> On Dec 17, 2019, at 10:13 AM, Michal Hocko <mhocko@xxxxxxxxxx> wrote: > > I do understand the general purpose of the warning. I am simply not sure > the kernel tree is a good candidate with a huge number of different > config combinations that might easily result in warnings which would > tend to result in even more ifdeferry than we have. Yes, compiling test without real-world use case is evil. Once we ignore the evil, it becomes much more manageable.
