[PATCH] regulator: Add option for machine drivers to enable the dummy regulator

[PATCH] regulator: Add option for machine drivers to enable the dummy regulator

[Mark Brown]

Allow machine drivers to explicitly enable the use of the dummy regulator,
enabling simpler support for systems with only a few specific supplies
visible to software.

It is strongly recommended that this is not used on systems with
substantial software control over their PMICs, for maximum functionality
constrints should be as fully specified as possible.

Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
---
 drivers/regulator/Makefile        |    3 +--
 drivers/regulator/core.c          |   22 ++++++++++++++++++++++
 include/linux/regulator/machine.h |    5 +++++
 3 files changed, 28 insertions(+), 2 deletions(-)

diff --git a/drivers/regulator/Makefile b/drivers/regulator/Makefile
index 8285fd8..f58bfdb 100644
--- a/drivers/regulator/Makefile
+++ b/drivers/regulator/Makefile
@@ -3,14 +3,13 @@
 #
 
 
-obj-$(CONFIG_REGULATOR) += core.o
+obj-$(CONFIG_REGULATOR) += core.o dummy.o
 obj-$(CONFIG_REGULATOR_FIXED_VOLTAGE) += fixed.o
 obj-$(CONFIG_REGULATOR_VIRTUAL_CONSUMER) += virtual.o
 obj-$(CONFIG_REGULATOR_USERSPACE_CONSUMER) += userspace-consumer.o
 
 obj-$(CONFIG_REGULATOR_AD5398) += ad5398.o
 obj-$(CONFIG_REGULATOR_BQ24022) += bq24022.o
-obj-$(CONFIG_REGULATOR_DUMMY) += dummy.o
 obj-$(CONFIG_REGULATOR_LP3971) += lp3971.o
 obj-$(CONFIG_REGULATOR_MAX1586) += max1586.o
 obj-$(CONFIG_REGULATOR_TWL4030) += twl-regulator.o
diff --git a/drivers/regulator/core.c b/drivers/regulator/core.c
index 422a709..bf82d01 100644
--- a/drivers/regulator/core.c
+++ b/drivers/regulator/core.c
@@ -33,6 +33,7 @@ static DEFINE_MUTEX(regulator_list_mutex);
 static LIST_HEAD(regulator_list);
 static LIST_HEAD(regulator_map_list);
 static int has_full_constraints;
+static bool board_wants_dummy_regulator;
 
 /*
  * struct regulator_map
@@ -1108,6 +1109,11 @@ static struct regulator *_regulator_get(struct device *dev, const char *id,
 		}
 	}
 
+	if (board_wants_dummy_regulator) {
+		rdev = dummy_regulator_rdev;
+		goto found;
+	}
+
 #ifdef CONFIG_REGULATOR_DUMMY
 	if (!devname)
 		devname = "deviceless";
@@ -2461,6 +2467,22 @@ void regulator_has_full_constraints(void)
 EXPORT_SYMBOL_GPL(regulator_has_full_constraints);
 
 /**
+ * regulator_use_dummy_regulator - Provide a dummy regulator when none is found
+ *
+ * Calling this function will cause the regulator API to provide a
+ * dummy regulator to consumers if no physical regulator is found,
+ * allowing most consumers to proceed as though a regulator were
+ * configured.  This allows systems such as those with software
+ * controllable regulators for the CPU core only to be brought up more
+ * readily.
+ */
+void regulator_use_dummy_regulator(void)
+{
+	board_wants_dummy_regulator = true;
+}
+EXPORT_SYMBOL_GPL(regulator_use_dummy_regulator);
+
+/**
  * rdev_get_drvdata - get rdev regulator driver data
  * @rdev: regulator
  *
diff --git a/include/linux/regulator/machine.h b/include/linux/regulator/machine.h
index e298028..761c745 100644
--- a/include/linux/regulator/machine.h
+++ b/include/linux/regulator/machine.h
@@ -189,10 +189,15 @@ int regulator_suspend_prepare(suspend_state_t state);
 
 #ifdef CONFIG_REGULATOR
 void regulator_has_full_constraints(void);
+void regulator_use_dummy_regulator(void);
 #else
 static inline void regulator_has_full_constraints(void)
 {
 }
+
+static inline void regulator_use_dummy_regulator(void)
+{
+}
 #endif
 
 #endif
-- 
1.7.1

Re: [PATCH] regulator: Add option for machine drivers to enable the dummy regulator

[Marek Vasut]

Dne St 6. října 2010 04:18:32 Mark Brown napsal(a):
> Allow machine drivers to explicitly enable the use of the dummy regulator,
> enabling simpler support for systems with only a few specific supplies
> visible to software.
> 
> It is strongly recommended that this is not used on systems with
> substantial software control over their PMICs, for maximum functionality
> constrints should be as fully specified as possible.

This still solves nothing ... This leaves boards broken and the granularity is 
too big. Consider there are boards that use regulator for MMC, but not for 
ADS7846 for example.

