> Hm, it just looks like you quoted my message and didn't write anything > new. Did something not go through correctly? :-/ ... I should learn to use my mailer correctly ... Should have been that one: ==============================[ cut here ]============================== To: Greg KH <greg@xxxxxxxxx> Subject: Re: [RFC/PATCH] sysfs'ify video4linux In-Reply-To: <20030717214907.GA3255@xxxxxxxxx> > > Version (1) can be done without breaking the build, with a hack along > > the lines "if (no release callback) printk(KERN_WARN please fix your > > driver)", so the drivers can be fixed step-by-step afterwards. > > That sounds like a nice way to start. > I don't think it will be that bad for them. Just have them change the > v4l device from being a structure included in their structure, into a > pointer, and then create it before registering, and free it in the > release() callback. Good point. So the mandatory ->release() callback version is also the more flexible one. I like that :) > Breaking the build is a very good thing to do at times, to ensure that > stuff gets fixed properly. Users might go for a while without realizing > that there really is a problem in their driver. > > But in the end, it's up to you... Breaking the build _right now_ with 2.6 becoming stable is IMHO not a good idea, I think I better try to avoid that until 2.7. New patch will come later today or early next week. Gerd -- sigfault