summaryrefslogtreecommitdiff
path: root/drivers/media/platform/qcom/camss/camss-vfe.c
diff options
context:
space:
mode:
authorMarek Vasut <marex@denx.de>2023-05-18 15:36:49 +0200
committerHans Verkuil <hverkuil-cisco@xs4all.nl>2023-05-25 16:21:21 +0200
commit26ae58f65e64fa7ba61d64bae752e59e08380c6a (patch)
tree954e35f61f7bce96d0b7b9d10eaa757349ba2b09 /drivers/media/platform/qcom/camss/camss-vfe.c
parentc82d4d9353b02bf531ff43654863e7497a1caebe (diff)
media: videodev2.h: Fix struct v4l2_input tuner index comment
VIDIOC_ENUMINPUT documentation describes the tuner field of struct v4l2_input as index: Documentation/userspace-api/media/v4l/vidioc-enuminput.rst " * - __u32 - ``tuner`` - Capture devices can have zero or more tuners (RF demodulators). When the ``type`` is set to ``V4L2_INPUT_TYPE_TUNER`` this is an RF connector and this field identifies the tuner. It corresponds to struct :c:type:`v4l2_tuner` field ``index``. For details on tuners see :ref:`tuner`. " Drivers I could find also use the 'tuner' field as an index, e.g.: drivers/media/pci/bt8xx/bttv-driver.c bttv_enum_input() drivers/media/usb/go7007/go7007-v4l2.c vidioc_enum_input() However, the UAPI comment claims this field is 'enum v4l2_tuner_type': include/uapi/linux/videodev2.h This field being 'enum v4l2_tuner_type' is unlikely as it seems to be never used that way in drivers, and documentation confirms it. It seem this comment got in accidentally in the commit which this patch fixes. Fix the UAPI comment to stop confusion. This was pointed out by Dmitry while reviewing VIDIOC_ENUMINPUT support for strace. Fixes: 6016af82eafc ("[media] v4l2: use __u32 rather than enums in ioctl() structs") Signed-off-by: Marek Vasut <marex@denx.de> Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Diffstat (limited to 'drivers/media/platform/qcom/camss/camss-vfe.c')
0 files changed, 0 insertions, 0 deletions