netstack/fdbased: add generic segmentation offload (GSO) support

The linux packet socket can handle GSO packets, so we can segment packets to
64K instead of the MTU which is usually 1500.

Here are numbers for the nginx-1m test:
runsc:		579330.01 [Kbytes/sec] received
runsc-gso:	1794121.66 [Kbytes/sec] received
runc:		2122139.06 [Kbytes/sec] received

and for tcp_benchmark:

$ tcp_benchmark  --duration 15   --ideal
[  4]  0.0-15.0 sec  86647 MBytes  48456 Mbits/sec

$ tcp_benchmark --client --duration 15   --ideal
[  4]  0.0-15.0 sec  2173 MBytes  1214 Mbits/sec

$ tcp_benchmark --client --duration 15   --ideal --gso 65536
[  4]  0.0-15.0 sec  19357 MBytes  10825 Mbits/sec

PiperOrigin-RevId: 240809103
Change-Id: I2637f104db28b5d4c64e1e766c610162a195775a

6 files changed, 30 insertions, 14 deletions

diff --git a/pkg/tcpip/network/arp/arp.go b/pkg/tcpip/network/arp/arp.go
index 5ab542f2c..975919e80 100644
--- a/pkg/tcpip/network/arp/arp.go
+++ b/pkg/tcpip/network/arp/arp.go
@@ -79,7 +79,7 @@ func (e *endpoint) MaxHeaderLength() uint16 {
 
 func (e *endpoint) Close() {}
 
-func (e *endpoint) WritePacket(*stack.Route, buffer.Prependable, buffer.VectorisedView, tcpip.TransportProtocolNumber, uint8, stack.PacketLooping) *tcpip.Error {
+func (e *endpoint) WritePacket(*stack.Route, *stack.GSO, buffer.Prependable, buffer.VectorisedView, tcpip.TransportProtocolNumber, uint8, stack.PacketLooping) *tcpip.Error {
 	return tcpip.ErrNotSupported
 }
 
@@ -103,7 +103,7 @@ func (e *endpoint) HandlePacket(r *stack.Route, vv buffer.VectorisedView) {
 		copy(pkt.HardwareAddressSender(), r.LocalLinkAddress[:])
 		copy(pkt.ProtocolAddressSender(), h.ProtocolAddressTarget())
 		copy(pkt.ProtocolAddressTarget(), h.ProtocolAddressSender())
-		e.linkEP.WritePacket(r, hdr, buffer.VectorisedView{}, ProtocolNumber)
+		e.linkEP.WritePacket(r, nil /* gso */, hdr, buffer.VectorisedView{}, ProtocolNumber)
 		fallthrough // also fill the cache from requests
 	case header.ARPReply:
 		addr := tcpip.Address(h.ProtocolAddressSender())
@@ -155,7 +155,7 @@ func (*protocol) LinkAddressRequest(addr, localAddr tcpip.Address, linkEP stack.
 	copy(h.ProtocolAddressSender(), localAddr)
 	copy(h.ProtocolAddressTarget(), addr)
 
-	return linkEP.WritePacket(r, hdr, buffer.VectorisedView{}, ProtocolNumber)
+	return linkEP.WritePacket(r, nil /* gso */, hdr, buffer.VectorisedView{}, ProtocolNumber)
 }
 
