Re: [PATCH v6 1/3] iio: adc: rockchip_saradc: move all of probe to devm-functions

From: Jonathan Cameron <jic23@kernel.org>
To: Heiko Stuebner <heiko@sntech.de>
Cc: knaack.h@gmx.de, lars@metafoo.de, pmeerw@pmeerw.net,
	linux-iio@vger.kernel.org, linux-kernel@vger.kernel.org,
	linux-rockchip@lists.infradead.org, xxm@rock-chips.com
Subject: Re: [PATCH v6 1/3] iio: adc: rockchip_saradc: move all of probe to devm-functions
Date: Sat, 4 Jul 2020 16:38:17 +0100	[thread overview]
Message-ID: <20200704163817.646ee01a@archlinux> (raw)
In-Reply-To: <3183602.Ou33jPcMzM@phil>

On Fri, 03 Jul 2020 09:48:42 +0200
Heiko Stuebner <heiko@sntech.de> wrote:

> Am Samstag, 27. Juni 2020, 14:03:29 CEST schrieb Jonathan Cameron:
> > On Wed, 24 Jun 2020 01:30:09 +0200
> > Heiko Stuebner <heiko@sntech.de> wrote:
> >   
> > > From: Heiko Stuebner <heiko.stuebner@theobroma-systems.com>
> > > 
> > > Parts of the saradc probe rely on devm functions and later parts do not.
> > > This makes it more difficult to for example enable triggers via their
> > > devm-functions and would need more undo-work in remove.
> > > 
> > > So to make life easier for the driver, move the rest of probe calls
> > > also to their devm-equivalents.
> > > 
> > > This includes moving the clk- and regulator-disabling to a devm_action
> > > so that they gets disabled both during remove and in the error case
> > > in probe, after the action is registered.
> > > 
> > > Signed-off-by: Heiko Stuebner <heiko.stuebner@theobroma-systems.com>  
> > Looks good to me. Applied to the togreg branch of iio.git with one small
> > tweak. See inline.
> > 
> > One other thing whilst we are here. Why do we need the build dependence
> > on ARM?  
> 
> I guess originally it was there simply, because Rockchip only makes
> ARM-SoCs so as not to pollute the config with too many unusable
> options and it had the COMPILE_TEST for the test-builds.

I'd have no problem with
depends on ARCH_ROCKCHIP || COMPILE_TEST
which would mean the unusable option only applied to those of us
deliberately doing test builds rather than anyone configuring a
real platform (where COMPILE_TEST makes no sense).

> 
> But if one driver more doesn't really matter, it can of course also just
> go away.
> 
> > I just scrapped it and the driver builds fine on x86 so would
> > be good to get the additional build coverage if we can.  
> 
> So I guess you'd like a patch removing that part, right?
> Because I guess the "I just scrapped it" meant locally, as I can't find
> a commit of that sort in the testing branch ;-)

Indeed I did this locally.

Patch would be idea, or I can do it if you prefer.

