系统调用socket 的kernel实现

这个函数初始化socket的主要数据结构：

socket --> inode

tcp_sock --> inet_connection_sock --> inet_sock --> sock --> sock_common

 

linux版本：4.9.51

在APP中， listenfd = socket(AF_INET, SOCK_STREAM, 0);

int socket(int domain, int type, int protocol);

因此 type=SOCK_STREAM, protocol=0,是使用默认值。

 SYSCALL_DEFINE3(socket, int, family, int, type, int, protocol) -- sock_create -- __sock_create

实现函数在 __sock_create 中

 

1  sock_alloc

分配结构体，关联inode，并设置fops

struct socket *sock_alloc(void)
{
	struct inode *inode;
	struct socket *sock;

	inode = new_inode_pseudo(sock_mnt->mnt_sb);
	if (!inode)
		return NULL;

	sock = SOCKET_I(inode);

	inode->i_op = &sockfs_inode_ops; // 设置ops
	return sock;
}

static const struct inode_operations sockfs_inode_ops = {
	.listxattr = sockfs_listxattr,
};


// new_inode_pseudo --> 会调用到这里: 
static struct inode *sock_alloc_inode(struct super_block *sb)
{
	struct socket_alloc *ei;
	struct socket_wq *wq;

	ei = kmem_cache_alloc(sock_inode_cachep, GFP_KERNEL);
	if (!ei)
		return NULL;
	wq = kmalloc(sizeof(*wq), GFP_KERNEL);
	if (!wq) {
		kmem_cache_free(sock_inode_cachep, ei);
		return NULL;
	}
	init_waitqueue_head(&wq->wait);
	wq->fasync_list = NULL;
	wq->flags = 0;
	RCU_INIT_POINTER(ei->socket.wq, wq);

	ei->socket.state = SS_UNCONNECTED;
	ei->socket.flags = 0;
	ei->socket.ops = NULL;
	ei->socket.sk = NULL;
	ei->socket.file = NULL;

	return &ei->vfs_inode;
}

设置socket的类型

// socket是面向VFS的结构体
/**
 *  struct socket - general BSD socket
 *  @state: socket state (%SS_CONNECTED, etc)
 *  @type: socket type (%SOCK_STREAM, etc)
 *  @flags: socket flags (%SOCK_NOSPACE, etc)
 *  @ops: protocol specific socket operations
 *  @file: File back pointer for gc
 *  @sk: internal networking protocol agnostic socket representation
 *  @wq: wait queue for several uses
 */
struct socket {
	socket_state		state;

	kmemcheck_bitfield_begin(type);
	short			type;
	kmemcheck_bitfield_end(type);

	unsigned long		flags;

	struct socket_wq __rcu	*wq;

	struct file		*file;
	struct sock		*sk;
	const struct proto_ops	*ops;
};

sock->type = type;

2 由入参family 找到inet_family_ops ，进而调用create

struct net_proto_family {
	int		family;
	int		(*create)(struct net *net, struct socket *sock,
				  int protocol, int kern);
	struct module	*owner;
};

static const struct net_proto_family __rcu *net_families[NPROTO] __read_mostly;

int __sock_create(struct net *net, int family, int type, int protocol,
			 struct socket **res, int kern)
{
	const struct net_proto_family *pf;

	pf = rcu_dereference(net_families[family]); // pf = inet_family_ops 

	err = pf->create(net, sock, protocol, kern);// inet_create
}

/*
net_families 的初始化在 sock_register 完成
*/ 
static const struct net_proto_family inet_family_ops = {
	.family = PF_INET,
	.create = inet_create,
	.owner	= THIS_MODULE,
};

static int __init inet_init(void)
{
    (void)sock_register(&inet_family_ops);
}

3 第二步通过net_families 完成分流后，进入主体 inet_create

static struct inet_protosw inetsw_array[] =
{
	{
		.type =       SOCK_STREAM,
		.protocol =   IPPROTO_TCP,
		.prot =       &tcp_prot,
		.ops =        &inet_stream_ops,
		.flags =      INET_PROTOSW_PERMANENT | INET_PROTOSW_ICSK,
	}, // 这个结构体就是下面的for循环要找的

