On Tue 20-02-24 15:22:07, Huang, Ying wrote: [...] > This isn't an issue now, because mpol_misplaced() are always called with > PTL held. And, we can still keep thiscpu local variable. yes, this is the case but it would be better if we made that assumption official by lockdep_assert_held -- Michal Hocko SUSE Labs