On Wed, Feb 05, 2020 at 06:42:41PM +0200, Felipe Balbi wrote: > > Overall, supporting O_NONBLOCK might be a useful feature for people > > who are doing something else other than fuzzing, We can account for > > potential future extensions that'll support it, so detecting > > O_NONBLOCK and returning an error for now makes sense. > > > > WDYT? > > If that's the way you want to go, that's okay. But let's, then, prepare > the code for extension later on. For example, let's add an IOCTL which > returns the "version" of the ABI. Based on that, userspace can detect > features and so on. Ick, no, no version mess. If you have a new api, just add a new ioctl and away you go, userspace can easily test for that. Don't go down the path of trying to version your api, that way never works. Trust me, been there, got the t-shirt, lived to regret it. greg k-h
