Re: [PATCH v2 2/2] iio: proximity: add support for PulsedLight LIDAR

[Tristan Lelong <tristan@lelong.xyz>]

On Mon, Aug 03, 2015 at 01:21:08AM -0700, Matt Ranostay wrote:
> On Sun, Aug 2, 2015 at 11:48 PM, Peter Meerwald <pmeerw@pmeerw.net> wrote:
> > On Sun, 2 Aug 2015, Matt Ranostay wrote:
> >
> >> +
> >> +static int lidar_get_measurement(struct lidar_data *data, u16 *reg)
> >> +{
> >> +     struct i2c_client *client = data->client;
> >> +     int tries = 10;
> >> +     int ret;
> >> +
> >> +     /* start sample */
> >> +     ret = lidar_write_control(data, LIDAR_REG_CONTROL_ACQUIRE);
> >> +     if (ret < 0) {
> >> +             dev_err(&client->dev, "cannot send start measurement command");
> >> +             return ret;
> >> +     }
> >> +
> >> +     while (tries--) {
> >> +             usleep_range(1000, 2000);
> >> +
> >> +             ret = lidar_read_byte(data, LIDAR_REG_STATUS);
> >> +             if (ret < 0)
> >> +                     break;
> >> +
> >> +             /* return 0 since laser is likely pointed out of range */
> >> +             if (ret & LIDAR_REG_STATUS_INVALID) {
> >> +                     *reg = 0;
> >> +                     ret = 0;
> >> +                     break;
> >> +             }
> >> +
> >> +             /* sample ready to read */
> >> +             if (!(ret & LIDAR_REG_STATUS_READY)) {
> >> +                     ret = lidar_read_measurement(data, reg);
> >> +                     break;
> >> +             }
> >> +     }
> >> +
> >
> > what happens when the timeout occurs?
> > no error code is set
> 
> Ah yes it should return an error code.  Would a -EIO be a logical choice?
> 

I would think -ETIMEDOUT to make more sense here.