From 3ef22dfff2390e75b379f9715388a852aa56e0d5 Mon Sep 17 00:00:00 2001 From: Michal Hocko Date: Thu, 19 May 2016 17:13:12 -0700 Subject: oom, oom_reaper: try to reap tasks which skip regular OOM killer path If either the current task is already killed or PF_EXITING or a selected task is PF_EXITING then the oom killer is suppressed and so is the oom reaper. This patch adds try_oom_reaper which checks the given task and queues it for the oom reaper if that is safe to be done meaning that the task doesn't share the mm with an alive process. This might help to release the memory pressure while the task tries to exit. [akpm@linux-foundation.org: fix nommu build] Signed-off-by: Michal Hocko Cc: Raushaniya Maksudova Cc: Michael S. Tsirkin Cc: Paul E. McKenney Cc: David Rientjes Cc: Tetsuo Handa Cc: Daniel Vetter Cc: Oleg Nesterov Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/oom.h | 8 ++++++++ 1 file changed, 8 insertions(+) (limited to 'include/linux/oom.h') diff --git a/include/linux/oom.h b/include/linux/oom.h index 628a43242a34..83b9c39bd8b7 100644 --- a/include/linux/oom.h +++ b/include/linux/oom.h @@ -72,6 +72,14 @@ static inline bool oom_task_origin(const struct task_struct *p) extern void mark_oom_victim(struct task_struct *tsk); +#ifdef CONFIG_MMU +extern void try_oom_reaper(struct task_struct *tsk); +#else +static inline void try_oom_reaper(struct task_struct *tsk) +{ +} +#endif + extern unsigned long oom_badness(struct task_struct *p, struct mem_cgroup *memcg, const nodemask_t *nodemask, unsigned long totalpages); -- cgit v1.2.3 From 98748bd722005be9de2662bd4f7e41ad8148bdbd Mon Sep 17 00:00:00 2001 From: Michal Hocko Date: Fri, 20 May 2016 16:57:24 -0700 Subject: oom: consider multi-threaded tasks in task_will_free_mem task_will_free_mem is a misnomer for a more complex PF_EXITING test for early break out from the oom killer because it is believed that such a task would release its memory shortly and so we do not have to select an oom victim and perform a disruptive action. Currently we make sure that the given task is not participating in the core dumping because it might get blocked for a long time - see commit d003f371b270 ("oom: don't assume that a coredumping thread will exit soon"). The check can still do better though. We shouldn't consider the task unless the whole thread group is going down. This is rather unlikely but not impossible. A single exiting thread would surely leave all the address space behind. If we are really unlucky it might get stuck on the exit path and keep its TIF_MEMDIE and so block the oom killer. Link: http://lkml.kernel.org/r/1460452756-15491-1-git-send-email-mhocko@kernel.org Signed-off-by: Michal Hocko Cc: Oleg Nesterov Cc: David Rientjes Cc: Tetsuo Handa Cc: Joonsoo Kim Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/oom.h | 15 +++++++++++++-- 1 file changed, 13 insertions(+), 2 deletions(-) (limited to 'include/linux/oom.h') diff --git a/include/linux/oom.h b/include/linux/oom.h index 83b9c39bd8b7..d3f533f2f481 100644 --- a/include/linux/oom.h +++ b/include/linux/oom.h @@ -110,13 +110,24 @@ extern struct task_struct *find_lock_task_mm(struct task_struct *p); static inline bool task_will_free_mem(struct task_struct *task) { + struct signal_struct *sig = task->signal; + /* * A coredumping process may sleep for an extended period in exit_mm(), * so the oom killer cannot assume that the process will promptly exit * and release memory. */ - return (task->flags & PF_EXITING) && - !(task->signal->flags & SIGNAL_GROUP_COREDUMP); + if (sig->flags & SIGNAL_GROUP_COREDUMP) + return false; + + if (!(task->flags & PF_EXITING)) + return false; + + /* Make sure that the whole thread group is going down */ + if (!thread_group_empty(task) && !(sig->flags & SIGNAL_GROUP_EXIT)) + return false; + + return true; } /* sysctls */ -- cgit v1.2.3 From c96fc2d85f4a827e3bb2abe7de2394a1fb8a0fe7 Mon Sep 17 00:00:00 2001 From: Tetsuo Handa Date: Mon, 23 May 2016 16:23:57 -0700 Subject: signal: make oom_flags a bool Currently the size of "struct signal_struct"->oom_flags member is sizeof(unsigned) bytes, but only one flag OOM_FLAG_ORIGIN which is updated by current thread is defined. We can convert OOM_FLAG_ORIGIN into a bool, and reuse the saved bytes for updating from the OOM killer and/or the OOM reaper thread. By the way, do we care about a race window between run_store() and swapoff() because it would be theoretically possible that two threads sharing the "struct signal_struct" concurrently call respective functions? If we care, we can make oom_flags an atomic_t. Signed-off-by: Tetsuo Handa Cc: Hugh Dickins Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/oom.h | 9 +++------ 1 file changed, 3 insertions(+), 6 deletions(-) (limited to 'include/linux/oom.h') diff --git a/include/linux/oom.h b/include/linux/oom.h index d3f533f2f481..83469522690a 100644 --- a/include/linux/oom.h +++ b/include/linux/oom.h @@ -50,24 +50,21 @@ enum oom_scan_t { OOM_SCAN_SELECT, /* always select this thread first */ }; -/* Thread is the potential origin of an oom condition; kill first on oom */ -#define OOM_FLAG_ORIGIN ((__force oom_flags_t)0x1) - extern struct mutex oom_lock; static inline void set_current_oom_origin(void) { - current->signal->oom_flags |= OOM_FLAG_ORIGIN; + current->signal->oom_flag_origin = true; } static inline void clear_current_oom_origin(void) { - current->signal->oom_flags &= ~OOM_FLAG_ORIGIN; + current->signal->oom_flag_origin = false; } static inline bool oom_task_origin(const struct task_struct *p) { - return !!(p->signal->oom_flags & OOM_FLAG_ORIGIN); + return p->signal->oom_flag_origin; } extern void mark_oom_victim(struct task_struct *tsk); -- cgit v1.2.3