	... ... 为了看着清晰，删掉了其他没用到的。
};

enum sock_type {
 	SOCK_STREAM	= 1,
    ... ...
	SOCK_PACKET	= 10,
};

#define SOCK_MAX (SOCK_PACKET + 1)
static struct list_head inetsw[SOCK_MAX];


struct inet_sock {
	/* sk and pinet6 has to be the first two members of inet_sock */
	struct sock		sk;

	/* Socket demultiplex comparisons on incoming packets. */
#define inet_daddr		sk.__sk_common.skc_daddr
#define inet_rcv_saddr		sk.__sk_common.skc_rcv_saddr
#define inet_dport		sk.__sk_common.skc_dport
#define inet_num		sk.__sk_common.skc_num

	__be32			inet_saddr;
	__s16			uc_ttl;
	__u16			cmsg_flags;
	__be16			inet_sport;
	__u16			inet_id;

	struct ip_options_rcu __rcu	*inet_opt;
	int			rx_dst_ifindex;
	__u8			tos;
	__u8			min_ttl;
	__u8			mc_ttl;
	__u8			pmtudisc;
	__u8			recverr:1,
				is_icsk:1,
				freebind:1,
				hdrincl:1,
				mc_loop:1,
				transparent:1,
				mc_all:1,
				nodefrag:1;
	__u8			bind_address_no_port:1;
	__u8			rcv_tos;
	__u8			convert_csum;
	int			uc_index;
	int			mc_index;
	__be32			mc_addr;
	struct ip_mc_socklist __rcu	*mc_list;
	struct inet_cork_full	cork;
};



// 这个函数创建了sock, inet_sock, 并对inet_sock做了基本的初始化
/*
 *	Create an inet socket.
 */
static int inet_create(struct net *net, struct socket *sock, int protocol,
		       int kern)
{
	struct sock *sk;
	struct inet_protosw *answer;
	struct inet_sock *inet;
	struct proto *answer_prot;
	unsigned char answer_flags;
	int try_loading_module = 0;
	int err;
	
	sock->state = SS_UNCONNECTED; // 先初始化socket的状态

	/* 找出合适的 inet_protosw 结构体 */
	/* Look for the requested type/protocol pair. */
lookup_protocol:

	list_for_each_entry_rcu(answer, &inetsw[sock->type], list) {
		// 此处的sock->type 是在上一步设定的，其实是入参type. 是SOCK_STREAM
		err = 0;
		/* Check the non-wild match. */
		if (protocol == answer->protocol) {
			if (protocol != IPPROTO_IP)
				break;
		} else {
			/* Check for the two wild cases (野生情况, 什么意思？). */
			if (IPPROTO_IP == protocol) {
				protocol = answer->protocol;
				break;
			}
			if (IPPROTO_IP == answer->protocol)
				break;
		}
		err = -EPROTONOSUPPORT;
	}

	// 用找到的 inet_protosw 赋值
	sock->ops = answer->ops;    //inet_stream_ops
	answer_prot = answer->prot;
	answer_flags = answer->flags;

	// 开始分配 sock. 这是一个核心的数据结构，并且要与socket区分开
	// tcp_sock --> inet_connection_sock --> inet_sock --> sock --> sock_common
	// 参考 https://blog.csdn.net/yinming4u/article/details/84312871
	sk = sk_alloc(net, PF_INET, GFP_KERNEL, answer_prot, kern); //answer_prot是 tcp_prot 

	inet = inet_sk(sk);
	inet->is_icsk = (INET_PROTOSW_ICSK & answer_flags) != 0;

	inet->nodefrag = 0;

	if (net->ipv4.sysctl_ip_no_pmtu_disc)
		inet->pmtudisc = IP_PMTUDISC_DONT;
	else
		inet->pmtudisc = IP_PMTUDISC_WANT;

	inet->inet_id = 0;

	sock_init_data(sock, sk);

	sk->sk_destruct	   = inet_sock_destruct;
	sk->sk_protocol	   = protocol;
	sk->sk_backlog_rcv = sk->sk_prot->backlog_rcv; // tcp_v4_do_rcv

	inet->uc_ttl	= -1;
	inet->mc_loop	= 1;
	inet->mc_ttl	= 1;
	inet->mc_all	= 1;
	inet->mc_index	= 0;
	inet->mc_list	= NULL;
	inet->rcv_tos	= 0;

	sk_refcnt_debug_inc(sk);

