rcutorture: Manually clean up after rcu_barrier() failure

Currently, if rcu_barrier() returns too soon, the test waits 100ms and
then does another instance of the test.  However, if rcu_barrier() were
to have waited for more than 100ms too short a time, this could cause
the test's rcu_head structures to be reused while they were still on
RCU's callback lists.  This can result in knock-on errors that obscure
the original rcu_barrier() test failure.

This commit therefore adds code that attempts to wait until all of
the test's callbacks have been invoked.  Of course, if RCU completely
lost track of the corresponding rcu_head structures, this wait could be
forever.  This commit therefore also complains if this attempted recovery
takes more than one second, and it also gives up when the test ends.

Signed-off-by: Paul E. McKenney <paulmck@kernel.org>

1 files changed, 15 insertions, 1 deletions

diff --git a/kernel/rcu/rcutorture.c b/kernel/rcu/rcutorture.c
index f82515cded34..5453bd557f43 100644
--- a/kernel/rcu/rcutorture.c
+++ b/kernel/rcu/rcutorture.c
@@ -2124,7 +2124,21 @@ static int rcu_torture_barrier(void *arg)
 			pr_err("barrier_cbs_invoked = %d, n_barrier_cbs = %d\n",
 			       atomic_read(&barrier_cbs_invoked),
 			       n_barrier_cbs);
-			WARN_ON_ONCE(1);
+			WARN_ON(1);
+			// Wait manually for the remaining callbacks
+			i = 0;
+			do {
+				if (WARN_ON(i++ > HZ))
+					i = INT_MIN;
+				schedule_timeout_interruptible(1);
+				cur_ops->cb_barrier();
+			} while (atomic_read(&barrier_cbs_invoked) !=
+				 n_barrier_cbs &&
+				 !torture_must_stop());
+			smp_mb(); // Can't trust ordering if broken.
+			if (!torture_must_stop())
+				pr_err("Recovered: barrier_cbs_invoked = %d\n",
+				       atomic_read(&barrier_cbs_invoked));
 		} else {
 			n_barrier_successes++;
 		}