From f89e49e79f839dbe79364ec6cd7e2274ad11b120 Mon Sep 17 00:00:00 2001
From: Anton Blanchard <anton@samba.org>
Date: Mon, 6 Sep 2010 18:21:41 -0700
Subject: ibmveth: Use lighter weight read memory barrier in ibmveth_poll

We want to order the read in ibmveth_rxq_pending_buffer and the read of
ibmveth_rxq_buffer_valid which are both cacheable memory. smp_rmb() is good
enough for this.

Signed-off-by: Anton Blanchard <anton@samba.org>
Signed-off-by: Santiago Leon <santil@linux.vnet.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ibmveth.c b/drivers/net/ibmveth.c
index c236c8a..aedcb18 100644
--- a/drivers/net/ibmveth.c
+++ b/drivers/net/ibmveth.c
@@ -1010,7 +1010,7 @@ static int ibmveth_poll(struct napi_struct *napi, int budget)
 		if (!ibmveth_rxq_pending_buffer(adapter))
 			break;
 
-		rmb();
+		smp_rmb();
 		if (!ibmveth_rxq_buffer_valid(adapter)) {
 			wmb(); /* suggested by larson1 */
 			adapter->rx_invalid_buffer++;
-- 
cgit v0.10.2