Reviewed-by: Crt Mori <cmo@xxxxxxxxxxx> We did not detect this issue on Linux chips, but there is a slight possibility that it might occur nevertheless. So I am proposing this to add to_fixes as well (maybe even for the upcoming 5.10?) Best regards, Crt On Fri, 11 Dec 2020 at 12:14, Slaveyko Slaveykov <sis@xxxxxxxxxxx> wrote: > > After an I2C reset command, the mlx90632 needs some time before > responding to other I2C commands. Without that delay, there is a chance > that the I2C command(s) after the reset will not be accepted > > Signed-off-by: Slaveyko Slaveykov <sis@xxxxxxxxxxx> > --- > drivers/iio/temperature/mlx90632.c | 1 + > 1 file changed, 1 insertion(+) > > diff --git a/drivers/iio/temperature/mlx90632.c b/drivers/iio/temperature/mlx90632.c > index 503fe54a0bb9..ee50f61e496a 100644 > --- a/drivers/iio/temperature/mlx90632.c > +++ b/drivers/iio/temperature/mlx90632.c > @@ -247,6 +247,7 @@ static int mlx90632_set_meas_type(struct regmap *regmap, u8 type) > ret = regmap_write(regmap, MLX90632_REG_I2C_CMD, MLX90632_RESET_CMD); > if (ret < 0) > return ret; > + usleep(150, 200); > > ret = regmap_write_bits(regmap, MLX90632_REG_CONTROL, > (MLX90632_CFG_MTYP_MASK | MLX90632_CFG_PWR_MASK), > -- > 2.16.2.windows.1 >
