Replying to both of your mails in one message here:
> > > +config DHT11_GPIO
> > > + tristate "DHT11 (and compatible sensors) driver"
> > > + help
> > > + This driver supports reading data via a single interrupts
>
> should probably depend on GENERIC_GPIO?
After thinking about this a bit, I believe a dependency on
OF_GPIO is necessary. Maybe also add an explicit dependency
on GPIOLIB?
I believe GPIO_GENERIC ist not a dependency as it seems to
mostly be support code for other gpio controler drivers.
Actually, no other drivers like leds-gpio depend on it.
> > > prefix with DHT11_
>
> > Do you think this is needed, even as all these defines are
> > local to this file? (It would force rather unpleasant line
> > breaks...) Is there some general rule when to prefix and
> > when not to?
>
> breakage occurs when some of your include file #define e.g. START_BIT or
> STARTUP in the future; I think the rule for iio is to prefix everything
Ok.
> > > > +#define EDGES_PREAMBLE 4
> > > > +#define BITS_PER_READ 40
> > > > +#define EDGES_PER_READ (2*BITS_PER_READ + EDGES_PREAMBLE + 1)
> > > > +
> > > > +/* Data transmission timing (nano seconds) */
> > > > +#define STARTUP 18 /* ms */
> > >
> > > just a line above nano seconds were promised, and now this: ms?! :)
> >
> > Sure, the exception from the rule is noted as end of line comment.
> > Is there some better way to do this?
>
> STARTUP is used exactly once; I'd just drop the #define and use the
> constant directly -- you'll get a checkpatch warning for msleep(18) though
>
> and startup is not transmission timing, right?
Actually it is in the sense that we need to pull down the gpio line
for at least this time to have the sensor start a transmission.
Maybe STARTTRANSMISSION is a better name than STARTUP.
It's not as critical as the other timing constants, but it is still
used whenever the driver wants any new data from the hardware.
Thanks for your help,
Harald
> > > > +#define SENSOR_RESPONSE 80000
> > > > +#define START_BIT 50000
> > > > +#define DATA_BIT_LOW 27000
> > > > +#define DATA_BIT_HIGH 70000
> > > > +
> > > > +/* TODO?: Support systems without DT? */
> > > > +
> > > > +struct dht11_gpio {
> > > > + struct device *dev;
> > > > +
> > > > + int gpio;
> > > > + int irq;
> > > > +
> > > > + struct completion completion;
> > > > +
> > > > + s64 timestamp;
> > > > + int temperature;
> > > > + int humidity;
> > > > +
> > > > + /* num_edges: -1 means "no transmission in progress" */
> > > > + int num_edges;
> > > > + struct {s64 ts; int value; } edges[EDGES_PER_READ];
> > > > +};
> > > > +
> > > > +/*
> > > > + * dht11_edges_print: show the data as actually received by the
> > > > + * driver.
> > > > + * This is dead code, keeping it for now just in case somebody needs
> > > > + * it for porting the driver to new sensor HW, etc.
