Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
From:   NeilBrown <neilb@suse.com>
To:     Thomas Graf <tgraf@suug.ch>,
        Herbert Xu <herbert@gondor.apana.org.au>
Date:   Thu, 14 Mar 2019 16:05:28 +1100
Subject: [PATCH 1/3] rhashtable: use cmpxchg() in nested_table_alloc()
Cc:     netdev@vger.kernel.org,
        "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>,
        linux-kernel@vger.kernel.org
Message-ID: <155253992824.5022.10318044793639169265.stgit@noble.brown>
In-Reply-To: <155253979234.5022.1840929790507376038.stgit@noble.brown>
References: <155253979234.5022.1840929790507376038.stgit@noble.brown>
User-Agent: StGit/0.17.1-dirty
MIME-Version: 1.0
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

nested_table_alloc() relies on the fact that there is
at most one spinlock allocated for every slot in the top
level nested table, so it is not possible for two threads
to try to allocate the same table at the same time.

This assumption is a little fragile (it is not explicit) and is
unnecessary.  We can instead protect against
a race using cmpxchg() - if it the cmp fails, free the table
that was just allocated.

Signed-off-by: NeilBrown <neilb@suse.com>
---
 lib/rhashtable.c |    8 +++++---
 1 file changed, 5 insertions(+), 3 deletions(-)

diff --git a/lib/rhashtable.c b/lib/rhashtable.c
index 0a105d4af166..c983c0ee15d5 100644
--- a/lib/rhashtable.c
+++ b/lib/rhashtable.c
@@ -131,9 +131,11 @@ static union nested_table *nested_table_alloc(struct rhashtable *ht,
 			INIT_RHT_NULLS_HEAD(ntbl[i].bucket);
 	}
 
-	rcu_assign_pointer(*prev, ntbl);
-
-	return ntbl;
+	if (cmpxchg(prev, NULL, ntbl) == NULL)
+		return ntbl;
+	/* Raced with another thread. */
+	kfree(ntbl);
+	return rcu_dereference(*prev);
 }
 
 static struct bucket_table *nested_bucket_table_alloc(struct rhashtable *ht,