IPoIB: Disable NAPI while CQ is being drained

If NAPI is enabled while IPoIB's CQ is being drained, it creates a
race on priv->ibwc between ipoib_poll() and ipoib_drain_cq(), leading
to memory corruption.

The solution is to enable/disable NAPI in ipoib_ib_dev_{open/stop}()
instead of in ipoib_{open/stop}(), and sync NAPI on the INITIALIZED
flag instead on the ADMIN_UP flag. This way NAPI will be disabled when
ipoib_drain_cq() is called.

This fixes <https://bugs.openfabrics.org/show_bug.cgi?id=1587>.

Signed-off-by: Yossi Etigin <yosefe@voltaire.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
diff --git a/drivers/infiniband/ulp/ipoib/ipoib_ib.c b/drivers/infiniband/ulp/ipoib/ipoib_ib.c
index da60827..e7e5adf 100644
--- a/drivers/infiniband/ulp/ipoib/ipoib_ib.c
+++ b/drivers/infiniband/ulp/ipoib/ipoib_ib.c
@@ -685,7 +685,8 @@
 	queue_delayed_work(ipoib_workqueue, &priv->ah_reap_task,
 			   round_jiffies_relative(HZ));
 
-	set_bit(IPOIB_FLAG_INITIALIZED, &priv->flags);
+	if (!test_and_set_bit(IPOIB_FLAG_INITIALIZED, &priv->flags))
+		napi_enable(&priv->napi);
 
 	return 0;
 }
@@ -804,7 +805,8 @@
 	struct ipoib_tx_buf *tx_req;
 	int i;
 
-	clear_bit(IPOIB_FLAG_INITIALIZED, &priv->flags);
+	if (test_and_clear_bit(IPOIB_FLAG_INITIALIZED, &priv->flags))
+		napi_disable(&priv->napi);
 
 	ipoib_cm_dev_stop(dev);