> > > > + *
> > > > +static void dht11_edges_print(struct dht11_gpio *dht11)
> > > > +{
> > > > + int i;
> > > > +
> > > > + for (i = 1; i < dht11->num_edges; ++i) {
> > > > + pr_err("dht11: %d: %lld ns %s\n", i,
> > >
> > > inconsistent driver name; dht11-gpio was used before
> > >
> > > > + dht11->edges[i].ts - dht11->edges[i-1].ts,
> > > > + dht11->edges[i-1].value ? "high" : "low");
> > > > + }
> > > > +}
> > > > +*/
> > > > +
> > > > +static unsigned char dht11_gpio_decode_byte(int *timing, int threshold
> )
> > > > +{
> > > > + unsigned char ret = 0;
> > > > + int i;
> > > > +
> > > > + for (i = 0; i < 8; ++i) {
> > > > + ret <<= 1;
> > > > + if (timing[i] >= threshold)
> > > > + ++ret;
> > > > + }
> > > > +
> > > > + return ret;
> > > > +}
> > > > +
> > > > +static int dht11_gpio_decode(struct dht11_gpio *dht11, int offset)
> > > > +{
> > > > + int i, t, timing[BITS_PER_READ], threshold, timeres = SENSOR_RE
> SPONSE;
> > > > + unsigned char temp_int, temp_dec, hum_int, hum_dec, checksum;
> > > > +
> > > > + /* Calculate timestamp resolution */
> > > > + for (i = 0; i < dht11->num_edges; ++i) {
> > > > + t = dht11->edges[i].ts - dht11->edges[i-1].ts;
> > > > + if (t > 0 && t < timeres)
> > > > + timeres = t;
> > > > + }
> > > > + if (2*timeres > DATA_BIT_HIGH) {
> > > > + pr_err("dht11-gpio: timeresolution %d too bad for decod
> ing\n",
> > > > + timeres);
> > > > + return -EIO;
> > > > + }
> > > > + threshold = DATA_BIT_HIGH/timeres;
> >
> > I'm inclined to change this to:
> > threshold = DATA_BIT_HIGH / timeres;
>
> this is what I had in mind
>
> > > > + if (DATA_BIT_LOW/timeres + 1 >= threshold)
> > > > + pr_err("dht11-gpio: WARNING: decoding ambiguous\n");
> > > > +
> > > > + /* scale down with timeres and check validity */
> > > > + for (i = 0; i < BITS_PER_READ; ++i) {
> > > > + t = dht11->edges[offset + 2*i + 2].ts -
> > > > + dht11->edges[offset + 2*i + 1].ts;
> > > > + if (!dht11->edges[offset + 2*i + 1].value)
> > > > + return -EIO; /* lost synchronisation */
> > > > + timing[i] = t / timeres;
> >
> > and leave this as is.
> >
> > > inconsistent whitespace around / operator
> >
> > Would that be consistent enough? (The rule being, that a / operator
> > on it's own gets spaces, but the same operator in a statement
> > together with operators with lower binding power gets none.)
> >
> > > > + }
> > > > +
> > > > + hum_int = dht11_gpio_decode_byte(timing, threshold);
> > > > + hum_dec = dht11_gpio_decode_byte(&timing[8], threshold);
> > > > + temp_int = dht11_gpio_decode_byte(&timing[16], threshold);
> > > > + temp_dec = dht11_gpio_decode_byte(&timing[24], threshold);
> > > > + checksum = dht11_gpio_decode_byte(&timing[32], threshold);
> > > > +
> > > > + if (((hum_int + hum_dec + temp_int + temp_dec) & 0x00ff) != che
> cksum)
> > >
> > > maybe 0xff instead of 0x00ff is clearer
> >
> > Ok, I'm happy either way.
> >
> > > > + return -EIO;
> > > > +
> > > > + dht11->timestamp = iio_get_time_ns();
> > > > + if (hum_int < 20) { /* DHT22 */
> > > > + dht11->temperature = (((temp_int & 0x7f) << 8) + temp_d
> ec) *
> > > > + ((temp_int & 0x80) ? -100 : 100
> );
> > > > + dht11->humidity = ((hum_int << 8) + hum_dec) * 100;
> > > > + } else if (temp_dec == 0 && hum_dec == 0) { /* DHT11 */
> > >
> > > extra space before comment
> >
> > Actually I think that code is much more readable if end of line
> > comments are visually separated by two or more spaces (and
> > check_patch.pl is happy with that). I see now, that I have not
> > been entirely consistent and would rather add extra spaces rather
> > then remove them. What do you think?
