From mboxrd@z Thu Jan 1 00:00:00 1970 From: Alexander Duyck Subject: [PATCH 3/3] etherdev: Use skb->data to retrieve Ethernet header instead of eth_hdr Date: Thu, 30 Apr 2015 14:53:59 -0700 Message-ID: <20150430215359.1798.79917.stgit@ahduyck-vm-fedora22> References: <20150430214917.1798.49769.stgit@ahduyck-vm-fedora22> Mime-Version: 1.0 Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: 7bit Cc: davem@davemloft.net To: netdev@vger.kernel.org Return-path: Received: from mx1.redhat.com ([209.132.183.28]:34091 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752192AbbD3VyA (ORCPT ); Thu, 30 Apr 2015 17:54:00 -0400 In-Reply-To: <20150430214917.1798.49769.stgit@ahduyck-vm-fedora22> Sender: netdev-owner@vger.kernel.org List-ID: Avoid recomputing the Ethernet header location and instead just use the pointer provided by skb->data. The problem with using eth_hdr is that the compiler wasn't smart enough to realize that skb->head + skb->mac_header was the same thing as skb->data before it added ETH_HLEN. By just caching it off before calling skb_pull_inline we can avoid a few unnecessary instructions. Signed-off-by: Alexander Duyck --- net/ethernet/eth.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/net/ethernet/eth.c b/net/ethernet/eth.c index 21c211e9fd5a..314e4c5a5a5e 100644 --- a/net/ethernet/eth.c +++ b/net/ethernet/eth.c @@ -156,8 +156,9 @@ __be16 eth_type_trans(struct sk_buff *skb, struct net_device *dev) skb->dev = dev; skb_reset_mac_header(skb); + + eth = (struct ethhdr *)skb->data; skb_pull_inline(skb, ETH_HLEN); - eth = eth_hdr(skb); if (unlikely(is_multicast_ether_addr_64bits(eth->h_dest))) { if (ether_addr_equal_64bits(eth->h_dest, dev->broadcast))