On Tue, Aug 5, 2014 at 4:38 PM, Jarno Rajahalme <jrajaha...@nicira.com> wrote:
> diff --git a/datapath/linux/compat/include/linux/openvswitch.h
> b/datapath/linux/compat/include/linux/openvswitch.h
> index 271a14e..64761b4 100644
> --- a/datapath/linux/compat/include/linux/openvswitch.h
> +++ b/datapath/linux/compat/include/linux/openvswitch.h
> @@ -590,6 +590,8 @@ struct ovs_action_hash {
> * indicate the new packet contents. This could potentially still be
> * %ETH_P_MPLS if the resulting MPLS label stack is not empty. If there
> * is no MPLS label stack, as determined by ethertype, no action is taken.
> + * @OVS_ACTION_ATTR_METER: Run packet through a meter, which may drop the
> + * packet, or modify the packet (e.g., change the DSCP field).
> *
> * Only a single header can be set with a single %OVS_ACTION_ATTR_SET. Not
> all
> * fields within a header are modifiable, e.g. the IPv4 protocol and fragment
> @@ -608,6 +610,7 @@ enum ovs_action_attr {
> OVS_ACTION_ATTR_HASH, /* struct ovs_action_hash. */
> OVS_ACTION_ATTR_PUSH_MPLS, /* struct ovs_action_push_mpls. */
> OVS_ACTION_ATTR_POP_MPLS, /* __be16 ethertype. */
> + OVS_ACTION_ATTR_METER, /* u32 meter number. */
One problem with sharing a single interface file across platforms is
that we can't do this unless there is a Linux implementation. We can
pick a high number like before and change it later but that only works
for tightly coupled interfaces like dpif-netdev. It wouldn't really be
OK if you had an implementation on say, Windows but not Linux.
_______________________________________________
dev mailing list
dev@openvswitch.org
http://openvswitch.org/mailman/listinfo/dev
