linux驱动_loopback源码注释

背景

linux 网络驱动中，loopback驱动算是最为简单的，用处却很大，随便找个linux系统，在shell里面敲"ifconfig" 或者 “ip -s link list” 命令可以看到：

# ifconfig
lo        Link encap:Local Loopback  
          inet addr:127.0.0.1  Mask:255.0.0.0
          UP LOOPBACK RUNNING  MTU:65536  Metric:1
          RX packets:593365 errors:0 dropped:0 overruns:0 frame:0
          TX packets:593365 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:1 
          RX bytes:86365483 (82.3 MiB)  TX bytes:86365483 (82.3 MiB)

# 或
# ip -s link list
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN mode DEFAULT group default qlen 1000
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
    RX: bytes  packets  errors  dropped overrun mcast   
    197217691  2769876  0       0       0       0       
    TX: bytes  packets  errors  dropped carrier collsns 
    197217691  2769876  0       0       0       0  

从上面的结果看RX和TX的统计，收发的网络报文还是挺多的。只要是给"127.0.0.1"这个IP发的报文都会走这个网口，很多本地的ip server或者进程间通信就会使用到。

源码

基本原理是用 register_netdev 函数注册一个struct net_device对象，在回调函数 loopback_setup里面初始化好一些操作的回调函数。其中发送的回调函数是： “.ndo_start_xmit= loopback_xmit”， 在这个函数里面会把要发送的内容发到自己的接收队列去，调用的接口是：“netif_rx”。源码及注释如下：

/*
    统计信息结构体，包括收发报文个数及字节数
    因为传输会发生在软中断(soft irq)中，syncp 用于多CPU核同步
*/
struct pcpu_lstats {
	u64			packets;
	u64			bytes;
	struct u64_stats_sync	syncp;
};

/*
 * The higher levels take care of making this non-reentrant (it's
 * called with bh's disabled).
 * 这个函数是可重入的，但由调用者保证他不可重入。需要在bottom half disabled 的情况下才能调用
 */
static netdev_tx_t loopback_xmit(struct sk_buff *skb,
				 struct net_device *dev)
{
	struct pcpu_lstats *lb_stats;
	int len;

    // 将一个缓冲结构体变成孤立的skb
	skb_orphan(skb);

	/* Before queueing this packet to netif_rx(),
	 * make sure dst is refcounted.
     * 强制对 skb 的目标缓存进行引用计数
	 */
	skb_dst_force(skb);

    // 获取mac数据帧的协议类型，三层(IP)根据这个字段决定使用哪个协议实例处理
	skb->protocol = eth_type_trans(skb, dev);

	/* it's OK to use per_cpu_ptr() because BHs are off */
    // 获取统计计数（当前CPU核的），因为下半部已经关闭(soft irq内)，这里调用是安全的
	lb_stats = this_cpu_ptr(dev->lstats);

	len = skb->len;
	if (likely(netif_rx(skb) == NET_RX_SUCCESS)) { // 放入接收协议栈队列
		u64_stats_update_begin(&lb_stats->syncp);  // 防止在更新的时候其它上下文会读这个变量值
		lb_stats->bytes += len;
		lb_stats->packets++;
		u64_stats_update_end(&lb_stats->syncp);
	}

	return NETDEV_TX_OK;
}

// 获取统计信息
static struct rtnl_link_stats64 *loopback_get_stats64(struct net_device *dev,
						      struct rtnl_link_stats64 *stats)
{
	u64 bytes = 0;
	u64 packets = 0;
	int i;

	for_each_possible_cpu(i) { // 遍历所有核，把计数加起来
		const struct pcpu_lstats *lb_stats;
		u64 tbytes, tpackets;
		unsigned int start;

		lb_stats = per_cpu_ptr(dev->lstats, i);
		do {
			start = u64_stats_fetch_begin_irq(&lb_stats->syncp);
			tbytes = lb_stats->bytes;
			tpackets = lb_stats->packets;
		} while (u64_stats_fetch_retry_irq(&lb_stats->syncp, start));
		bytes   += tbytes;
		packets += tpackets;
	}
	stats->rx_packets = packets;
	stats->tx_packets = packets;
	stats->rx_bytes   = bytes;
	stats->tx_bytes   = bytes;
	return stats;
}

static u32 always_on(struct net_device *dev)
{
	return 1;
}

static const struct ethtool_ops loopback_ethtool_ops = {
	.get_link		= always_on,
};

// 初始化统计结构体
static int loopback_dev_init(struct net_device *dev)
{
	dev->lstats = netdev_alloc_pcpu_stats(struct pcpu_lstats);
	if (!dev->lstats)
		return -ENOMEM;
	return 0;
}

// 注销设备，回收资源
static void loopback_dev_free(struct net_device *dev)
{
	dev_net(dev)->loopback_dev = NULL;
	free_percpu(dev->lstats);
	free_netdev(dev);
}

static const struct net_device_ops loopback_ops = {
	.ndo_init      = loopback_dev_init,
	.ndo_start_xmit= loopback_xmit,
	.ndo_get_stats64 = loopback_get_stats64,
	.ndo_set_mac_address = eth_mac_addr,
};

/*
 * The loopback device is special. There is only one instance
 * per network namespace.
 */
static void loopback_setup(struct net_device *dev)
{
	dev->mtu		= 64 * 1024; // 与最大数据帧有关
	dev->hard_header_len	= ETH_HLEN;	/* 14	*/
	dev->min_header_len	= ETH_HLEN;	/* 14	*/
	dev->addr_len		= ETH_ALEN;	/* 6	*/
	dev->type		= ARPHRD_LOOPBACK;	/* 0x0001*/
	dev->flags		= IFF_LOOPBACK;
	dev->priv_flags		|= IFF_LIVE_ADDR_CHANGE | IFF_NO_QUEUE;
	netif_keep_dst(dev);
	dev->hw_features	= NETIF_F_GSO_SOFTWARE;
	dev->features 		= NETIF_F_SG | NETIF_F_FRAGLIST
		| NETIF_F_GSO_SOFTWARE
		| NETIF_F_HW_CSUM
		| NETIF_F_RXCSUM
		| NETIF_F_SCTP_CRC
		| NETIF_F_HIGHDMA
		| NETIF_F_LLTX
		| NETIF_F_NETNS_LOCAL
		| NETIF_F_VLAN_CHALLENGED
		| NETIF_F_LOOPBACK;
	dev->ethtool_ops	= &loopback_ethtool_ops;  // ethtool 是用于查询和设置网卡参数的命令
	dev->header_ops		= &eth_header_ops;
	dev->netdev_ops		= &loopback_ops;          // 网络设备的操作
	dev->destructor		= loopback_dev_free;      // 注销设备，回收资源
}

/* Setup and register the loopback device. */
static __net_init int loopback_net_init(struct net *net)
{
	struct net_device *dev;
	int err;

	err = -ENOMEM;
	dev = alloc_netdev(0, "lo", NET_NAME_UNKNOWN, loopback_setup); // 申请空间，成功就回调 loopback_setup
	if (!dev)
		goto out;

	dev_net_set(dev, net);
	err = register_netdev(dev); // 注册 net_device
	if (err)
		goto out_free_netdev;

	BUG_ON(dev->ifindex != LOOPBACK_IFINDEX);
	net->loopback_dev = dev;
	return 0;


out_free_netdev:
	free_netdev(dev);
out:
	if (net_eq(net, &init_net))
		panic("loopback: Failed to register netdevice: %d\n", err);
	return err;
}

/* Registered in net/core/dev.c */
struct pernet_operations __net_initdata loopback_net_ops = {
       .init = loopback_net_init,
};

end