linux-kernel - Re: [PATCH v1 1/2] iio: mtk-auxadc: add support IIO_CHAN_INFO

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>] [day] [month] [year] [list]
Message-ID: <20210817165418.4b1cfdce@jic23-huawei>
Date:   Tue, 17 Aug 2021 16:54:18 +0100
From:   Jonathan Cameron <jic23@...nel.org>
To:     "hui.liu" <hui.liu@...iatek.com>
Cc:     Jonathan Cameron <Jonathan.Cameron@...wei.com>,
        <robh+dt@...nel.org>, <lars@...afoo.de>, <pmeerw@...erw.net>,
        <srv_heupstream@...iatek.com>, <zhiyong.tao@...iatek.com>,
        <chun-hung.wu@...iatek.com>, <yingjoe.chen@...iatek.com>,
        <seiya.wang@...iatek.com>, <matthias.bgg@...il.com>,
        <s.hauer@...gutronix.de>, <devicetree@...r.kernel.org>,
        <linux-kernel@...r.kernel.org>,
        <linux-arm-kernel@...ts.infradead.org>,
        <linux-iio@...r.kernel.org>, <linux-mediatek@...ts.infradead.org>
Subject: Re: [PATCH v1 1/2] iio: mtk-auxadc: add support IIO_CHAN_INFO_RAW
 case

On Tue, 17 Aug 2021 16:37:52 +0800
hui.liu <hui.liu@...iatek.com> wrote:

> On Sat, 2021-08-14 at 17:10 +0100, Jonathan Cameron wrote:
> > On Fri, 13 Aug 2021 11:46:24 +0800
> > hui.liu <hui.liu@...iatek.com> wrote:
> >   
> > > On Thu, 2021-08-12 at 19:07 +0100, Jonathan Cameron wrote:  
> > > > On Thu, 12 Aug 2021 13:48:43 +0800
> > > > Hui Liu <hui.liu@...iatek.com> wrote:
> > > >     
> > > > > Add support IIO_CHAN_INFO_RAW case.    
> > > > 
> > > > Why?
> > > > 
> > > > We almost never support both RAW and PROCESSED as userspace
> > > > should be
> > > > fine to use either.  There are a few reasons we've let drivers do
> > > > this but I would like know why it matters to you and it
> > > > definitely
> > > > needs to be in the patch description.
> > > >     
> > > 
> > > Hi Jonathan,
> > > 
> > > 1. To support ADC consumers' different types of requirement: some
> > > consumers want to call iio_read_channel_raw to get raw data, the
> > > others
> > > use iio_read_channel_processed to get voltage.  
> > 
> > Give an example of the consumer using the raw channel readback
> > (without
> > acess to any scaling information?)
> >   
> 
> > > 2. In our origin driver, if consumer call
> > > iio_read_channel_processed,
> > > read back value is raw data. 
> > > 
> > > Could we use SCALE instead of PROCESSED in patch for next version,
> > > or
> > > what's your suggestion?  
> > 
> > That would unfortunately be a userspace ABI change.  We can add
> > interfaces
> > but taking them away is normally a problem :( 
> > 
> > Your reasons here are fine, subject to information on what consumer
> > cares
> > about having _RAW, please resend the patch with this information
> > added
> > to the description.
> > 
> > Thanks,
> > 
> > Jonathan  
> 
> 1. We found afe/iio-rescale.c, dac/dpot-dac.c and multiplexer/iio-mux.c 
> call iio_read_channel_raw to get raw data. If they use our ADC driver,
> I think we should support _RAW case. 
> If we support _RAW case, we will add more information in v2
> description.

iio-rescale has recently gained support for processed.
https://git.kernel.org/pub/scm/linux/kernel/git/jic23/iio.git/commit/drivers/iio/afe/iio-rescale.c?h=testing&id=53ebee9499805add3eef630d998c40812e6a1c39
dpot-dac is a rather obscure special case, so I doubt you actually have one of those.
If iio-mux is relevant then we should add processed support to that driver as well.

I would rather see those users of the interface fixed than us
having to tweak lots of drivers to provide _raw when it isn't appropriate for
that piece of hardware.

Jonathan

> 
> 2. Since we change _PROCESSED readback value from raw data to voltage,
> our consumer will make the changes synchronously. 
> 
> > > 
> > > Thanks.
> > >   
> > > > > 
> > > > > Signed-off-by: Hui Liu <hui.liu@...iatek.com>
> > > > > ---
> > > > >  drivers/iio/adc/mt6577_auxadc.c | 16 +++++++++++++++-
> > > > >  1 file changed, 15 insertions(+), 1 deletion(-)
> > > > > 
> > > > > diff --git a/drivers/iio/adc/mt6577_auxadc.c
> > > > > b/drivers/iio/adc/mt6577_auxadc.c
> > > > > index 79c1dd68b909..e995d43287b2 100644
> > > > > --- a/drivers/iio/adc/mt6577_auxadc.c
> > > > > +++ b/drivers/iio/adc/mt6577_auxadc.c
> > > > > @@ -60,7 +60,8 @@ static const struct mtk_auxadc_compatible
> > > > > mt6765_compat = {
> > > > >  		.type = IIO_VOLTAGE,				
> > > > >    
> > > > >  \
> > > > >  		.indexed = 1,					
> > > > >    
> > > > >  \
> > > > >  		.channel = (idx),				
> > > > >     \
> > > > > -		.info_mask_separate =
> > > > > BIT(IIO_CHAN_INFO_PROCESSED), \
> > > > > +		.info_mask_separate = BIT(IIO_CHAN_INFO_RAW) |	
> > > > >    
> > > > >  \
> > > > > +				      BIT(IIO_CHAN_INFO_PROCESS
> > > > > ED), \
> > > > >  }
> > > > >  
> > > > >  static const struct iio_chan_spec mt6577_auxadc_iio_channels[]
> > > > > = {
> > > > > @@ -181,6 +182,19 @@ static int mt6577_auxadc_read_raw(struct
> > > > > iio_dev *indio_dev,
> > > > >  	struct mt6577_auxadc_device *adc_dev =
> > > > > iio_priv(indio_dev);
> > > > >  
> > > > >  	switch (info) {
> > > > > +	case IIO_CHAN_INFO_RAW:
> > > > > +		*val = mt6577_auxadc_read(indio_dev, chan);
> > > > > +		if (*val < 0) {
> > > > > +			dev_notice(indio_dev->dev.parent,
> > > > > +				"failed to sample data on
> > > > > channel[%d]\n",
> > > > > +				chan->channel);
> > > > > +			return *val;
> > > > > +		}
> > > > > +		if (adc_dev->dev_comp->sample_data_cali)
> > > > > +			*val = mt_auxadc_get_cali_data(*val,
> > > > > true);
> > > > > +
> > > > > +		return IIO_VAL_INT;
> > > > > +
> > > > >  	case IIO_CHAN_INFO_PROCESSED:
> > > > >  		*val = mt6577_auxadc_read(indio_dev, chan);
> > > > >  		if (*val < 0) {    
> > > > 
> > > >     
> > 
> >