Blob Blame History Raw
From: "Paul E. McKenney" <paulmck@kernel.org>
Date: Fri, 3 Jan 2020 12:12:06 -0800
Subject: rcu: Add WRITE_ONCE to rcu_node ->exp_seq_rq store
Patch-mainline: v5.7-rc1
Git-commit: b0c18c87730a4de6da0303fa99aea43e814233f9
References: bsc#1171828

The rcu_node structure's ->exp_seq_rq field is read locklessly, so
this commit adds the WRITE_ONCE() to a load in order to provide proper
documentation and READ_ONCE()/WRITE_ONCE() pairing.

This data race was reported by KCSAN.  Not appropriate for backporting
due to failure being unlikely.

Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Acked-by: Daniel Wagner <dwagner@suse.de>
---
 kernel/rcu/tree_exp.h |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

--- a/kernel/rcu/tree_exp.h
+++ b/kernel/rcu/tree_exp.h
@@ -309,7 +309,7 @@ static bool exp_funnel_lock(unsigned lon
 				   sync_exp_work_done(s));
 			return true;
 		}
-		rnp->exp_seq_rq = s; /* Followers can wait on us. */
+		WRITE_ONCE(rnp->exp_seq_rq, s); /* Followers can wait on us. */
 		spin_unlock(&rnp->exp_lock);
 		trace_rcu_exp_funnel_lock(rcu_state.name, rnp->level,
 					  rnp->grplo, rnp->grphi, TPS("nxtlvl"));