On Tue 08-10-19 06:04:32, Qian Cai wrote: > > > > On Oct 8, 2019, at 4:40 AM, Michal Hocko <mhocko@xxxxxxxxxx> wrote: > > > > Does this tip point to a real deadlock or merely a class of lockdep > > false dependencies? > > I lean towards it is a real deadlock given how trivial to generate those lock orders everywhere. Have you actually triggered any real deadlock? With a zone->lock in place it would be pretty clear with hard lockups detected. -- Michal Hocko SUSE Labs
