udp通信的Connect 和 Read 结束明天看Waiter 这相当于linux内核的事件驱动机制

当有某种事件就绪后通知waiter 监听着waiter的监听者就能通过waiter得知事件已经发生从而不再阻塞
2025-10-06 21:32:59 +08:00 · 2022-12-01 22:36:40 +08:00
parent 3d8ca3c0c8
commit be40f904fc
11 changed files with 741 additions and 123 deletions
--- a/cmd/netstack/main.go
+++ b/cmd/netstack/main.go
@@ -12,36 +12,31 @@ import (
 	"netstack/tcpip/link/tuntap"
 	"netstack/tcpip/network/arp"
 	"netstack/tcpip/network/ipv4"
-	"netstack/tcpip/network/ipv6"
 	"netstack/tcpip/stack"
 	"netstack/tcpip/transport/udp"
 	"netstack/waiter"
 	"os"
 	"strings"
+	"time"
 )

-var mac = flag.String("mac", "01:01:01:01:01:01", "mac address to use in tap device")
-
 func main() {
 	flag.Parse()
-	if len(flag.Args()) != 2 {
-		log.Fatal("Usage: ", os.Args[0], " <tap-device> <listen-address>")
+	if len(flag.Args()) < 2 {
+		log.Fatal("Usage: ", os.Args[0], " <tap-device> <local-address/mask>")
 	}

 	log.SetFlags(log.Lshortfile | log.LstdFlags)
 	tapName := flag.Arg(0)
-	listeAddr := flag.Arg(1)
+	cidrName := flag.Arg(1)

-	log.Printf("tap: %v, listeAddr: %v", tapName, listeAddr)
+	log.Printf("tap: %v, cidrName: %v", tapName, cidrName)

-	// Parse the mac address.
-	maddr, err := net.ParseMAC(*mac)
+	parsedAddr, cidr, err := net.ParseCIDR(cidrName)
 	if err != nil {
-		log.Fatalf("Bad MAC address: %v", *mac)
+		log.Fatalf("Bad cidr: %v", cidrName)
 	}

-	parsedAddr := net.ParseIP(listeAddr)
-
 	// 解析地址ip地址，ipv4或者ipv6地址都支持
 	var addr tcpip.Address
 	var proto tcpip.NetworkProtocolNumber
@@ -50,7 +45,7 @@ func main() {
 		proto = ipv4.ProtocolNumber
 	} else if parsedAddr.To16() != nil {
 		addr = tcpip.Address(parsedAddr.To16())
-		proto = ipv6.ProtocolNumber
+		//proto = ipv6.ProtocolNumber
 	} else {
 		log.Fatalf("Unknown IP type: %v", parsedAddr)
 	}
@@ -69,17 +64,22 @@ func main() {
 	}

 	// 启动tap网卡
-	_ = tuntap.SetLinkUp(tapName)
-	// 设置tap网卡IP地址
-	_ = tuntap.AddIP(tapName, listeAddr)
+	tuntap.SetLinkUp(tapName)
+	// 设置路由
+	tuntap.SetRoute(tapName, cidr.String())
+
+	// 获取mac地址
+	mac, err := tuntap.GetHardwareAddr(tapName)
+	if err != nil {
+		panic(err)
+	}

 	// 抽象网卡的文件接口
 	linkID := fdbased.New(&fdbased.Options{
 		FD:      fd,
 		MTU:     1500,
-		Address: tcpip.LinkAddress(maddr),
+		Address: tcpip.LinkAddress(mac),
 	})
-
 	// 新建相关协议的协议栈
 	s := stack.New([]string{ipv4.ProtocolName, arp.ProtocolName},
 		[]string{ /*tcp.ProtocolName, */ udp.ProtocolName}, stack.Options{})
@@ -109,37 +109,33 @@ func main() {
 		},
 	})

-	conn, _ := net.Listen("tcp", "0.0.0.0:9999")
-	TCPServer(conn, &RCV{s, nil, nil})
-	// 同时监听tcp和udp localPort端口
-	//tcpEp := tcpListen(s, proto, localPort)
-	//udpEp := udpListen(s, proto, localPort)
-	// 关闭监听服务，此时会释放端口
-	//tcpEp.Close()
-	//udpEp.Close()
-}
+	go func() {
+		// 监听udp localPort端口
+		udpEp := udpListen(s, proto, 9999)

-//func tcpListen(s *stack.Stack, proto tcpip.NetworkProtocolNumber, localPort int) tcpip.Endpoint {
-//	var wq waiter.Queue
-//	// 新建一个tcp端
-//	ep, err := s.NewEndpoint(tcp.ProtocolNumber, proto, &wq)
-//	if err != nil {
-//		log.Fatal(err)
-//	}
-//
-//	// 绑定IP和端口，这里的IP地址为空，表示绑定任何IP
-//	// 此时就会调用端口管理器
-//	if err := ep.Bind(tcpip.FullAddress{0, "", uint16(localPort)}, nil); err != nil {
-//		log.Fatal("Bind failed: ", err)
-//	}
-//
-//	// 开始监听
-//	if err := ep.Listen(10); err != nil {
-//		log.Fatal("Listen failed: ", err)
-//	}
-//
-//	return ep
-//}
+		for {
+			buf, _, err := udpEp.Read(nil)
+			if err != nil {
+				if err == tcpip.ErrWouldBlock {
+					time.Sleep(100 * time.Millisecond)
+					log.Println("阻塞中")
+					continue
+				}
+			}
+			log.Println(buf)
+			break
+		}
+		// 关闭监听服务，此时会释放端口
+		udpEp.Close()
+	}()
+
+	conn, _ := net.Listen("tcp", "0.0.0.0:9999")
+	rcv := &RCV{
+		Stack: s,
+		addr:  tcpip.FullAddress{},
+	}
+	TCPServer(conn, rcv)
+}

 func udpListen(s *stack.Stack, proto tcpip.NetworkProtocolNumber, localPort int) tcpip.Endpoint {
 	var wq waiter.Queue
@@ -156,10 +152,6 @@ func udpListen(s *stack.Stack, proto tcpip.NetworkProtocolNumber, localPort int)
 		log.Fatal("Bind failed: ", err)
 	}

