iio: accel: kionix-kx022a: use iio_push_to_buffers_with_ts()
Subsystems (2)
PATCH 1/1 iio: accel: kionix-kx022a: use iio_push_to_buffers_with_ts()
Replace the deprecated iio_push_to_buffers_with_timestamp() with iio_push_to_buffers_with_ts(), which takes the destination buffer size and checks it against scan_bytes at runtime.
Both push sites already use suitably sized and aligned storage: the FIFO flush path pushes data->scan ({ __le16 channels[3]; aligned_s64 ts; }), and the trigger handler pushes data->buffer (__le16[8], aligned to IIO_DMA_MINALIGN). Both are 16 bytes, which matches scan_bytes for the 3-axis plus s64 timestamp layout used by all supported variants, so pass sizeof() for each.
--- drivers/iio/accel/kionix-kx022a.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-)
drivers/iio/accel/kionix-kx022a.c +4 -2
@@ -864,7 +864,8 @@ static int __kx022a_fifo_flush(struct iio_dev *idev, unsigned int samples,for_each_set_bit (bit, idev->active_scan_mask,AXIS_MAX ) chs[bit] = sam[bit]; -iio_push_to_buffers_with_timestamp (idev, &data->scan, tstamp); +iio_push_to_buffers_with_ts (idev, &data->scan, +sizeof (data->scan), tstamp); tstamp += sample_period; }
@@ -1034,7 +1035,8 @@ static irqreturn_t kx022a_trigger_handler(int irq, void *p)if (ret <0 ) goto err_read; -iio_push_to_buffers_with_timestamp (idev, data->buffer, data->timestamp); +iio_push_to_buffers_with_ts (idev, data->buffer,sizeof (data->buffer), + data->timestamp); err_read:iio_trigger_notify_done (idev->trig);
Signature · 1 line
2.50.1 (Apple Git-155)
Jonathan Cameron
maintainerQuoted 10 lines · On Tue, 18 Aug 2026 16:56:35 +0100 Gabriel Rondon <[email protected]> wrote:
Replace the deprecated
iio_push_to_buffers_with_timestamp()withiio_push_to_buffers_with_ts(), which takes the destination buffer size and checks it againstscan_bytesat runtime.Both push sites already use suitably sized and aligned storage: the FIFO flush path pushes
data->scan({__le16channels[3];aligned_s64ts; }), and the trigger handler pushesdata->buffer(__le16[8], aligned toIIO_DMA_MINALIGN). Both are 16 bytes, which matchesscan_bytesfor the 3-axis plus s64 timestamp layout used by all supported variants, so passsizeof()for each.
As is quite common with this sort of cleanup, it got me looking at code that has evolved in an odd direction. In this case it was you pointing out both buffers are same size and have same content!
Why do we have both scan and buffer? The kx022a_trigger_handler() could just as easily use scan (which is the cleaner solution here)
That leaves get_axis which uses st->buffer as well. Whilst it is a little ugly, that can just use the first channel of scan as well.
If you are happy to do so could you make that change as a precursor to this set? I don't particularly want to leave that mess in place now I've noticed it.
Thanks,
Jonathan
Quoted 30 lines
Signed-off-by Gabriel Rondon <[email protected]>--- drivers/iio/accel/kionix-kx022a.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-)drivers/iio/accel/kionix-kx022a.c +4
-2@@ -864,7 +864,8 @@ static int __kx022a_fifo_flush(struct iio_dev *idev, unsigned int samples,for_each_set_bit (bit, idev->active_scan_mask,AXIS_MAX ) chs[bit] = sam[bit]; -iio_push_to_buffers_with_timestamp (idev, &data->scan, tstamp); +iio_push_to_buffers_with_ts (idev, &data->scan, +sizeof (data->scan), tstamp); tstamp += sample_period; }@@ -1034,7 +1035,8 @@ static irqreturn_t kx022a_trigger_handler(int irq, void *p)if (ret <0 ) goto err_read; -iio_push_to_buffers_with_timestamp (idev, data->buffer, data->timestamp); +iio_push_to_buffers_with_ts (idev, data->buffer,sizeof (data->buffer), + data->timestamp); err_read:iio_trigger_notify_done (idev->trig);
Quoted 8 lines · On Tue, 18 Aug 2026 17:16:57 +0100 Jonathan Cameron <[email protected]> wrote:
Why do we have both scan and buffer? The
kx022a_trigger_handler()could just as easily use scan (which is the cleaner solution here)That leaves
get_axiswhich usesst->bufferas well. Whilst it is a little ugly, that can just use the first channel of scan as well.If you are happy to do so could you make that change as a precursor to this set?
Happy to. v2 splits it into a precursor that drops buffer and routes both the one-shot read and the triggered handler through scan (moving the IIO_DMA_MINALIGN alignment onto scan), then the conversion on top.
Thanks,
Gabriel