[PATCHv3 0/2] regulator: of: Add support for pasing regulator suspend state

[PATCHv3 0/2] regulator: of: Add support for pasing regulator suspend state

[Chanwoo Choi]

The regulators would set different state/mode according to the kind of suspend
state. So regulation_constraints structure has already regulator suspend state filed.
This patch parse regulator suspend state from devicetree file.

For example:

	ldoX_reg: LDOx {
		regulator-name = "VAP_XXX_1.2V";
		regulator-min-microvolt = <1200000>;
		regulator-max-microvolt = <1200000>;
		regulator-always-on;

		regulator-initial-state = <3>;	/* PM_SUSPEND_MEM */
		regulator-state-standby {
			regulator-volt = <1200000>;
		};

		regulator-state-mem {
			regulator-volt = <1200000>;
			regulator-off-in-suspend;
		};

		regulator-state-disk {
			regulator-volt = <1200000>;
			regulator-on-in-suspend;
		};
	};

Changes from v2:
- Fix over 80 lines by using checkpatch script
- Rebase this patchset on latest for-next branch of regulator.git

Changes from v1:
- Check whether regulator-initial-state and regulator-mode is correct or not
- Add more detailed description about regulator-initial-state, regulator-mode
  and regulator-state-[standby/mem/disk] for devicetree bindings
- Modify example of regulator suspend state in bindings documentation

Chanwoo Choi (2):
  regulator: of: Add support for parsing regulator_state for suspend state
  dt-bindings: regulator: Add regulator suspend state for PM state

 .../devicetree/bindings/regulator/regulator.txt    | 32 +++++++++
 drivers/regulator/of_regulator.c                   | 79 +++++++++++++++++++++-
 2 files changed, 109 insertions(+), 2 deletions(-)

-- 
1.8.0

[PATCHv3 1/2] regulator: of: Add support for parsing regulator_state for suspend state

[Chanwoo Choi]

The regulation_constraints structure includes specific field to support
suspend state for global PMIC STANDBY/HIBERNATE mode. This patch add support
for parsing regulator_state for suspend state.

Signed-off-by: Chanwoo Choi <cw00.choi@samsung.com>
Acked-by: Kyungmin Park <kyungmin.park@samsung.com>
---
 drivers/regulator/of_regulator.c | 79 +++++++++++++++++++++++++++++++++++++++-
 1 file changed, 77 insertions(+), 2 deletions(-)

diff --git a/drivers/regulator/of_regulator.c b/drivers/regulator/of_regulator.c
index ee5e67b..24d5952 100644
--- a/drivers/regulator/of_regulator.c
+++ b/drivers/regulator/of_regulator.c
@@ -16,12 +16,20 @@
 #include <linux/regulator/machine.h>
 #include <linux/regulator/of_regulator.h>
 
+const char *const regulator_states[PM_SUSPEND_MAX + 1] = {
+	[PM_SUSPEND_STANDBY]	= "regulator-state-standby",
+	[PM_SUSPEND_MEM]	= "regulator-state-mem",
+	[PM_SUSPEND_MAX]	= "regulator-state-disk",
+};
+
 static void of_get_regulation_constraints(struct device_node *np,
 					struct regulator_init_data **init_data)
 {
-	const __be32 *min_uV, *max_uV;
+	const __be32 *min_uV, *max_uV, *suspend_uV;
 	struct regulation_constraints *constraints = &(*init_data)->constraints;
-	int ret;
+	struct regulator_state *suspend_state;
+	struct device_node *suspend_np;
+	int ret, i;
 	u32 pval;
 
 	constraints->name = of_get_property(np, "regulator-name", NULL);
@@ -70,6 +78,73 @@ static void of_get_regulation_constraints(struct device_node *np,
 	ret = of_property_read_u32(np, "regulator-enable-ramp-delay", &pval);
 	if (!ret)
 		constraints->enable_time = pval;
+
+	ret = of_property_read_u32(np, "regulator-initial-state", &pval);
+	if (!ret) {
+		switch (pval) {
+		case PM_SUSPEND_STANDBY:
+		case PM_SUSPEND_MEM:
+		case PM_SUSPEND_MAX:
+			constraints->initial_state = pval;
+			break;
+		default:
+			break;
+		};
+	}
+
+	for (i = 0; i < ARRAY_SIZE(regulator_states); i++) {
+		switch (i) {
+		case PM_SUSPEND_STANDBY:
+			suspend_state = &constraints->state_standby;
+			break;
+		case PM_SUSPEND_MEM:
+			suspend_state = &constraints->state_mem;
+			break;
+		case PM_SUSPEND_MAX:
+			suspend_state = &constraints->state_disk;
+			break;
+		case PM_SUSPEND_ON:
+		case PM_SUSPEND_FREEZE:
+		default:
+			continue;
+		};
+
+		suspend_np = of_get_child_by_name(np, regulator_states[i]);
+		if (!suspend_np || !suspend_state)
+			continue;
+
+		suspend_uV = of_get_property(suspend_np, "regulator-volt",
+						NULL);
+		if (suspend_uV) {
+			suspend_state->uV = be32_to_cpu(*suspend_uV);
+
+			if (suspend_state->uV < constraints->min_uV)
+				suspend_state->uV = constraints->min_uV;
+			if (suspend_state->uV > constraints->max_uV)
+				suspend_state->uV = constraints->max_uV;
+		}
+
+		ret = of_property_read_u32(suspend_np, "regulator-mode", &pval);
+		if (!ret) {
+			u32 regulator_mode = REGULATOR_MODE_FAST
+					| REGULATOR_MODE_NORMAL
+					| REGULATOR_MODE_IDLE
+					| REGULATOR_MODE_STANDBY;
+			if (pval <= regulator_mode)
+				suspend_state->mode = pval;
+		}
+
+		if (of_property_read_bool(suspend_np,
+					"regulator-on-in-suspend"))
+			suspend_state->enabled = true;
+
+		if (of_property_read_bool(suspend_np,
+					"regulator-off-in-suspend"))
+			suspend_state->disabled = true;
+
+		suspend_state = NULL;
+		suspend_np = NULL;
+	}
 }
 
