| Message ID | 20180117180225.25604-2-blp@ovn.org |
|---|---|
| State | Accepted |
| Headers | show |
| Series | Fix GRE tunnel MTU issues. | expand |
On Wed, Jan 17, 2018 at 10:02:24AM -0800, Ben Pfaff wrote: > Most of the existing tunnels accept 65535 for MTU and internally reduce it > to the maximum value actually supported. However, in RTM_SETLINK calls, > at least GRE tunnels reject MTU larger than actually supported. This > commit changes the MTU used in RTM_NEWLINK calls to use a value that should > be acceptable to all tunnels and yet does not noticeably reduce > performance. > > (This code doesn't actually use RTM_SETLINK to change MTU yet, but that's > coming up.) > > Suggested-by: Eric Garver <e@erig.me> > Suggested-at: https://mail.openvswitch.org/pipermail/ovs-dev/2018-January/343304.html > Signed-off-by: Ben Pfaff <blp@ovn.org> > --- > lib/dpif-netlink-rtnl.c | 11 ++++++++++- > 1 file changed, 10 insertions(+), 1 deletion(-) > > diff --git a/lib/dpif-netlink-rtnl.c b/lib/dpif-netlink-rtnl.c > index fe9c8ed7104f..37451b80de0e 100644 > --- a/lib/dpif-netlink-rtnl.c > +++ b/lib/dpif-netlink-rtnl.c > @@ -277,6 +277,15 @@ dpif_netlink_rtnl_create(const struct netdev_tunnel_config *tnl_cfg, > const char *name, enum ovs_vport_type type, > const char *kind, uint32_t flags) > { > + enum { > + /* For performance, we want to use the largest MTU that the system > + * supports. Most existing tunnels will accept UINT16_MAX, treating it > + * as the actual max MTU, but some do not. Thus, we use a slightly > + * smaller value, that should always be safe yet does not noticeably > + * reduce performance. */ > + MAX_MTU = 65000 > + }; > + > size_t linkinfo_off, infodata_off; > struct ifinfomsg *ifinfo; > struct ofpbuf request; > @@ -287,7 +296,7 @@ dpif_netlink_rtnl_create(const struct netdev_tunnel_config *tnl_cfg, > ifinfo = ofpbuf_put_zeros(&request, sizeof(struct ifinfomsg)); > ifinfo->ifi_change = ifinfo->ifi_flags = IFF_UP; > nl_msg_put_string(&request, IFLA_IFNAME, name); > - nl_msg_put_u32(&request, IFLA_MTU, UINT16_MAX); > + nl_msg_put_u32(&request, IFLA_MTU, MAX_MTU); > linkinfo_off = nl_msg_start_nested(&request, IFLA_LINKINFO); > nl_msg_put_string(&request, IFLA_INFO_KIND, kind); > infodata_off = nl_msg_start_nested(&request, IFLA_INFO_DATA); > -- > 2.10.2 Thanks Ben! Acked-by: Eric Garver <e@erig.me>
diff --git a/lib/dpif-netlink-rtnl.c b/lib/dpif-netlink-rtnl.c index fe9c8ed7104f..37451b80de0e 100644 --- a/lib/dpif-netlink-rtnl.c +++ b/lib/dpif-netlink-rtnl.c @@ -277,6 +277,15 @@ dpif_netlink_rtnl_create(const struct netdev_tunnel_config *tnl_cfg, const char *name, enum ovs_vport_type type, const char *kind, uint32_t flags) { + enum { + /* For performance, we want to use the largest MTU that the system + * supports. Most existing tunnels will accept UINT16_MAX, treating it + * as the actual max MTU, but some do not. Thus, we use a slightly + * smaller value, that should always be safe yet does not noticeably + * reduce performance. */ + MAX_MTU = 65000 + }; + size_t linkinfo_off, infodata_off; struct ifinfomsg *ifinfo; struct ofpbuf request; @@ -287,7 +296,7 @@ dpif_netlink_rtnl_create(const struct netdev_tunnel_config *tnl_cfg, ifinfo = ofpbuf_put_zeros(&request, sizeof(struct ifinfomsg)); ifinfo->ifi_change = ifinfo->ifi_flags = IFF_UP; nl_msg_put_string(&request, IFLA_IFNAME, name); - nl_msg_put_u32(&request, IFLA_MTU, UINT16_MAX); + nl_msg_put_u32(&request, IFLA_MTU, MAX_MTU); linkinfo_off = nl_msg_start_nested(&request, IFLA_LINKINFO); nl_msg_put_string(&request, IFLA_INFO_KIND, kind); infodata_off = nl_msg_start_nested(&request, IFLA_INFO_DATA);
Most of the existing tunnels accept 65535 for MTU and internally reduce it to the maximum value actually supported. However, in RTM_SETLINK calls, at least GRE tunnels reject MTU larger than actually supported. This commit changes the MTU used in RTM_NEWLINK calls to use a value that should be acceptable to all tunnels and yet does not noticeably reduce performance. (This code doesn't actually use RTM_SETLINK to change MTU yet, but that's coming up.) Suggested-by: Eric Garver <e@erig.me> Suggested-at: https://mail.openvswitch.org/pipermail/ovs-dev/2018-January/343304.html Signed-off-by: Ben Pfaff <blp@ovn.org> --- lib/dpif-netlink-rtnl.c | 11 ++++++++++- 1 file changed, 10 insertions(+), 1 deletion(-)