-	if err := ep.Connect(tcpip.FullAddress{NIC: 0, Addr: "", Port: uint16(localPort)}); err != nil {
-		log.Fatal("Conn failed: ", err)
-	}
-
 	// 注意UDP是无连接的，它不需要Listen
 	return ep
 }
@@ -167,6 +159,7 @@ func udpListen(s *stack.Stack, proto tcpip.NetworkProtocolNumber, localPort int)
 type RCV struct {
 	*stack.Stack
 	ep     tcpip.Endpoint
+	addr   tcpip.FullAddress
 	rcvBuf []byte
 }

@@ -189,6 +182,7 @@ func (r *RCV) Handle(conn net.Conn) {
 		}
 		r.ep = ep
 		r.Bind()
+		r.Connect()
 		r.Close()
 	case "tcp":
 	default:
@@ -202,12 +196,16 @@ func (r *RCV) Bind() {
 		return
 	}
 	port := binary.BigEndian.Uint16(r.rcvBuf[7:9])
-	addr := tcpip.FullAddress{
-		NIC:  0,
+	r.addr = tcpip.FullAddress{
+		NIC:  1,
 		Addr: tcpip.Address(r.rcvBuf[3:7]),
 		Port: port,
 	}
-	r.ep.Bind(addr, nil)
+	r.ep.Bind(r.addr, nil)
+}
+
+func (r *RCV) Connect() {
+	r.ep.Connect(tcpip.FullAddress{NIC: 1, Addr: "\xc0\xa8\x01\x02", Port: 8888})
 }

 func (r *RCV) Close() {
--- a/cmd/udp_client/main.go
+++ b/cmd/udp_client/main.go
@@ -0,0 +1,42 @@
+package main
+
+import (
+	"flag"
+	"log"
+	"net"
+)
+
+func main() {
+	var (
+		addr = flag.String("a", "192.168.1.1:9999", "udp dst address")
+	)
+
+	log.SetFlags(log.Lshortfile | log.LstdFlags)
+
+	udpAddr, err := net.ResolveUDPAddr("udp", *addr)
+	if err != nil {
+		panic(err)
+	}
+	log.Println("解析地址")
+
+	// 建立UDP连接（只是填息了目的IP和端口，并未真正的建立连接）
+	conn, err := net.DialUDP("udp", nil, udpAddr)
+	if err != nil {
+		panic(err)
+	}
+	log.Println("TEST")
+
+	send := []byte("hello")
+
+	if _, err := conn.Write(send); err != nil {
+		panic(err)
+	}
+	log.Printf("send: %s", string(send))
+
+	//recv := make([]byte, 10)
+	//rn, _, err := conn.ReadFrom(recv)
+	//if err != nil {
+	//	panic(err)
+	//}
+	//log.Printf("recv: %s", string(recv[:rn]))
+}
--- a/tcpip/header/checksum.go
+++ b/tcpip/header/checksum.go
@@ -2,7 +2,10 @@ package header

 import "netstack/tcpip"

-// 校验和的计算
+// Checksum 校验和的计算
+// UDP 检验和的计算方法是： 按每 16 位求和得出一个 32 位的数；
+// 如果这个 32 位的数，高 16 位不为 0，则高 16 位加低 16 位再得到一个 32 位的数；
+// 重复第 2 步直到高 16 位为 0，将低 16 位取反，得到校验和。
 func Checksum(buf []byte, initial uint16) uint16 {
 	v := uint32(initial)

--- a/tcpip/header/udp.go
+++ b/tcpip/header/udp.go
@@ -1,6 +1,10 @@
 package header

-import "netstack/tcpip"
+import (
+	"encoding/binary"
+	"fmt"
+	"netstack/tcpip"
+)

