diff --git a/fs/exec.c b/fs/exec.c
index b8ee842..fe2873b 100644
--- a/fs/exec.c
+++ b/fs/exec.c
@@ -1604,6 +1604,8 @@
 
 	init_completion(&mm->core_done);
 	init_completion(&core_state->startup);
+	core_state->dumper.task = tsk;
+	core_state->dumper.next = NULL;
 	core_waiters = zap_threads(tsk, mm, core_state, exit_code);
 	up_write(&mm->mmap_sem);
 
diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h
index ae99a28..4d0d0ab 100644
--- a/include/linux/mm_types.h
+++ b/include/linux/mm_types.h
@@ -159,8 +159,14 @@
 #endif
 };
 
+struct core_thread {
+	struct task_struct *task;
+	struct core_thread *next;
+};
+
 struct core_state {
 	atomic_t nr_threads;
+	struct core_thread dumper;
 	struct completion startup;
 };
 
diff --git a/kernel/exit.c b/kernel/exit.c
index 63d8295..b66f0d5 100644
--- a/kernel/exit.c
+++ b/kernel/exit.c
@@ -664,6 +664,7 @@
 static void exit_mm(struct task_struct * tsk)
 {
 	struct mm_struct *mm = tsk->mm;
+	struct core_state *core_state;
 
 	mm_release(tsk, mm);
 	if (!mm)
@@ -676,11 +677,19 @@
 	 * group with ->mm != NULL.
 	 */
 	down_read(&mm->mmap_sem);
-	if (mm->core_state) {
+	core_state = mm->core_state;
+	if (core_state) {
+		struct core_thread self;
 		up_read(&mm->mmap_sem);
 
-		if (atomic_dec_and_test(&mm->core_state->nr_threads))
-			complete(&mm->core_state->startup);
+		self.task = tsk;
+		self.next = xchg(&core_state->dumper.next, &self);
+		/*
+		 * Implies mb(), the result of xchg() must be visible
+		 * to core_state->dumper.
+		 */
+		if (atomic_dec_and_test(&core_state->nr_threads))
+			complete(&core_state->startup);
 
 		wait_for_completion(&mm->core_done);
 		down_read(&mm->mmap_sem);