lkml.org 
[lkml]   [2016]   [Apr]   [8]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: [PATCH] iio: potentiometer: add driver for Maxim Integrated DS1803

> > maybe a #define to explain the name of register 0
>
> Well this zero is not register nor any kind of command that DS1803 needs to send
> back pots values.
> DS1803 only requires the standard R/W bit to be set as read.
> I used _word_ function series to get back a word (16 bits) as this poti returns
> 16 bits.
>
> How should I named it?
>
> #define NON_COMMAND 0
> ?

maybe i2c_transfer() is more appropriate then

u8 databuf[2];
struct i2c_msg msgs[2] = {
{ .addr = client->addr, .len = sizeof(databuf), .buf = databuf,
.flags = I2C_M_RD } };
ret = i2c_transfer(client->adapter, msgs, 2);

this does not send any data to the chip but only reads two bytes

> Or should I use different function? (2x i2c_smbus_read_byte?)
>
> The i2c_smbus_read_byte() function also used 0 as command
> for its transfers...

> > > +static int ds1803_write_raw(struct iio_dev *indio_dev,
> > > + struct iio_chan_spec const *chan,
> > > + int val, int val2, long mask)
> > > +{
> > > + struct ds1803_data *data = iio_priv(indio_dev);
> > > + int pot = chan->channel;
> > > +
> > > + switch (mask) {
> > > + case IIO_CHAN_INFO_RAW:
> > > + if (val > DS1803_MAX_POS || val < 0)
> >
> > check that val2 is 0 or use .write_raw_get_fmt
>
> At this point I do not know why should I do it, but I will look into that.

write_raw expects a VAL_INT_PLUS_MICROS per default, passed
in val and val2

you can change that with .write_raw_get_fmt (e.g. to expect VAL_INT),
or just make sure that the micros are 0

--

Peter Meerwald-Stadler
+43-664-2444418 (mobile)

\
 
 \ /
  Last update: 2016-04-08 18:21    [W:0.077 / U:1.164 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site