Use PacketBuffers, rather than VectorisedViews, in netstack.

PacketBuffers are analogous to Linux's sk_buff. They hold all information about
a packet, headers, and payload. This is important for:

* iptables to access various headers of packets
* Preventing the clutter of passing different net and link headers along with
  VectorisedViews to packet handling functions.

This change only affects the incoming packet path, and a future change will
change the outgoing path.

Benchmark               Regular         PacketBufferPtr  PacketBufferConcrete
--------------------------------------------------------------------------------
BM_Recvmsg             400.715MB/s      373.676MB/s      396.276MB/s
BM_Sendmsg             361.832MB/s      333.003MB/s      335.571MB/s
BM_Recvfrom            453.336MB/s      393.321MB/s      381.650MB/s
BM_Sendto              378.052MB/s      372.134MB/s      341.342MB/s
BM_SendmsgTCP/0/1k     353.711MB/s      316.216MB/s      322.747MB/s
BM_SendmsgTCP/0/2k     600.681MB/s      588.776MB/s      565.050MB/s
BM_SendmsgTCP/0/4k     995.301MB/s      888.808MB/s      941.888MB/s
BM_SendmsgTCP/0/8k     1.517GB/s        1.274GB/s        1.345GB/s
BM_SendmsgTCP/0/16k    1.872GB/s        1.586GB/s        1.698GB/s
BM_SendmsgTCP/0/32k    1.017GB/s        1.020GB/s        1.133GB/s
BM_SendmsgTCP/0/64k    475.626MB/s      584.587MB/s      627.027MB/s
BM_SendmsgTCP/0/128k   416.371MB/s      503.434MB/s      409.850MB/s
BM_SendmsgTCP/0/256k   323.449MB/s      449.599MB/s      388.852MB/s
BM_SendmsgTCP/0/512k   243.992MB/s      267.676MB/s      314.474MB/s
BM_SendmsgTCP/0/1M     95.138MB/s       95.874MB/s       95.417MB/s
BM_SendmsgTCP/0/2M     96.261MB/s       94.977MB/s       96.005MB/s
BM_SendmsgTCP/0/4M     96.512MB/s       95.978MB/s       95.370MB/s
BM_SendmsgTCP/0/8M     95.603MB/s       95.541MB/s       94.935MB/s
BM_SendmsgTCP/0/16M    94.598MB/s       94.696MB/s       94.521MB/s
BM_SendmsgTCP/0/32M    94.006MB/s       94.671MB/s       94.768MB/s
BM_SendmsgTCP/0/64M    94.133MB/s       94.333MB/s       94.746MB/s
BM_SendmsgTCP/0/128M   93.615MB/s       93.497MB/s       93.573MB/s
BM_SendmsgTCP/0/256M   93.241MB/s       95.100MB/s       93.272MB/s
BM_SendmsgTCP/1/1k     303.644MB/s      316.074MB/s      308.430MB/s
BM_SendmsgTCP/1/2k     537.093MB/s      584.962MB/s      529.020MB/s
BM_SendmsgTCP/1/4k     882.362MB/s      939.087MB/s      892.285MB/s
BM_SendmsgTCP/1/8k     1.272GB/s        1.394GB/s        1.296GB/s
BM_SendmsgTCP/1/16k    1.802GB/s        2.019GB/s        1.830GB/s
BM_SendmsgTCP/1/32k    2.084GB/s        2.173GB/s        2.156GB/s
BM_SendmsgTCP/1/64k    2.515GB/s        2.463GB/s        2.473GB/s
BM_SendmsgTCP/1/128k   2.811GB/s        3.004GB/s        2.946GB/s
BM_SendmsgTCP/1/256k   3.008GB/s        3.159GB/s        3.171GB/s
BM_SendmsgTCP/1/512k   2.980GB/s        3.150GB/s        3.126GB/s
BM_SendmsgTCP/1/1M     2.165GB/s        2.233GB/s        2.163GB/s
BM_SendmsgTCP/1/2M     2.370GB/s        2.219GB/s        2.453GB/s
BM_SendmsgTCP/1/4M     2.005GB/s        2.091GB/s        2.214GB/s
BM_SendmsgTCP/1/8M     2.111GB/s        2.013GB/s        2.109GB/s
BM_SendmsgTCP/1/16M    1.902GB/s        1.868GB/s        1.897GB/s
BM_SendmsgTCP/1/32M    1.655GB/s        1.665GB/s        1.635GB/s
BM_SendmsgTCP/1/64M    1.575GB/s        1.547GB/s        1.575GB/s
BM_SendmsgTCP/1/128M   1.524GB/s        1.584GB/s        1.580GB/s
BM_SendmsgTCP/1/256M   1.579GB/s        1.607GB/s        1.593GB/s

