Re: [PATCH v3 3/4] drivers: iio: chemical: Add support for Sensirion SCD4x CO2 sensor

[Roan van Dijk <roan@protonic.nl>]

On 25-09-2021 18:23, Jonathan Cameron wrote:
> On Wed, 22 Sep 2021 12:39:24 +0200
> Roan van Dijk <roan@protonic.nl> wrote:
> 
>> This is a driver for the SCD4x CO2 sensor from Sensirion. The sensor is
>> able to measure CO2 concentration, temperature and relative humdity.
>> The sensor uses a photoacoustic principle for measuring CO2 concentration.
>> An I2C interface is supported by this driver in order to communicate with
>> the sensor.
>>
>> Signed-off-by: Roan van Dijk <roan@protonic.nl>
> 
> Hi Roan,
> 
> Only thing in here of significance is that the format for available attribute
> is wrong + it needs adding to the ABI docs.
> 
> Given we are going to have a v4, I noted a few other minor things to tidy up.
> 
> Thanks,
> 
> Jonathan
> 

Thank you for your feedback. I will make a new patch with these fixes.

The documentation for the _available and the fixes you suggested, have 
already been added in patch [PATCH v3 4/4]. Do I still need to add 
something to the ABI documentation or is it fine like this?

However, I will change the KernelVersion to 5.15 of the _available 
description in patch v4. I left this at 5.8 and that's not right, 
because this information is newly added.

Thanks,

Roan

> 
>> +static int scd4x_read(struct scd4x_state *state, enum scd4x_cmd cmd,
>> +			void *response, int response_sz)
>> +{
>> +	struct i2c_client *client = state->client;
>> +	char buf[SCD4X_READ_BUF_SIZE];
>> +	char *rsp = response;
>> +	int i, ret;
>> +	char crc;
>> +
>> +	/*
>> +	 * Measurement needs to be stopped before sending commands.
>> +	 * Except for reading measurement and data ready command.
>> +	 */
>> +	if ((cmd != CMD_GET_DATA_READY) && (cmd != CMD_READ_MEAS)) {
>> +		ret = scd4x_send_command(state, CMD_STOP_MEAS);
>> +		if (ret)
>> +			return ret;
>> +
>> +		/* execution time for stopping measurement */
>> +		msleep_interruptible(500);
>> +	}
>> +
>> +	/*CRC byte for every 2 bytes of data */
> 
> /* CRC..
> 
> Please check for similar as otherwise we'll get 'cleanup' patches the moment various
> scripts hit this new code and it'll waste our time!
> 
>> +	response_sz += response_sz / 2;
>> +
>> +	put_unaligned_be16(cmd, buf);
>> +	ret = scd4x_i2c_xfer(state, buf, 2, buf, response_sz);
>> +	if (ret)
>> +		return ret;
>> +
>> +	for (i = 0; i < response_sz; i += 3) {
>> +		crc = crc8(scd4x_crc8_table, buf + i, 2, CRC8_INIT_VALUE);
>> +		if (crc != buf[i + 2]) {
>> +			dev_err(&client->dev, "CRC error\n");
>> +			return -EIO;
>> +		}
>> +
>> +		*rsp++ = buf[i];
>> +		*rsp++ = buf[i + 1];
>> +	}
>> +
>> +	/* start measurement */
>> +	if ((cmd != CMD_GET_DATA_READY) && (cmd != CMD_READ_MEAS)) {
>> +		ret = scd4x_send_command(state, CMD_START_MEAS);
>> +		if (ret)
>> +			return ret;
>> +	}
>> +
>> +	return 0;
>> +}
>> +
> 
> ...
> 
>> +
>> +static IIO_DEVICE_ATTR_RW(calibration_auto_enable, 0);
>> +static IIO_DEVICE_ATTR_WO(calibration_forced_value, 0);
>> +
>> +static IIO_CONST_ATTR(calibration_forced_value_available,
>> +	       __stringify(SCD4X_FRC_MIN_PPM 1 SCD4X_FRC_MAX_PPM));
> 
> Ah, I wasn't completely clear on this.  See the main ABI doc for
> _available
> 
> Format for this needs to include brackets to indicate it's a range
> rather than 3 numbers
> "[MIN 1  MAX]"
> 
> 
> Having added this it also needs to be in the ABI documentation.
> Whilst somewhat trivial, all ABI should be documented there.
> 
>> +
>> +static struct attribute *scd4x_attrs[] = {
>> +	&iio_dev_attr_calibration_auto_enable.dev_attr.attr,
>> +	&iio_dev_attr_calibration_forced_value.dev_attr.attr,
>> +	&iio_const_attr_calibration_forced_value_available.dev_attr.attr,
>> +	NULL
>> +};
>> +
>> +static const struct attribute_group scd4x_attr_group = {
>> +	.attrs = scd4x_attrs,
>> +};
>> +
>> +static const struct iio_info scd4x_info = {
>> +	.attrs = &scd4x_attr_group,
>> +	.read_raw = scd4x_read_raw,
>> +	.write_raw = scd4x_write_raw,
>> +};
>> +
> 
> ...
> 
>> +
>> +static irqreturn_t scd4x_trigger_handler(int irq, void *p)
>> +{
>> +	struct iio_poll_func *pf = p;
>> +	struct iio_dev *indio_dev = pf->indio_dev;
>> +	struct scd4x_state *state = iio_priv(indio_dev);
>> +	struct {
>> +		uint16_t data[3];
>> +		int64_t ts __aligned(8);
>> +	} scan;
>> +	int ret;
>> +	uint16_t buf[3];
>> +
>> +	mutex_lock(&state->lock);
>> +	ret = scd4x_read_poll(state, buf);
>> +	mutex_unlock(&state->lock);
>> +	if (ret)
>> +		goto out;
>> +
>> +	memset(&scan, 0, sizeof(scan));
>> +	memcpy(scan.data, buf, sizeof(buf));
> 
> I missed this before, but why not do the scd4x_read_poll() directly into scan->data after
> you've done the memset?  That way you avoid the need for a memcpy.
> 
> i.e.
> 
> 	memset(&scan, 0, sizeof(scan));
> 	mutex_lock(&state->lock)
> 	ret = scd4x_read_poll(state, scan->data);
> 	mutex_unlock(&state->lock);
> 	if (ret)
> 	...
> 
> 
>> +
>> +	iio_push_to_buffers_with_timestamp(indio_dev, &scan, iio_get_time_ns(indio_dev));
>> +out:
>> +	iio_trigger_notify_done(indio_dev->trig);
>> +	return IRQ_HANDLED;
>> +}
>> +