 /**
-- 
1.8.0

[PATCHv3 2/2] dt-bindings: regulator: Add regulator suspend state for PM state

[Chanwoo Choi]

This patch add regulator suspend state to constraint in dt file. The regulation_
constraints structure already has regulator suspend state field as following.
The regulator suspend state control the state of regulator according to
PM (Power Management) state.
- struct regulator_state state_disk
- struct regulator_state state_mem
- struct regulator_state state_standby

Signed-off-by: Chanwoo Choi <cw00.choi@samsung.com>
Acked-by: Kyungmin Park <kyungmin.park@samsung.com>
---
 .../devicetree/bindings/regulator/regulator.txt    | 32 ++++++++++++++++++++++
 1 file changed, 32 insertions(+)

diff --git a/Documentation/devicetree/bindings/regulator/regulator.txt b/Documentation/devicetree/bindings/regulator/regulator.txt
index 8607433..c4f566a 100644
--- a/Documentation/devicetree/bindings/regulator/regulator.txt
+++ b/Documentation/devicetree/bindings/regulator/regulator.txt
@@ -19,6 +19,33 @@ Optional properties:
   design requires. This property describes the total system ramp time
   required due to the combination of internal ramping of the regulator itself,
   and board design issues such as trace capacitance and load on the supply.
+- regulator-initial-state: initial state for suspend state, cnd set initial
+  state among following defined suspend states:
+  <2>: PM_SUSPEND_STANDBY - Setup regulator according to regulator-state-standby
+  <3>: PM_SUSPEND_MEM - Setup regulator according to regulator-state-mem
+  <4>: PM_SUSPEND_MAX - Setup regulator according to regulator-state-disk
+- regulator-state-standby sub-root node for Standby mode
+  : the device is in a power-saving state, but can also receive certain events,
+  specific behavior depends on the specific device.
+- regulator-state-mem sub-root node for Suspend-to-RAM mode
+  : suspend to memory, the device goes to sleep, but all data stored in memory,
+  only some external interrupt can wake the device.
+- regulator-state-disk sub-root node for Suspend-to-disk mode
+  : suspend to disk, this state operates similarly to Suspend-to-RAM,
+  but includes a final step of writing memory contents to disk.
+- regulator-state-[standby/mem/disk] node has following common properties:
+	- regulator-volt: voltage consumers may set in suspend state.
+	- regulator-mode: voltage mode in suspend state, can set mode among
+	following defined regulator modes:
+	0x1: REGULATOR_MODE_FAST, Regulator can handle fast changes.
+	0x2: REGULATOR_MODE_NORMAL, Normal regulator power supply mode.
+	0x4: REGULATOR_MODE_IDLE, Regulator runs in a more efficient mode.
+	0x8: REGULATOR_MODE_STANDBY, Regulator runs in the most efficient mode.
+	- regulator-on-in-suspend: regulator should be on in suspend state.
+	- regulator-off-in-suspend: regulator should be off in suspend state.
+	If node don't include regulator-[on/off]-in-suspend, can't change
+	regulator state in suspend mode and only should sustain the regulator
+	state of normal state.
 
