Re: [PATCH 2/2] iio: imu: st_lsm6dsx: add iio trigger and buffer support

[Sean Nyekjaer <sean@geanix.com>]

On 11/05/2019 13.37, Jonathan Cameron wrote:
> On Tue,  7 May 2019 10:02:25 +0200
> Sean Nyekjaer <sean@geanix.com> wrote:
> 
>> This adds support for using iio triggers, this is needed because
>> our hardware guys forgot to connect the irq pins from imu device
>> to the SoC.
>>
>> Signed-off-by: Sean Nyekjaer <sean@geanix.com>
> Hi Sean
> 
> A small question on the size of the buffer needed inline.  Otherwise looks
> good to me.
>> ---
>>   drivers/iio/imu/st_lsm6dsx/st_lsm6dsx_core.c | 37 ++++++++++++++++++++
>>   1 file changed, 37 insertions(+)
>>
>> diff --git a/drivers/iio/imu/st_lsm6dsx/st_lsm6dsx_core.c b/drivers/iio/imu/st_lsm6dsx/st_lsm6dsx_core.c
>> index 1ca69598678f..65ab853202de 100644
>> --- a/drivers/iio/imu/st_lsm6dsx/st_lsm6dsx_core.c
>> +++ b/drivers/iio/imu/st_lsm6dsx/st_lsm6dsx_core.c
>> @@ -40,8 +40,11 @@
>>   #include <linux/kernel.h>
>>   #include <linux/module.h>
>>   #include <linux/delay.h>
>> +#include <linux/iio/buffer.h>
>>   #include <linux/iio/iio.h>
>>   #include <linux/iio/sysfs.h>
>> +#include <linux/iio/trigger_consumer.h>
>> +#include <linux/iio/triggered_buffer.h>
>>   #include <linux/pm.h>
>>   #include <linux/regmap.h>
>>   #include <linux/bitfield.h>
>> @@ -945,6 +948,30 @@ static int st_lsm6dsx_init_hw_timer(struct st_lsm6dsx_hw *hw)
>>   	return 0;
>>   }
>>   
>> +static irqreturn_t st_lsm6dsx_trigger_handler(int irq, void *p)
>> +{
>> +	struct iio_poll_func *pf = p;
>> +	struct iio_dev *indio_dev = pf->indio_dev;
>> +	struct st_lsm6dsx_sensor *sensor = iio_priv(indio_dev);
>> +	u16 buffer[4 * sizeof(s64)/sizeof(u16)];
> The size may well be correct, but that particular way of expressing it doesn't make
> it terribly clear.
> 
> It seems a bit large... You need the space for the timestamp, so it needs to be
> padded to a u64, but that would still require your other channels to need
> more than 2*sizeof(s64)/sizeof(u16) u16s which is 9 or more of them.
Right now we have u16[16], so that quite at bit large :-)

So maybe something like:
/* 3x s16 + 1x s16(padding) + 1x s64 timestamp */
u16 buffer[2 * sizeof(s64)/sizeof(u16)];
Would do it?

> 
>> +	int tmp, bit;
>> +
>> +	memset(buffer, 0, sizeof(buffer));
>> +
>> +	for_each_set_bit(bit, indio_dev->active_scan_mask, indio_dev->masklength) {
>> +		st_lsm6dsx_read_oneshot(sensor,
>> +					indio_dev->channels[bit].address, &tmp);
>> +		buffer[bit] = tmp;
>> +	}
As you pointed out in the
"[PATCH 1/2] iio: adc: ti-ads8688: save values correct in buffer"
When taking this approach it will not shift down.
>> +
>> +	iio_push_to_buffers_with_timestamp(indio_dev, buffer,
>> +					   iio_get_time_ns(indio_dev));
>> +
>> +	iio_trigger_notify_done(indio_dev->trig);
>> +
>> +	return IRQ_HANDLED;
>> +}
>> +
>>   static int st_lsm6dsx_init_device(struct st_lsm6dsx_hw *hw)
>>   {
>>   	u8 drdy_int_reg;
>> @@ -1093,6 +1120,16 @@ int st_lsm6dsx_probe(struct device *dev, int irq, int hw_id, const char *name,
>>   		err = st_lsm6dsx_fifo_setup(hw);
>>   		if (err < 0)
>>   			return err;
>> +	} else {
>> +		for (i = 0; i < ST_LSM6DSX_ID_MAX; i++) {
>> +			err = devm_iio_triggered_buffer_setup(hw->dev,
>> +					hw->iio_devs[i], NULL,
>> +					st_lsm6dsx_trigger_handler, NULL);
>> +			if (err < 0) {
>> +				dev_err(hw->dev, "iio triggered buffer setup failed\n");
>> +				return err;
>> +			}
>> +		}
>>   	}
>>   
>>   	for (i = 0; i < ST_LSM6DSX_ID_MAX; i++) {
>