Linux Kernel
3.7.1
|
struct sk_buff - socket buffer : Next buffer in list : Previous buffer in list : Time we arrived : Socket we are owned by : Device we arrived on/are leaving by : Control buffer. Free for use by every layer. Put private vars here : destination entry (with norefcount bit) : the security path, used for xfrm : Length of actual data : Data length : Length of link layer header : writable header length of cloned skb : Checksum (must include start/offset pair) : Offset from skb->head where checksumming should start : Offset from csum_start where checksum should be stored : Packet queueing priority : allow local fragmentation : Head may be cloned (check refcnt to be sure) : Driver fed us an IP checksum : Payload reference only, must not modify header : Relationship of this skb to the connection : Packet class : skbuff clone status : skbuff is owned by ipvs : this packet has been seen already, so stats have been done for it, don't do them again : netfilter packet trace flag driver : Destruct function : Associated connection, if any : netfilter conntrack re-assembly pointer : Saved data about a bridged frame - see br_netfilter.c : ifindex of device we arrived on : Traffic control index : traffic control verdict : the packet hash computed on receive : Queue mapping for multiqueue devices : router type (from link layer) : allow the mapping of a socket to a queue to be changed : indicate rxhash is a canonical 4-tuple hash over transport ports. : wifi_acked was set : whether frame was acked on wifi or not : Request NIC to treat last 4 bytes as Ethernet FCS : a cookie to one of several possible DMA operations done by skb DMA functions : security marking : Generic packet mark : total number of sk_receive_queue overflows : vlan tag control information : Transport layer header : Network layer header : Link layer header : Tail pointer : End pointer : Head of buffer : Data head pointer : Buffer size : User count - see {datagram,tcp}.c