author reviewer
#7 lore raw
Quoted 16 lines · Sakari Ailus
Quoted 15 lines · Andy Shevchenko

Dunno what happened to my previous reply to this. Okay, trying again...

+ Cc: Petr, Sergey

Quoted 2 lines · Andy Shevchenko

Use acpi_fwnode_handle() instead of dereferencing an fwnode handle directly, which is a better coding practice.

It appears that this series depends on fd070e8ceb90 ("test_printf: Refactor fwnode_pointer() to make it more readable") which is in PRINTK tree.

Sakari, Mauro, if you are okay to route this via that tree, can we get your tags for that? Otherwise we need to postpone this till v6.2-rc1 (but I would like to decrease the chances to appear a new user of the to be removed API).

Quoted 5 lines
Quoted 1 line

Note, that Greg Acked v1 of the swnode patches (which are the same in v2).

Sorry for the late reply. Feel free to do that if it's not too late, with:

Thank you!
I think it's a bit late for printk tree to consume this. If it's the case (Petr?) then I will submit a new version after v6.2-rc1 is out.

Quoted 1 line

I don't think the linkelihood for having a new user for this API is high.

Signature · 2 lines
With Best Regards,
Andy Shevchenko
maintainer
#8 lore raw
Quoted 17 lines · Andy Shevchenko
Quoted 16 lines · Sakari Ailus
Quoted 15 lines · Andy Shevchenko

Dunno what happened to my previous reply to this. Okay, trying again...

+ Cc: Petr, Sergey

Quoted 2 lines · Andy Shevchenko

Use acpi_fwnode_handle() instead of dereferencing an fwnode handle directly, which is a better coding practice.

It appears that this series depends on fd070e8ceb90 ("test_printf: Refactor fwnode_pointer() to make it more readable") which is in PRINTK tree.

Sakari, Mauro, if you are okay to route this via that tree, can we get your tags for that? Otherwise we need to postpone this till v6.2-rc1 (but I would like to decrease the chances to appear a new user of the to be removed API).

Quoted 5 lines
Quoted 5 lines
Quoted 1 line

Note, that Greg Acked v1 of the swnode patches (which are the same in v2).

Sorry for the late reply. Feel free to do that if it's not too late, with:

I intended to add this applies to the set.

Quoted 6 lines

Thank you!
I think it's a bit late for printk tree to consume this. If it's the case (Petr?) then I will submit a new version after v6.2-rc1 is out.

Quoted 1 line

I don't think the linkelihood for having a new user for this API is high.

Signature · 1 line
Sakari Ailus
#9 lore raw
In reply to Andy Shevchenko
Quoted 19 lines · Andy Shevchenko
Quoted 18 lines · Sakari Ailus
Quoted 16 lines · Andy Shevchenko

Dunno what happened to my previous reply to this. Okay, trying again...

+ Cc: Petr, Sergey

Quoted 2 lines · Andy Shevchenko

Use acpi_fwnode_handle() instead of dereferencing an fwnode handle directly, which is a better coding practice.

It appears that this series depends on fd070e8ceb90 ("test_printf: Refactor fwnode_pointer() to make it more readable") which is in PRINTK tree.

Sakari, Mauro, if you are okay to route this via that tree, can we get your tags for that? Otherwise we need to postpone this till v6.2-rc1 (but I would like to decrease the chances to appear a new user of the to be removed API).

Note, that Greg Acked v1 of the swnode patches (which are the same in v2).

Quoted 7 lines
Quoted 3 lines

Sorry for the late reply. Feel free to do that if it's not too late, with:

Thank you!
I think it's a bit late for printk tree to consume this. If it's the case (Petr?) then I will submit a new version after v6.2-rc1 is out.

Yes, I am sorry but it is too late for the printk tree. I am going to send the pull request for 6.2 today or tomorrow. Linus explicitly asked to send the pull request early this time because the merge window will be overlapping with the holidays.

On the positive side. There is a high chance that the changes from the printk tree will be in the mainline early enough so that you could manage to send this still during the merge window.

Best Regards,
Petr