Thanks,

Jonathan

> 
> 
> Heiko
> 
> 
> > > ---
> > > changes in v6:
> > > - move devm actions into separate functions as suggested by Jonathan
> > > changes in v5:
> > > - none
> > > changes in v4:
> > > - new patch as suggested by Jonathan
> > > 
> > >  drivers/iio/adc/rockchip_saradc.c | 72 ++++++++++++++++++++-----------
> > >  1 file changed, 46 insertions(+), 26 deletions(-)
> > > 
> > > diff --git a/drivers/iio/adc/rockchip_saradc.c b/drivers/iio/adc/rockchip_saradc.c
> > > index 582ba047c4a6..1a7990d60f9f 100644
> > > --- a/drivers/iio/adc/rockchip_saradc.c
> > > +++ b/drivers/iio/adc/rockchip_saradc.c
> > > @@ -193,6 +193,27 @@ static void rockchip_saradc_reset_controller(struct reset_control *reset)
> > >  	reset_control_deassert(reset);
> > >  }
> > >  
> > > +static void rockchip_saradc_clk_disable(void *data)
> > > +{
> > > +	struct rockchip_saradc *info = data;
> > > +
> > > +	clk_disable_unprepare(info->clk);
> > > +}
> > > +
> > > +static void rockchip_saradc_pclk_disable(void *data)
> > > +{
> > > +	struct rockchip_saradc *info = data;
> > > +
> > > +	clk_disable_unprepare(info->pclk);
> > > +}
> > > +
> > > +static void rockchip_saradc_regulator_disable(void *data)
> > > +{
> > > +	struct rockchip_saradc *info = data;
> > > +
> > > +	regulator_disable(info->vref);
> > > +}
> > > +
> > >  static int rockchip_saradc_probe(struct platform_device *pdev)
> > >  {
> > >  	struct rockchip_saradc *info = NULL;
> > > @@ -291,17 +312,38 @@ static int rockchip_saradc_probe(struct platform_device *pdev)
> > >  		dev_err(&pdev->dev, "failed to enable vref regulator\n");
> > >  		return ret;
> > >  	}
> > > +	ret = devm_add_action_or_reset(&pdev->dev,
> > > +				       rockchip_saradc_regulator_disable, info);
> > > +	if (ret) {
> > > +		dev_err(&pdev->dev, "failed to register devm action, %d\n",
> > > +			ret);
> > > +		return ret;
> > > +	}
> > >  
> > >  	ret = clk_prepare_enable(info->pclk);
> > >  	if (ret < 0) {
> > >  		dev_err(&pdev->dev, "failed to enable pclk\n");
> > > -		goto err_reg_voltage;
> > > +		return ret;
> > > +	}
> > > +	ret = devm_add_action_or_reset(&pdev->dev,
> > > +				       rockchip_saradc_pclk_disable, info);
> > > +	if (ret) {
> > > +		dev_err(&pdev->dev, "failed to register devm action, %d\n",
> > > +			ret);
> > > +		return ret;
> > >  	}
> > >  
> > >  	ret = clk_prepare_enable(info->clk);
> > >  	if (ret < 0) {
> > >  		dev_err(&pdev->dev, "failed to enable converter clock\n");
> > > -		goto err_pclk;
> > > +		return ret;
> > > +	}
> > > +	ret = devm_add_action_or_reset(&pdev->dev,
> > > +				       rockchip_saradc_clk_disable, info);
> > > +	if (ret) {
> > > +		dev_err(&pdev->dev, "failed to register devm action, %d\n",
> > > +			ret);
> > > +		return ret;
> > >  	}
> > >  
> > >  	platform_set_drvdata(pdev, indio_dev);
> > > @@ -315,30 +357,9 @@ static int rockchip_saradc_probe(struct platform_device *pdev)
> > >  	indio_dev->channels = info->data->channels;
> > >  	indio_dev->num_channels = info->data->num_channels;
> > >  
> > > -	ret = iio_device_register(indio_dev);
> > > +	ret = devm_iio_device_register(&pdev->dev, indio_dev);
> > >  	if (ret)
> > > -		goto err_clk;
> > > -
> > > -	return 0;
> > > -
> > > -err_clk:
> > > -	clk_disable_unprepare(info->clk);
> > > -err_pclk:
> > > -	clk_disable_unprepare(info->pclk);
> > > -err_reg_voltage:
> > > -	regulator_disable(info->vref);
> > > -	return ret;
> > > -}
> > > -
> > > -static int rockchip_saradc_remove(struct platform_device *pdev)
> > > -{
> > > -	struct iio_dev *indio_dev = platform_get_drvdata(pdev);
> > > -	struct rockchip_saradc *info = iio_priv(indio_dev);
> > > -
> > > -	iio_device_unregister(indio_dev);
> > > -	clk_disable_unprepare(info->clk);
> > > -	clk_disable_unprepare(info->pclk);
> > > -	regulator_disable(info->vref);
> > > +		return ret;  
> > Small tweak rather hidden by how diff presents this but
> > might as well just
> > 
> > return devm_iio_device_register
> >   
> > >  
> > >  	return 0;
> > >  }
> > > @@ -383,7 +404,6 @@ static SIMPLE_DEV_PM_OPS(rockchip_saradc_pm_ops,
> > >  
> > >  static struct platform_driver rockchip_saradc_driver = {
> > >  	.probe		= rockchip_saradc_probe,
> > > -	.remove		= rockchip_saradc_remove,
> > >  	.driver		= {
> > >  		.name	= "rockchip-saradc",
> > >  		.of_match_table = rockchip_saradc_match,  
> > 
> >   
> 
> 
> 
> 