 const (
 	udpSrcPort  = 0
@@ -36,3 +40,99 @@ const (
 	// UDPProtocolNumber is UDP's transport protocol number.
 	UDPProtocolNumber tcpip.TransportProtocolNumber = 17
 )
+
+/*
+UDP 是 User Datagram Protocol 的简称，中文名是用户数据报协议。UDP 只在 IP 数据报服务上增加了一点功能，就是复用和分用的功能以及差错检测，UDP 主要的特点是：
+
+1. UDP 是无连接的，即发送数据之前不需要建立连接，发送结束也不需要连接释放，因此减少了开销和发送数据之间的延时。
+2. UDP 是不可靠传输，尽最大努力交付，因此不需要维护复杂的连接状态。
+3. UDP 的数据报是有消息边界的，发送方发送一个报文，接收方就会完整的收到一个报文。
+4. UDP 没有拥塞控制，网络出现阻塞，UDP 是无感知的，也就不会降低发送速度。
+5. UDP 支持一对一，一对多，多对一，多对多的通信。
+*/
+
+/*
+|source Port|destination Port|
+|    Length  |  UDP Checksum |
+|            Data            |
+*/
+
+// SourcePort returns the "source port" field of the udp header.
+func (b UDP) SourcePort() uint16 {
+	return binary.BigEndian.Uint16(b[udpSrcPort:])
+}
+
+// DestinationPort returns the "destination port" field of the udp header.
+func (b UDP) DestinationPort() uint16 {
+	return binary.BigEndian.Uint16(b[udpDstPort:])
+}
+
+// Length returns the "length" field of the udp header.
+func (b UDP) Length() uint16 {
+	return binary.BigEndian.Uint16(b[udpLength:])
+}
+
+// Payload returns the data contained in the UDP datagram.
+func (b UDP) Payload() []byte {
+	return b[UDPMinimumSize:]
+}
+
+// Checksum returns the "checksum" field of the udp header.
+func (b UDP) Checksum() uint16 {
+	return binary.BigEndian.Uint16(b[udpChecksum:])
+}
+
+// SetSourcePort sets the "source port" field of the udp header.
+func (b UDP) SetSourcePort(port uint16) {
+	binary.BigEndian.PutUint16(b[udpSrcPort:], port)
+}
+
+// SetDestinationPort sets the "destination port" field of the udp header.
+func (b UDP) SetDestinationPort(port uint16) {
+	binary.BigEndian.PutUint16(b[udpDstPort:], port)
+}
+
+// SetChecksum sets the "checksum" field of the udp header.
+func (b UDP) SetChecksum(checksum uint16) {
+	binary.BigEndian.PutUint16(b[udpChecksum:], checksum)
+}
+
+// CalculateChecksum calculates the checksum of the udp packet, given the total
+// length of the packet and the checksum of the network-layer pseudo-header
+// (excluding the total length) and the checksum of the payload.
+func (b UDP) CalculateChecksum(partialChecksum uint16, totalLen uint16) uint16 {
+	// Add the length portion of the checksum to the pseudo-checksum.
+	tmp := make([]byte, 2)
+	binary.BigEndian.PutUint16(tmp, totalLen)
+	checksum := Checksum(tmp, partialChecksum)
+
+	// Calculate the rest of the checksum.
+	return Checksum(b[:UDPMinimumSize], checksum)
+}
+
+// Encode encodes all the fields of the udp header.
+func (b UDP) Encode(u *UDPFields) {
+	binary.BigEndian.PutUint16(b[udpSrcPort:], u.SrcPort)
+	binary.BigEndian.PutUint16(b[udpDstPort:], u.DstPort)
+	binary.BigEndian.PutUint16(b[udpLength:], u.Length)
+	binary.BigEndian.PutUint16(b[udpChecksum:], u.Checksum)
+}
+
+var udpFmt string = `
+|% 16s|% 16s|
+|% 16s|% 16s|
+%v
+`
+
+func (b UDP) String() string {
+	for i := range b.Payload() {
+		if i != int(b.Length()-8-1) && b.Payload()[i]^b.Payload()[i+1] != 0 {
+			return fmt.Sprintf(udpFmt, atoi(b.SourcePort()), atoi(b.DestinationPort()),
+				atoi(b.Length()), atoi(b.Checksum()),
+				b.Payload())
+		}
+	}
+	return fmt.Sprintf(udpFmt, atoi(b.SourcePort()), atoi(b.DestinationPort()),
+		atoi(b.Length()), atoi(b.Checksum()),
+		fmt.Sprintf("%v x %d", b.Payload()[0], b.Length()-8))
+}
--- a/tcpip/stack/nic.go
+++ b/tcpip/stack/nic.go
@@ -60,7 +60,7 @@ func newNIC(stack *Stack, id tcpip.NICID, name string, ep LinkEndpoint) *NIC {
 		id:        id,
 		name:      name,
 		linkEP:    ep,
-		demux:     nil, // TODO 需要处理
+		demux:     newTransportDemuxer(stack), // NOTE 注册网卡自己的传输层分流器
 		primary:   make(map[tcpip.NetworkProtocolNumber]*ilist.List),
 		endpoints: make(map[NetworkEndpointID]*referencedNetworkEndpoint),
 	}
@@ -302,6 +302,75 @@ func (n *NIC) Subnets() []tcpip.Subnet {
 	return append(sns, n.subnets...)
 }

+// DeliverNetworkPacket 当 NIC 从物理接口接收数据包时，将调用函数 DeliverNetworkPacket，用来分发网络层数据包。
+// 比如 protocol 是 arp 协议号，那么会找到arp.HandlePacket来处理数据报。
+// 简单来说就是根据网络层协议和目的地址来找到相应的网络层端，将网络层数据发给它，
+// 当前实现的网络层协议有 arp、ipv4 和 ipv6。
+func (n *NIC) DeliverNetworkPacket(linkEP LinkEndpoint, remoteLinkAddr, localLinkAddr tcpip.LinkAddress,
+	protocol tcpip.NetworkProtocolNumber, vv buffer.VectorisedView) {
+	netProto, ok := n.stack.networkProtocols[protocol]
+	if !ok {
+		n.stack.stats.UnknownProtocolRcvdPackets.Increment()
+		return
+	}
+
+	if netProto.Number() == header.IPv4ProtocolNumber || netProto.Number() == header.IPv6ProtocolNumber {
+		n.stack.stats.IP.PacketsReceived.Increment()
+	}
+
+	if len(vv.First()) < netProto.MinimumPacketSize() {
+		n.stack.stats.MalformedRcvdPackets.Increment()
+		return
+	}
+	src, dst := netProto.ParseAddresses(vv.First())
+	log.Printf("设备[%v]准备从 [%s] 向 [%s] 分发数据: %v\n", linkEP.LinkAddress(), src, dst, func() []byte {
+		if len(vv.ToView()) > 64 {
+			return vv.ToView()[:64]
+		}
+		return vv.ToView()
+	}())
+	// 根据网络协议和数据包的目的地址，找到网络端
+	// 然后将数据包分发给网络层
+	if ref := n.getRef(protocol, dst); ref != nil {
+		r := makeRoute(protocol, dst, src, linkEP.LinkAddress(), ref)
+		r.RemoteLinkAddress = remoteLinkAddr
+		ref.ep.HandlePacket(&r, vv)
+		ref.decRef()
+		return
+	}
+
+	if n.stack.Forwarding() {
+		r, err := n.stack.FindRoute(0, "", dst, protocol)
+		if err != nil {
+			n.stack.stats.IP.InvalidAddressesReceived.Increment()
+			return
+		}
+		defer r.Release()
+
+		r.LocalLinkAddress = n.linkEP.LinkAddress()
+		r.RemoteLinkAddress = remoteLinkAddr
+
+		// Found a NIC.
+		n := r.ref.nic
+		n.mu.RLock()
+		ref, ok := n.endpoints[NetworkEndpointID{dst}]
+		n.mu.RUnlock()
+		if ok && ref.tryIncRef() {
+			ref.ep.HandlePacket(&r, vv)
+			ref.decRef()
+		} else {
+			// n doesn't have a destination endpoint.
+			// Send the packet out of n.
+			hdr := buffer.NewPrependableFromView(vv.First())
+			vv.RemoveFirst()
+			n.linkEP.WritePacket(&r, hdr, vv, protocol)
+		}
+		return
+	}
+
+	n.stack.stats.IP.InvalidAddressesReceived.Increment()
+}
+
 // 根据协议类型和目标地址，找出关联的Endpoint
 func (n *NIC) getRef(protocol tcpip.NetworkProtocolNumber, dst tcpip.Address) *referencedNetworkEndpoint {
 	id := NetworkEndpointID{dst}
@@ -344,57 +413,49 @@ func (n *NIC) getRef(protocol tcpip.NetworkProtocolNumber, dst tcpip.Address) *r
 	return nil
 }

