On 3/15/19 8:43 PM, Derek Lesho wrote:
Ah, I see. Wouldn't it be cleaner to put a whitelist inside of device.c and see if the incoming object type is on that whitelist? This seems like it would be a more self-contained solution to the problem.
I don't see a clean way to do that. Besides, if we consider being able to be associated with kernel object be a feature of object type, having it inside object type seems right.
We could consider moving the list back to object header, but keep the new op and limit its functionality to allow or not the association. This would allow sharing slightly more code at expense of making all object use more memory (even those not supporting kernel objects). It's probably not a big deal through.
Jacek