Re: [PATCH v3 2/3] iio: adc: Add driver for Renesas RZ/G2L A/D converter

[Jonathan Cameron <jic23@kernel.org>]

On Sat, 31 Jul 2021 19:31:52 +0100
"Lad, Prabhakar" <prabhakar.csengg@gmail.com> wrote:

> Hi Jonathan,
> 
> Thank you for the review.
> 

...

> > > +#define DRIVER_NAME          "rzg2l-adc"  
> >
> > As only used in one place, just put it inline there so we don't need
> > to go find if we want to know the value - I'm lazy.
> >  
> Its being used in two places
> 1: indio_dev->name = DRIVER_NAME #In probe call
> 2: .name = DRIVER_NAME # In platform_driver struct
> 
> Let me know if you want me to replace them inline and drop the above macro.

oops.  Searching apparently failed me ;)  Fine as is.

...


> > > +static const struct iio_info rzg2l_adc_iio_info = {
> > > +     .read_raw = rzg2l_adc_read_raw,
> > > +     .read_label = rzg2l_adc_read_label,
> > > +};
> > > +
> > > +static irqreturn_t rzg2l_adc_isr(int irq, void *dev_id)
> > > +{
> > > +     struct rzg2l_adc *adc = (struct rzg2l_adc *)dev_id;  
> >
> > No need for explicit cast from void * to another pointer type.
> > This is always valid without in C.
> >  
> Agreed.
> 
> > > +     unsigned long intst;
> > > +     u32 reg;
> > > +     int ch;
> > > +
> > > +     reg = rzg2l_adc_readl(adc, RZG2L_ADSTS);
> > > +
> > > +     /* A/D conversion channel select error interrupt */
> > > +     if (reg & RZG2L_ADSTS_CSEST) {
> > > +             rzg2l_adc_writel(adc, RZG2L_ADSTS, reg);
> > > +             return IRQ_HANDLED;
> > > +     }
> > > +
> > > +     intst = reg & RZG2L_ADSTS_INTST_MASK;
> > > +     if (!intst)
> > > +             return IRQ_NONE;
> > > +
> > > +     for_each_set_bit(ch, &intst, RZG2L_ADC_MAX_CHANNELS) {
> > > +             if (intst & BIT(ch))  
> >
> > I'm missing how this if can fail given we only end up in here when the bit is
> > set.
> >  
> ADC has 8 channels RZG2L_ADSTS register bits [0:7] will be set to 1
> when the given channel ADC conversion has been completed. So the above
> if condition checks if the bit is set to 1 and then reads the
> corresponding value of that channel.

Just looking at the two lines of code above
for_each_set_bit(ch, &intst, RZG2L_ADC_MAX_CHANNELS) 
will only call the the next line if the bit is set.  E.g. It will only call
it
if (intst & BIT(ch))

So you can only get into the body of the for loop if this bit is set and the
condition is always true.  Hence drop 
if (intst & BIT(ch)) 

> 
> > > +                     adc->last_val[ch] = rzg2l_adc_readl(adc, RZG2L_ADCR(ch)) &
> > > +                                         RZG2L_ADCR_AD_MASK;
> > > +     }
> > > +
> > > +     /* clear the channel interrupt */
> > > +     rzg2l_adc_writel(adc, RZG2L_ADSTS, reg);
> > > +
> > > +     complete(&adc->completion);
> > > +
> > > +     return IRQ_HANDLED;
> > > +}
> > > +

...

> > > +
> > > +     pm_runtime_enable(dev);  
> >
> > I think you also want to set the state to suspended to ensure the resume is
> > called on get.
> >  
> pm_runtime_set_suspended() should only be called when runtime is
> disabled or is it that I am missing something ?

If you want the runtime pm code to assume your device is suspended initially
then you set the state before you call pm_runtime_enable(dev).


J