-// DeliverNetworkPacket 当 NIC 从物理接口接收数据包时，将调用函数 DeliverNetworkPacket，用来分发网络层数据包。
-// 比如 protocol 是 arp 协议号，那么会找到arp.HandlePacket来处理数据报。
-// 简单来说就是根据网络层协议和目的地址来找到相应的网络层端，将网络层数据发给它，
-// 当前实现的网络层协议有 arp、ipv4 和 ipv6。
-func (n *NIC) DeliverNetworkPacket(linkEP LinkEndpoint, remoteLinkAddr, localLinkAddr tcpip.LinkAddress,
-	protocol tcpip.NetworkProtocolNumber, vv buffer.VectorisedView) {
-	netProto, ok := n.stack.networkProtocols[protocol]
-	if !ok {
-		n.stack.stats.UnknownProtocolRcvdPackets.Increment()
-		return
-	}
-
-	if netProto.Number() == header.IPv4ProtocolNumber || netProto.Number() == header.IPv6ProtocolNumber {
-		n.stack.stats.IP.PacketsReceived.Increment()
-	}
-
-	if len(vv.First()) < netProto.MinimumPacketSize() {
-		n.stack.stats.MalformedRcvdPackets.Increment()
-		return
-	}
-	src, dst := netProto.ParseAddresses(vv.First())
-	log.Printf("设备[%v]准备从 [%s] 向 [%s] 分发数据: %v\n", linkEP.LinkAddress(), src, dst, func() []byte {
-		if len(vv.ToView()) > 64 {
-			return vv.ToView()[:64]
-		}
-		return vv.ToView()
-	}())
-	// 根据网络协议和数据包的目的地址，找到网络端
-	// 然后将数据包分发给网络层
-	if ref := n.getRef(protocol, dst); ref != nil {
-		r := makeRoute(protocol, dst, src, linkEP.LinkAddress(), ref)
-		r.RemoteLinkAddress = remoteLinkAddr
-		ref.ep.HandlePacket(&r, vv)
-		ref.decRef()
-
-		return
-	}
-	n.stack.stats.IP.InvalidAddressesReceived.Increment()
-}
-
 // DeliverTransportPacket delivers packets to the appropriate
 // transport protocol endpoint.
 func (n *NIC) DeliverTransportPacket(r *Route, protocol tcpip.TransportProtocolNumber, vv buffer.VectorisedView) {
 	// 先查找协议栈是否注册了该传输层协议
-	_, ok := n.stack.transportProtocols[protocol]
+	state, ok := n.stack.transportProtocols[protocol]
 	if !ok {
 		n.stack.stats.UnknownProtocolRcvdPackets.Increment()
 		return
 	}
-	log.Println("准备分发传输层数据报", n.stack.transportProtocols)
+	transProto := state.proto
+	// 如果报文长度比该协议最小报文长度还小，那么丢弃它
+	if len(vv.First()) < transProto.MinimumPacketSize() {
+		n.stack.stats.MalformedRcvdPackets.Increment()
+		return
+	}
+	// 解析报文得到源端口和目的端口
+	srcPort, dstPort, err := transProto.ParsePorts(vv.First())
+	if err != nil {
+		n.stack.stats.MalformedRcvdPackets.Increment()
+		return
+	}
+	log.Println("准备分发传输层数据报", n.stack.transportProtocols, srcPort, dstPort)
+	id := TransportEndpointID{dstPort, r.LocalAddress, srcPort, r.RemoteAddress}
+	// 调用分流器，根据传输层协议和传输层id分发数据报文
+	if n.demux.deliverPacket(r, protocol, vv, id) {
+		return
+	}
+	if n.stack.demux.deliverPacket(r, protocol, vv, id) {
+		return
+	}

+	// Try to deliver to per-stack default handler.
+	if state.defaultHandler != nil {
+		if state.defaultHandler(r, id, vv) {
+			return
+		}
+	}
+
+	// We could not find an appropriate destination for this packet, so
+	// deliver it to the global handler.
+	if !transProto.HandleUnknownDestinationPacket(r, id, vv) {
+		n.stack.stats.MalformedRcvdPackets.Increment()
+	}
 }

 // DeliverTransportControlPacket delivers control packets to the
--- a/tcpip/stack/registration.go
+++ b/tcpip/stack/registration.go
@@ -185,7 +185,7 @@ type NetworkEndpointID struct {
 type TransportEndpointID struct {
 	LocalPort     uint16
 	LocalAddress  tcpip.Address
-	remotePort    uint16
+	RemotePort    uint16
 	RemoteAddress tcpip.Address
 }

--- a/tcpip/stack/stack.go
+++ b/tcpip/stack/stack.go
@@ -116,12 +116,87 @@ func New(network []string, transport []string, opts Options) *Stack {
 			proto: transProto,
 		}
 	}
-	// 添加传输层分流器
+	// NOTE 添加协议栈全局传输层分流器
 	s.demux = newTransportDemuxer(s)

