On Fri, 23 Aug 2024 19:20:17 +0200 Vasileios Amoiridis <vassilisamir@xxxxxxxxx> wrote: > As it was reported and discussed here [1], storing the sensor data in an > endian aware s32 buffer is not optimal. Advertising the timestamp as an > addition of 2 s32 variables which is also implied is again not the best > practice. For that reason, change the s32 sensor_data buffer to a char > buffer with an extra value for the timestamp (as it is common practice). > > [1]: https://lore.kernel.org/linux-iio/73d13cc0-afb9-4306-b498-5d821728c3ba@stanley.mountain/ > > Signed-off-by: Vasileios Amoiridis <vassilisamir@xxxxxxxxx> Hi. Thanks for poking. Looks like I'd accidentally marked this one handled without actually doing so :( > --- > drivers/iio/pressure/bmp280-core.c | 43 +++++++++++++++++------------- > drivers/iio/pressure/bmp280.h | 5 +++- > 2 files changed, 28 insertions(+), 20 deletions(-) > > diff --git a/drivers/iio/pressure/bmp280-core.c b/drivers/iio/pressure/bmp280-core.c > index 3deaa57bb3f5..71e481c2f30d 100644 > --- a/drivers/iio/pressure/bmp280-core.c > +++ b/drivers/iio/pressure/bmp280-core.c > @@ -1035,7 +1035,8 @@ static irqreturn_t bmp280_trigger_handler(int irq, void *p) > goto out; > } > > - data->sensor_data[1] = bmp280_compensate_temp(data, adc_temp); > + ret = bmp280_compensate_temp(data, adc_temp); > + memcpy(&data->buffer.buf[4], &ret, 4); The 4's in here are magic indexes. Probably use sizeof(s32) in both cases. Maybe for the first one make the fact it's after an s32 explicit with &data->buffer.buf[1 * sizeof(s32)] Of carry an offset through the code that you update when you fill in part of the buffer. That is probably cleaner. int offset = 0; stuff offset += sizeof(s32); mempcy(&data->bufer.buf[offset], &ret, 4); if offset is used for a later entry then add the new element. > > /* Pressure calculations */ > adc_press = FIELD_GET(BMP280_MEAS_TRIM_MASK, get_unaligned_be24(&data->buf[0])); > @@ -1045,10 +1046,10 @@ static irqreturn_t bmp280_trigger_handler(int irq, void *p) > } > > t_fine = bmp280_calc_t_fine(data, adc_temp); > + ret = bmp280_compensate_press(data, adc_press, t_fine); > + memcpy(&data->buffer.buf[0], &ret, 4); > > - data->sensor_data[0] = bmp280_compensate_press(data, adc_press, t_fine); > - > - iio_push_to_buffers_with_timestamp(indio_dev, &data->sensor_data, > + iio_push_to_buffers_with_timestamp(indio_dev, &data->buffer, > iio_get_time_ns(indio_dev)); > > out: > @@ -1148,7 +1149,8 @@ static irqreturn_t bme280_trigger_handler(int irq, void *p) > goto out; > } > > - data->sensor_data[1] = bmp280_compensate_temp(data, adc_temp); > + ret = bmp280_compensate_temp(data, adc_temp); > + memcpy(&data->buffer.buf[4], &ret, 4); > > /* Pressure calculations */ > adc_press = FIELD_GET(BMP280_MEAS_TRIM_MASK, get_unaligned_be24(&data->buf[0])); > @@ -1158,8 +1160,8 @@ static irqreturn_t bme280_trigger_handler(int irq, void *p) > } > > t_fine = bmp280_calc_t_fine(data, adc_temp); > - > - data->sensor_data[0] = bmp280_compensate_press(data, adc_press, t_fine); > + ret = bmp280_compensate_press(data, adc_press, t_fine); > + memcpy(&data->buffer.buf[0], &ret, 4); > > /* Humidity calculations */ > adc_humidity = get_unaligned_be16(&data->buf[6]); > @@ -1168,9 +1170,11 @@ static irqreturn_t bme280_trigger_handler(int irq, void *p) > dev_err(data->dev, "reading humidity skipped\n"); > goto out; > } > - data->sensor_data[2] = bme280_compensate_humidity(data, adc_humidity, t_fine); > > - iio_push_to_buffers_with_timestamp(indio_dev, &data->sensor_data, > + ret = bme280_compensate_humidity(data, adc_humidity, t_fine); > + memcpy(&data->buffer.buf[8], &ret, 4); as above, but here it is I guess after 2 other elements, so 2 * sizeof(s32) > + > + iio_push_to_buffers_with_timestamp(indio_dev, &data->buffer, > iio_get_time_ns(indio_dev)); > > out: > @@ -1628,7 +1632,8 @@ static irqreturn_t bmp380_trigger_handler(int irq, void *p) > goto out; > } > > - data->sensor_data[1] = bmp380_compensate_temp(data, adc_temp); > + ret = bmp380_compensate_temp(data, adc_temp); > + memcpy(&data->buffer.buf[4], &ret, 4); > > /* Pressure calculations */ > adc_press = get_unaligned_le24(&data->buf[0]); > @@ -1638,10 +1643,10 @@ static irqreturn_t bmp380_trigger_handler(int irq, void *p) > } > > t_fine = bmp380_calc_t_fine(data, adc_temp); > + ret = bmp380_compensate_press(data, adc_press, t_fine); > + memcpy(&data->buffer.buf[0], &ret, 4); > > - data->sensor_data[0] = bmp380_compensate_press(data, adc_press, t_fine); > - > - iio_push_to_buffers_with_timestamp(indio_dev, &data->sensor_data, > + iio_push_to_buffers_with_timestamp(indio_dev, &data->buffer, > iio_get_time_ns(indio_dev)); > > out: > @@ -2203,12 +2208,12 @@ static irqreturn_t bmp580_trigger_handler(int irq, void *p) > } > > /* Temperature calculations */ > - memcpy(&data->sensor_data[1], &data->buf[0], 3); > + memcpy(&data->buffer.buf[4], &data->buf[0], 3); > > /* Pressure calculations */ > - memcpy(&data->sensor_data[0], &data->buf[3], 3); > + memcpy(&data->buffer.buf[0], &data->buf[3], 3); > - iio_push_to_buffers_with_timestamp(indio_dev, &data->sensor_data, > + iio_push_to_buffers_with_timestamp(indio_dev, &data->buffer, > iio_get_time_ns(indio_dev)); > > out: > @@ -2522,15 +2527,15 @@ static irqreturn_t bmp180_trigger_handler(int irq, void *p) > if (ret) > goto out; > > - data->sensor_data[1] = chan_value; > + memcpy(&data->buffer.buf[4], &chan_value, 4); > > ret = bmp180_read_press(data, &chan_value); > if (ret) > goto out; > > - data->sensor_data[0] = chan_value; > + memcpy(&data->buffer.buf[0], &chan_value, 4); > > - iio_push_to_buffers_with_timestamp(indio_dev, &data->sensor_data, > + iio_push_to_buffers_with_timestamp(indio_dev, &data->buffer, > iio_get_time_ns(indio_dev)); > > out: > diff --git a/drivers/iio/pressure/bmp280.h b/drivers/iio/pressure/bmp280.h > index ccacc67c1473..a853b6d5bdfa 100644 > --- a/drivers/iio/pressure/bmp280.h > +++ b/drivers/iio/pressure/bmp280.h > @@ -411,7 +411,10 @@ struct bmp280_data { > * Data to push to userspace triggered buffer. Up to 3 channels and > * s64 timestamp, aligned. > */ > - s32 sensor_data[6] __aligned(8); > + struct { > + u8 buf[12]; > + s64 ts __aligned(8); We have a new aligned_s64 that is cleaner for these cases. It's only in my togreg branch currently though so you'll need to base on that to use it. Jonathan > + } buffer; > > /* > * DMA (thus cache coherency maintenance) may require the