Cheers
> 
> Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
> ---
>  drivers/regulator/Makefile        |    3 +--
>  drivers/regulator/core.c          |   22 ++++++++++++++++++++++
>  include/linux/regulator/machine.h |    5 +++++
>  3 files changed, 28 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/regulator/Makefile b/drivers/regulator/Makefile
> index 8285fd8..f58bfdb 100644
> --- a/drivers/regulator/Makefile
> +++ b/drivers/regulator/Makefile
> @@ -3,14 +3,13 @@
>  #
> 
> 
> -obj-$(CONFIG_REGULATOR) += core.o
> +obj-$(CONFIG_REGULATOR) += core.o dummy.o
>  obj-$(CONFIG_REGULATOR_FIXED_VOLTAGE) += fixed.o
>  obj-$(CONFIG_REGULATOR_VIRTUAL_CONSUMER) += virtual.o
>  obj-$(CONFIG_REGULATOR_USERSPACE_CONSUMER) += userspace-consumer.o
> 
>  obj-$(CONFIG_REGULATOR_AD5398) += ad5398.o
>  obj-$(CONFIG_REGULATOR_BQ24022) += bq24022.o
> -obj-$(CONFIG_REGULATOR_DUMMY) += dummy.o
>  obj-$(CONFIG_REGULATOR_LP3971) += lp3971.o
>  obj-$(CONFIG_REGULATOR_MAX1586) += max1586.o
>  obj-$(CONFIG_REGULATOR_TWL4030) += twl-regulator.o
> diff --git a/drivers/regulator/core.c b/drivers/regulator/core.c
> index 422a709..bf82d01 100644
> --- a/drivers/regulator/core.c
> +++ b/drivers/regulator/core.c
> @@ -33,6 +33,7 @@ static DEFINE_MUTEX(regulator_list_mutex);
>  static LIST_HEAD(regulator_list);
>  static LIST_HEAD(regulator_map_list);
>  static int has_full_constraints;
> +static bool board_wants_dummy_regulator;
> 
>  /*
>   * struct regulator_map
> @@ -1108,6 +1109,11 @@ static struct regulator *_regulator_get(struct
> device *dev, const char *id, }
>  	}
> 
> +	if (board_wants_dummy_regulator) {
> +		rdev = dummy_regulator_rdev;
> +		goto found;
> +	}
> +
>  #ifdef CONFIG_REGULATOR_DUMMY
>  	if (!devname)
>  		devname = "deviceless";
> @@ -2461,6 +2467,22 @@ void regulator_has_full_constraints(void)
>  EXPORT_SYMBOL_GPL(regulator_has_full_constraints);
> 
>  /**
> + * regulator_use_dummy_regulator - Provide a dummy regulator when none is
> found + *
> + * Calling this function will cause the regulator API to provide a
> + * dummy regulator to consumers if no physical regulator is found,
> + * allowing most consumers to proceed as though a regulator were
> + * configured.  This allows systems such as those with software
> + * controllable regulators for the CPU core only to be brought up more
> + * readily.
> + */
> +void regulator_use_dummy_regulator(void)
> +{
> +	board_wants_dummy_regulator = true;
> +}
> +EXPORT_SYMBOL_GPL(regulator_use_dummy_regulator);
> +
> +/**
>   * rdev_get_drvdata - get rdev regulator driver data
>   * @rdev: regulator
>   *
> diff --git a/include/linux/regulator/machine.h
> b/include/linux/regulator/machine.h index e298028..761c745 100644
> --- a/include/linux/regulator/machine.h
> +++ b/include/linux/regulator/machine.h
> @@ -189,10 +189,15 @@ int regulator_suspend_prepare(suspend_state_t state);
> 
>  #ifdef CONFIG_REGULATOR
>  void regulator_has_full_constraints(void);
> +void regulator_use_dummy_regulator(void);
>  #else
>  static inline void regulator_has_full_constraints(void)
>  {
>  }
> +
> +static inline void regulator_use_dummy_regulator(void)
> +{
> +}
>  #endif
> 
>  #endif

Re: [PATCH] regulator: Add option for machine drivers to enable the dummy regulator

[Mark Brown]

On Wed, Oct 06, 2010 at 06:09:17AM +0200, Marek Vasut wrote:

> This still solves nothing ... This leaves boards broken and the granularity is 
We've already got a compatibility option which allows this to be enabled
with a config change, and the actual code update required has been
reduced to a single line.  This does not seem terribly onerous.

> too big. Consider there are boards that use regulator for MMC, but not for 
> ADS7846 for example.

You can't have it both ways.  Either drivers just ignore missing
regulators and just run as though everything is fine (which is what
you've been requesting enthusiastically) or they trust that they can get
accurate information about what is and isn't mapped.  All this is doing
is providing a way to enable the former behaviour in the core.

Re: [PATCH] regulator: Add option for machine drivers to enable the dummy regulator

[Liam Girdwood]

On Tue, 2010-10-05 at 19:18 -0700, Mark Brown wrote:
> Allow machine drivers to explicitly enable the use of the dummy regulator,
> enabling simpler support for systems with only a few specific supplies
> visible to software.
> 
> It is strongly recommended that this is not used on systems with
> substantial software control over their PMICs, for maximum functionality
> constrints should be as fully specified as possible.
> 
> Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>

Applied.

Thanks

Liam
-- 
Freelance Developer, SlimLogic Ltd
ASoC and Voltage Regulator Maintainer.
http://www.slimlogic.co.uk