memcg: avoid deadlock caused by race between oom and cpuset_attach

mpol_rebind_mm(), which can be called from cpuset_attach(), does
down_write(mm->mmap_sem).  This means down_write(mm->mmap_sem) can be
called under cgroup_mutex.

OTOH, page fault path does down_read(mm->mmap_sem) and calls
mem_cgroup_try_charge_xxx(), which may eventually calls
mem_cgroup_out_of_memory().  And mem_cgroup_out_of_memory() calls
cgroup_lock().  This means cgroup_lock() can be called under
down_read(mm->mmap_sem).

If those two paths race, deadlock can happen.

This patch avoid this deadlock by:
  - remove cgroup_lock() from mem_cgroup_out_of_memory().
  - define new mutex (memcg_tasklist) and serialize mem_cgroup_move_task()
    (->attach handler of memory cgroup) and mem_cgroup_out_of_memory.

Signed-off-by: Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp>
Reviewed-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Acked-by: Balbir Singh <balbir@linux.vnet.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index 435f08dac8bf6c02a2b71580f8de9dc652fa914f..861037070f6672b3fd9e69d4aa0a796980fb2c7c 100644 (file)
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -51,6 +51,7 @@ static int really_do_swap_account __initdata = 1; /* for remember boot option*/
 #define do_swap_account                (0)
 #endif
 
+static DEFINE_MUTEX(memcg_tasklist);   /* can be hold under cgroup_mutex */
 
 /*
  * Statistics for memory cgroup.
@@ -827,7 +828,9 @@ static int __mem_cgroup_try_charge(struct mm_struct *mm,
 
                if (!nr_retries--) {
                        if (oom) {
+                               mutex_lock(&memcg_tasklist);
                                mem_cgroup_out_of_memory(mem_over_limit, gfp_mask);
+                               mutex_unlock(&memcg_tasklist);
                                mem_over_limit->last_oom_jiffies = jiffies;
                        }
                        goto nomem;
@@ -2211,10 +2214,12 @@ static void mem_cgroup_move_task(struct cgroup_subsys *ss,
                                struct cgroup *old_cont,
                                struct task_struct *p)
 {
+       mutex_lock(&memcg_tasklist);
        /*
         * FIXME: It's better to move charges of this process from old
         * memcg to new memcg. But it's just on TODO-List now.
         */
+       mutex_unlock(&memcg_tasklist);
 }
 
 struct cgroup_subsys mem_cgroup_subsys = {

diff --git a/mm/oom_kill.c b/mm/oom_kill.c
index fd150e3a2567a137eef87c6c76772fd0c34fd112..40ba05061a4fbc8a43d3eb046a58fe756748cd06 100644 (file)
--- a/mm/oom_kill.c
+++ b/mm/oom_kill.c
@@ -429,7 +429,6 @@ void mem_cgroup_out_of_memory(struct mem_cgroup *mem, gfp_t gfp_mask)
        unsigned long points = 0;
        struct task_struct *p;
 
-       cgroup_lock();
        read_lock(&tasklist_lock);
 retry:
        p = select_bad_process(&points, mem);
@@ -444,7 +443,6 @@ retry:
                goto retry;
 out:
        read_unlock(&tasklist_lock);
-       cgroup_unlock();
 }
 #endif