Commit 9497df88 authored by Herbert Xu's avatar Herbert Xu Committed by David S. Miller

rhashtable: Fix reader/rehash race

There is a potential race condition between readers and the rehasher.
In particular, the rehasher could have started a rehash while the
reader finishes a scan of the old table but fails to see the new
table pointer.

This patch closes this window by adding smp_wmb/smp_rmb.
Signed-off-by: default avatarHerbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
parent 5ff0d16a
...@@ -271,6 +271,9 @@ static void rhashtable_rehash(struct rhashtable *ht, ...@@ -271,6 +271,9 @@ static void rhashtable_rehash(struct rhashtable *ht,
*/ */
rcu_assign_pointer(ht->future_tbl, new_tbl); rcu_assign_pointer(ht->future_tbl, new_tbl);
/* Ensure the new table is visible to readers. */
smp_wmb();
for (old_hash = 0; old_hash < old_tbl->size; old_hash++) for (old_hash = 0; old_hash < old_tbl->size; old_hash++)
rhashtable_rehash_chain(ht, old_hash); rhashtable_rehash_chain(ht, old_hash);
...@@ -618,6 +621,9 @@ void *rhashtable_lookup_compare(struct rhashtable *ht, const void *key, ...@@ -618,6 +621,9 @@ void *rhashtable_lookup_compare(struct rhashtable *ht, const void *key,
return rht_obj(ht, he); return rht_obj(ht, he);
} }
/* Ensure we see any new tables. */
smp_rmb();
old_tbl = tbl; old_tbl = tbl;
tbl = rht_dereference_rcu(ht->future_tbl, ht); tbl = rht_dereference_rcu(ht->future_tbl, ht);
if (unlikely(tbl != old_tbl)) if (unlikely(tbl != old_tbl))
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment