On 02/25/15 14:57, Sakari Ailus wrote: > Hi Hans, > > On Wed, Feb 25, 2015 at 02:45:58PM +0100, Hans Verkuil wrote: > ... >>> @@ -1535,15 +1529,15 @@ static int smiapp_set_stream(struct v4l2_subdev *subdev, int enable) >>> if (sensor->streaming == enable) >>> goto out; >>> >>> - if (enable) { >>> - sensor->streaming = true; >>> + if (enable) >>> rval = smiapp_start_streaming(sensor); >>> - if (rval < 0) >>> - sensor->streaming = false; >>> - } else { >>> + else >>> rval = smiapp_stop_streaming(sensor); >>> - sensor->streaming = false; >>> - } >>> + >>> + sensor->streaming = enable; >>> + __v4l2_ctrl_grab(sensor->hflip, enable); >>> + __v4l2_ctrl_grab(sensor->vflip, enable); >>> + __v4l2_ctrl_grab(sensor->link_freq, enable); >> >> Just checking: is it really not possible to change these controls >> while streaming? Most devices I know of allow changing this on the fly. >> >> If it is really not possible, then you can add my Ack for this series: > > I'm not sure what the sensors would do in practice, but the problem is that > changing the values of these control affect the pixel order. That's why > changing them has been prevented while streaming. Ah, OK. Can you add a comment explaining why this is done? BTW, I understand that HFLIP will cause changes in the pixel order, but VFLIP and link_freq should be OK, I would expect. Regards, Hans -- To unsubscribe from this list: send the line "unsubscribe linux-media" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html