>
> I'd use just one space, just a matter of taste; no strong feelings :)
>
> > > > + dht11->temperature = temp_int * 1000;
> > > > + dht11->humidity = hum_int * 1000;
> > > > + } else {
> > > > + dev_err(dht11->dev,
> > > > + "Don't know how to decode data: %d %d %d %d\n",
> > > > + hum_int, hum_dec, temp_int, temp_dec);
> > > > + return -EIO;
> > > > + }
> > > > +
> > > > + return 0;
> > > > +}
> > > > +
> > > > +static int dht11_gpio_read_raw(struct iio_dev *iio_dev,
> > > > + const struct iio_chan_spec *chan,
> > > > + int *val, int *val2, long m)
> > > > +{
> > > > + struct dht11_gpio *dht11 = iio_priv(iio_dev);
> > > > + int ret = 0;
> > >
> > > the initialization of ret is not used; don't or init with -EINVAL to save
>
> > > an assignment
> > >
> > > > +
> > > > + if (dht11->timestamp + DATA_VALID_TIME < iio_get_time_ns()) {
> > > > + INIT_COMPLETION(dht11->completion);
> > > > +
> > > > + dht11->num_edges = 0;
> > > > + ret = gpio_direction_output(dht11->gpio, 0);
> > > > + if (ret)
> > > > + goto err;
> > > > + msleep(STARTUP);
> > > > + ret = gpio_direction_input(dht11->gpio);
> > > > + if (ret)
> > > > + goto err;
> > > > +
> > > > + ret = wait_for_completion_killable_timeout(&dht11->comp
> letion,
> > > > + HZ);
> > > > + if (ret == 0 && dht11->num_edges < EDGES_PER_READ - 1)
> {
> > > > + dev_err(&iio_dev->dev,
> > > > + "Only %d signal edges detected\
> n",
> > > > + dht11->num_edges);
> > > > + ret = -ETIMEDOUT;
> > > > + }
> > > > + if (ret < 0)
> > > > + goto err;
> > > > +
> > > > + ret = dht11_gpio_decode(dht11,
> > > > + dht11->num_edges == EDGES_PER_READ ?
> > > > + EDGES_PREAMBLE : EDGES_PREAMBLE
> - 2);
> > > > + if (ret)
> > > > + goto err;
> > > > + }
> > > > +
> > > > + ret = IIO_VAL_INT;
> > > > + if (chan->channel == 0)
> > > > + *val = dht11->temperature;
> > > > + else if (chan->channel == 1)
> > > > + *val = dht11->humidity;
> > >
> > > use channel type to discriminate channels
> >
> > Yes, very good point.
> >
> > I agree with the rest of the comments. Thanks for the review.
> > Harald
> >
> > > > + else
> > > > + ret = -EINVAL;
> > > > +err:
> > > > + dht11->num_edges = -1;
> > > > + return ret;
> > > > +}
> > > > +
> > > > +static const struct iio_info dht11_gpio_iio_info = {
> > > > + .driver_module = THIS_MODULE,
> > > > + .read_raw = dht11_gpio_read_raw,
> > > > +};
> > > > +
> > > > +/*
> > > > + * IRQ handler called on GPIO edges
> > > > +*/
> > > > +static irqreturn_t dht11_gpio_handle_irq(int irq, void *data)
> > > > +{
> > > > + struct iio_dev *iio = data;
> > > > + struct dht11_gpio *dht11 = iio_priv(iio);
> > > > +
> > > > + /* TODO: Consider making the handler safe for IRQ sharing */
> > > > + if (dht11->num_edges < EDGES_PER_READ && dht11->num_edges >= 0)
> {
> > > > + dht11->edges[dht11->num_edges].ts = iio_get_time_ns();
> > > > + dht11->edges[dht11->num_edges++].value =
> > > > + gpio_get_value(dht11->g
> pio);
> > > > +
> > > > + if (dht11->num_edges >= EDGES_PER_READ)
> > > > + complete(&dht11->completion);
> > > > + }
> > > > +
> > > > + return IRQ_HANDLED;
> > > > +}
> > > > +
> > > > +static const struct iio_chan_spec dht11_gpio_chan_spec[] = {
> > > > + { .type = IIO_TEMP, .channel = 0,
> > > > + .info_mask_separate = BIT(IIO_CHAN_INFO_PROCESSED), },
> > > > + { .type = IIO_HUMIDITYRELATIVE, .channel = 1,
> > > > + .info_mask_separate = BIT(IIO_CHAN_INFO_PROCESSED),
> > >
> > > .channel is not needed; there aren't multiple channels of the same type
> > >
> > > > +}
> > >
> > > weird indentation level
> > >
> > > > +};
> > > > +
> > > > +static const struct of_device_id dht11_gpio_dt_ids[] = {
