| Message ID | 20200803080217.391850-3-liuhangbin@gmail.com |
|---|---|
| State | Changes Requested |
| Delegated to: | David Miller |
| Headers | show |
| Series | Add IP_DSCP_MASK and fix vxlan tos value before xmit | expand |
On Mon, Aug 03, 2020 at 04:02:17PM +0800, Hangbin Liu wrote: > In commit 71130f29979c ("vxlan: fix tos value before xmit") we strict > the vxlan tos value before xmit. But as IP tos field has been obsoleted > by RFC2474, and updated by RFC3168 later. We should use new DSCP field, > or we will lost the first 3 bits value when xmit. > Why sending this patch to net-next? Commit 71130f29979c ("vxlan: fix tos value before xmit") broke setups where the high TOS bits were used. This needs to be fixed in net (and probably pushed to -stable too). > Fixes: 71130f29979c ("vxlan: fix tos value before xmit") > Signed-off-by: Hangbin Liu <liuhangbin@gmail.com> > --- > drivers/net/vxlan.c | 4 ++-- > 1 file changed, 2 insertions(+), 2 deletions(-) > > diff --git a/drivers/net/vxlan.c b/drivers/net/vxlan.c > index 77658425db8a..fe051ed0f6db 100644 > --- a/drivers/net/vxlan.c > +++ b/drivers/net/vxlan.c > @@ -2722,7 +2722,7 @@ static void vxlan_xmit_one(struct sk_buff *skb, struct net_device *dev, > ndst = &rt->dst; > skb_tunnel_check_pmtu(skb, ndst, VXLAN_HEADROOM); > > - tos = ip_tunnel_ecn_encap(RT_TOS(tos), old_iph, skb); > + tos = ip_tunnel_ecn_encap(RT_DSCP(tos), old_iph, skb); > ttl = ttl ? : ip4_dst_hoplimit(&rt->dst); > err = vxlan_build_skb(skb, ndst, sizeof(struct iphdr), > vni, md, flags, udp_sum); > @@ -2762,7 +2762,7 @@ static void vxlan_xmit_one(struct sk_buff *skb, struct net_device *dev, > > skb_tunnel_check_pmtu(skb, ndst, VXLAN6_HEADROOM); > > - tos = ip_tunnel_ecn_encap(RT_TOS(tos), old_iph, skb); > + tos = ip_tunnel_ecn_encap(RT_DSCP(tos), old_iph, skb); > ttl = ttl ? : ip6_dst_hoplimit(ndst); > skb_scrub_packet(skb, xnet); > err = vxlan_build_skb(skb, ndst, sizeof(struct ipv6hdr), > -- > 2.25.4 >
diff --git a/drivers/net/vxlan.c b/drivers/net/vxlan.c index 77658425db8a..fe051ed0f6db 100644 --- a/drivers/net/vxlan.c +++ b/drivers/net/vxlan.c @@ -2722,7 +2722,7 @@ static void vxlan_xmit_one(struct sk_buff *skb, struct net_device *dev, ndst = &rt->dst; skb_tunnel_check_pmtu(skb, ndst, VXLAN_HEADROOM); - tos = ip_tunnel_ecn_encap(RT_TOS(tos), old_iph, skb); + tos = ip_tunnel_ecn_encap(RT_DSCP(tos), old_iph, skb); ttl = ttl ? : ip4_dst_hoplimit(&rt->dst); err = vxlan_build_skb(skb, ndst, sizeof(struct iphdr), vni, md, flags, udp_sum); @@ -2762,7 +2762,7 @@ static void vxlan_xmit_one(struct sk_buff *skb, struct net_device *dev, skb_tunnel_check_pmtu(skb, ndst, VXLAN6_HEADROOM); - tos = ip_tunnel_ecn_encap(RT_TOS(tos), old_iph, skb); + tos = ip_tunnel_ecn_encap(RT_DSCP(tos), old_iph, skb); ttl = ttl ? : ip6_dst_hoplimit(ndst); skb_scrub_packet(skb, xnet); err = vxlan_build_skb(skb, ndst, sizeof(struct ipv6hdr),
In commit 71130f29979c ("vxlan: fix tos value before xmit") we strict the vxlan tos value before xmit. But as IP tos field has been obsoleted by RFC2474, and updated by RFC3168 later. We should use new DSCP field, or we will lost the first 3 bits value when xmit. Fixes: 71130f29979c ("vxlan: fix tos value before xmit") Signed-off-by: Hangbin Liu <liuhangbin@gmail.com> --- drivers/net/vxlan.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)