[GIT net] Open vSwitch

[GIT net] Open vSwitch

[Jesse Gross]

A few different bug fixes, including several for issues with userspace
communication that have gone unnoticed up until now.  These are intended
for net/3.9.

The following changes since commit 19f949f52599ba7c3f67a5897ac6be14bfcb1200:

  Linux 3.8 (2013-02-18 15:58:34 -0800)

are available in the git repository at:

  git://git.kernel.org/pub/scm/linux/kernel/git/jesse/openvswitch.git fixes

for you to fetch changes up to d176ca2a48ff2b5d7becfacdcbd1d72c73bd22d1:

  openvswitch: remove some useless comments (2013-02-22 17:02:49 -0800)

----------------------------------------------------------------
Cong Wang (2):
      openvswitch: fix the calculation of checksum for vlan header
      openvswitch: remove some useless comments

Rich Lane (4):
      openvswitch: Fix ovs_vport_cmd_del return value on success
      openvswitch: Fix ovs_vport_cmd_new return value on success
      openvswitch: Call genlmsg_end in queue_userspace_packet
      openvswitch: Fix parsing invalid LLC/SNAP ethertypes

 net/openvswitch/actions.c      |    4 ++--
 net/openvswitch/datapath.c     |    3 +++
 net/openvswitch/flow.c         |    6 +++++-
 net/openvswitch/vport-netdev.c |    3 +--
 net/openvswitch/vport.c        |    3 +--
 5 files changed, 12 insertions(+), 7 deletions(-)

[PATCH net 1/6] openvswitch: Fix ovs_vport_cmd_del return value on success

[Jesse Gross]

From: Rich Lane <rlane@bigswitch.com>

If the pointer does not represent an error then the PTR_ERR macro may still
return a nonzero value. The fix is the same as in ovs_vport_cmd_set.

Signed-off-by: Rich Lane <rlane@bigswitch.com>
Signed-off-by: Jesse Gross <jesse@nicira.com>
---
 net/openvswitch/datapath.c |    1 +
 1 file changed, 1 insertion(+)

diff --git a/net/openvswitch/datapath.c b/net/openvswitch/datapath.c
index f996db3..5e275b9 100644
--- a/net/openvswitch/datapath.c
+++ b/net/openvswitch/datapath.c
@@ -1772,6 +1772,7 @@ static int ovs_vport_cmd_del(struct sk_buff *skb, struct genl_info *info)
 	if (IS_ERR(reply))
 		goto exit_unlock;
 
+	err = 0;
 	ovs_dp_detach_port(vport);
 
 	genl_notify(reply, genl_info_net(info), info->snd_portid,
-- 
1.7.10.4

[PATCH net 2/6] openvswitch: Fix ovs_vport_cmd_new return value on success

[Jesse Gross]

From: Rich Lane <rlane-oPdFNf9WwoK1Z/+hSey0Gg@public.gmane.org>

If the pointer does not represent an error then the PTR_ERR
macro may still return a nonzero value.

Signed-off-by: Rich Lane <rlane-oPdFNf9WwoK1Z/+hSey0Gg@public.gmane.org>
Signed-off-by: Jesse Gross <jesse-l0M0P4e3n4LQT0dZR+AlfA@public.gmane.org>
---
 net/openvswitch/datapath.c |    1 +
 1 file changed, 1 insertion(+)

diff --git a/net/openvswitch/datapath.c b/net/openvswitch/datapath.c
index 5e275b9..a2cd3e6 100644
--- a/net/openvswitch/datapath.c
+++ b/net/openvswitch/datapath.c
@@ -1691,6 +1691,7 @@ static int ovs_vport_cmd_new(struct sk_buff *skb, struct genl_info *info)
 	if (IS_ERR(vport))
 		goto exit_unlock;
 
+	err = 0;
 	reply = ovs_vport_cmd_build_info(vport, info->snd_portid, info->snd_seq,
 					 OVS_VPORT_CMD_NEW);
 	if (IS_ERR(reply)) {
-- 
1.7.10.4

[PATCH net 3/6] openvswitch: Call genlmsg_end in queue_userspace_packet

[Jesse Gross]

From: Rich Lane <rlane-oPdFNf9WwoK1Z/+hSey0Gg@public.gmane.org>

Without genlmsg_end the upcall message ends (according to nlmsg_len)
after the struct ovs_header.

