linux-kernel - Re: [PATCH] hwmon: ina238: Add support for shunt voltage scaling

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <db04ca68-dea3-4c1d-9378-cb0316759d16@roeck-us.net>
Date: Mon, 20 Jan 2025 17:45:50 -0800
From: Guenter Roeck <linux@...ck-us.net>
To: Potin Lai <potin.lai.pt@...il.com>
Cc: Jean Delvare <jdelvare@...e.com>, Patrick Williams <patrick@...cx.xyz>,
 linux-hwmon@...r.kernel.org, linux-kernel@...r.kernel.org,
 Cosmo Chou <cosmo.chou@...ntatw.com>, Potin Lai <potin.lai@...ntatw.com>
Subject: Re: [PATCH] hwmon: ina238: Add support for shunt voltage scaling

On 1/20/25 17:18, Potin Lai wrote:
> On Tue, Jan 21, 2025 at 12:38 AM Guenter Roeck <linux@...ck-us.net> wrote:
>>
>> On 1/20/25 08:23, Potin Lai wrote:
>>> The INA238 sensor reports shunt voltage with microvolt precision.
>>> However, the hwmon driver currently exposes this value only in
>>> millivolts via `in0_input`, which results in a loss of precision for
>>> readings within the range of ±1 mV.
>>>
>>> This patch introduces an `in0_scale` attribute to provide the scaling
>>> factor applied to the shunt voltage reading. By exposing this attribute,
>>> users can accurately interpret the in0_input values in microvolts,
>>> preserving the sensor's full precision.
>>>
>>> Signed-off-by: Potin Lai <potin.lai.pt@...il.com>
>>
>> Sorry, this is an ABI violation and thus a no-go. in0_input is and will
>> always be reported in mV.
>>
>> Guenter
>>
> 
> Hi Guenter,
> In our use case, the shunt voltage is less than 1 mv most of the time.
> I would like to get your advice on the correct way of getting
> microvolt reading from the driver? Thank you.
> 

Given that V = I * R, assuming you know the shunt resistor value,
it is an easy calculation which does not require any code changes.

Possible alternatives:
- Create a debugfs file to display the voltage in uV.
   See drivers/hwmon/isl28022.c for an example.
- Implement a hwmon -> iio bridge and use IIO to report
   the voltage.

Having said that, do you have an actual use case ? What is the point
of knowing the voltage across the shunt resistor ? "Because the chip
reports it" is not a use case. Literally _every_ current sensor chip
ultimately reports the shunt resistor voltage, after all (typically
as current; calculation see above). Commit 63fb21afc1f5 ("hwmon:
(ina2xx) Use shunt voltage to calculate current") is a case in point.

Guenter