mbox series

[V5,0/4] hwmon: ina3221: Add selective summation support

Message ID 20230929103650.86074-1-jonathanh@nvidia.com (mailing list archive)
Headers show
Series hwmon: ina3221: Add selective summation support | expand

Message

Jon Hunter Sept. 29, 2023, 10:36 a.m. UTC
The current INA3221 driver always sums the shunt voltage for all enabled
channels regardless of the shunt-resistor used for each channel. Summing
the shunt-voltage for channels is only meaningful if the shunt resistor
is the same for each channel. This series adds device-tree support to
allow which channels are summed in device-tree.

Changes since V4:
- Moved dt-binding comment added in V4 from patch #2 to patch #1.

Changes since V3:
- Added missing descriptions for new structure members that was reported
  by the kernel-test-bot.
- Added comment in the ina3221 dt-binding doc example to explain why we
  need to explicitly disable channels.
- Added more commentary in the commit message for the new DT property
  to explain that this property does not change the behaviour of the
  driver unless it is populated.

Changes since V2:
- Added note to binding-doc to indicate that input channels must be
  explicitly disabled.
- Corrected ordering of properties in the binding-doc
- Updated license for the binding-doc to be dual licensed.
- Changed newly added property from 'summation-bypass' to
  summation-disable'.
- Documented type for the new 'summation-disable' property.
- Corrected spelling and comments as per the feedback received.
- Used debugfs instead of sysfs for exposing the 'summation-disable'
  status for each input channel.
- Populated missing instances for the ina3221 device for Tegra234
  boards.
- Populated ina219 device for the NVIDIA IGX board (not strictly
  related to this series but related to populating all
  power-sensors for Tegra234 boards)

Changes since V1:
- Added yaml conversion patch for binding-doc
- Added binding-doc documentation patch for new property
- Added patch to populate ina3221 devices for Tegra234.

Jon Hunter (2):
  dt-bindings: hwmon: ina3221: Add ti,summation-disable
  arm64: tegra: Add power-sensors for Tegra234 boards

Ninad Malwade (2):
  dt-bindings: hwmon: ina3221: Convert to json-schema
  hwmon: ina3221: Add support for channel summation disable

 .../devicetree/bindings/hwmon/ina3221.txt     |  54 --------
 .../devicetree/bindings/hwmon/ti,ina3221.yaml | 121 ++++++++++++++++++
 .../boot/dts/nvidia/tegra234-p3701-0008.dtsi  |  33 +++++
 .../arm64/boot/dts/nvidia/tegra234-p3701.dtsi |  53 ++++++++
 .../arm64/boot/dts/nvidia/tegra234-p3767.dtsi |  29 +++++
 drivers/hwmon/ina3221.c                       |  33 ++++-
 6 files changed, 266 insertions(+), 57 deletions(-)
 delete mode 100644 Documentation/devicetree/bindings/hwmon/ina3221.txt
 create mode 100644 Documentation/devicetree/bindings/hwmon/ti,ina3221.yaml

Comments

Thierry Reding Oct. 11, 2023, 8:28 p.m. UTC | #1
On Fri, Sep 29, 2023 at 11:36:46AM +0100, Jon Hunter wrote:
> The current INA3221 driver always sums the shunt voltage for all enabled
> channels regardless of the shunt-resistor used for each channel. Summing
> the shunt-voltage for channels is only meaningful if the shunt resistor
> is the same for each channel. This series adds device-tree support to
> allow which channels are summed in device-tree.
> 
> Changes since V4:
> - Moved dt-binding comment added in V4 from patch #2 to patch #1.
> 
> Changes since V3:
> - Added missing descriptions for new structure members that was reported
>   by the kernel-test-bot.
> - Added comment in the ina3221 dt-binding doc example to explain why we
>   need to explicitly disable channels.
> - Added more commentary in the commit message for the new DT property
>   to explain that this property does not change the behaviour of the
>   driver unless it is populated.
> 
> Changes since V2:
> - Added note to binding-doc to indicate that input channels must be
>   explicitly disabled.
> - Corrected ordering of properties in the binding-doc
> - Updated license for the binding-doc to be dual licensed.
> - Changed newly added property from 'summation-bypass' to
>   summation-disable'.
> - Documented type for the new 'summation-disable' property.
> - Corrected spelling and comments as per the feedback received.
> - Used debugfs instead of sysfs for exposing the 'summation-disable'
>   status for each input channel.
> - Populated missing instances for the ina3221 device for Tegra234
>   boards.
> - Populated ina219 device for the NVIDIA IGX board (not strictly
>   related to this series but related to populating all
>   power-sensors for Tegra234 boards)
> 
> Changes since V1:
> - Added yaml conversion patch for binding-doc
> - Added binding-doc documentation patch for new property
> - Added patch to populate ina3221 devices for Tegra234.
> 
> Jon Hunter (2):
>   dt-bindings: hwmon: ina3221: Add ti,summation-disable
>   arm64: tegra: Add power-sensors for Tegra234 boards
> 
> Ninad Malwade (2):
>   dt-bindings: hwmon: ina3221: Convert to json-schema
>   hwmon: ina3221: Add support for channel summation disable

