linux-kernel - Re: [PATCH v3 2/2] iio: adc: Add TI ADS1100 and ADS1000

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]

Date:   Sat, 4 Mar 2023 17:19:50 +0000
From:   Jonathan Cameron <jic23@...nel.org>
To:     Lars-Peter Clausen <lars@...afoo.de>
Cc:     Mike Looijmans <mike.looijmans@...ic.nl>,
        Andy Shevchenko <andriy.shevchenko@...ux.intel.com>,
        devicetree@...r.kernel.org, linux-iio@...r.kernel.org,
        Caleb Connolly <caleb.connolly@...aro.org>,
        ChiYuan Huang <cy_huang@...htek.com>,
        ChiaEn Wu <chiaen_wu@...htek.com>,
        Cosmin Tanislav <demonsingur@...il.com>,
        Ibrahim Tilki <Ibrahim.Tilki@...log.com>,
        Ramona Bolboaca <ramona.bolboaca@...log.com>,
        William Breathitt Gray <william.gray@...aro.org>,
        linux-kernel@...r.kernel.org
Subject: Re: [PATCH v3 2/2] iio: adc: Add TI ADS1100 and ADS1000

On Thu, 2 Mar 2023 05:20:38 -0800
Lars-Peter Clausen <lars@...afoo.de> wrote:

> On 3/2/23 05:16, Lars-Peter Clausen wrote:
> > On 3/1/23 23:49, Mike Looijmans wrote:  
> >>
> >>  
> >>> ...
> >>> ...
> >>>  
> >>>> +static int ads1100_runtime_suspend(struct device *dev)
> >>>> +{
> >>>> +    struct iio_dev *indio_dev = 
> >>>> i2c_get_clientdata(to_i2c_client(dev));
> >>>> +    struct ads1100_data *data = iio_priv(indio_dev);
> >>>> +
> >>>> +    ads1100_set_config_bits(data, ADS1100_CFG_SC, 
> >>>> ADS1100_SINGLESHOT);
> >>>> +    regulator_disable(data->reg_vdd);  
> >>> Wrong devm / non-devm ordering.  
> >>
> >> Don't understand your remark, can you explain further please?
> >>
> >> devm / non-devm ordering would be related to the "probe" function. As 
> >> far as I can tell, I'm not allocating resources after the devm calls. 
> >> And the "remove" is empty.  
> >
> > Strictly speaking we need to unregister the IIO device before 
> > disabling the regulator, otherwise there is a small window where the 
> > IIO device still exists, but doesn't work anymore. This is a very 
> > theoretical scenario though.
> >
> > You are lucky :) There is a new function 
> > `devm_regulator_get_enable()`[1], which will manage the 
> > regulator_disable() for you. Using that will also reduce the 
> > boilerplate in `probe()` a bit
> >
> > - Lars
> >
> > [1] https://lwn.net/Articles/904383/
> >  
> Sorry, just saw that Andy's comment was on the suspend() function, not 
> remove(). In that case there is of course no need for any devm things. 
> But still a good idea to use `devm_regulator_get_enable()` in probe for 
> the boiler plate.
> 
You can't because (annoyingly) devem_regulator_get_enable() doesn't
provide you access to the struct regulator that you need to be able
to turn it of for power management.
That case only works for the leave the power on all the time cases.

Jonathan