On Thu, Dec 13, 2018 at 11:25 PM Nick Desaulniers <ndesaulni...@google.com> wrote: > > Moving the __KERNEL__ guard should not affect the kernel, only what > userspace sees. __gnu_inline only affects which > implementation/definition you get, so even if userspace doesn't know > what the kernel's inline is redefined to, it should not matter as > userspace should only ever care about the function signature, which > does not change between our definitions of inline.
Hm... I am unsure what you mean by this. Were you replying to me or to the original patch? Cheers, Miguel