 Deprecated properties:
 - regulator-compatible: If a regulator chip contains multiple
@@ -34,6 +61,11 @@ Example:
 		regulator-max-microvolt = <2500000>;
 		regulator-always-on;
 		vin-supply = <&vin>;
+
+		regulator-state-mem {
+			regulator-volt = <1000000>;
+			regulator-off-in-suspend;
+		};
 	};
 
 Regulator Consumers:
-- 
1.8.0

Re: [PATCHv3 2/2] dt-bindings: regulator: Add regulator suspend state for PM state

[Mark Brown]

[-- Attachment #1: Type: text/plain, Size: 2119 bytes --]

On Thu, Aug 14, 2014 at 09:40:14AM +0900, Chanwoo Choi wrote:

> +- regulator-initial-state: initial state for suspend state, cnd set initial
> +  state among following defined suspend states:
> +  <2>: PM_SUSPEND_STANDBY - Setup regulator according to regulator-state-standby
> +  <3>: PM_SUSPEND_MEM - Setup regulator according to regulator-state-mem
> +  <4>: PM_SUSPEND_MAX - Setup regulator according to regulator-state-disk
> +- regulator-state-standby sub-root node for Standby mode
> +  : the device is in a power-saving state, but can also receive certain events,
> +  specific behavior depends on the specific device.

These are all Linux internal descriptions of the states but the device
tree is supposed to be OS neutral.  For suspend to memory and suspend to
disk that's probably adequately clear but _STANDBY is really unclear.
I would suggest just dropping this without a clearer defintion, it's
something that I'd expect to emerge organically from low power modes
rather than having a specific definition anyway.

> +- regulator-state-[standby/mem/disk] node has following common properties:
> +	- regulator-volt: voltage consumers may set in suspend state.
> +	- regulator-mode: voltage mode in suspend state, can set mode among
> +	following defined regulator modes:
> +	0x1: REGULATOR_MODE_FAST, Regulator can handle fast changes.
> +	0x2: REGULATOR_MODE_NORMAL, Normal regulator power supply mode.
> +	0x4: REGULATOR_MODE_IDLE, Regulator runs in a more efficient mode.
> +	0x8: REGULATOR_MODE_STANDBY, Regulator runs in the most efficient mode.
> +	- regulator-on-in-suspend: regulator should be on in suspend state.
> +	- regulator-off-in-suspend: regulator should be off in suspend state.
> +	If node don't include regulator-[on/off]-in-suspend, can't change
> +	regulator state in suspend mode and only should sustain the regulator
> +	state of normal state.

Modes are a similarly problematic thing - their definition is really
unclear even within Linux and we don't support them at all at present.
I'd just drop them initially and then add them in as a part of adding
mode support in general.

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 819 bytes --]

Re: [PATCHv3 2/2] dt-bindings: regulator: Add regulator suspend state for PM state

[Chanwoo Choi]

Dear Mark,

On Fri, Aug 15, 2014 at 7:56 PM, Mark Brown <broonie@kernel.org> wrote:
> On Thu, Aug 14, 2014 at 09:40:14AM +0900, Chanwoo Choi wrote:
>
>> +- regulator-initial-state: initial state for suspend state, cnd set initial
>> +  state among following defined suspend states:
>> +  <2>: PM_SUSPEND_STANDBY - Setup regulator according to regulator-state-standby
>> +  <3>: PM_SUSPEND_MEM - Setup regulator according to regulator-state-mem
>> +  <4>: PM_SUSPEND_MAX - Setup regulator according to regulator-state-disk
>> +- regulator-state-standby sub-root node for Standby mode
>> +  : the device is in a power-saving state, but can also receive certain events,
>> +  specific behavior depends on the specific device.
>
> These are all Linux internal descriptions of the states but the device
> tree is supposed to be OS neutral.  For suspend to memory and suspend to
> disk that's probably adequately clear but _STANDBY is really unclear.
> I would suggest just dropping this without a clearer defintion, it's
> something that I'd expect to emerge organically from low power modes
> rather than having a specific definition anyway.

OK, I'll drop PM_SUSPEND_STANDBY mode on next patchset.

>
>> +- regulator-state-[standby/mem/disk] node has following common properties:
>> +     - regulator-volt: voltage consumers may set in suspend state.
>> +     - regulator-mode: voltage mode in suspend state, can set mode among
>> +     following defined regulator modes:
>> +     0x1: REGULATOR_MODE_FAST, Regulator can handle fast changes.
>> +     0x2: REGULATOR_MODE_NORMAL, Normal regulator power supply mode.
>> +     0x4: REGULATOR_MODE_IDLE, Regulator runs in a more efficient mode.
>> +     0x8: REGULATOR_MODE_STANDBY, Regulator runs in the most efficient mode.
>> +     - regulator-on-in-suspend: regulator should be on in suspend state.
>> +     - regulator-off-in-suspend: regulator should be off in suspend state.
>> +     If node don't include regulator-[on/off]-in-suspend, can't change
>> +     regulator state in suspend mode and only should sustain the regulator
>> +     state of normal state.
>
> Modes are a similarly problematic thing - their definition is really
> unclear even within Linux and we don't support them at all at present.
> I'd just drop them initially and then add them in as a part of adding
> mode support in general.

OK, I'll drop 'regulator-mode' property on next patchset.

Thanks for your review.

Best Regards,
Chanwoo Choi