> Christian's version using the copy_struct_{from,to}_user()
> aims to avoid most of the problems. The main downside I see
> here is the extra complexity in the kernel. As far as I can
> tell, this has mainly led to extra kernel bugs but has not
> actually resulted in any structure getting seamlessly
> extended.
We extended ioctls multiple times seemlessly and other than this bug
right here I'm not aware of anything serious. Not liking it is fine of
course but saying "this caused a bug so go away" I won't take all too
seriously, sorry.
I don't want to go down the road of structure revisions for stuff in the
generic layer. Others can do whatever they see fit ofc and userspace can
then have its usualy ifdeffery and structure layout detection party
instead of a clean generic solution. I'd rather clean up the necessary
vetting bits and properly document how this can be done.