> > > > + { .compatible = "dht11-gpio", },
> > > > + { }
> > > > +};
> > > > +MODULE_DEVICE_TABLE(of, dht11_gpio_dt_ids);
> > > > +
> > > > +static int dht11_gpio_probe(struct platform_device *pdev)
> > > > +{
> > > > + struct device *dev = &pdev->dev;
> > > > + struct device_node *node = dev->of_node;
> > > > + struct dht11_gpio *dht11;
> > > > + struct iio_dev *iio;
> > > > + int ret = 0;
> > >
> > > don't initialize ret
> > >
> > > > +
> > > > + /* Allocate the IIO device. */
> > >
> > > obvious comment
> > >
> > > > + iio = devm_iio_device_alloc(dev, sizeof(*dht11));
> > > > + if (!iio) {
> > > > + dev_err(dev, "Failed to allocate IIO device\n");
> > > > + return -ENOMEM;
> > > > + }
> > > > +
> > > > + dht11 = iio_priv(iio);
> > > > + dht11->dev = dev;
> > > > +
> > > > + dht11->gpio = ret = of_get_gpio(node, 0);
> > > > + if (ret < 0)
> > > > + return ret;
> > > > + ret = devm_gpio_request_one(dev, dht11->gpio, GPIOF_IN, pdev->n
> ame);
> > > > + if (ret)
> > > > + return ret;
> > > > +
> > > > + dht11->irq = gpio_to_irq(dht11->gpio);
> > > > + if (dht11->irq < 0) {
> > > > + dev_err(dev, "GPIO %d has no interrupt\n", dht11->gpio)
> ;
> > > > + return -EINVAL;
> > > > + }
> > > > + ret = devm_request_irq(dev, dht11->irq, dht11_gpio_handle_irq,
> > > > + IRQF_TRIGGER_RISING | IRQF_TRIGGER_FALL
> ING,
> > > > + pdev->name, iio);
> > > > + if (ret)
> > > > + return ret;
> > > > +
> > > > + dht11->timestamp = iio_get_time_ns() - DATA_VALID_TIME - 1;
> > > > + dht11->num_edges = -1;
> > > > +
> > > > + platform_set_drvdata(pdev, iio);
> > > > +
> > > > + init_completion(&dht11->completion);
> > > > + iio->name = pdev->name;
> > > > + iio->dev.parent = &pdev->dev;
> > > > + iio->info = &dht11_gpio_iio_info;
> > > > + iio->modes = INDIO_DIRECT_MODE;
> > > > + iio->channels = dht11_gpio_chan_spec;
> > > > + iio->num_channels = ARRAY_SIZE(dht11_gpio_chan_spec);
> > > > +
> > > > + /* Register IIO device. */
> > >
> > > the function name is pretty obvious, no comment needed
> > >
> > > > + ret = iio_device_register(iio);
> > >
> > > maybe just return here, the extra output is not needed
> > >
> > > > + if (ret) {
> > > > + dev_err(dev, "Failed to register IIO device\n");
> > > > + return ret;
> > > > + }
> > > > +
> > > > + return 0;
> > > > +}
> > > > +
> > > > +static int dht11_gpio_remove(struct platform_device *pdev)
> > > > +{
> > > > + struct iio_dev *iio = platform_get_drvdata(pdev);
> > > > +
> > > > + iio_device_unregister(iio);
> > > > +
> > > > + return 0;
> > > > +}
> > > > +
> > > > +static struct platform_driver dht11_gpio_driver = {
> > > > + .driver = {
> > > > + .name = DRIVER_NAME,
> > > > + .owner = THIS_MODULE,
> > > > + .of_match_table = dht11_gpio_dt_ids,
> > > > + },
> > > > + .probe = dht11_gpio_probe,
> > > > + .remove = dht11_gpio_remove,
> > > > +};
> > > > +
> > > > +module_platform_driver(dht11_gpio_driver);
> > > > +
> > > > +MODULE_AUTHOR("Harald Geyer <harald@xxxxxxxxx>");
> > > > +MODULE_DESCRIPTION("DHT11 humidity/temperature sensor driver");
> > > > +MODULE_LICENSE("GPL v2");
> > > > +
> > > >
> > >
> > > --
> > >
> > > Peter Meerwald
> > > +43-664-2444418 (mobile)
> > > --
> > > To unsubscribe from this list: send the line "unsubscribe linux-iio" in
> > > the body of a message to majordomo@xxxxxxxxxxxxxxx
> > > More majordomo info at http://vger.kernel.org/majordomo-info.html
> >
>
> --
>
> Peter Meerwald
> +43-664-2444418 (mobile)
--
To unsubscribe from this list: send the line "unsubscribe linux-iio" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
