linux-kernel - Re: [PATCH] hwmon: (ina2xx) make regulator 'vs' support optional

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: <661f972b-cd0f-4d9c-b429-5990edf8f13f@roeck-us.net>
Date: Tue, 8 Apr 2025 11:07:48 -0700
From: Guenter Roeck <linux@...ck-us.net>
To: Ciprian Marian Costea <ciprianmarian.costea@....nxp.com>
Cc: Jean Delvare <jdelvare@...e.com>, linux-hwmon@...r.kernel.org,
 linux-kernel@...r.kernel.org, NXP S32 Linux <s32@....com>,
 imx@...ts.linux.dev, Christophe Lizzi <clizzi@...hat.com>,
 Alberto Ruiz <aruizrui@...hat.com>, Enric Balletbo <eballetb@...hat.com>,
 Eric Chanudet <echanude@...hat.com>, Florin Buica <florin.buica@....com>
Subject: Re: [PATCH] hwmon: (ina2xx) make regulator 'vs' support optional

On 4/4/25 01:36, Ciprian Marian Costea wrote:
> On 4/3/2025 7:06 PM, Guenter Roeck wrote:
>> On Thu, Apr 03, 2025 at 05:29:26PM +0300, Ciprian Marian Costea wrote:
>>> On 4/3/2025 3:15 PM, Guenter Roeck wrote:
>>>> On 4/3/25 03:15, Ciprian Costea wrote:
>>>>> From: Ciprian Marian Costea <ciprianmarian.costea@....nxp.com>
>>>>>
>>>>> S32G2/S32G3 based boards which integrate the ina231 sensor do not have a
>>>>> dedicated voltage regulator.
>>>>>
>>>>> Co-developed-by: Florin Buica <florin.buica@....com>
>>>>> Signed-off-by: Florin Buica <florin.buica@....com>
>>>>> Signed-off-by: Ciprian Marian Costea <ciprianmarian.costea@....nxp.com>
>>>>> ---
>>>>>    drivers/hwmon/ina2xx.c | 4 ++--
>>>>>    1 file changed, 2 insertions(+), 2 deletions(-)
>>>>>
>>>>> diff --git a/drivers/hwmon/ina2xx.c b/drivers/hwmon/ina2xx.c
>>>>> index 345fe7db9de9..ab4972f94a8c 100644
>>>>> --- a/drivers/hwmon/ina2xx.c
>>>>> +++ b/drivers/hwmon/ina2xx.c
>>>>> @@ -959,8 +959,8 @@ static int ina2xx_probe(struct i2c_client *client)
>>>>>            return PTR_ERR(data->regmap);
>>>>>        }
>>>>> -    ret = (dev, "vs");
>>>>> -    if (ret)
>>>>> +    ret = devm_regulator_get_enable_optional(dev, "vs");
>>>>
>>>> devm_regulator_get_enable() should provide a dummy regulator if there is
>>>> no explicit regulator. Why does this not work ?
>>>>
>>>>> +    if (ret < 0 && ret != -ENODEV)
>>>>
>>>> Why this added check ?
>>>>
>>>> I know it used to be necessary if regulator support is disabled,
>>>> but that is no longer the case.
>>>>
>>>> Guenter
>>>>
>>>
>>> Hello Guenter,
>>>
>>> I've just tested and devm_regulator_get_enable() does work as you've
>>> described, providing a dummy regulator.
>>>
>>> But, according to the 'ti,ina2xx' binding [1] I see that the `vs-supply`
>>> property is not required. Hence wouldn't it be correct for `vs-supply` to be
>>> optional ? Using 'devm_regulator_get_enable_optional()'
>>>
>> Yes, but the reasoning you provided is different and suggested that the
>> current code would not work. Since that is not the case, the change would
>> be purely cosmetic. Also, I still don't see why the -ENODEV check would be
>> necessary.
>>
>> Guenter
> 
> For boards such as S32G274A-EVB, S32G274A-RDB2 and S32G399A-RDB3 which do not have a voltage regulator, 'devm_regulator_get_enable_optional()' would return error value -19 (-ENODEV). Also, other usages from the Linux Kernel seem to perform the same error check when using 'devm_regulator_get_enable_optional()' [1], [2] and [3].
> 
> This patch would help in S32G2 and S32G3 to not print an unnecessary kernel log warning hinting usage of a dummy regulator when such a regulator is not required according to the binding.
> 
> Would you like me to send a V2 with the commit title updated as follows ?
> 
> "
> hwmon: (ina2xx) make regulator 'vs' support optional
> 
> According to the 'ti,ina2xx' binding, the 'vs-supply' property is optional. Furthermore, S32G2/S32G3 based boards which integrate the ina231 sensor do not have a dedicated voltage regulator. Thus, making regulator support optional would help in avoiding any unnecessary kernel log warnings during boot.
> "

Make it:

"According to the 'ti,ina2xx' binding, the 'vs-supply' property is optional.
  Use devm_regulator_get_enable_optional() to avoid a kernel warning message
  if the property is not provided.
"

Then add a note to the code explaining that the check for -ENODEV is necessary
because the regulator core returns -ENODEV if the regulator is not available.

Why it makes sense for this function to return -ENODEV if an _optional_ regulator
is not available escapes me, but that is a different issue.

Guenter