	// inet->inet_num = 0
	if (inet->inet_num) {
		/* It assumes that any protocol which allows
		 * the user to assign a number at socket
		 * creation time automatically
		 * shares.
		 */
		inet->inet_sport = htons(inet->inet_num);
		/* Add to protocol hash chains. */
		err = sk->sk_prot->hash(sk);
		if (err) {
			sk_common_release(sk);
			goto out;
		}
	}

	// tcp_v4_init_sock
	if (sk->sk_prot->init) {
		err = sk->sk_prot->init(sk);
		if (err)
			sk_common_release(sk);
	}
out:
	return err;
out_rcu_unlock:
	rcu_read_unlock();
	goto out;
}

初始化 prot->slab

static int __init inet_init(void)
{
	rc = proto_register(&tcp_prot, 1);
}


int proto_register(struct proto *prot, int alloc_slab)
{
	if (alloc_slab) {
		prot->slab = kmem_cache_create(prot->name, prot->obj_size, 0,
					SLAB_HWCACHE_ALIGN | prot->slab_flags,
					NULL);		
	}

	list_add(&prot->node, &proto_list);
	return 0;
}


struct proto tcp_prot = {
	.name			= "TCP",
	.owner			= THIS_MODULE,
	.close			= tcp_close,
	.connect		= tcp_v4_connect,
	.disconnect		= tcp_disconnect,
	.accept			= inet_csk_accept,
	.ioctl			= tcp_ioctl,
	.init			= tcp_v4_init_sock,
	.destroy		= tcp_v4_destroy_sock,
	.shutdown		= tcp_shutdown,
	.setsockopt		= tcp_setsockopt,
	.getsockopt		= tcp_getsockopt,
	.recvmsg		= tcp_recvmsg,
	.sendmsg		= tcp_sendmsg,
	.sendpage		= tcp_sendpage,
	.backlog_rcv		= tcp_v4_do_rcv,
	.release_cb		= tcp_release_cb,
	.hash			= inet_hash,
	.unhash			= inet_unhash,
	.get_port		= inet_csk_get_port,
	.enter_memory_pressure	= tcp_enter_memory_pressure,
	.stream_memory_free	= tcp_stream_memory_free,
	.sockets_allocated	= &tcp_sockets_allocated,
	.orphan_count		= &tcp_orphan_count,
	.memory_allocated	= &tcp_memory_allocated,
	.memory_pressure	= &tcp_memory_pressure,
	.sysctl_mem		= sysctl_tcp_mem,
	.sysctl_wmem		= sysctl_tcp_wmem,
	.sysctl_rmem		= sysctl_tcp_rmem,
	.max_header		= MAX_TCP_HEADER,
	.obj_size		= sizeof(struct tcp_sock),
	.slab_flags		= SLAB_DESTROY_BY_RCU,
	.twsk_prot		= &tcp_timewait_sock_ops,
	.rsk_prot		= &tcp_request_sock_ops,
	.h.hashinfo		= &tcp_hashinfo,
	.no_autobind		= true,
#ifdef CONFIG_COMPAT
	.compat_setsockopt	= compat_tcp_setsockopt,
	.compat_getsockopt	= compat_tcp_getsockopt,
#endif
	.diag_destroy		= tcp_abort,
};

分配sock

/* Networking protocol blocks we attach to sockets.
 * socket layer -> transport layer interface
 */
struct proto {
	void			(*close)(struct sock *sk,
					long timeout);
	int			(*connect)(struct sock *sk,
					struct sockaddr *uaddr,
					int addr_len);
	int			(*disconnect)(struct sock *sk, int flags);

	struct sock *		(*accept)(struct sock *sk, int flags, int *err);

