On Tue 14-07-20 13:06:48, Tetsuo Handa wrote: > On 2020/07/14 11:58, Yafang Shao wrote: [...] > > But I find that we doesn't sched it out in pagefault path, > > > > pagefault_out_of_memory > > if (!mutex_trylock(&oom_lock)) > > return; > > > > I haven't thought deeply what the difference is ... > > David Rientjes is proposing it for avoiding soft lockup, and Michal Hocko is refusing it. > How to give the OOM killer/reaper enough CPU time for reclaiming memory is a dogfight. :-( > > https://lkml.kernel.org/r/alpine.DEB.2.21.2003181458100.70237@xxxxxxxxxxxxxxxxxxxxxxxxx I believe there was a review feedback to be addressed and never followed up. -- Michal Hocko SUSE Labs
