Arkadiusz reported that enabling memcg's group oom killing causes strange memcg statistics where there is no task in a memcg despite the number of tasks in that memcg is not 0. It turned out that there is a bug in wake_oom_reaper() which allows enqueuing same task twice which makes impossible to decrease the number of tasks in that memcg due to a refcount leak. This bug existed since the OOM reaper became invokable from task_will_free_mem(current) path in out_of_memory() in Linux 4.7, T1@P1 |T2@P1 |T3@P1 |OOM reaper ----------+----------+----------+------------ # Processing an OOM victim in a different memcg domain. try_charge() mem_cgroup_out_of_memory() mutex_lock(&oom_lock) try_charge() mem_cgroup_out_of_memory() mutex_lock(&oom_lock) try_charge() mem_cgroup_out_of_memory() mutex_lock(&oom_lock) out_of_memory() oom_kill_process(P1) do_send_sig_info(SIGKILL, @P1) mark_oom_victim(T1@P1) wake_oom_reaper(T1@P1) # T1@P1 is enqueued. mutex_unlock(&oom_lock) out_of_memory() mark_oom_victim(T2@P1) wake_oom_reaper(T2@P1) # T2@P1 is enqueued. mutex_unlock(&oom_lock) out_of_memory() mark_oom_victim(T1@P1) wake_oom_reaper(T1@P1) # T1@P1 is enqueued again due to oom_reaper_list == T2@P1 && T1@P1->oom_reaper_list == NULL. mutex_unlock(&oom_lock) # Completed processing an OOM victim in a different memcg domain. spin_lock(&oom_reaper_lock) # T1P1 is dequeued. spin_unlock(&oom_reaper_lock) but memcg's group oom killing made it easier to trigger this bug by calling wake_oom_reaper() on the same task from one out_of_memory() request. Fix this bug using an approach used by commit 855b018325737f76 ("oom, oom_reaper: disable oom_reaper for oom_kill_allocating_task"). As a side effect of this patch, this patch also avoids enqueuing multiple threads sharing memory via task_will_free_mem(current) path.
Unified · Split
| @@ -71,6 +71,7 @@ static inline int get_dumpable(struct mm_struct *mm) · 70 unmodified lines | |||
| 71 | #define MMF_HUGE_ZERO_PAGE 23 /* mm has ever used the global huge zero page */ | 71 | #define MMF_HUGE_ZERO_PAGE 23 /* mm has ever used the global huge zero page */ |
| 72 | #define MMF_DISABLE_THP 24 /* disable THP for all VMAs */ | 72 | #define MMF_DISABLE_THP 24 /* disable THP for all VMAs */ |
| 73 | #define MMF_OOM_VICTIM 25 /* mm is the oom victim */ | 73 | #define MMF_OOM_VICTIM 25 /* mm is the oom victim */ |
| 74 | #define MMF_OOM_REAP_QUEUED 26 /* mm was queued for oom_reaper */ | ||
| 74 | #define MMF_DISABLE_THP_MASK (1 << MMF_DISABLE_THP) | 75 | #define MMF_DISABLE_THP_MASK (1 << MMF_DISABLE_THP) |
| 75 | 76 | ||
| 76 | #define MMF_INIT_MASK (MMF_DUMPABLE_MASK | MMF_DUMP_FILTER_MASK |\ | 77 | #define MMF_INIT_MASK (MMF_DUMPABLE_MASK | MMF_DUMP_FILTER_MASK |\ |
| @@ -647,8 +647,8 @@ static int oom_reaper(void *unused) · 646 unmodified lines | |||
| 647 | 647 | ||
| 648 | static void wake_oom_reaper(struct task_struct *tsk) | 648 | static void wake_oom_reaper(struct task_struct *tsk) |
| 649 | { | 649 | { |
| 650 | /* tsk is already queued? */ | 650 | /* mm is already queued? */ |
| 651 | if (tsk == oom_reaper_list || tsk->oom_reaper_list) | 651 | if (test_and_set_bit(MMF_OOM_REAP_QUEUED, &tsk->signal->oom_mm->flags)) |
| 652 | return; | 652 | return; |
| 653 | 653 | ||
| 654 | get_task_struct(tsk); | 654 | get_task_struct(tsk); |