Skip to contentSkip to conversation

regulator: core: Use devres for releasing of_regulator_match of_nodes

PATCH In review Charles Keepax 2 participants 2 messages Apr 15, 2014 – Apr 18, 2014 Voltage and current regulator framework

lore
Share & tools
lore

Charles Keepax

author
#0 lore raw

PATCH 1/1 regulator: core: Use devres for releasing of_regulator_match of_nodes

Rather than requiring individual drivers to put the of_nodes returned from of_regulator_match use devres to put them. This also has the benefit it makes the life-time of the of_nodes match the lifetime of the init data also contained in the of_regulator_match structure, which seems more consistent.

Signed-off-by Charles Keepax <[email protected]>
---
 drivers/regulator/of_regulator.c       |   47 +++++++++++++++++--------------
 include/linux/regulator/of_regulator.h |    7 -----
 2 files changed, 26 insertions(+), 28 deletions(-)
drivers/regulator/of_regulator.c +26 -21
index 188e0cb..4672cd2 100644
@@ -106,6 +106,20 @@ struct regulator_init_data *of_get_regulator_init_data(struct device *dev,
 }
 EXPORT_SYMBOL_GPL(of_get_regulator_init_data);
 
+struct devm_of_regulator_matches {
+	struct of_regulator_match *matches;
+	unsigned int num_matches;
+};
+
+static void devm_of_regulator_put_matches(struct device *dev, void *res)
+{
+	struct devm_of_regulator_matches *devm_matches = res;
+	int i;
+
+	for (i = 0; i < devm_matches->num_matches; i++)
+		of_node_put(devm_matches->matches[i].of_node);
+}
+
 /**
  * of_regulator_match - extract multiple regulator init data from device tree.
  * @dev: device requesting the data

@@ -132,10 +146,22 @@ int of_regulator_match(struct device *dev, struct device_node *node,
 	unsigned int i;
 	const char *name;
 	struct device_node *child;
+	struct devm_of_regulator_matches *devm_matches;
 
 	if (!dev || !node)
 		return -EINVAL;
 
+	devm_matches = devres_alloc(devm_of_regulator_put_matches,
+				    sizeof(struct devm_of_regulator_matches),
+				    GFP_KERNEL);
+	if (!devm_matches)
+		return -ENOMEM;
+
+	devm_matches->matches = matches;
+	devm_matches->num_matches = num_matches;
+
+	devres_add(dev, devm_matches);
+
 	for (i = 0; i < num_matches; i++) {
 		struct of_regulator_match *match = &matches[i];
 		match->init_data = NULL;

@@ -172,24 +198,3 @@ int of_regulator_match(struct device *dev, struct device_node *node,
 	return count;
 }
 EXPORT_SYMBOL_GPL(of_regulator_match);
-
-/**
- * of_regulator_put_match - put the of_node references from an
- *                          of_regulator_match structure
- * @matches: match table for the regulators
- * @num_matches: number of entries in match table
- *
- * This function goes through a match table and calls of_node_put on each
- * of_node.
- */
-int of_regulator_put_match(struct of_regulator_match *matches,
-			   unsigned int num_matches)
-{
-	int i;
-
-	for (i = 0; i < num_matches; i++)
-		of_node_put(matches[i].of_node);
-
-	return 0;
-}
-EXPORT_SYMBOL_GPL(of_regulator_put_match);
include/linux/regulator/of_regulator.h +0 -7
index a5abd83..f921796 100644
@@ -20,8 +20,6 @@ extern struct regulator_init_data
 extern int of_regulator_match(struct device *dev, struct device_node *node,
 			      struct of_regulator_match *matches,
 			      unsigned int num_matches);
-extern int of_regulator_put_match(struct of_regulator_match *matches,
-				  unsigned int num_matches);
 #else
 static inline struct regulator_init_data
 	*of_get_regulator_init_data(struct device *dev,

@@ -37,11 +35,6 @@ static inline int of_regulator_match(struct device *dev,
 {
 	return 0;
 }
-static inline int of_regulator_put_match(struct of_regulator_match *matches,
-					 unsigned int num_matches)
-{
-	return 0;
-}
 #endif /* CONFIG_OF */
 
 #endif /* __LINUX_OF_REG_H */
Signature · 1 line
1.7.2.5
maintainer
#1 lore raw
Quoted 5 lines · Charles Keepax

Rather than requiring individual drivers to put the of_nodes returned from of_regulator_match use devres to put them. This also has the benefit it makes the life-time of the of_nodes match the lifetime of the init data also contained in the of_regulator_match structure, which seems more consistent.

Applied, thanks. It'd have been good to have explained that we were already using managed allocations - allocating things without an explict method for freeing if required is normally bad form, though it's harmless and like I say already done here.