On Fri 09-11-18 19:24:48, Tetsuo Handa wrote: > On 2018/11/09 19:10, Vlastimil Babka wrote:>>>> + * reclaim >= MAX_ORDER areas which will never succeed. Callers may > >>>> + * be using allocators in order of preference for an area that is > >>>> + * too large. > >>>> + */ > >>>> + if (order >= MAX_ORDER) { > >>> > >>> Also, why not to add BUG_ON(gfp_mask & __GFP_NOFAIL); here? > >> > >> Because we do not want to blow up the kernel just because of a stupid > >> usage of the allocator. Can you think of an example where it would > >> actually make any sense? > >> > >> I would argue that such a theoretical abuse would blow up on an > >> unchecked NULL ptr access. Isn't that enough? > > > > Agreed. > > > > If someone has written a module with __GFP_NOFAIL for an architecture > where PAGE_SIZE == 2048KB, and someone else tried to use that module on > another architecture where PAGE_SIZE == 4KB. You are saying that > triggering NULL pointer dereference is a fault of that user's ignorance > about MM. You are saying that everyone knows internal of MM. Sad... What kind of argument is this? Seriously! We do consider GFP_NOFAIL problematic even for !order-0 requests and warn appropriately. Talking about anything getting close to MAX_ORDER is just a crazy talk. In any case this is largely tangential to the issue reported here. -- Michal Hocko SUSE Labs