Hi Hans, On Thu, Nov 16, 2023 at 04:29:58PM +0100, Hans Verkuil wrote: > Hi Sakari, > > On 12/11/2023 21:14, Sakari Ailus wrote: > > Hi Mauro, Hans, > > > > Here's a bunch of patches from myself and Laurent, cleaning up V4L2 > > sub-device state information access API, drivers using it plus fixing IVSC > > driver firmware parsing and IMX319 driver. A few CCS driver fixes and > > documentation improvement are included, too. > > > > Please pull. > > > > since v2: > > > > - Fix sub-device state access conditional compilation (was dependent on > > sub-device UAPI). > > > > since v1: > > > > - Fix IVSC driver patch error handling. > > > > > > The following changes since commit 3e238417254bfdcc23fe207780b59cbb08656762: > > > > media: nuvoton: VIDEO_NPCM_VCD_ECE should depend on ARCH_NPCM (2023-10-27 11:44:19 +0200) > > > > are available in the Git repository at: > > > > git://linuxtv.org/sailus/media_tree.git tags/for-6.8-1.2-signed > > > > for you to fetch changes up to 61ab89c1916219ccbeca9f6cefb1680212de8d1c: > > > > media: v4l: subdev: Return NULL from pad access functions on error (2023-11-12 20:51:35 +0200) > > As mentioned on irc, this PR no longer applies after the starfive driver > was merged. > > I did a run for this PR with the starfive patches dropped, and I got a few > kerneldoc warnings: > > include/media/v4l2-subdev.h:1425: warning: Excess function parameter 'state' description in 'v4l2_subdev_state_get_format' > include/media/v4l2-subdev.h:1425: warning: Excess function parameter 'pad' description in 'v4l2_subdev_state_get_format' > include/media/v4l2-subdev.h:1445: warning: Excess function parameter 'state' description in 'v4l2_subdev_state_get_crop' > include/media/v4l2-subdev.h:1445: warning: Excess function parameter 'pad' description in 'v4l2_subdev_state_get_crop' > include/media/v4l2-subdev.h:1465: warning: Excess function parameter 'state' description in 'v4l2_subdev_state_get_compose' > include/media/v4l2-subdev.h:1465: warning: Excess function parameter 'pad' description in 'v4l2_subdev_state_get_compose' > > Can you fix this for the next version? It seems how the macro was designed in the last version of the sub-device state information access functions caused this. I'll figure out how to address this. Thanks. -- Regards, Sakari Ailus
