DPDK-实战之skeleton(basicfwd)
来源:互联网 发布:淘宝客服每日知识总结 编辑:程序博客网 时间:2024/06/04 18:00
0x01 缘由
继续学习DPDK示例,主要熟悉每个环节和设计理念,今天要学习的是一个相对简答的服务,仅仅从一个网口抓取数据包转发到另外一个网口,这样做双向转发,相当于桥的功能。其他基础业务都不做。
0x02 直接上源码分析
#include <stdint.h>#include <inttypes.h>#include <rte_eal.h>#include <rte_ethdev.h>#include <rte_cycles.h>#include <rte_lcore.h>#include <rte_mbuf.h>#define RX_RING_SIZE 128#define TX_RING_SIZE 512#define NUM_MBUFS 8191#define MBUF_CACHE_SIZE 250#define BURST_SIZE 32static const struct rte_eth_conf port_conf_default = { .rxmode = { .max_rx_pkt_len = ETHER_MAX_LEN }};/* basicfwd.c: Basic DPDK skeleton forwarding example. *//* * 用全局默认的配置来初始化网口。在内存池中分配接收队列和发送队列。 */static inline intport_init(uint8_t port, struct rte_mempool *mbuf_pool){ struct rte_eth_conf port_conf = port_conf_default; //全部配置模式 const uint16_t rx_rings = 1, tx_rings = 1; //每个网口队列多少,这里都为1 int retval; uint16_t q; if (port >= rte_eth_dev_count()) //检查网口编号 return -1; /* 配置以太网设备 */ retval = rte_eth_dev_configure(port, rx_rings, tx_rings, &port_conf); if (retval != 0) return retval; /* 为每个以太网网口分配和设置1个接收队列 */ for (q = 0; q < rx_rings; q++) { retval = rte_eth_rx_queue_setup(port, q, RX_RING_SIZE, rte_eth_dev_socket_id(port), NULL, mbuf_pool); if (retval < 0) return retval; } /* 为每个以太网网口分配和设置1个发送队列 */ for (q = 0; q < tx_rings; q++) { retval = rte_eth_tx_queue_setup(port, q, TX_RING_SIZE, rte_eth_dev_socket_id(port), NULL); if (retval < 0) return retval; } /* 开启以太网网口 */ retval = rte_eth_dev_start(port); if (retval < 0) return retval; /* 打印网卡地址 MAC address. */ struct ether_addr addr; rte_eth_macaddr_get(port, &addr); printf("Port %u MAC: %02" PRIx8 " %02" PRIx8 " %02" PRIx8 " %02" PRIx8 " %02" PRIx8 " %02" PRIx8 "\n", (unsigned)port, addr.addr_bytes[0], addr.addr_bytes[1], addr.addr_bytes[2], addr.addr_bytes[3], addr.addr_bytes[4], addr.addr_bytes[5]); /* 网口开启混杂模式 */ rte_eth_promiscuous_enable(port); return 0;}/* * 一个线程处理一对网口,数据包一进一出。 */static __attribute__((noreturn)) voidlcore_main(void){ const uint8_t nb_ports = rte_eth_dev_count(); //网口数 uint8_t port; /* * 当有NUMA结构时,检查网口是否在同一个NUMA node节点上,只有在一个NUMA node上时轮询线程效率最好。 */ for (port = 0; port < nb_ports; port++) if (rte_eth_dev_socket_id(port) > 0 && rte_eth_dev_socket_id(port) != (int)rte_socket_id()) printf("WARNING, port %u is on remote NUMA node to " "polling thread.\n\tPerformance will " "not be optimal.\n", port); printf("\nCore %u forwarding packets. [Ctrl+C to quit]\n", rte_lcore_id()); /* 死循环处理数据包的搬运 */ for (;;) { /* * 一个端口接收包,然后直接转发到这对数据包上。 * 如 0 -> 1, 1 -> 0, 2 -> 3, 3 -> 2 等。 */ for (port = 0; port < nb_ports; port++) { /* 从一个端口接收数据包 */ struct rte_mbuf *bufs[BURST_SIZE]; const uint16_t nb_rx = rte_eth_rx_burst(port, 0, bufs, BURST_SIZE); if (unlikely(nb_rx == 0)) continue; /* Send burst of TX packets, to second port of pair. */ const uint16_t nb_tx = rte_eth_tx_burst(port ^ 1, 0, bufs, nb_rx); /* 释放没有发送的数据包。 */ if (unlikely(nb_tx < nb_rx)) { uint16_t buf; for (buf = nb_tx; buf < nb_rx; buf++) rte_pktmbuf_free(bufs[buf]); } } }}/* * The main function, which does initialization and calls the per-lcore * functions. */intmain(int argc, char *argv[]){ struct rte_mempool *mbuf_pool; //使用内存池空间来容纳ring队列,接收和发送数据包 unsigned nb_ports; //网卡数 uint8_t portid; //网卡编号 /* Initialize the Environment Abstraction Layer (EAL). */ int ret = rte_eal_init(argc, argv); if (ret < 0) rte_exit(EXIT_FAILURE, "Error with EAL initialization\n"); argc -= ret; argv += ret; /* 检查网卡是否为偶数 send/receive */ nb_ports = rte_eth_dev_count(); if (nb_ports < 2 || (nb_ports & 1)) rte_exit(EXIT_FAILURE, "Error: number of ports must be even\n"); /* 创建一个内存池来容纳mbufs。大小按网口多少来分配。相当于每个网口都要有一个接受和发送队列。*/ mbuf_pool = rte_pktmbuf_pool_create("MBUF_POOL", NUM_MBUFS * nb_ports, MBUF_CACHE_SIZE, 0, RTE_MBUF_DEFAULT_BUF_SIZE, rte_socket_id()); if (mbuf_pool == NULL) rte_exit(EXIT_FAILURE, "Cannot create mbuf pool\n"); /* 初始化所有网口配置 */ for (portid = 0; portid < nb_ports; portid++) if (port_init(portid, mbuf_pool) != 0) rte_exit(EXIT_FAILURE, "Cannot init port %"PRIu8 "\n", portid); if (rte_lcore_count() > 1) printf("\nWARNING: Too many lcores enabled. Only 1 used.\n"); /* 仅仅用一个主线程调用,相当与一个线程处理一对网口的数据包。*/ lcore_main(); return 0;}
0x03 运行实例
[root@Huawei x86_64-native-linuxapp-gcc]# ./basicfwd -l 1 -n 4EAL: Detected lcore 0 as core 0 on socket 0EAL: Detected lcore 1 as core 1 on socket 0EAL: Detected lcore 2 as core 2 on socket 0EAL: Detected lcore 3 as core 3 on socket 0EAL: Detected lcore 4 as core 4 on socket 0EAL: Detected lcore 5 as core 5 on socket 0EAL: Detected lcore 6 as core 6 on socket 0EAL: Detected lcore 7 as core 7 on socket 0EAL: Detected lcore 8 as core 0 on socket 1EAL: Detected lcore 9 as core 1 on socket 1EAL: Detected lcore 10 as core 2 on socket 1EAL: Detected lcore 11 as core 3 on socket 1EAL: Detected lcore 12 as core 4 on socket 1EAL: Detected lcore 13 as core 5 on socket 1EAL: Detected lcore 14 as core 6 on socket 1EAL: Detected lcore 15 as core 7 on socket 1EAL: Detected lcore 16 as core 0 on socket 0EAL: Detected lcore 17 as core 1 on socket 0EAL: Detected lcore 18 as core 2 on socket 0EAL: Detected lcore 19 as core 3 on socket 0EAL: Detected lcore 20 as core 4 on socket 0EAL: Detected lcore 21 as core 5 on socket 0EAL: Detected lcore 22 as core 6 on socket 0EAL: Detected lcore 23 as core 7 on socket 0EAL: Detected lcore 24 as core 0 on socket 1EAL: Detected lcore 25 as core 1 on socket 1EAL: Detected lcore 26 as core 2 on socket 1EAL: Detected lcore 27 as core 3 on socket 1EAL: Detected lcore 28 as core 4 on socket 1EAL: Detected lcore 29 as core 5 on socket 1EAL: Detected lcore 30 as core 6 on socket 1EAL: Detected lcore 31 as core 7 on socket 1EAL: Support maximum 128 logical core(s) by configuration.EAL: Detected 32 lcore(s)EAL: Setting up physically contiguous memory...EAL: Ask a virtual area of 0x200000 bytesEAL: Virtual area found at 0x7f45f8000000 (size = 0x200000)EAL: Ask a virtual area of 0x7c00000 bytesEAL: Virtual area found at 0x7f45f0200000 (size = 0x7c00000)EAL: Ask a virtual area of 0x200000 bytesEAL: Virtual area found at 0x7f45efe00000 (size = 0x200000)EAL: Ask a virtual area of 0x7c00000 bytesEAL: Virtual area found at 0x7f45e8000000 (size = 0x7c00000)EAL: Ask a virtual area of 0x400000 bytesEAL: Virtual area found at 0x7f45e7a00000 (size = 0x400000)EAL: Requesting 64 pages of size 2MB from socket 0EAL: Requesting 64 pages of size 2MB from socket 1EAL: TSC frequency is ~2593994 KHzEAL: Master lcore 1 is ready (tid=f83f2880;cpuset=[1])EAL: PCI device 0000:04:00.0 on NUMA socket 0EAL: probe driver: 8086:1521 rte_igb_pmdEAL: Not managed by a supported kernel driver, skippedEAL: PCI device 0000:04:00.1 on NUMA socket 0EAL: probe driver: 8086:1521 rte_igb_pmdEAL: Not managed by a supported kernel driver, skippedEAL: PCI device 0000:04:00.2 on NUMA socket 0EAL: probe driver: 8086:1521 rte_igb_pmdEAL: PCI memory mapped at 0x7f45f7f00000EAL: PCI memory mapped at 0x7f45f7efc000PMD: eth_igb_dev_init(): port_id 0 vendorID=0x8086 deviceID=0x1521EAL: PCI device 0000:04:00.3 on NUMA socket 0EAL: probe driver: 8086:1521 rte_igb_pmdEAL: PCI memory mapped at 0x7f45f0100000EAL: PCI memory mapped at 0x7f45f8408000PMD: eth_igb_dev_init(): port_id 1 vendorID=0x8086 deviceID=0x1521EAL: PCI device 0000:05:00.0 on NUMA socket 0EAL: probe driver: 8086:1522 rte_igb_pmdEAL: Not managed by a supported kernel driver, skippedEAL: PCI device 0000:05:00.1 on NUMA socket 0EAL: probe driver: 8086:1522 rte_igb_pmdEAL: Not managed by a supported kernel driver, skippedEAL: PCI device 0000:05:00.2 on NUMA socket 0EAL: probe driver: 8086:1522 rte_igb_pmdEAL: Not managed by a supported kernel driver, skippedEAL: PCI device 0000:05:00.3 on NUMA socket 0EAL: probe driver: 8086:1522 rte_igb_pmdEAL: Not managed by a supported kernel driver, skippedPMD: eth_igb_rx_queue_setup(): sw_ring=0x7f45eff90fc0 hw_ring=0x7f45eff91400 dma_addr=0x67391400PMD: eth_igb_tx_queue_setup(): sw_ring=0x7f45eff7ee40 hw_ring=0x7f45eff80e80 dma_addr=0x67380e80PMD: eth_igb_start(): <<Port 0 MAC: a0 36 9f 03 a8 caPMD: eth_igb_rx_queue_setup(): sw_ring=0x7f45eff6e7c0 hw_ring=0x7f45eff6ec00 dma_addr=0x6736ec00PMD: eth_igb_tx_queue_setup(): sw_ring=0x7f45eff5c640 hw_ring=0x7f45eff5e680 dma_addr=0x6735e680PMD: eth_igb_start(): <<Port 1 MAC: a0 36 9f 03 a8 cbCore 1 forwarding packets. [Ctrl+C to quit]
0x04 总结
思路比较简单,一个网口用两个缓存来缓存数据包,然后将数据包转发出去。此处用到了内存池结构。
阅读全文
0 0
- DPDK-实战之skeleton(basicfwd)
- DPDK-实战之helloworld
- DPDK Skeleton代码详读
- DPDK-实战之load_balancer(负载均衡)
- DPDK-实战之ipsec-secgw(安全网关)
- 实战录 | 云端卫士之DPDK技术简介
- DPDK-实战之rxtx_callbacks(回调函数)
- DPDK-实战之l3fwd-vf(虚拟化环境)
- DPDK-实战之Virtio/vhost(虚拟技术)
- LDD3之usb-skeleton
- LDD3之usb-skeleton
- Skeleton
- DPDK简介之二
- DPDK 之ubuntu安装
- usb-skeleton.c 之 08---skel_open
- usb-skeleton.c 之 09---skel_release
- usb-skeleton.c 之 10---skel_flush
- usb-skeleton.c 之 11---skel_read
- 【Java相关】二进制角度计算~0x99^0x66
- 【Java基础知识】 Annotation03_注解(Annotation)--注解处理器
- 常用电平标准——LVTTL、LVCMOS、LVDS等
- Superuser creation skipped due to not running in a TTY创建超级管理员报错
- Vuex2.0初识
- DPDK-实战之skeleton(basicfwd)
- C语言条件编译与预处理
- lua
- Mysql下进行经纬度的距离测算
- 第十一章pose_graph_g2o_SE3
- Android GLSurfaceView 显示YUV数据的坑,图像偏蓝色
- PHP五种设计模式
- JAVA LIST分批处理
- Qt样式表