......
}

/**
 *	sk_alloc - All socket objects are allocated here
 *	@net: the applicable net namespace
 *	@family: protocol family
 *	@priority: for allocation (%GFP_KERNEL, %GFP_ATOMIC, etc)
 *	@prot: struct proto associated with this new sock instance
 *	@kern: is this to be a kernel socket?
 */
struct sock *sk_alloc(struct net *net, int family, gfp_t priority,
		      struct proto *prot, int kern)
{
	struct sock *sk;

	sk = sk_prot_alloc(prot, priority | __GFP_ZERO, family);
	if (sk) {
		sk->sk_family = family;

		// 设置prot
		sk->sk_prot = sk->sk_prot_creator = prot;

		sock_net_set(sk, net);
	}

	return sk;
}


static struct sock *sk_prot_alloc(struct proto *prot, gfp_t priority,
		int family)
{
	struct sock *sk;
	struct kmem_cache *slab;

	slab = prot->slab;
	if (slab != NULL) {
		sk = kmem_cache_alloc(slab, priority & ~__GFP_ZERO);
		if (!sk)
			return sk;
		if (priority & __GFP_ZERO)
			sk_prot_clear_nulls(sk, prot->obj_size);
	} else
		sk = kmalloc(prot->obj_size, priority);

	return sk;

}

 

sock 是 inet_sock的父类，所以可以这样转换

struct inet_sock {
	/* sk and pinet6 has to be the first two members of inet_sock */
	struct sock		sk;         // sock是第一个元素，所以inet_sk可以这样转换这两个指针

	... ... // 其他元素
};


static inline struct inet_sock *inet_sk(const struct sock *sk)
{
	return (struct inet_sock *)sk;
}

初始化sock

static int tcp_v4_init_sock(struct sock *sk)
{
	struct inet_connection_sock *icsk = inet_csk(sk);

	tcp_init_sock(sk);

	// 初始化 inet_connection_sock 
	icsk->icsk_af_ops = &ipv4_specific;

	return 0;
}


// 初始化 tcp_sock, sock

/* Address-family independent initialization for a tcp_sock.
 *
 * NOTE: A lot of things set to zero explicitly by call to
 *       sk_alloc() so need not be done here.
 */
void tcp_init_sock(struct sock *sk)
{
	struct inet_connection_sock *icsk = inet_csk(sk);
	struct tcp_sock *tp = tcp_sk(sk);

	tp->out_of_order_queue = RB_ROOT;
	tcp_init_xmit_timers(sk);
	tcp_prequeue_init(tp);
	INIT_LIST_HEAD(&tp->tsq_node);

	icsk->icsk_rto = TCP_TIMEOUT_INIT;
	tp->mdev_us = jiffies_to_usecs(TCP_TIMEOUT_INIT);
	minmax_reset(&tp->rtt_min, tcp_time_stamp, ~0U);

	/* So many TCP implementations out there (incorrectly) count the
	 * initial SYN frame in their delayed-ACK and congestion control
	 * algorithms that we must have the following bandaid to talk
	 * efficiently to them.  -DaveM
	 */
	tp->snd_cwnd = TCP_INIT_CWND;

	/* There's a bubble in the pipe until at least the first ACK. */
	tp->app_limited = ~0U;

	/* See draft-stevens-tcpca-spec-01 for discussion of the
	 * initialization of these values.
	 */
	tp->snd_ssthresh = TCP_INFINITE_SSTHRESH;
	tp->snd_cwnd_clamp = ~0;
	tp->mss_cache = TCP_MSS_DEFAULT;
	u64_stats_init(&tp->syncp);

	tp->reordering = sock_net(sk)->ipv4.sysctl_tcp_reordering;
	tcp_enable_early_retrans(tp);
	tcp_assign_congestion_control(sk);

	tp->tsoffset = 0;

	sk->sk_state = TCP_CLOSE;

	sk->sk_write_space = sk_stream_write_space;
	sock_set_flag(sk, SOCK_USE_WRITE_QUEUE);

	icsk->icsk_sync_mss = tcp_sync_mss;

	sk->sk_sndbuf = sysctl_tcp_wmem[1];
	sk->sk_rcvbuf = sysctl_tcp_rmem[1];

	local_bh_disable();
	sk_sockets_allocated_inc(sk);
	local_bh_enable();
}