Signed-off-by: Rich Lane <rlane-oPdFNf9WwoK1Z/+hSey0Gg@public.gmane.org>
Signed-off-by: Jesse Gross <jesse-l0M0P4e3n4LQT0dZR+AlfA@public.gmane.org>
---
 net/openvswitch/datapath.c |    1 +
 1 file changed, 1 insertion(+)

diff --git a/net/openvswitch/datapath.c b/net/openvswitch/datapath.c
index a2cd3e6..cae1062 100644
--- a/net/openvswitch/datapath.c
+++ b/net/openvswitch/datapath.c
@@ -395,6 +395,7 @@ static int queue_userspace_packet(struct net *net, int dp_ifindex,
 
 	skb_copy_and_csum_dev(skb, nla_data(nla));
 
+	genlmsg_end(user_skb, upcall);
 	err = genlmsg_unicast(net, user_skb, upcall_info->portid);
 
 out:
-- 
1.7.10.4

[PATCH net 4/6] openvswitch: Fix parsing invalid LLC/SNAP ethertypes

[Jesse Gross]

From: Rich Lane <rlane-oPdFNf9WwoK1Z/+hSey0Gg@public.gmane.org>

Before this patch, if an LLC/SNAP packet with OUI 00:00:00 had an
ethertype less than 1536 the flow key given to userspace in the upcall
would contain the invalid ethertype (for example, 3). If userspace
attempted to insert a kernel flow for this key it would be rejected
by ovs_flow_from_nlattrs.

This patch allows OVS to pass the OFTest pktact.DirectBadLlcPackets.

Signed-off-by: Rich Lane <rlane-oPdFNf9WwoK1Z/+hSey0Gg@public.gmane.org>
Signed-off-by: Jesse Gross <jesse-l0M0P4e3n4LQT0dZR+AlfA@public.gmane.org>
---
 net/openvswitch/flow.c |    6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/net/openvswitch/flow.c b/net/openvswitch/flow.c
index c3294ce..0c98d40 100644
--- a/net/openvswitch/flow.c
+++ b/net/openvswitch/flow.c
@@ -484,7 +484,11 @@ static __be16 parse_ethertype(struct sk_buff *skb)
 		return htons(ETH_P_802_2);
 
 	__skb_pull(skb, sizeof(struct llc_snap_hdr));
-	return llc->ethertype;
+
+	if (ntohs(llc->ethertype) >= 1536)
+		return llc->ethertype;
+
+	return htons(ETH_P_802_2);
 }
 
 static int parse_icmpv6(struct sk_buff *skb, struct sw_flow_key *key,
-- 
1.7.10.4

[PATCH net 5/6] openvswitch: fix the calculation of checksum for vlan header

[Jesse Gross]

From: Cong Wang <amwang-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>

In vlan_insert_tag(), we insert a 4-byte VLAN header _after_
mac header:

        memmove(skb->data, skb->data + VLAN_HLEN, 2 * ETH_ALEN);
        ...
        veth->h_vlan_proto = htons(ETH_P_8021Q);
        ...
        veth->h_vlan_TCI = htons(vlan_tci);

so after it, we should recompute the checksum to include these 4 bytes.
skb->data still points to the mac header, therefore VLAN header is at
(2 * ETH_ALEN = 12) bytes after it, not (ETH_HLEN = 14) bytes.

This can also be observed via tcpdump:

         0x0000:  ffff ffff ffff 5254 005d 6f6e 8100 000a
         0x0010:  0806 0001 0800 0604 0001 5254 005d 6f6e
         0x0020:  c0a8 026e 0000 0000 0000 c0a8 0282

Similar for __pop_vlan_tci(), the vlan header we remove is the one
overwritten in:

	memmove(skb->data + VLAN_HLEN, skb->data, 2 * ETH_ALEN);

Therefore the VLAN_HLEN = 4 bytes after 2 * ETH_ALEN is the part
we want to sub from checksum.