 	return s
 }

+// SetNetworkProtocolOption allows configuring individual protocol level
+// options. This method returns an error if the protocol is not supported or
+// option is not supported by the protocol implementation or the provided value
+// is incorrect.
+func (s *Stack) SetNetworkProtocolOption(network tcpip.NetworkProtocolNumber, option interface{}) *tcpip.Error {
+	netProto, ok := s.networkProtocols[network]
+	if !ok {
+		return tcpip.ErrUnknownProtocol
+	}
+	return netProto.SetOption(option)
+}
+
+// NetworkProtocolOption allows retrieving individual protocol level option
+// values. This method returns an error if the protocol is not supported or
+// option is not supported by the protocol implementation.
+// e.g.
+// var v ipv4.MyOption
+// err := s.NetworkProtocolOption(tcpip.IPv4ProtocolNumber, &v)
+//
+//	if err != nil {
+//	  ...
+//	}
+func (s *Stack) NetworkProtocolOption(network tcpip.NetworkProtocolNumber, option interface{}) *tcpip.Error {
+	netProto, ok := s.networkProtocols[network]
+	if !ok {
+		return tcpip.ErrUnknownProtocol
+	}
+	return netProto.Option(option)
+}
+
+// SetTransportProtocolOption allows configuring individual protocol level
+// options. This method returns an error if the protocol is not supported or
+// option is not supported by the protocol implementation or the provided value
+// is incorrect.
+func (s *Stack) SetTransportProtocolOption(transport tcpip.TransportProtocolNumber, option interface{}) *tcpip.Error {
+	transProtoState, ok := s.transportProtocols[transport]
+	if !ok {
+		return tcpip.ErrUnknownProtocol
+	}
+	return transProtoState.proto.SetOption(option)
+}
+
+// TransportProtocolOption allows retrieving individual protocol level option
+// values. This method returns an error if the protocol is not supported or
+// option is not supported by the protocol implementation.
+// var v tcp.SACKEnabled
+//
+//	if err := s.TransportProtocolOption(tcpip.TCPProtocolNumber, &v); err != nil {
+//	  ...
+//	}
+func (s *Stack) TransportProtocolOption(transport tcpip.TransportProtocolNumber, option interface{}) *tcpip.Error {
+	transProtoState, ok := s.transportProtocols[transport]
+	if !ok {
+		return tcpip.ErrUnknownProtocol
+	}
+	return transProtoState.proto.Option(option)
+}
+
+// SetTransportProtocolHandler sets the per-stack default handler for the given
+// protocol.
+//
+// It must be called only during initialization of the stack. Changing it as the
+// stack is operating is not supported.
+func (s *Stack) SetTransportProtocolHandler(p tcpip.TransportProtocolNumber, h func(*Route, TransportEndpointID, buffer.VectorisedView) bool) {
+	state := s.transportProtocols[p]
+	if state != nil {
+		state.defaultHandler = h
+	}
+}
+
+// NowNanoseconds implements tcpip.Clock.NowNanoseconds.
+func (s *Stack) NowNanoseconds() int64 {
+	return s.clock.NowNanoseconds()
+}
+
 func (s *Stack) Stats() tcpip.Stats {
 	return s.stats
 }
@@ -260,19 +335,19 @@ func (s *Stack) ContainsSubnet(id tcpip.NICID, subnet tcpip.Subnet) (bool, *tcpi
 	return false, tcpip.ErrUnknownNICID
 }

-// 路由查找实现，比如当tcp建立连接时，会用该函数得到路由信息
+// FindRoute 路由查找实现，比如当tcp建立连接时，会用该函数得到路由信息
 func (s *Stack) FindRoute(id tcpip.NICID, localAddr, remoteAddr tcpip.Address,
 	netProto tcpip.NetworkProtocolNumber) (Route, *tcpip.Error) {
 	s.mu.RLock()
 	defer s.mu.RUnlock()

 	for i := range s.routeTable {
-		if (id != 0 && id != s.routeTable[i].NIC) ||
+		if (id != 0 && id != s.routeTable[i].NIC) || // 检查是否是对应的网卡
 			(len(remoteAddr) != 0 && !s.routeTable[i].Match(remoteAddr)) {
 			continue
 		}

-		nic := s.nics[s.routeTable[i].NIC]
+		nic := s.nics[s.routeTable[i].NIC] // 在协议栈里找到这张网卡
 		if nic == nil {
 			continue
 		}
@@ -372,14 +447,34 @@ func (s *Stack) RemoveWaker(nicid tcpip.NICID, addr tcpip.Address, waker *sleep.
 // 最终调用 demuxer.registerEndpoint 函数来实现注册。
 func (s *Stack) RegisterTransportEndpoint(nicID tcpip.NICID, netProtos []tcpip.NetworkProtocolNumber,
 	protocol tcpip.TransportProtocolNumber, id TransportEndpointID, ep TransportEndpoint) *tcpip.Error {
-	// TODO 需要实现
-	return nil
+	log.Println("往", nicID, "网卡注册新的传输端")
+	if nicID == 0 {
+		return s.demux.registerEndpoint(netProtos, protocol, id, ep) // 给协议栈的所有网卡注册传输端
+	}
+	s.mu.RLock()
+	defer s.mu.RUnlock()
+
+	nic := s.nics[nicID]
+	if nic == nil {
+		return tcpip.ErrUnknownNICID
+	}
+	return nic.demux.registerEndpoint(netProtos, protocol, id, ep) // 给这张网卡注册传输端
 }

 // UnregisterTransportEndpoint removes the endpoint with the given id from the
 // stack transport dispatcher.
 func (s *Stack) UnregisterTransportEndpoint(nicID tcpip.NICID, netProtos []tcpip.NetworkProtocolNumber,
 	protocol tcpip.TransportProtocolNumber, id TransportEndpointID) {
+	if nicID == 0 {
+		s.demux.unregisterEndpoint(netProtos, protocol, id)
+		return
+	}
+	s.mu.RLock()
+	defer s.mu.RUnlock()
+	nic := s.nics[nicID]
+	if nic != nil {
+		nic.demux.unregisterEndpoint(netProtos, protocol, id)
+	}

 }

--- a/tcpip/stack/transport_demuxer.go
+++ b/tcpip/stack/transport_demuxer.go
@@ -2,6 +2,7 @@ package stack

 import (
 	"netstack/tcpip"
+	"netstack/tcpip/buffer"
 	"sync"
 )

