Le 09/11/2023 à 10:43, Tomasz Figa a écrit :

On Tue, Oct 31, 2023 at 05:31:04PM +0100, Benjamin Gaignard wrote:

quoted

Allow test drivers to use DELETE_BUFS by adding vb2_ioctl_delete_bufs() helper.

Signed-off-by: Benjamin Gaignard <benjamin.gaignard@collabora.com>
---
  drivers/media/test-drivers/vicodec/vicodec-core.c |  2 ++
  drivers/media/test-drivers/vimc/vimc-capture.c    |  2 ++
  drivers/media/test-drivers/visl/visl-video.c      |  2 ++
  drivers/media/test-drivers/vivid/vivid-core.c     | 13 ++++++++++---
  4 files changed, 16 insertions(+), 3 deletions(-)

diff --git a/drivers/media/test-drivers/vicodec/vicodec-core.c b/drivers/media/test-drivers/vicodec/vicodec-core.c
index 69cbe2c094e1..f14a8fd506d0 100644
--- a/drivers/media/test-drivers/vicodec/vicodec-core.c
+++ b/drivers/media/test-drivers/vicodec/vicodec-core.c

@@ -1339,6 +1339,7 @@ static const struct v4l2_ioctl_ops vicodec_ioctl_ops = {
  	.vidioc_prepare_buf	= v4l2_m2m_ioctl_prepare_buf,
  	.vidioc_create_bufs	= v4l2_m2m_ioctl_create_bufs,
  	.vidioc_expbuf		= v4l2_m2m_ioctl_expbuf,
+	.vidioc_delete_bufs	= v4l2_m2m_ioctl_delete_bufs,
  
  	.vidioc_streamon	= v4l2_m2m_ioctl_streamon,
  	.vidioc_streamoff	= v4l2_m2m_ioctl_streamoff,

@@ -1725,6 +1726,7 @@ static int queue_init(void *priv, struct vb2_queue *src_vq,
  	dst_vq->mem_ops = &vb2_vmalloc_memops;
  	dst_vq->timestamp_flags = V4L2_BUF_FLAG_TIMESTAMP_COPY;
  	dst_vq->lock = src_vq->lock;
+	dst_vq->supports_delete_bufs = true;

Since we have to explicitly provide the vidioc_delete_bufs callback anyway,
is there any value in having a separate supports_delete_bufs flag? Or we
envision that some drivers would support deleting buffers only for some
queues?

That exactly the case for Hantro driver, it can support deleting buffers on
capture queue but not on output queue.

Best regards,
Tomasz

_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel