From: Frederic Weisbecker <fweisbec@gmail.com>
Date: Thu, 15 Apr 2010 21:10:42 +0000 (+0200)
Subject: lockdep: Fix redundant_hardirqs_on incremented with irqs enabled
X-Git-Url: https://git.stricted.de/?a=commitdiff_plain;h=8795d7717c467bea7b0a0649d44a258e09f34db2;p=GitHub%2FLineageOS%2FG12%2Fandroid_kernel_amlogic_linux-4.9.git

lockdep: Fix redundant_hardirqs_on incremented with irqs enabled

When a path restore the flags while irqs are already enabled, we
update the per cpu var redundant_hardirqs_on in a racy fashion
and debug_atomic_inc() warns about this situation.

In this particular case, loosing a few hits in a stat is not a big
deal, so increment it without protection.

v2: Don't bother with disabling irq, we can miss one count in
    rare situations

Reported-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: David Miller <davem@davemloft.net>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---

diff --git a/kernel/lockdep.c b/kernel/lockdep.c
index 78325f8f1139..1b58a1bbcc87 100644
--- a/kernel/lockdep.c
+++ b/kernel/lockdep.c
@@ -2298,7 +2298,12 @@ void trace_hardirqs_on_caller(unsigned long ip)
 		return;
 
 	if (unlikely(curr->hardirqs_enabled)) {
-		debug_atomic_inc(redundant_hardirqs_on);
+		/*
+		 * Neither irq nor preemption are disabled here
+		 * so this is racy by nature but loosing one hit
+		 * in a stat is not a big deal.
+		 */
+		this_cpu_inc(lockdep_stats.redundant_hardirqs_on);
 		return;
 	}
 	/* we'll do an OFF -> ON transition: */