Cc: David S. Miller <davem-fT/PcQaiUtIeIZ0/mPfg9Q@public.gmane.org>
Cc: Jesse Gross <jesse-l0M0P4e3n4LQT0dZR+AlfA@public.gmane.org>
Signed-off-by: Cong Wang <amwang-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
Signed-off-by: Jesse Gross <jesse-l0M0P4e3n4LQT0dZR+AlfA@public.gmane.org>
---
 net/openvswitch/actions.c |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/net/openvswitch/actions.c b/net/openvswitch/actions.c
index ac2defe..d4d5363 100644
--- a/net/openvswitch/actions.c
+++ b/net/openvswitch/actions.c
@@ -58,7 +58,7 @@ static int __pop_vlan_tci(struct sk_buff *skb, __be16 *current_tci)
 
 	if (skb->ip_summed == CHECKSUM_COMPLETE)
 		skb->csum = csum_sub(skb->csum, csum_partial(skb->data
-					+ ETH_HLEN, VLAN_HLEN, 0));
+					+ (2 * ETH_ALEN), VLAN_HLEN, 0));
 
 	vhdr = (struct vlan_hdr *)(skb->data + ETH_HLEN);
 	*current_tci = vhdr->h_vlan_TCI;
@@ -115,7 +115,7 @@ static int push_vlan(struct sk_buff *skb, const struct ovs_action_push_vlan *vla
 
 		if (skb->ip_summed == CHECKSUM_COMPLETE)
 			skb->csum = csum_add(skb->csum, csum_partial(skb->data
-					+ ETH_HLEN, VLAN_HLEN, 0));
+					+ (2 * ETH_ALEN), VLAN_HLEN, 0));
 
 	}
 	__vlan_hwaccel_put_tag(skb, ntohs(vlan->vlan_tci) & ~VLAN_TAG_PRESENT);
-- 
1.7.10.4

[PATCH net 6/6] openvswitch: remove some useless comments

[Jesse Gross]

From: Cong Wang <amwang-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>

These comments are useless in upstream kernel.

Cc: David S. Miller <davem-fT/PcQaiUtIeIZ0/mPfg9Q@public.gmane.org>
Cc: Jesse Gross <jesse-l0M0P4e3n4LQT0dZR+AlfA@public.gmane.org>
Signed-off-by: Cong Wang <amwang-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
Signed-off-by: Jesse Gross <jesse-l0M0P4e3n4LQT0dZR+AlfA@public.gmane.org>
---
 net/openvswitch/vport-netdev.c |    3 +--
 net/openvswitch/vport.c        |    3 +--
 2 files changed, 2 insertions(+), 4 deletions(-)

diff --git a/net/openvswitch/vport-netdev.c b/net/openvswitch/vport-netdev.c
index 670cbc3..2130d61 100644
--- a/net/openvswitch/vport-netdev.c
+++ b/net/openvswitch/vport-netdev.c
@@ -43,8 +43,7 @@ static void netdev_port_receive(struct vport *vport, struct sk_buff *skb)
 
 	/* Make our own copy of the packet.  Otherwise we will mangle the
 	 * packet for anyone who came before us (e.g. tcpdump via AF_PACKET).
-	 * (No one comes after us, since we tell handle_bridge() that we took
-	 * the packet.) */
+	 */
 	skb = skb_share_check(skb, GFP_ATOMIC);
 	if (unlikely(!skb))
 		return;
diff --git a/net/openvswitch/vport.c b/net/openvswitch/vport.c
index 70af0be..6255e48 100644
--- a/net/openvswitch/vport.c
+++ b/net/openvswitch/vport.c
@@ -326,8 +326,7 @@ int ovs_vport_get_options(const struct vport *vport, struct sk_buff *skb)
  * @skb: skb that was received
  *
  * Must be called with rcu_read_lock.  The packet cannot be shared and
- * skb->data should point to the Ethernet header.  The caller must have already
- * called compute_ip_summed() to initialize the checksumming fields.
+ * skb->data should point to the Ethernet header.
  */
 void ovs_vport_receive(struct vport *vport, struct sk_buff *skb)
 {
-- 
1.7.10.4

Re: [GIT net] Open vSwitch

[David Miller]

From: Jesse Gross <jesse@nicira.com>
Date: Thu, 14 Mar 2013 17:15:00 -0700

> A few different bug fixes, including several for issues with userspace
> communication that have gone unnoticed up until now.  These are intended
> for net/3.9.

Pulled, thanks.