@@ -23,6 +24,112 @@ type transportDemuxer struct {
 }

 // 新建一个分流器
-func newTransportDemuxer(stacl *Stack) *transportDemuxer {
+func newTransportDemuxer(stack *Stack) *transportDemuxer {
+	d := &transportDemuxer{protocol: make(map[protocolIDs]*transportEndpoints)}
+
+	for netProto := range stack.networkProtocols {
+		for tranProto := range stack.transportProtocols {
+			d.protocol[protocolIDs{network: netProto, transport: tranProto}] = &transportEndpoints{
+				endpoints: make(map[TransportEndpointID]TransportEndpoint),
+			}
+		}
+	}
+	return d
+}
+
+// registerEndpoint 向分发器注册给定端点，以便将与端点ID匹配的数据包传递给它
+func (d *transportDemuxer) registerEndpoint(netProtos []tcpip.NetworkProtocolNumber,
+	protocol tcpip.TransportProtocolNumber, id TransportEndpointID, ep TransportEndpoint) *tcpip.Error {
+	for i, n := range netProtos {
+		if err := d.singleRegisterEndpoint(n, protocol, id, ep); err != nil {
+			d.unregisterEndpoint(netProtos[:i], protocol, id) // 把刚才注册的注销掉
+			return err
+		}
+	}
 	return nil
 }
+
+func (d *transportDemuxer) singleRegisterEndpoint(netProto tcpip.NetworkProtocolNumber,
+	protocol tcpip.TransportProtocolNumber, id TransportEndpointID, ep TransportEndpoint) *tcpip.Error {
+	eps, ok := d.protocol[protocolIDs{netProto, protocol}] // IPv4:udp
+	if !ok {                                               // 未曾注册过这个传输端集合
+		return nil
+	}
+
+	eps.mu.Lock()
+	defer eps.mu.Unlock()
+
+	if _, ok := eps.endpoints[id]; ok { // 遍历传输端集合
+		return tcpip.ErrPortInUse
+	}
+	eps.endpoints[id] = ep
+	return nil
+}
+
+// unregisterEndpoint 使用给定的id注销端点，使其不再接收任何数据包
+func (d *transportDemuxer) unregisterEndpoint(netProtos []tcpip.NetworkProtocolNumber,
+	protocol tcpip.TransportProtocolNumber, id TransportEndpointID) {
+	for _, n := range netProtos {
+		if eps, ok := d.protocol[protocolIDs{n, protocol}]; ok {
+			eps.mu.Lock()
+			delete(eps.endpoints, id)
+			eps.mu.Unlock()
+		}
+	}
+}
+
+// 根据传输层的id来找到对应的传输端，再将数据包交给这个传输端处理
+func (d *transportDemuxer) deliverPacket(r *Route, protocol tcpip.TransportProtocolNumber, vv buffer.VectorisedView, id TransportEndpointID) bool {
+	// 先看看分流器里有没有注册相关协议端，如果没有则返回false
+	eps, ok := d.protocol[protocolIDs{r.NetProto, protocol}]
+	if !ok {
+		return false
+	}
+	// 从 eps 中找符合 id 的传输端
+	eps.mu.RLock()
+	ep := d.findEndpointLocked(eps, vv, id)
+	eps.mu.RUnlock()
+
+	if ep == nil {
+		return false
+	}
+
+	// Deliver the packet
+	ep.HandlePacket(r, id, vv)
+
+	return true
+}
+
+func (d *transportDemuxer) deliverControlPacket(net tcpip.NetworkProtocolNumber,
+	trans tcpip.TransportProtocolNumber, typ ControlType, extra uint32, vv buffer.VectorisedView, id TransportEndpointID) bool {
+	return false
+}
+
+// 根据传输层id来找到相应的传输层端
+func (d *transportDemuxer) findEndpointLocked(eps *transportEndpoints,
+	vv buffer.VectorisedView, id TransportEndpointID) TransportEndpoint {
+	if ep := eps.endpoints[id]; ep != nil { // IPv4:udp
+		return ep
+	}
+	// Try to find a match with the id minus the local address.
+	nid := id
+	// 如果上面的 endpoints 没有找到，那么去掉本地ip地址，看看有没有相应的传输层端
+	// 因为有时候传输层监听的时候没有绑定本地ip，也就是 any address，此时的 LocalAddress
+	// 为空。
+	nid.LocalAddress = ""
+	if ep := eps.endpoints[nid]; ep != nil {
+		return ep
+	}
+
+	// Try to find a match with the id minus the remote part.
+	nid.LocalAddress = id.LocalAddress
+	nid.RemoteAddress = ""
+	nid.RemotePort = 0
+	if ep := eps.endpoints[nid]; ep != nil {
+		return ep
+	}
+
+	// Try to find a match with only the local port.
+	nid.LocalAddress = ""
+	return eps.endpoints[nid]
+}
--- a/tcpip/transport/udp/README.md
+++ b/tcpip/transport/udp/README.md
@@ -3,6 +3,7 @@
 ![img](https://doc.shiyanlou.com/document-uid949121labid10418timestamp1555488741384.png)

 传输层是整个网络体系结构中的关键之一，我们很多编程都是直接和传输层打交道的，我们需要了解以下的概念：
+
 1. 端口的意义 - 上一章已经介绍过了
 2. 无连接 UDP 协议及特点 - 本章介绍
 3. 面向连接 TCP 协议及特点 - 下章会介绍
--- a/tcpip/transport/udp/endpoint.go
+++ b/tcpip/transport/udp/endpoint.go
@@ -13,7 +13,13 @@ import (
 // udp报文结构 当收到udp报文时 会用这个结构来保存udp报文数据
 type udpPacket struct {
 	udpPacketEntry // 链表实现
-	// TODO 需要添加
+	senderAddress  tcpip.FullAddress
+	data           buffer.VectorisedView
+	timestamp      int64
+	hasTimestamp   bool
+	// views is used as buffer for data when its length is large
+	// enough to store a VectorisedView.
+	views [8]buffer.View
 }

 type endpointState int
@@ -40,7 +46,7 @@ type endpoint struct {
 	rcvBufSizeMax int
 	rcvBufSize    int
 	rcvClosed     bool
-	rcvTimestamp  bool
+	rcvTimestamp  bool // 通过SetSocket进行设置 是否开启时间戳

 	// The following fields are protected by the mu mutex.
 	mu           sync.RWMutex
@@ -57,7 +63,7 @@ type endpoint struct {
 	// shutdownFlags represent the current shutdown state of the endpoint.
 	shutdownFlags tcpip.ShutdownFlags

-	// TODO
+	multicastMemberships []multicastMembership

 	// effectiveNetProtos contains the network protocols actually in use. In
 	// most cases it will only contain "netProto", but in cases like IPv6
@@ -68,6 +74,12 @@ type endpoint struct {
 	effectiveNetProtos []tcpip.NetworkProtocolNumber
 }

