On Fri, 2019-09-06 at 22:41 +0900, Tetsuo Handa wrote: > On 2019/09/06 22:08, Qian Cai wrote: > > Yes, mlocked is troublesome. I have other incidents where crond and systemd- > > udevd were killed by mistake, > > Yes. How to mitigate this regression is a controversial topic. > Michal thinks that we should make mlocked pages reclaimable, but > we haven't reached there. I think that we can monitor whether > counters decay over time (with timeout), but Michal refuses any > timeout based approach. We are deadlocked there. > > > and it even tried to kill kworker/0. > > No. The OOM killer never tries to kill kernel threads like kworker/0. I meant, [40040.401575] kworker/0:1 invoked oom-killer: gfp_mask=0x40cc0(GFP_KERNEL|__GFP_COMP), order=2, oom_score_adj=0 > > > > > https://cailca.github.io/files/dmesg.txt > >