 // ResolveStaticAddress implements stack.LinkAddressResolver.
diff --git a/pkg/tcpip/network/ip_test.go b/pkg/tcpip/network/ip_test.go
index 7eb0e697d..d79eba4b0 100644
--- a/pkg/tcpip/network/ip_test.go
+++ b/pkg/tcpip/network/ip_test.go
@@ -145,7 +145,7 @@ func (*testObject) LinkAddress() tcpip.LinkAddress {
 // WritePacket is called by network endpoints after producing a packet and
 // writing it to the link endpoint. This is used by the test object to verify
 // that the produced packet is as expected.
-func (t *testObject) WritePacket(_ *stack.Route, hdr buffer.Prependable, payload buffer.VectorisedView, protocol tcpip.NetworkProtocolNumber) *tcpip.Error {
+func (t *testObject) WritePacket(_ *stack.Route, _ *stack.GSO, hdr buffer.Prependable, payload buffer.VectorisedView, protocol tcpip.NetworkProtocolNumber) *tcpip.Error {
 	var prot tcpip.TransportProtocolNumber
 	var srcAddr tcpip.Address
 	var dstAddr tcpip.Address
@@ -221,7 +221,7 @@ func TestIPv4Send(t *testing.T) {
 	if err != nil {
 		t.Fatalf("could not find route: %v", err)
 	}
-	if err := ep.WritePacket(&r, hdr, payload.ToVectorisedView(), 123, 123, stack.PacketOut); err != nil {
+	if err := ep.WritePacket(&r, nil /* gso */, hdr, payload.ToVectorisedView(), 123, 123, stack.PacketOut); err != nil {
 		t.Fatalf("WritePacket failed: %v", err)
 	}
 }
@@ -450,7 +450,7 @@ func TestIPv6Send(t *testing.T) {
 	if err != nil {
 		t.Fatalf("could not find route: %v", err)
 	}
-	if err := ep.WritePacket(&r, hdr, payload.ToVectorisedView(), 123, 123, stack.PacketOut); err != nil {
+	if err := ep.WritePacket(&r, nil /* gso */, hdr, payload.ToVectorisedView(), 123, 123, stack.PacketOut); err != nil {
 		t.Fatalf("WritePacket failed: %v", err)
 	}
 }
diff --git a/pkg/tcpip/network/ipv4/icmp.go b/pkg/tcpip/network/ipv4/icmp.go
index 8f94246c9..a9650de03 100644
--- a/pkg/tcpip/network/ipv4/icmp.go
+++ b/pkg/tcpip/network/ipv4/icmp.go
@@ -76,7 +76,7 @@ func (e *endpoint) handleICMP(r *stack.Route, netHeader buffer.View, vv buffer.V
 		copy(pkt, h)
 		pkt.SetType(header.ICMPv4EchoReply)
 		pkt.SetChecksum(^header.Checksum(pkt, header.ChecksumVV(vv, 0)))
-		r.WritePacket(hdr, vv, header.ICMPv4ProtocolNumber, r.DefaultTTL())
+		r.WritePacket(nil /* gso */, hdr, vv, header.ICMPv4ProtocolNumber, r.DefaultTTL())
 
 	case header.ICMPv4EchoReply:
 		if len(v) < header.ICMPv4EchoMinimumSize {
diff --git a/pkg/tcpip/network/ipv4/ipv4.go b/pkg/tcpip/network/ipv4/ipv4.go
index c2f9a1bcf..cbdca98a5 100644
--- a/pkg/tcpip/network/ipv4/ipv4.go
+++ b/pkg/tcpip/network/ipv4/ipv4.go
@@ -99,8 +99,16 @@ func (e *endpoint) MaxHeaderLength() uint16 {
 	return e.linkEP.MaxHeaderLength() + header.IPv4MinimumSize
 }
 
+// GSOMaxSize returns the maximum GSO packet size.
+func (e *endpoint) GSOMaxSize() uint32 {
+	if gso, ok := e.linkEP.(stack.GSOEndpoint); ok {
+		return gso.GSOMaxSize()
+	}
+	return 0
+}
+
 // WritePacket writes a packet to the given destination address and protocol.
-func (e *endpoint) WritePacket(r *stack.Route, hdr buffer.Prependable, payload buffer.VectorisedView, protocol tcpip.TransportProtocolNumber, ttl uint8, loop stack.PacketLooping) *tcpip.Error {
+func (e *endpoint) WritePacket(r *stack.Route, gso *stack.GSO, hdr buffer.Prependable, payload buffer.VectorisedView, protocol tcpip.TransportProtocolNumber, ttl uint8, loop stack.PacketLooping) *tcpip.Error {
 	ip := header.IPv4(hdr.Prepend(header.IPv4MinimumSize))
 	length := uint16(hdr.UsedLength() + payload.Size())
 	id := uint32(0)
@@ -132,7 +140,7 @@ func (e *endpoint) WritePacket(r *stack.Route, hdr buffer.Prependable, payload b
 	}
 
 	r.Stats().IP.PacketsSent.Increment()
-	return e.linkEP.WritePacket(r, hdr, payload, ProtocolNumber)
+	return e.linkEP.WritePacket(r, gso, hdr, payload, ProtocolNumber)
 }
 
