Re: [RFC PATCH -v2] mm, oom: introduce oom reaper

[Tetsuo Handa]

Tetsuo Handa wrote:
> > Users of mmap_sem which need it for write should be carefully reviewed
> > to use _killable waiting as much as possible and reduce allocations
> > requests done with the lock held to absolute minimum to reduce the risk
> > even further.
> 
> It will be nice if we can have down_write_killable()/down_read_killable().

It will be nice if we can also have __GFP_KILLABLE. Although currently it can't
be perfect because reclaim functions called from __alloc_pages_slowpath() use
unkillable waits, starting from just bail out as with __GFP_NORETRY when
fatal_signal_pending(current) is true will be helpful.

So far I'm hitting no problem with testers except the one using mmap()/munmap().

I think that cmpxchg() was not needed.

diff --git a/mm/oom_kill.c b/mm/oom_kill.c
index c2ab7f9..1a65739 100644
--- a/mm/oom_kill.c
+++ b/mm/oom_kill.c
@@ -483,8 +483,6 @@ static int oom_reaper(void *unused)
 
 static void wake_oom_reaper(struct mm_struct *mm)
 {
-	struct mm_struct *old_mm;
-
 	if (!oom_reaper_th)
 		return;
 
@@ -492,14 +490,15 @@ static void wake_oom_reaper(struct mm_struct *mm)
 	 * Make sure that only a single mm is ever queued for the reaper
 	 * because multiple are not necessary and the operation might be
 	 * disruptive so better reduce it to the bare minimum.
+	 * Caller is serialized by oom_lock mutex.
 	 */
-	old_mm = cmpxchg(&mm_to_reap, NULL, mm);
-	if (!old_mm) {
+	if (!mm_to_reap) {
 		/*
 		 * Pin the given mm. Use mm_count instead of mm_users because
 		 * we do not want to delay the address space tear down.
 		 */
 		atomic_inc(&mm->mm_count);
+		mm_to_reap = mm;
 		wake_up(&oom_reaper_wait);
 	}
 }
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/