Jean, Guenter,

do you mind if I pick up patches 1, 2 and 4 into the Tegra tree? It's
usually convenient to keep the DT bindings and DT additions in the same
tree for validation.

Thanks,
Thierry
Thierry Reding Oct. 13, 2023, 12:36 p.m. UTC | #2
From: Thierry Reding <treding@nvidia.com>


On Fri, 29 Sep 2023 11:36:46 +0100, Jon Hunter wrote:
> The current INA3221 driver always sums the shunt voltage for all enabled
> channels regardless of the shunt-resistor used for each channel. Summing
> the shunt-voltage for channels is only meaningful if the shunt resistor
> is the same for each channel. This series adds device-tree support to
> allow which channels are summed in device-tree.
> 
> Changes since V4:
> - Moved dt-binding comment added in V4 from patch #2 to patch #1.
> 
> [...]

Applied, thanks!

[4/4] arm64: tegra: Add power-sensors for Tegra234 boards
      commit: 9152ed09309de1a876680e6309c8eccb509b44b0

Best regards,
Jon Hunter Oct. 25, 2023, 10:54 a.m. UTC | #3
Hi Jean, Guenter,

On 11/10/2023 21:28, Thierry Reding wrote:
> On Fri, Sep 29, 2023 at 11:36:46AM +0100, Jon Hunter wrote:
>> The current INA3221 driver always sums the shunt voltage for all enabled
>> channels regardless of the shunt-resistor used for each channel. Summing
>> the shunt-voltage for channels is only meaningful if the shunt resistor
>> is the same for each channel. This series adds device-tree support to
>> allow which channels are summed in device-tree.
>>
>> Changes since V4:
>> - Moved dt-binding comment added in V4 from patch #2 to patch #1.
>>
>> Changes since V3:
>> - Added missing descriptions for new structure members that was reported
>>    by the kernel-test-bot.
>> - Added comment in the ina3221 dt-binding doc example to explain why we
>>    need to explicitly disable channels.
>> - Added more commentary in the commit message for the new DT property
>>    to explain that this property does not change the behaviour of the
>>    driver unless it is populated.
>>
>> Changes since V2:
>> - Added note to binding-doc to indicate that input channels must be
>>    explicitly disabled.
>> - Corrected ordering of properties in the binding-doc
>> - Updated license for the binding-doc to be dual licensed.
>> - Changed newly added property from 'summation-bypass' to
>>    summation-disable'.
>> - Documented type for the new 'summation-disable' property.
>> - Corrected spelling and comments as per the feedback received.
>> - Used debugfs instead of sysfs for exposing the 'summation-disable'
>>    status for each input channel.
>> - Populated missing instances for the ina3221 device for Tegra234
>>    boards.
>> - Populated ina219 device for the NVIDIA IGX board (not strictly
>>    related to this series but related to populating all
>>    power-sensors for Tegra234 boards)
>>
>> Changes since V1:
>> - Added yaml conversion patch for binding-doc
>> - Added binding-doc documentation patch for new property
>> - Added patch to populate ina3221 devices for Tegra234.
>>
>> Jon Hunter (2):
>>    dt-bindings: hwmon: ina3221: Add ti,summation-disable
>>    arm64: tegra: Add power-sensors for Tegra234 boards
>>
>> Ninad Malwade (2):
>>    dt-bindings: hwmon: ina3221: Convert to json-schema
>>    hwmon: ina3221: Add support for channel summation disable
> 
> Jean, Guenter,
> 
> do you mind if I pick up patches 1, 2 and 4 into the Tegra tree? It's
> usually convenient to keep the DT bindings and DT additions in the same
> tree for validation.


I have not seen any feedback on this from you? Please let me know if 
this version is now OK with you?

Thanks
Jon