PiperOrigin-RevId: 278940079

1 files changed, 24 insertions, 24 deletions

diff --git a/pkg/tcpip/stack/nic.go b/pkg/tcpip/stack/nic.go
index 12969c74e..28a28ae6e 100644
--- a/pkg/tcpip/stack/nic.go
+++ b/pkg/tcpip/stack/nic.go
@@ -723,10 +723,10 @@ func (n *NIC) leaveGroupLocked(addr tcpip.Address) *tcpip.Error {
 	return nil
 }
 
-func handlePacket(protocol tcpip.NetworkProtocolNumber, dst, src tcpip.Address, localLinkAddr, remotelinkAddr tcpip.LinkAddress, ref *referencedNetworkEndpoint, vv buffer.VectorisedView) {
+func handlePacket(protocol tcpip.NetworkProtocolNumber, dst, src tcpip.Address, localLinkAddr, remotelinkAddr tcpip.LinkAddress, ref *referencedNetworkEndpoint, pkt tcpip.PacketBuffer) {
 	r := makeRoute(protocol, dst, src, localLinkAddr, ref, false /* handleLocal */, false /* multicastLoop */)
 	r.RemoteLinkAddress = remotelinkAddr
-	ref.ep.HandlePacket(&r, vv)
+	ref.ep.HandlePacket(&r, pkt)
 	ref.decRef()
 }
 
@@ -736,9 +736,9 @@ func handlePacket(protocol tcpip.NetworkProtocolNumber, dst, src tcpip.Address,
 // Note that the ownership of the slice backing vv is retained by the caller.
 // This rule applies only to the slice itself, not to the items of the slice;
 // the ownership of the items is not retained by the caller.
-func (n *NIC) DeliverNetworkPacket(linkEP LinkEndpoint, remote, local tcpip.LinkAddress, protocol tcpip.NetworkProtocolNumber, vv buffer.VectorisedView, linkHeader buffer.View) {
+func (n *NIC) DeliverNetworkPacket(linkEP LinkEndpoint, remote, local tcpip.LinkAddress, protocol tcpip.NetworkProtocolNumber, pkt tcpip.PacketBuffer) {
 	n.stats.Rx.Packets.Increment()
-	n.stats.Rx.Bytes.IncrementBy(uint64(vv.Size()))
+	n.stats.Rx.Bytes.IncrementBy(uint64(pkt.Data.Size()))
 
 	netProto, ok := n.stack.networkProtocols[protocol]
 	if !ok {
@@ -763,22 +763,22 @@ func (n *NIC) DeliverNetworkPacket(linkEP LinkEndpoint, remote, local tcpip.Link
 	}
 	n.mu.RUnlock()
 	for _, ep := range packetEPs {
-		ep.HandlePacket(n.id, local, protocol, vv.Clone(nil), linkHeader)
+		ep.HandlePacket(n.id, local, protocol, pkt.Clone())
 	}
 
 	if netProto.Number() == header.IPv4ProtocolNumber || netProto.Number() == header.IPv6ProtocolNumber {
 		n.stack.stats.IP.PacketsReceived.Increment()
 	}
 
-	if len(vv.First()) < netProto.MinimumPacketSize() {
+	if len(pkt.Data.First()) < netProto.MinimumPacketSize() {
 		n.stack.stats.MalformedRcvdPackets.Increment()
 		return
 	}
 
-	src, dst := netProto.ParseAddresses(vv.First())
+	src, dst := netProto.ParseAddresses(pkt.Data.First())
 
 	if ref := n.getRef(protocol, dst); ref != nil {
-		handlePacket(protocol, dst, src, linkEP.LinkAddress(), remote, ref, vv)
+		handlePacket(protocol, dst, src, linkEP.LinkAddress(), remote, ref, pkt)
 		return
 	}
 
@@ -806,20 +806,20 @@ func (n *NIC) DeliverNetworkPacket(linkEP LinkEndpoint, remote, local tcpip.Link
 		if ok {
 			r.RemoteAddress = src
 			// TODO(b/123449044): Update the source NIC as well.
-			ref.ep.HandlePacket(&r, vv)
+			ref.ep.HandlePacket(&r, pkt)
 			ref.decRef()
 		} else {
 			// n doesn't have a destination endpoint.
 			// Send the packet out of n.
-			hdr := buffer.NewPrependableFromView(vv.First())
-			vv.RemoveFirst()
+			hdr := buffer.NewPrependableFromView(pkt.Data.First())
+			pkt.Data.RemoveFirst()
 
