On 11/19/2011 3:06 PM, Jesse Gross wrote:
> On Fri, Nov 18, 2011 at 9:30 PM, John Fastabend
> <john.r.fastab...@intel.com> wrote:
>> On 11/18/2011 3:12 PM, Jesse Gross wrote:
>>> + */
>>> +enum ovs_frag_type {
>>> + OVS_FRAG_TYPE_NONE,
>>> + OVS_FRAG_TYPE_FIRST,
>>> + OVS_FRAG_TYPE_LATER,
>>> + __OVS_FRAG_TYPE_MAX
>>> +};
>>> +
>>> +#define OVS_FRAG_TYPE_MAX (__OVS_FRAG_TYPE_MAX - 1)
>>> +
>>> +struct ovs_key_ethernet {
>>> + __u8 eth_src[6];
>>> + __u8 eth_dst[6];
>>> +};
>>> +
>>> +struct ovs_key_ipv4 {
>>> + __be32 ipv4_src;
>>> + __be32 ipv4_dst;
>>> + __u8 ipv4_proto;
>>> + __u8 ipv4_tos;
>>> + __u8 ipv4_ttl;
>>> + __u8 ipv4_frag; /* One of OVS_FRAG_TYPE_*. */
>>> +};
>>> +
>>> +struct ovs_key_ipv6 {
>>> + __be32 ipv6_src[4];
>>> + __be32 ipv6_dst[4];
>>> + __be32 ipv6_label; /* 20-bits in least-significant bits. */
>>> + __u8 ipv6_proto;
>>> + __u8 ipv6_tclass;
>>> + __u8 ipv6_hlimit;
>>> + __u8 ipv6_frag; /* One of OVS_FRAG_TYPE_*. */
>>> +};
>>> +
>>> +struct ovs_key_tcp {
>>> + __be16 tcp_src;
>>> + __be16 tcp_dst;
>>> +};
>>> +
>>> +struct ovs_key_udp {
>>> + __be16 udp_src;
>>> + __be16 udp_dst;
>>> +};
>>> +
>>> +struct ovs_key_icmp {
>>> + __u8 icmp_type;
>>> + __u8 icmp_code;
>>> +};
>>> +
>>> +struct ovs_key_icmpv6 {
>>> + __u8 icmpv6_type;
>>> + __u8 icmpv6_code;
>>> +};
>>> +
>>> +struct ovs_key_arp {
>>> + __be32 arp_sip;
>>> + __be32 arp_tip;
>>> + __be16 arp_op;
>>> + __u8 arp_sha[6];
>>> + __u8 arp_tha[6];
>>> +};
>>> +
>>> +struct ovs_key_nd {
>>> + __u32 nd_target[4];
>>> + __u8 nd_sll[6];
>>> + __u8 nd_tll[6];
>>> +};
>>> +
>>
>> We already have defines for many of these headers
>> struct arphdr {
>> __be16 ar_hrd; /* format of hardware address */
>> __be16 ar_pro; /* format of protocol address */
>> unsigned char ar_hln; /* length of hardware address */
>> unsigned char ar_pln; /* length of protocol address */
>> __be16 ar_op; /* ARP opcode (command) */
>>
>> #if 0
>> /*
>> * Ethernet looks like this : This bit is variable sized
>> however...
>> */
>> unsigned char ar_sha[ETH_ALEN]; /* sender hardware
>> address */
>> unsigned char ar_sip[4]; /* sender IP address
>> */
>> unsigned char ar_tha[ETH_ALEN]; /* target hardware
>> address */
>> unsigned char ar_tip[4]; /* target IP address
>> */
>> #endif
>>
>> };
>>
>> Do we have to redefine them here?
>
> These aren't packet format definitions, they're keys for describing a
> flow to userspace. For example, they don't contain checksums or
> lengths because those vary on a per-packet basis.
OK thanks for explaining. I get it now, the other responses seem reasonable
as well.
Thanks,
John.
_______________________________________________
dev mailing list
dev@openvswitch.org
http://openvswitch.org/mailman/listinfo/dev
