[PATCH v3] RDMA/core: Fix race when resolving IP address

[Dag Moxnes]

Use neighbour lock when copying MAC address from neighbour data struct
in dst_fetch_ha.

When not using the lock, it is possible for the function to race with
neigh_update, causing it to copy an invalid MAC address.

It is possible to provoke this error by calling rdma_resolve_addr in a
tight loop, while deleting the corresponding ARP entry in another tight
loop.

This will cause the race shown it the following sample trace:

rdma_resolve_addr()
  rdma_resolve_ip()
    addr_resolve()
      addr_resolve_neigh()
        fetch_ha()
          dst_fetch_ha()
            n->nud_state == NUD_VALID

and

net_ioctl()
  arp_ioctl()
    arp_rec_delete()
      arp_invalidate()
        neigh_update()
          __neigh_update()
            neigh->nud_state = new;

Signed-off-by: Dag Moxnes <dag.mox...@oracle.com>
Signed-off-by: Håkon Bugge <haakon.bu...@oracle.com>
Reviewed-by: Parav Pandit <pa...@mellanox.com>
---
v1 -> v2:
   * Modified implementation to improve readability

v2 -> v3:
   * Added sample trace as suggested by Parav Pandit
   * Added Reviewed-by


---
 drivers/infiniband/core/addr.c | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)

diff --git a/drivers/infiniband/core/addr.c b/drivers/infiniband/core/addr.c
index 2f7d141598..51323ffbc5 100644
--- a/drivers/infiniband/core/addr.c
+++ b/drivers/infiniband/core/addr.c
@@ -333,11 +333,14 @@ static int dst_fetch_ha(const struct dst_entry *dst,
        if (!n)
                return -ENODATA;
 
-       if (!(n->nud_state & NUD_VALID)) {
+       read_lock_bh(&n->lock);
+       if (n->nud_state & NUD_VALID) {
+               memcpy(dev_addr->dst_dev_addr, n->ha, MAX_ADDR_LEN);
+               read_unlock_bh(&n->lock);
+       } else {
+               read_unlock_bh(&n->lock);
                neigh_event_send(n, NULL);
                ret = -ENODATA;
-       } else {
-               memcpy(dev_addr->dst_dev_addr, n->ha, MAX_ADDR_LEN);
        }
 
        neigh_release(n);
-- 
2.20.1