+// 多播的成员关系，包括多播地址和网卡ID
+type multicastMembership struct {
+	nicID         tcpip.NICID
+	multicastAddr tcpip.Address
+}
+
 func newEndpoint(stack *stack.Stack, netProto tcpip.NetworkProtocolNumber,
 	waiterQueue *waiter.Queue) *endpoint {
 	log.Println("新建一个udp端")
@@ -76,8 +88,32 @@ func newEndpoint(stack *stack.Stack, netProto tcpip.NetworkProtocolNumber,
 		netProto:      netProto,
 		waiterQueue:   waiterQueue,
 		multicastTTL:  1,
-		rcvBufSizeMax: 32 * 1024,
-		sndBufSize:    32 * 1024}
+		rcvBufSizeMax: 32 * 1024, // 接收缓存 32k
+		sndBufSize:    32 * 1024, // 发送缓存 32k
+	}
+}
+
+// NewConnectedEndpoint creates a new endpoint in the connected state using the
+// provided route.
+func NewConnectedEndpoint(stack *stack.Stack, r *stack.Route, id stack.TransportEndpointID,
+	waiterQueue *waiter.Queue) (tcpip.Endpoint, *tcpip.Error) {
+	ep := newEndpoint(stack, r.NetProto, waiterQueue)
+
+	// Register new endpoint so that packets are routed to it.
+	if err := stack.RegisterTransportEndpoint(r.NICID(),
+		[]tcpip.NetworkProtocolNumber{r.NetProto}, ProtocolNumber, id, ep); err != nil {
+		ep.Close()
+		return nil, err
+	}
+
+	ep.id = id
+	ep.route = r.Clone()
+	ep.dstPort = id.RemotePort
+	ep.regNICID = r.NICID()
+
+	ep.state = stateConnected
+
+	return ep, nil
 }

 // Close UDP端的关闭，释放相应的资源
@@ -98,8 +134,37 @@ func (e *endpoint) Close() {
 	e.mu.Unlock()
 }

-func (e *endpoint) Read(*tcpip.FullAddress) (buffer.View, tcpip.ControlMessages, *tcpip.Error) {
-	return nil, tcpip.ControlMessages{}, nil
+func (e *endpoint) Read(addr *tcpip.FullAddress) (buffer.View, tcpip.ControlMessages, *tcpip.Error) {
+	e.rcvMu.Lock()
+
+	// 如果接收链表为空，即没有任何数据
+	if e.rcvList.Empty() {
+		err := tcpip.ErrWouldBlock
+		if e.rcvClosed {
+			err = tcpip.ErrClosedForReceive
+		}
+		e.rcvMu.Unlock()
+		return buffer.View{}, tcpip.ControlMessages{}, err
+	}
+	// 从接收链表中取出最前面的数据报，接着从链表中删除该数据报
+	// 然后减少接收缓存的大小
+	p := e.rcvList.Front()
+	e.rcvList.Remove(p)
+	e.rcvBufSize -= p.data.Size()
+	ts := e.rcvTimestamp
+
+	e.rcvMu.Unlock()
+
+	if ts && !p.hasTimestamp {
+		// Linux uses the current time.
+		p.timestamp = e.stack.NowNanoseconds()
+	}
+	if addr != nil {
+		// 赋值发送地址
+		*addr = p.senderAddress
+	}
+
+	return p.data.ToView(), tcpip.ControlMessages{HasTimestamp: ts, Timestamp: p.timestamp}, nil
 }

 func (e *endpoint) Write(tcpip.Payload, tcpip.WriteOptions) (uintptr, <-chan struct{}, *tcpip.Error) {
@@ -141,8 +206,95 @@ func (e *endpoint) checkV4Mapped(addr *tcpip.FullAddress, allowMismatch bool) (t
 	return netProto, nil
 }

-func (e *endpoint) Connect(address tcpip.FullAddress) *tcpip.Error {
-	log.Println("连接")
+// Connect UDP中调用connect内核仅仅把对端ip&port记录下来. 这样在发送数据的时候无需再次指定
+// UDP多次调用connect有两种用途:1,指定一个新的ip&port连结. 2,断开和之前的ip&port的连结
+func (e *endpoint) Connect(addr tcpip.FullAddress) *tcpip.Error {
+	// 目标端口为0是错误的
+	if addr.Port == 0 {
+		// We don't support connecting to port zero.
+		return tcpip.ErrInvalidEndpointState
+	}
+	e.mu.Lock()
+	defer e.mu.Unlock()
+
+	nicid := addr.NIC
+	var localPort uint16
+	// 判断UDP端的状态
+	switch e.state {
+	case stateInitial:
+		// 如果是初始状态，直接下一步
+	case stateBound, stateConnected:
+		localPort = e.id.LocalPort
+		log.Printf("绑定了 %d 的udp端 向[%d]网卡发起连接\n", localPort, nicid)
+		if e.bindNICID == 0 {
+			break
+		}
+		if nicid != 0 && nicid != e.bindNICID {
+			return tcpip.ErrInvalidEndpointState
+		}
+		nicid = e.bindNICID
+	default:
+		return tcpip.ErrInvalidEndpointState
+	}
+
+	// 检查地址的映射，得到相应的协议
+	netProto, err := e.checkV4Mapped(&addr, false)
+	if err != nil {
+		return err
+	}
+	// Find a route to the desired destination.
+	// 在全局协议栈中查找路由
+	r, err := e.stack.FindRoute(nicid, e.id.LocalAddress, addr.Addr, netProto)
+	if err != nil {
+		return err
+	}
+	defer r.Release()
+
+	// 新建一个传输端的标识，包括源IP、源端口、目的IP、目的端口
+	id := stack.TransportEndpointID{
+		LocalAddress:  r.LocalAddress,
+		LocalPort:     localPort,
+		RemotePort:    addr.Port,
+		RemoteAddress: r.RemoteAddress,
+	}
+
+	// 设置网络层协议，IPV4或IPV6，或两者都有
+	netProtos := []tcpip.NetworkProtocolNumber{netProto}
+	if netProto == header.IPv6ProtocolNumber && !e.v6only {
+		netProtos = []tcpip.NetworkProtocolNumber{
+			header.IPv4ProtocolNumber,
+			header.IPv6ProtocolNumber,
+		}
+	}
+
+	// 将该UDP端注册到协议栈中
+	id, err = e.registerWithStack(nicid, netProtos, id)
+	if err != nil {
+		return err
+	}
+	// Remove the old registration.
+	// 如果源端口不为0，则尝试在传输层端中删除老的UDP端
+	if e.id.LocalPort != 0 {
+		e.stack.UnregisterTransportEndpoint(e.regNICID, e.effectiveNetProtos, ProtocolNumber, e.id)
+	}
+
+	log.Println(e.id, id)
+
+	// 赋值UDP端的属性
+	e.id = id
+	e.route = r.Clone()
+	e.dstPort = addr.Port
+	e.regNICID = nicid
+	e.effectiveNetProtos = netProtos
+
+	// 更改该UDP端的状态为已连接
+	e.state = stateConnected
+
+	// 标志该UDP端可以接收数据了
+	e.rcvMu.Lock()
+	e.rcvReady = true
+	e.rcvMu.Unlock()
+
 	return nil
 }

