On 06/11, Rob Herring wrote:

On Thu, Jun 11, 2015 at 7:37 PM, Stephen Boyd [off-list ref] wrote:

quoted

Add support for over current protection (OCP), pin control
selection, soft start and soft start strength, auto-mode, input
current limiting, and pull down.

Cc: <redacted>
Signed-off-by: Stephen Boyd <redacted>
---

Changes from v1:
* New patch split from original SPMI regulator driver

 .../bindings/regulator/qcom,spmi-regulator.txt     |  62 +++++
 drivers/regulator/qcom_spmi-regulator.c            | 298 ++++++++++++++++++++-
 2 files changed, 358 insertions(+), 2 deletions(-)

diff --git a/Documentation/devicetree/bindings/regulator/qcom,spmi-regulator.txt b/Documentation/devicetree/bindings/regulator/qcom,spmi-regulator.txt
index 75b4604bad07..ab01a152e930 100644
--- a/Documentation/devicetree/bindings/regulator/qcom,spmi-regulator.txt
+++ b/Documentation/devicetree/bindings/regulator/qcom,spmi-regulator.txt

@@ -99,6 +99,68 @@ see regulator.txt - with additional custom properties described below:
                    soft start are active all the time. 0 = Set initial mode to
                    low power mode (LPM).

+- qcom,auto-mode-enable:

Auto regulator modes are fairly common. Can't we have a common property here?

quoted

+       Usage: optional
+       Value type: <u32>

This can be bool. Unless we want to have a mode property with "auto"
being one possible value.

From what I can tell we use this property as qcom,auto-mode-enable = <0>.

Put another way, that's qcom,auto-mode-disable. Should we have two
properties? Or is it better to mark all regulators explicitly with
either auto-mode-enable or the absence of such a property? The bindings
take an approach of giving all these "boolean" properties a 0 or 1 value
so that the absence of the property indicates that we should use the
default value in the hardware.

quoted

+       Description: 1 = Enable automatic hardware selection of regulator
+                        mode (HPM vs LPM); not available on boost type
+                        regulators. 0 = Disable auto mode selection.
+
+- qcom,ocp-enable:
+       Usage: optional
+       Value type: <u32>
+       Description: 1 = Allow over current protection (OCP) to be enabled for
+                    voltage switch type regulators so that they latch off
+                    automatically when over current is detected. OCP is
+                    enabled when in HPM or auto mode.  0 = Disable OCP.

This seems common too.

Ok. I'll make up another patch for this one. Perhaps call it
regulator-over-current-protection?

quoted

+
+- qcom,ocp-max-retries:
+       Usage: optional
+       Value type: <u32>
+       Description: Maximum number of times to try toggling a voltage switch
+                    off and back on as a result of consecutive over current
+                    events.
+
+- qcom,ocp-retry-delay:
+       Usage: optional
+       Value type: <u32>
+       Description: Time to delay in milliseconds between each voltage switch
+                    toggle after an over current event takes place.

These 2 are perhaps less common. I hope I don't have a device that does this. :)

:)

-- 
Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum,
a Linux Foundation Collaborative Project