regulator: arizona-micsupp: Fix build errors with CONFIG_OF not defined
PATCH 1/2 regulator: arizona-micsupp: Fix build errors with CONFIG_OF not defined
Some of the OF handling functions do not have empty alternatives defined if CONFIG_OF is not defined. This patch ifdefs out the offending code.
--- drivers/regulator/arizona-micsupp.c | 20 +++++++++++++++++++- 1 file changed, 19 insertions(+), 1 deletion(-)
drivers/regulator/arizona-micsupp.c +19 -1
@@ -196,6 +196,7 @@ static const struct regulator_init_data arizona_micsupp_ext_default = { .num_consumer_supplies = 1, }; +#if IS_ENABLED (CONFIG_OF )static int arizona_micsupp_of_get_pdata (struct arizona *arizona,struct regulator_config *config) {
@@ -222,6 +223,22 @@ static int arizona_micsupp_of_get_pdata(struct arizona *arizona, return 0; } +static void arizona_micsupp_of_put_pdata (struct regulator_config *config) +{ +of_node_put (config->of_node); +} +#else +static int arizona_micsupp_of_get_pdata (struct arizona *arizona, +struct regulator_config *config) +{ +return 0 ; +} + +static void arizona_micsupp_of_put_pdata (struct regulator_config *config) +{ +} +#endif +static int arizona_micsupp_probe (struct platform_device *pdev) {struct arizona *arizona =dev_get_drvdata (pdev->dev.parent);
@@ -288,7 +305,8 @@ static int arizona_micsupp_probe(struct platform_device *pdev)return ret; } -of_node_put (config.of_node); +if (IS_ENABLED (CONFIG_OF )) +arizona_micsupp_of_put_pdata (&config);platform_set_drvdata (pdev, micsupp);
Signature · 1 line
1.7.9.5
PATCH 2/2 regulator: arizona-ldo1: Fix build errors for when CONFIG_OF not defined
Some of the OF handling functions do not have empty alternatives defined if CONFIG_OF is not defined. This patch ifdefs out the offending code.
--- drivers/regulator/arizona-ldo1.c | 20 +++++++++++++++++++- 1 file changed, 19 insertions(+), 1 deletion(-)
drivers/regulator/arizona-ldo1.c +19 -1
@@ -179,6 +179,7 @@ static const struct regulator_init_data arizona_ldo1_default = { .num_consumer_supplies = 1, }; +#if IS_ENABLED (CONFIG_OF )static int arizona_ldo1_of_get_pdata (struct arizona *arizona,struct regulator_config *config) {
@@ -215,6 +216,22 @@ static int arizona_ldo1_of_get_pdata(struct arizona *arizona, return 0; } +static void arizona_ldo1_of_put_pdata (struct regulator_config *config) +{ +of_node_put (config->of_node); +} +#else +static int arizona_ldo1_of_get_pdata (struct arizona *arizona, +struct regulator_config *config) +{ +return 0 ; +} + +static void arizona_ldo1_of_put_pdata (struct regulator_config *config) +{ +} +#endif +static int arizona_ldo1_probe (struct platform_device *pdev) {struct arizona *arizona =dev_get_drvdata (pdev->dev.parent);
@@ -286,7 +303,8 @@ static int arizona_ldo1_probe(struct platform_device *pdev)return ret; } -of_node_put (config.of_node); +if (IS_ENABLED (CONFIG_OF )) +arizona_ldo1_of_put_pdata (&config);platform_set_drvdata (pdev, ldo1);
Signature · 1 line
1.7.9.5
Mark Brown
maintainerQuoted 2 lines · Charles Keepax
Some of the OF handling functions do not have empty alternatives defined if
CONFIG_OFis not defined. This patch ifdefs out the offending code.
Which functions and why is it not better to provide the stubs?
Quoted 6 lines · Mark Brown
Quoted 2 lines · Charles Keepax
Some of the OF handling functions do not have empty alternatives defined if
CONFIG_OFis not defined. This patch ifdefs out the offending code.Which functions and why is it not better to provide the stubs?
Specifically of_get_child_by_name, of_parse_handle and of_node_put, I guess stub would probably be better. I will do a patch for that instead.
Thanks,
Charles
Mark Brown
maintainerQuoted 11 lines · Charles Keepax
Quoted 6 lines · Mark Brown
Quoted 2 lines · Charles Keepax
Some of the OF handling functions do not have empty alternatives defined if
CONFIG_OFis not defined. This patch ifdefs out the offending code.Which functions and why is it not better to provide the stubs?
Specifically
of_get_child_by_name,of_parse_handleandof_node_put, I guess stub would probably be better. I will do a patch for that instead.
There's already a stub for of_get_child_by_name() in current code. of_parse_handle() might not be ideal, I'd need to check the usage, but I'd definitely expect to see a stub for of_node_put(), it's quite common to need it even for fairly basic parsing.