Re: [PATCH v2.57] datapath: Add basic MPLS support to kernel

On Tue, May 20, 2014 at 07:30:38PM -0700, Jesse Gross wrote:

On Thu, May 15, 2014 at 4:07 PM, Simon Horman [off-list ref] wrote:

quoted

diff --git a/datapath/linux/compat/gso.c b/datapath/linux/compat/gso.c
index 9ded17c..d9b4485 100644
--- a/datapath/linux/compat/gso.c
+++ b/datapath/linux/compat/gso.c
-#if LINUX_VERSION_CODE < KERNEL_VERSION(2,6,37)
+#if LINUX_VERSION_CODE < KERNEL_VERSION(3,11,0)
 static bool dev_supports_vlan_tx(struct net_device *dev)
 {
-#if defined(HAVE_VLAN_BUG_WORKAROUND)
+#if LINUX_VERSION_CODE < KERNEL_VERSION(2,6,37)
+       return true;

Should this be greater than 2.6.37 instead of less than?

Yes, thanks.

quoted

diff --git a/datapath/linux/compat/include/linux/netdevice.h b/datapath/linux/compat/include/linux/netdevice.h
index d726390..0381002 100644
--- a/datapath/linux/compat/include/linux/netdevice.h
+++ b/datapath/linux/compat/include/linux/netdevice.h
+#if LINUX_VERSION_CODE < KERNEL_VERSION(2,6,38)
 #define netif_skb_features rpl_netif_skb_features
 netdev_features_t rpl_netif_skb_features(struct sk_buff *skb);

Currently this function doesn't look at mpls_features. Should it for
things other than TSO?

It seems to me that rpl_netif_skb_features is only used by
rpl_dev_queue_xmit().

If that is the case then I don't think it is necessary to examine
mpls_features, which was introduced in v3.11, as rpl_dev_queue_xmit()
only exists for older kernel versions.

quoted

diff --git a/datapath/linux/compat/netdevice.c b/datapath/linux/compat/netdevice.c
index 1dc5abf..d22fced 100644
--- a/datapath/linux/compat/netdevice.c
+++ b/datapath/linux/compat/netdevice.c

@@ -78,6 +83,9 @@ struct sk_buff *rpl_skb_gso_segment(struct sk_buff *skb,
        __be16 skb_proto;
        struct sk_buff *skb_gso;

+       if (eth_p_mpls(skb->protocol))
+               type = ovs_skb_get_inner_protocol(skb);
+
        while (type == htons(ETH_P_8021Q)) {
                struct vlan_hdr *vh;

Is the ordering on this backwards given that we expect that VLANs can
be on the outside but not the other way around?

Yes, I think so.

I will change this around as follows:

diff --git a/datapath/linux/compat/netdevice.c b/datapath/linux/compat/netdevice.c
index d22fced..d63f1f5 100644
--- a/datapath/linux/compat/netdevice.c
+++ b/datapath/linux/compat/netdevice.c

@@ -83,9 +83,6 @@ struct sk_buff *rpl_skb_gso_segment(struct sk_buff *skb,
 	__be16 skb_proto;
 	struct sk_buff *skb_gso;
 
-	if (eth_p_mpls(skb->protocol))
-		type = ovs_skb_get_inner_protocol(skb);
-
 	while (type == htons(ETH_P_8021Q)) {
 		struct vlan_hdr *vh;
 

@@ -97,6 +94,9 @@ struct sk_buff *rpl_skb_gso_segment(struct sk_buff *skb,
 		vlan_depth += VLAN_HLEN;
 	}
 
+	if (eth_p_mpls(type))
+		type = ovs_skb_get_inner_protocol(skb);
+
 	/* this hack needed to get regular skb_gso_segment() */
 #undef skb_gso_segment
 	skb_proto = skb->protocol;