Now that synchronize_rcu() waits for preempt-disable regions of code
in addition to explicitly marked RCU read-side critical sections,
synchronize_rcu() can be used in place of synchronize_sched(). This
commit therefore makes that change.
Signed-off-by: Paul E. McKenney <paul...@linux.ibm.com>
Cc: Robert Richter <r...@kernel.org>
Cc: "David S. Miller" <da...@davemloft.net>
Cc: <oprofile-l...@lists.sf.net>
Cc: <sparcli...@vger.kernel.org>
---
arch/sparc/oprofile/init.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/arch/sparc/oprofile/init.c b/arch/sparc/oprofile/init.c
index f9024bccff16..43730c9b1c86 100644
--- a/arch/sparc/oprofile/init.c
+++ b/arch/sparc/oprofile/init.c
@@ -53,7 +53,7 @@ static void timer_stop(void)
{
nmi_adjust_hz(1);
unregister_die_notifier(&profile_timer_exceptions_nb);
- synchronize_sched(); /* Allow already-started NMIs to complete. */
+ synchronize_rcu(); /* Allow already-started NMIs to complete. */
}
static int op_nmi_timer_init(struct oprofile_operations *ops)
--
2.17.1
