linux-kernel - Re: [PATCH v5 5/5] hwmon: Add support for Amphenol ChipCap 2

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: <30b9ab0c-f3cb-4b5a-a726-de9f7c61769b@gmail.com>
Date: Thu, 18 Jan 2024 16:30:37 +0100
From: Javier Carrasco <javier.carrasco.cruz@...il.com>
To: Mark Brown <broonie@...nel.org>
Cc: Rob Herring <robh+dt@...nel.org>,
 Krzysztof Kozlowski <krzysztof.kozlowski+dt@...aro.org>,
 Conor Dooley <conor+dt@...nel.org>, Jean Delvare <jdelvare@...e.com>,
 Guenter Roeck <linux@...ck-us.net>, Jonathan Corbet <corbet@....net>,
 Liam Girdwood <lgirdwood@...il.com>, Rob Herring <robh@...nel.org>,
 devicetree@...r.kernel.org, linux-kernel@...r.kernel.org,
 linux-hwmon@...r.kernel.org, linux-doc@...r.kernel.org
Subject: Re: [PATCH v5 5/5] hwmon: Add support for Amphenol ChipCap 2

Hello Mark,

On 18.01.24 14:49, Mark Brown wrote:
> On Mon, Jan 15, 2024 at 09:02:25PM +0100, Javier Carrasco wrote:
> 
>> +static int cc2_enable(struct cc2_data *data)
>> +{
>> +	int ret;
>> +
>> +	if (regulator_is_enabled(data->regulator))
>> +		return 0;
> 
> This is generally a sign that the regulator API usage is not good, the
> driver should not rely on references to the regulator held by anything
> else since whatever else is holding the regulator on could turn it off
> at any time.  If the driver did the enable itself then it should know
> that it did so and not need to query.
> 

The driver handles a dedicated regulator, but I wanted to account for
the cases where the attempts to enable and disable the regulator fail
and keep parity. If the disabling attempt fails, will the regulator not
stay enabled? In that case, an additional call to regulator_enable would
not be required, right?
That is the only reason I am using regulator_is_enabled(), but maybe
things don't work like that.

>> +	ret = regulator_enable(data->regulator);
>> +	if (ret < 0)
>> +		return ret;
>> +
>> +	/*
>> +	 * TODO: the startup-delay-us property of the regulator might be
>> +	 * added to the delay (if provided).
>> +	 * Currently there is no interface to read its value apart from
>> +	 * a direct access to regulator->rdev->constraints->enable_time,
>> +	 * which is discouraged like any direct access to the regulator_dev
>> +	 * structure. This would be relevant in cases where the startup delay
>> +	 * is in the range of milliseconds.
>> +	 */
>> +	usleep_range(CC2_STARTUP_TIME_US, CC2_STARTUP_TIME_US + 125);
> 
> Note that the regulator startup delay is the time taken for the
> regulator to power up so if the device needs additional delay then that
> will always need to be in addition to whatever the regulator is doing.

What I mean by that is that the device cannot be ready until the
regulator powers it up (obvious) plus the start up time of the device
itself once it gets powered up. So if a regulator takes for example 1 ms
to power up, the sleep function could (and should) wait for 1 ms longer.

I could define a longer start up time to account for "slow" regulators
while still staying in the command window range. Retries have already
been taken into account for longer sleeps.

Thank you for your feedback.

Best regards,
Javier Carrasco