 			// TODO(b/128629022): use route.WritePacket.
-			if err := n.linkEP.WritePacket(&r, nil /* gso */, hdr, vv, protocol); err != nil {
+			if err := n.linkEP.WritePacket(&r, nil /* gso */, hdr, pkt.Data, protocol); err != nil {
 				r.Stats().IP.OutgoingPacketErrors.Increment()
 			} else {
 				n.stats.Tx.Packets.Increment()
-				n.stats.Tx.Bytes.IncrementBy(uint64(hdr.UsedLength() + vv.Size()))
+				n.stats.Tx.Bytes.IncrementBy(uint64(hdr.UsedLength() + pkt.Data.Size()))
 			}
 		}
 		return
@@ -833,7 +833,7 @@ func (n *NIC) DeliverNetworkPacket(linkEP LinkEndpoint, remote, local tcpip.Link
 
 // DeliverTransportPacket delivers the packets to the appropriate transport
 // protocol endpoint.
-func (n *NIC) DeliverTransportPacket(r *Route, protocol tcpip.TransportProtocolNumber, netHeader buffer.View, vv buffer.VectorisedView) {
+func (n *NIC) DeliverTransportPacket(r *Route, protocol tcpip.TransportProtocolNumber, pkt tcpip.PacketBuffer) {
 	state, ok := n.stack.transportProtocols[protocol]
 	if !ok {
 		n.stack.stats.UnknownProtocolRcvdPackets.Increment()
@@ -845,41 +845,41 @@ func (n *NIC) DeliverTransportPacket(r *Route, protocol tcpip.TransportProtocolN
 	// Raw socket packets are delivered based solely on the transport
 	// protocol number. We do not inspect the payload to ensure it's
 	// validly formed.
-	n.stack.demux.deliverRawPacket(r, protocol, netHeader, vv)
+	n.stack.demux.deliverRawPacket(r, protocol, pkt)
 
-	if len(vv.First()) < transProto.MinimumPacketSize() {
+	if len(pkt.Data.First()) < transProto.MinimumPacketSize() {
 		n.stack.stats.MalformedRcvdPackets.Increment()
 		return
 	}
 
-	srcPort, dstPort, err := transProto.ParsePorts(vv.First())
+	srcPort, dstPort, err := transProto.ParsePorts(pkt.Data.First())
 	if err != nil {
 		n.stack.stats.MalformedRcvdPackets.Increment()
 		return
 	}
 
 	id := TransportEndpointID{dstPort, r.LocalAddress, srcPort, r.RemoteAddress}
-	if n.stack.demux.deliverPacket(r, protocol, netHeader, vv, id) {
+	if n.stack.demux.deliverPacket(r, protocol, pkt, id) {
 		return
 	}
 
 	// Try to deliver to per-stack default handler.
 	if state.defaultHandler != nil {
-		if state.defaultHandler(r, id, netHeader, vv) {
+		if state.defaultHandler(r, id, pkt) {
 			return
 		}
 	}
 
 	// We could not find an appropriate destination for this packet, so
 	// deliver it to the global handler.
-	if !transProto.HandleUnknownDestinationPacket(r, id, netHeader, vv) {
+	if !transProto.HandleUnknownDestinationPacket(r, id, pkt) {
 		n.stack.stats.MalformedRcvdPackets.Increment()
 	}
 }
 
 // DeliverTransportControlPacket delivers control packets to the appropriate
 // transport protocol endpoint.
-func (n *NIC) DeliverTransportControlPacket(local, remote tcpip.Address, net tcpip.NetworkProtocolNumber, trans tcpip.TransportProtocolNumber, typ ControlType, extra uint32, vv buffer.VectorisedView) {
+func (n *NIC) DeliverTransportControlPacket(local, remote tcpip.Address, net tcpip.NetworkProtocolNumber, trans tcpip.TransportProtocolNumber, typ ControlType, extra uint32, pkt tcpip.PacketBuffer) {
 	state, ok := n.stack.transportProtocols[trans]
 	if !ok {
 		return
@@ -890,17 +890,17 @@ func (n *NIC) DeliverTransportControlPacket(local, remote tcpip.Address, net tcp
 	// ICMPv4 only guarantees that 8 bytes of the transport protocol will
 	// be present in the payload. We know that the ports are within the
 	// first 8 bytes for all known transport protocols.
-	if len(vv.First()) < 8 {
+	if len(pkt.Data.First()) < 8 {
 		return
 	}
 
-	srcPort, dstPort, err := transProto.ParsePorts(vv.First())
+	srcPort, dstPort, err := transProto.ParsePorts(pkt.Data.First())
 	if err != nil {
 		return
 	}
 
 	id := TransportEndpointID{srcPort, local, dstPort, remote}
-	if n.stack.demux.deliverControlPacket(n, net, trans, typ, extra, vv, id) {
+	if n.stack.demux.deliverControlPacket(n, net, trans, typ, extra, pkt, id) {
 		return
 	}
 }