Hi Sakari,
Since you are planning to PR this patch, I will continue to use
CUR_LINK_FREQ to initialize the link_freq v4l2_ctl.
Besides, I see your patch in intel/ipu6 that obtains the link frequency
from the sub-device instead of a control handler.
Maybe I don't need init link_freq v4l2_ctl anymore?
Thanks,
Dongcheng
On 12/12/2024 5:12 AM, Sakari Ailus wrote:
Hi Dongcheng,
On Wed, Dec 11, 2024 at 10:39:53PM +0800, Yan, Dongcheng wrote:
+static int lt6911uxe_init_controls(struct lt6911uxe *lt6911uxe)
+{
+ struct v4l2_ctrl_handler *ctrl_hdlr;
+ s64 pixel_rate;
+ int ret;
+
+ ctrl_hdlr = <6911uxe->ctrl_handler;
+ ret = v4l2_ctrl_handler_init(ctrl_hdlr, 8);
+ if (ret)
+ return ret;
+
+ lt6911uxe->link_freq =
+ v4l2_ctrl_new_int_menu(ctrl_hdlr, NULL, V4L2_CID_LINK_FREQ,
+ sizeof(lt6911uxe->cur_mode.link_freq),
There's just a single entry in the array. sizeof() will get the size in
bytes, which is 8! Use 1 here instead.
Hi Sakari, I used CUR_LINK_FREQ before, which refers to your patch:
<URL:https://lore.kernel.org/linux-media/20240426145538.654212-1-sakari.ailus@xxxxxxxxxxxxxxx/>.
There were a few iterations of the set and this is
where it eventually ended:
<URL:https://lore.kernel.org/linux-media/20241210075906.609490-1-sakari.ailus@xxxxxxxxxxxxxxx/T/#t>
I intend to include that in a PR soonish.
But I am worried about the progress of this patch.
If you have expectations for it, I can modify it to CUR_LINK_FREQ.