Towards the end of __oom_kill_task() we see the following comment/code:
/*
* We give our sacrificial lamb high priority and access to
* all the memory it needs. That way it should be able to
* exit() and clear out its resources quickly...
*/
p->time_slice = HZ;
set_tsk_thread_flag(p, TIF_MEMDIE);
force_sig(SIGKILL, p);
I see it getting a large timeslice and access to memory, but I don't
actually see the priority getting bumped at all. It appears the comment
is inaccurate.
Should the process actually get its priority bumped up as well so that
it can process its own death faster? The reason I ask is that we're
seeing some oom-killed processes taking a very long time (multiple
seconds) to actually die.
Chris