diff options
| author | Eric Dumazet <eric.dumazet@gmail.com> | 2012-07-31 05:08:23 +0400 |
|---|---|---|
| committer | David S. Miller <davem@davemloft.net> | 2012-08-01 01:41:38 +0400 |
| commit | 54764bb647b2e847c512acf8d443df965da35000 (patch) | |
| tree | 3918f6f42679d0ebdcef230f28285f99d178be62 /net/ipv4/fib_semantics.c | |
| parent | 5a0d513b622ee41e117fc37e26e27e8ef42e8dae (diff) | |
| download | linux-54764bb647b2e847c512acf8d443df965da35000.tar.xz | |
ipv4: Restore old dst_free() behavior.
commit 404e0a8b6a55 (net: ipv4: fix RCU races on dst refcounts) tried
to solve a race but added a problem at device/fib dismantle time :
We really want to call dst_free() as soon as possible, even if sockets
still have dst in their cache.
dst_release() calls in free_fib_info_rcu() are not welcomed.
Root of the problem was that now we also cache output routes (in
nh_rth_output), we must use call_rcu() instead of call_rcu_bh() in
rt_free(), because output route lookups are done in process context.
Based on feedback and initial patch from David Miller (adding another
call_rcu_bh() call in fib, but it appears it was not the right fix)
I left the inet_sk_rx_dst_set() helper and added __rcu attributes
to nh_rth_output and nh_rth_input to better document what is going on in
this code.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Diffstat (limited to 'net/ipv4/fib_semantics.c')
| -rw-r--r-- | net/ipv4/fib_semantics.c | 21 |
1 files changed, 17 insertions, 4 deletions
diff --git a/net/ipv4/fib_semantics.c b/net/ipv4/fib_semantics.c index e55171f184f9..625cf185c489 100644 --- a/net/ipv4/fib_semantics.c +++ b/net/ipv4/fib_semantics.c @@ -161,6 +161,21 @@ static void free_nh_exceptions(struct fib_nh *nh) kfree(hash); } +static void rt_nexthop_free(struct rtable __rcu **rtp) +{ + struct rtable *rt = rcu_dereference_protected(*rtp, 1); + + if (!rt) + return; + + /* Not even needed : RCU_INIT_POINTER(*rtp, NULL); + * because we waited an RCU grace period before calling + * free_fib_info_rcu() + */ + + dst_free(&rt->dst); +} + /* Release a nexthop info record */ static void free_fib_info_rcu(struct rcu_head *head) { @@ -171,10 +186,8 @@ static void free_fib_info_rcu(struct rcu_head *head) dev_put(nexthop_nh->nh_dev); if (nexthop_nh->nh_exceptions) free_nh_exceptions(nexthop_nh); - if (nexthop_nh->nh_rth_output) - dst_release(&nexthop_nh->nh_rth_output->dst); - if (nexthop_nh->nh_rth_input) - dst_release(&nexthop_nh->nh_rth_input->dst); + rt_nexthop_free(&nexthop_nh->nh_rth_output); + rt_nexthop_free(&nexthop_nh->nh_rth_input); } endfor_nexthops(fi); release_net(fi->fib_net); |