[media] msp3400: fix mute audio regression

The switch to the new control framework caused a regression where the audio was
no longer unmuted after the carrier scan finished.

The original code attempted to set the volume control to its current value in
order to have the set-volume control code to be called that handles the volume
and muting. However, the framework will not call that code unless the new volume
value is different from the old.

Instead we now call msp_s_ctrl directly.

It is a bit of a hack: we really need a v4l2_ctrl_refresh_ctrl function for this
(or something along those lines).

Thanks to Andy Walls for bisecting this and to Shane Shrybman for reporting it!

Reported-by: Shane Shrybman <shrybman@teksavvy.com>
Thanks-to: Andy Walls <awalls@md.metrocast.net>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

diff --git a/drivers/media/video/msp3400-driver.c b/drivers/media/video/msp3400-driver.c
index fe18a0a..b1763ac 100644 (file)
--- a/drivers/media/video/msp3400-driver.c
+++ b/drivers/media/video/msp3400-driver.c
@@ -381,7 +381,12 @@ static int msp_s_ctrl(struct v4l2_ctrl *ctrl)
 
 void msp_update_volume(struct msp_state *state)
 {
-       v4l2_ctrl_s_ctrl(state->volume, v4l2_ctrl_g_ctrl(state->volume));
+       /* Force an update of the volume/mute cluster */
+       v4l2_ctrl_lock(state->volume);
+       state->volume->val = state->volume->cur.val;
+       state->muted->val = state->muted->cur.val;
+       msp_s_ctrl(state->volume);
+       v4l2_ctrl_unlock(state->volume);
 }
 
 /* --- v4l2 ioctls --- */