diff --git a/Blatt02/Blatt02.md b/Blatt02/Blatt02.md index c509bf6..9b84e18 100644 --- a/Blatt02/Blatt02.md +++ b/Blatt02/Blatt02.md @@ -362,3 +362,143 @@ ICMPv6 邻居通告消息由以下几部分组成: - 确保链路层地址和目标地址一致。 如果你需要进一步分析或有其他问题,请随时补充! +======= +是的,你可以在这个 `tuntap.c` 文件的基础上完成第二问和第三问。以下是如何扩展这个文件来实现处理 ARP 请求和生成 ARP 响应的建议步骤: + +--- + +### **扩展步骤** + +#### **1. 增加 ARP 数据包解析和生成的逻辑** +- 定义 ARP 数据包的结构。 +- 添加解析函数,用于从 `buf` 中提取 ARP 请求包的信息。 +- 添加响应生成函数,用于根据解析结果生成 ARP 响应包。 + +#### **2. 在现有的读取逻辑中嵌入 ARP 处理** +- 当前的代码已经实现了从 `TUN/TAP` 设备读取数据的功能(`read`)。 +- 在读取的数据中,检查是否为 ARP 请求包(根据协议类型和操作码)。 +- 如果是 ARP 请求,则生成响应数据,并通过设备写回。 + +#### **3. 实现发送 ARP 响应的逻辑** +- 使用 `write` 系统调用,将生成的 ARP 响应包通过 `tap` 接口发送回去。 + +--- + +### **代码实现(示例)** + +以下是如何在现有代码基础上扩展实现 ARP 请求解析和响应的代码示例。 + +#### **(1) 定义 ARP 数据包结构和处理函数** + +在文件顶部增加 ARP 相关的定义和函数: +```c +#include +#include + +// 定义 ARP 包结构 +typedef struct arp_packet { + uint16_t hardware_type; + uint16_t protocol_type; + uint8_t hardware_len; + uint8_t protocol_len; + uint16_t opcode; + uint8_t sender_mac[6]; + uint8_t sender_ip[4]; + uint8_t target_mac[6]; + uint8_t target_ip[4]; +} arp_packet_t; + +// 解析 ARP 请求 +int parse_arp_request(const uint8_t *buf, arp_packet_t *arp_req) { + memcpy(arp_req, buf + 14, sizeof(arp_packet_t)); // 跳过以太网头部 + return ntohs(arp_req->opcode) == 1; // 返回是否为 ARP 请求 +} + +// 生成 ARP 响应 +void create_arp_response(const arp_packet_t *arp_req, uint8_t *buf) { + memset(buf, 0, 42); // 清空响应缓冲区 + + // 填充以太网头部 + memcpy(buf, arp_req->sender_mac, 6); // 目标 MAC 地址(原请求的发送方) + uint8_t mac[6] = {0xDE, 0xAD, 0xBE, 0xEF, 0x00, 0x01}; // 响应方 MAC + memcpy(buf + 6, mac, 6); // 源 MAC 地址 + buf[12] = 0x08; buf[13] = 0x06; // ARP 协议类型 (0x0806) + + // 填充 ARP 数据包部分 + arp_packet_t *arp_resp = (arp_packet_t *)(buf + 14); + arp_resp->hardware_type = htons(1); // 硬件类型 (Ethernet) + arp_resp->protocol_type = htons(0x0800); // 协议类型 (IPv4) + arp_resp->hardware_len = 6; + arp_resp->protocol_len = 4; + arp_resp->opcode = htons(2); // ARP 响应 + memcpy(arp_resp->sender_mac, mac, 6); // 响应方 MAC + memcpy(arp_resp->sender_ip, arp_req->target_ip, 4); // 响应方 IP + memcpy(arp_resp->target_mac, arp_req->sender_mac, 6); // 请求方 MAC + memcpy(arp_resp->target_ip, arp_req->sender_ip, 4); // 请求方 IP +} +``` + +--- + +#### **(2) 修改 `main` 函数,添加 ARP 处理逻辑** + +在 `main` 函数的 `while` 循环中扩展读取和处理逻辑: +```c +uint8_t buf[BUF_LEN]; +uint8_t response_buf[BUF_LEN]; // 用于存储响应数据 +arp_packet_t arp_req; + +while (running) { + int nread; + if ((nread = read(tun_fd, buf, BUF_LEN)) < 0) { + perror("ERR: Read from tun_fd"); + break; + } + + printf("Read %d bytes from device %s\n", nread, dev); + + // 检查是否为 ARP 包 + if (nread >= 42 && buf[12] == 0x08 && buf[13] == 0x06) { + printf("ARP packet detected\n"); + + // 解析 ARP 请求 + if (parse_arp_request(buf, &arp_req)) { + printf("Processing ARP request from %d.%d.%d.%d\n", + arp_req.sender_ip[0], arp_req.sender_ip[1], + arp_req.sender_ip[2], arp_req.sender_ip[3]); + + // 生成 ARP 响应 + create_arp_response(&arp_req, response_buf); + + // 发送 ARP 响应 + if (write(tun_fd, response_buf, 42) < 0) { + perror("ERR: Write to tun_fd"); + } else { + printf("Sent ARP response\n"); + } + } + } +} +``` + +--- + +### **功能验证** +1. **启动 `tuntap.c` 程序**: + ```bash + sudo ./tuntap mytap0 + ``` + +2. **使用 `arping` 测试**: + 在另一台主机上运行以下命令: + ```bash + arping -I mytap0 + ``` + 程序应打印检测到 ARP 请求,并生成和发送 ARP 响应。 + +--- + +### **总结** +- 第二问通过定义 `arp_packet_t` 数据结构和解析/生成函数,设计了一个合适的接口。 +- 第三问通过扩展 `tuntap.c` 文件,增加了 ARP 请求处理和响应功能。 +- 使用 `arping` 工具可以测试实现的正确性。