@@ -167,7 +319,7 @@ func (e *endpoint) registerWithStack(nicid tcpip.NICID, netProtos []tcpip.Networ
 		}
 		id.LocalPort = port
 	}
-	err := e.stack.RegisterTransportEndpoint(nicid, netProtos, ProtocolNumber, id, e)
+	err := e.stack.RegisterTransportEndpoint(nicid, netProtos, ProtocolNumber, id, e) // 往网卡注册一个绑定了端口的udp端
 	if err != nil {
 		// 释放端口
 		e.stack.ReleasePort(netProtos, ProtocolNumber, id.LocalAddress, id.LocalPort)
@@ -206,6 +358,7 @@ func (e *endpoint) bindLocked(addr tcpip.FullAddress, commit func() *tcpip.Error
 		LocalAddress: addr.Addr,
 		LocalPort:    addr.Port,
 	}
+	log.Println("Bind", id)
 	// 在协议栈中注册该UDP端
 	id, err = e.registerWithStack(addr.NIC, netProtos, id)
 	if err != nil {
@@ -229,6 +382,10 @@ func (e *endpoint) bindLocked(addr tcpip.FullAddress, commit func() *tcpip.Error
 	// 标记状态为已绑定
 	e.state = stateBound

+	e.rcvMu.Lock()
+	e.rcvReady = true
+	e.rcvMu.Unlock()
+
 	return nil
 }

@@ -271,9 +428,64 @@ func (e *endpoint) GetSockOpt(opt interface{}) *tcpip.Error {
 	return nil
 }

-// 从网络层接收到UDP数据报时的处理函数
+// HandlePacket 从网络层接收到UDP数据报时的处理函数
+// 首先 UDP 端有接收队列的概念，不像网络层接收到数据包立马发送给传输层，
+// 对于协议栈来说，传输层是最后的一站，接下来的数据就需要交给用户层了，
+// 但是用户层的行为是不可预知的，不知道用户层何时将数据取走（也就是 UDP Read 过程），
+// 那么协议栈就实现一个接收队列，将接收的数据去掉 UDP 头部后保存在这个队列中，用户层需要的时候取走就可以了，
+// 但是队列存数据量是有限制的，这个限制叫接收缓存大小，当接收队列中的数据总和超过这个缓存，那么接下来的这些报文将会被直接丢包。
 func (e *endpoint) HandlePacket(r *stack.Route, id stack.TransportEndpointID, vv buffer.VectorisedView) {
+	// Get the header then trim it from the view.
+	hdr := header.UDP(vv.First())
+	if int(hdr.Length()) > vv.Size() {
+		// Malformed packet.
+		// 错误报文
+		e.stack.Stats().UDP.MalformedPacketsReceived.Increment()
+		return
+	}

+	log.Println("udp 正式接收数据", hdr)
+	// 去除UDP首部
+	vv.TrimFront(header.UDPMinimumSize)
+
+	e.rcvMu.Lock()
+	e.stack.Stats().UDP.PacketsReceived.Increment()
+
+	// Drop the packet if our buffer is currently full.
+	// 如果UDP的接收缓存已经满了，那么丢弃报文。
+	if !e.rcvReady || e.rcvClosed || e.rcvBufSize >= e.rcvBufSizeMax {
+		e.stack.Stats().UDP.ReceiveBufferErrors.Increment()
+		e.rcvMu.Unlock()
+		log.Println("udp 接收缓存不足 丢弃报文")
+		return
+	}
+
+	// 接收缓存是否为空
+	wasEmpty := e.rcvBufSize == 0
+	// 新建一个UDP数据包结构 插入到接收链表中
+	pkt := &udpPacket{
+		senderAddress: tcpip.FullAddress{
+			NIC:  r.NICID(),
+			Addr: id.RemoteAddress,
+			Port: hdr.SourcePort(),
+		},
+	}
+	// 复制UDP数据包的用户数据
+	pkt.data = vv.Clone(pkt.views[:]) // 当vv中的数据<=8时 无需再次分配内存
+	// 插入到接收链表中 并增加已经使用的缓存
+	e.rcvList.PushBack(pkt)
+	e.rcvBufSize += vv.Size()
+
+	if e.rcvTimestamp {
+		pkt.timestamp = e.stack.NowNanoseconds()
+		pkt.hasTimestamp = true
+	}
+
+	e.rcvMu.Unlock()
+	// TODO 通知用户层可以读取数据了
+	if wasEmpty {
+
+	}
 }

 // HandleControlPacket implements stack.TransportEndpoint.HandleControlPacket.
--- a/tcpip/transport/udp/protocol.go
+++ b/tcpip/transport/udp/protocol.go
@@ -38,9 +38,8 @@ func (*protocol) MinimumPacketSize() int {
 // ParsePorts returns the source and destination ports stored in the given udp
 // packet.
 func (*protocol) ParsePorts(v buffer.View) (src, dst uint16, err *tcpip.Error) {
-	//h := header.UDP(v)
-	//return h.SourcePort(), h.DestinationPort(), nil
-	return 0, 0, nil
+	h := header.UDP(v)
+	return h.SourcePort(), h.DestinationPort(), nil
 }

 // HandleUnknownDestinationPacket handles packets targeted at this protocol but