[PATCH net v2] lwt: fix rx checksum setting for lwt devices tunneling over ipv6

the commit 35e2d1152b22 ("tunnels: Allow IPv6 UDP checksums to be
correctly controlled.") changed the default xmit checksum setting
for lwt vxlan/geneve ipv6 tunnels, so that now the checksum is not
set into external UDP header.
This commit changes the rx checksum setting for both lwt vxlan/geneve
devices created by openvswitch accordingly, so that lwt over ipv6
tunnel pairs are again able to communicate with default values.

Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
v2: rebased against current -net
---
 drivers/net/geneve.c          | 3 ++-
 net/openvswitch/vport-vxlan.c | 2 +-
 2 files changed, 3 insertions(+), 2 deletions(-)

diff --git a/drivers/net/geneve.c b/drivers/net/geneve.c
index 028e387..3496d78 100644
--- a/drivers/net/geneve.c
+++ b/drivers/net/geneve.c

@@ -1452,7 +1452,8 @@ struct net_device *geneve_dev_create_fb(struct net *net, const char *name,
 		return dev;
 
 	err = geneve_configure(net, dev, &geneve_remote_unspec,
-			       0, 0, 0, htons(dst_port), true, 0);
+			       0, 0, 0, htons(dst_port), true,
+			       GENEVE_F_UDP_ZERO_CSUM6_RX);
 	if (err)
 		goto err;
 

diff --git a/net/openvswitch/vport-vxlan.c b/net/openvswitch/vport-vxlan.c
index de9cb19..5eb7694 100644
--- a/net/openvswitch/vport-vxlan.c
+++ b/net/openvswitch/vport-vxlan.c

@@ -90,7 +90,7 @@ static struct vport *vxlan_tnl_create(const struct vport_parms *parms)
 	int err;
 	struct vxlan_config conf = {
 		.no_share = true,
-		.flags = VXLAN_F_COLLECT_METADATA,
+		.flags = VXLAN_F_COLLECT_METADATA | VXLAN_F_UDP_ZERO_CSUM6_RX,
 		/* Don't restrict the packets that can be sent by MTU */
 		.mtu = IP_MAX_MTU,
 	};

-- 
1.8.3.1

On Wed, Feb 17, 2016 at 10:30 AM, Paolo Abeni [off-list ref] wrote:

the commit 35e2d1152b22 ("tunnels: Allow IPv6 UDP checksums to be
correctly controlled.") changed the default xmit checksum setting
for lwt vxlan/geneve ipv6 tunnels, so that now the checksum is not
set into external UDP header.
This commit changes the rx checksum setting for both lwt vxlan/geneve
devices created by openvswitch accordingly, so that lwt over ipv6
tunnel pairs are again able to communicate with default values.

Signed-off-by: Paolo Abeni <pabeni@redhat.com>

This is compat code that is being changed here. Based on the previous
discussion, shouldn't these be made the default for all users?

On Wed, 2016-02-17 at 11:19 -0800, Jesse Gross wrote:

On Wed, Feb 17, 2016 at 10:30 AM, Paolo Abeni [off-list ref] wrote:

quoted

the commit 35e2d1152b22 ("tunnels: Allow IPv6 UDP checksums to be
correctly controlled.") changed the default xmit checksum setting
for lwt vxlan/geneve ipv6 tunnels, so that now the checksum is not
set into external UDP header.
This commit changes the rx checksum setting for both lwt vxlan/geneve
devices created by openvswitch accordingly, so that lwt over ipv6
tunnel pairs are again able to communicate with default values.

Signed-off-by: Paolo Abeni <pabeni@redhat.com>

This is compat code that is being changed here. Based on the previous
discussion, shouldn't these be made the default for all users?

This change was initially aimed at a smaller scope: use self consistent
default for ovs lwt devices/tunnel terminations.

If there is agreement, I can post a separate patch changing the default
rx checksum setting for plain vxlan and geneve[1] devices tunneling over
ipv6.

Paolo

[1] according to the ietf draft the geneve devices have the same
checksum requirement than vxlan ones

On Wed, 17 Feb 2016 19:30:01 +0100, Paolo Abeni wrote:

the commit 35e2d1152b22 ("tunnels: Allow IPv6 UDP checksums to be
correctly controlled.") changed the default xmit checksum setting
for lwt vxlan/geneve ipv6 tunnels, so that now the checksum is not
set into external UDP header.
This commit changes the rx checksum setting for both lwt vxlan/geneve
devices created by openvswitch accordingly, so that lwt over ipv6
tunnel pairs are again able to communicate with default values.

Acked-by: Jiri Benc <redacted>

On Thu, Feb 18, 2016 at 1:42 AM, Paolo Abeni [off-list ref] wrote:

On Wed, 2016-02-17 at 11:19 -0800, Jesse Gross wrote:

quoted

On Wed, Feb 17, 2016 at 10:30 AM, Paolo Abeni [off-list ref] wrote:

quoted

the commit 35e2d1152b22 ("tunnels: Allow IPv6 UDP checksums to be
correctly controlled.") changed the default xmit checksum setting
for lwt vxlan/geneve ipv6 tunnels, so that now the checksum is not
set into external UDP header.
This commit changes the rx checksum setting for both lwt vxlan/geneve
devices created by openvswitch accordingly, so that lwt over ipv6
tunnel pairs are again able to communicate with default values.

Signed-off-by: Paolo Abeni <pabeni@redhat.com>

This is compat code that is being changed here. Based on the previous
discussion, shouldn't these be made the default for all users?

This change was initially aimed at a smaller scope: use self consistent
default for ovs lwt devices/tunnel terminations.

If there is agreement, I can post a separate patch changing the default
rx checksum setting for plain vxlan and geneve[1] devices tunneling over
ipv6.

OK. I think it's important to do the followup to keep things
consistent but this is fine to start.

Acked-by: Jesse Gross <jesse@kernel.org>

From: Paolo Abeni <pabeni@redhat.com>
Date: Wed, 17 Feb 2016 19:30:01 +0100

the commit 35e2d1152b22 ("tunnels: Allow IPv6 UDP checksums to be
correctly controlled.") changed the default xmit checksum setting
for lwt vxlan/geneve ipv6 tunnels, so that now the checksum is not
set into external UDP header.
This commit changes the rx checksum setting for both lwt vxlan/geneve
devices created by openvswitch accordingly, so that lwt over ipv6
tunnel pairs are again able to communicate with default values.

Signed-off-by: Paolo Abeni <pabeni@redhat.com>

Applied.