Hi Janusz, On Sun, Jun 02, 2019 at 11:58:23AM +0200, Janusz Krzysztofik wrote: > Hi Sakari, > > On Sunday, June 2, 2019 12:37:55 AM CEST Sakari Ailus wrote: > > > > ... I realised that the subtle effect of "media: > > ov6650: Register with asynchronous subdevice framework" is that the driver > > is now responsible for serialising the access to its own data structures > > now. > > Indeed, I must have been not thinking much while preparing it, only following > patterns from other implementations blindly, sorry.
No worries. I missed it at the time, too... > > > And it doesn't do that. Could you submit a fix, please? It'd be good to > > get that to 5.2 through the fixes branch. > > How about dropping that V4L2_SUBDEV_FL_HAS_DEVNODE flag for now? I think > that > will be the most safe approach for a quick fix. I'd then re-add it together > with proper locking in a separate patch later. What do yo think? Sure. Then we just re-introduce the flag when the driver is ready for that. -- Regards, Sakari Ailus