 // HandlePacket is called by the link layer when new ipv4 packets arrive for
diff --git a/pkg/tcpip/network/ipv6/icmp.go b/pkg/tcpip/network/ipv6/icmp.go
index cfc05d9e1..36d98caef 100644
--- a/pkg/tcpip/network/ipv6/icmp.go
+++ b/pkg/tcpip/network/ipv6/icmp.go
@@ -118,7 +118,7 @@ func (e *endpoint) handleICMP(r *stack.Route, netHeader buffer.View, vv buffer.V
 		defer r.Release()
 		r.LocalAddress = targetAddr
 		pkt.SetChecksum(icmpChecksum(pkt, r.LocalAddress, r.RemoteAddress, buffer.VectorisedView{}))
-		r.WritePacket(hdr, buffer.VectorisedView{}, header.ICMPv6ProtocolNumber, r.DefaultTTL())
+		r.WritePacket(nil /* gso */, hdr, buffer.VectorisedView{}, header.ICMPv6ProtocolNumber, r.DefaultTTL())
 
 		e.linkAddrCache.AddLinkAddress(e.nicid, r.RemoteAddress, r.RemoteLinkAddress)
 
@@ -143,7 +143,7 @@ func (e *endpoint) handleICMP(r *stack.Route, netHeader buffer.View, vv buffer.V
 		copy(pkt, h)
 		pkt.SetType(header.ICMPv6EchoReply)
 		pkt.SetChecksum(icmpChecksum(pkt, r.LocalAddress, r.RemoteAddress, vv))
-		r.WritePacket(hdr, vv, header.ICMPv6ProtocolNumber, r.DefaultTTL())
+		r.WritePacket(nil /* gso */, hdr, vv, header.ICMPv6ProtocolNumber, r.DefaultTTL())
 
 	case header.ICMPv6EchoReply:
 		if len(v) < header.ICMPv6EchoMinimumSize {
@@ -202,7 +202,7 @@ func (*protocol) LinkAddressRequest(addr, localAddr tcpip.Address, linkEP stack.
 		DstAddr:       r.RemoteAddress,
 	})
 
-	return linkEP.WritePacket(r, hdr, buffer.VectorisedView{}, ProtocolNumber)
+	return linkEP.WritePacket(r, nil /* gso */, hdr, buffer.VectorisedView{}, ProtocolNumber)
 }
 
 // ResolveStaticAddress implements stack.LinkAddressResolver.
diff --git a/pkg/tcpip/network/ipv6/ipv6.go b/pkg/tcpip/network/ipv6/ipv6.go
index df3b64c98..9a743ea80 100644
--- a/pkg/tcpip/network/ipv6/ipv6.go
+++ b/pkg/tcpip/network/ipv6/ipv6.go
@@ -83,8 +83,16 @@ func (e *endpoint) MaxHeaderLength() uint16 {
 	return e.linkEP.MaxHeaderLength() + header.IPv6MinimumSize
 }
 
+// GSOMaxSize returns the maximum GSO packet size.
+func (e *endpoint) GSOMaxSize() uint32 {
+	if gso, ok := e.linkEP.(stack.GSOEndpoint); ok {
+		return gso.GSOMaxSize()
+	}
+	return 0
+}
+
 // WritePacket writes a packet to the given destination address and protocol.
-func (e *endpoint) WritePacket(r *stack.Route, hdr buffer.Prependable, payload buffer.VectorisedView, protocol tcpip.TransportProtocolNumber, ttl uint8, loop stack.PacketLooping) *tcpip.Error {
+func (e *endpoint) WritePacket(r *stack.Route, gso *stack.GSO, hdr buffer.Prependable, payload buffer.VectorisedView, protocol tcpip.TransportProtocolNumber, ttl uint8, loop stack.PacketLooping) *tcpip.Error {
 	length := uint16(hdr.UsedLength() + payload.Size())
 	ip := header.IPv6(hdr.Prepend(header.IPv6MinimumSize))
 	ip.Encode(&header.IPv6Fields{
@@ -107,7 +115,7 @@ func (e *endpoint) WritePacket(r *stack.Route, hdr buffer.Prependable, payload b
 	}
 
 	r.Stats().IP.PacketsSent.Increment()
-	return e.linkEP.WritePacket(r, hdr, payload, ProtocolNumber)
+	return e.linkEP.WritePacket(r, gso, hdr, payload, ProtocolNumber)
 }
 
 // HandlePacket is called by the link layer when new ipv6 packets arrive for