Em Thu, 27 Jul 2023 08:42:43 +0300 Sakari Ailus <sakari.ailus@xxxxxxxxxxxxxxx> escreveu: > When an async sub-device is registered, it used to be that the first one > of its connections were matched when found. Continue looking for matches > until a notifier no longer has any. > > Signed-off-by: Sakari Ailus <sakari.ailus@xxxxxxxxxxxxxxx> > Tested-by: Philipp Zabel <p.zabel@xxxxxxxxxxxxxx> # imx6qp > Tested-by: Niklas Söderlund <niklas.soderlund@xxxxxxxxxxxx> # rcar + adv746x > Tested-by: Aishwarya Kothari <aishwarya.kothari@xxxxxxxxxxx> # Apalis i.MX6Q with TC358743 > Tested-by: Lad Prabhakar <prabhakar.mahadev-lad.rj@xxxxxxxxxxxxxx> # Renesas RZ/G2L SMARC > --- > drivers/media/v4l2-core/v4l2-async.c | 4 ++-- > 1 file changed, 2 insertions(+), 2 deletions(-) > > diff --git a/drivers/media/v4l2-core/v4l2-async.c b/drivers/media/v4l2-core/v4l2-async.c > index 44f72aa75c19..6b3c02d27ebf 100644 > --- a/drivers/media/v4l2-core/v4l2-async.c > +++ b/drivers/media/v4l2-core/v4l2-async.c > @@ -789,6 +789,7 @@ int v4l2_async_register_subdev(struct v4l2_subdev *sd) > if (!v4l2_dev) > continue; > > +again: > asc = v4l2_async_find_match(notifier, sd); > if (!asc) > continue; > @@ -801,13 +802,12 @@ int v4l2_async_register_subdev(struct v4l2_subdev *sd) > if (ret) > goto err_unbind; > > - goto out_unlock; > + goto again; > } Using a goto here instead of a do {} while is not nice. I'll merge it as-is, but please send a followup patch using a proper loop. Regards Thanks, Mauro