accept 系统调用流程

分析下accept系统调用流程，基于linux 6.1.4内核

SYSCALL_DEFINE3

< net/socket.c >

SYSCALL_DEFINE4(accept4, int, fd, struct sockaddr __user *, upeer_sockaddr,
		int __user *, upeer_addrlen, int, flags)
{
	return __sys_accept4(fd, upeer_sockaddr, upeer_addrlen, flags);
}

SYSCALL_DEFINE3(accept, int, fd, struct sockaddr __user *, upeer_sockaddr,
		int __user *, upeer_addrlen)
{
	return __sys_accept4(fd, upeer_sockaddr, upeer_addrlen, 0);
}

__sys_accept4()

< net/socket.c >

int __sys_accept4(int fd, struct sockaddr __user *upeer_sockaddr,
		  int __user *upeer_addrlen, int flags)
{
	int ret = -EBADF;
	struct fd f;

	f = fdget(fd);
	if (f.file) {
		ret = __sys_accept4_file(f.file, upeer_sockaddr,
					 upeer_addrlen, flags);
		fdput(f);
	}

	return ret;
}

__sys_accept4_file()

< net/socket.c >

static int __sys_accept4_file(struct file *file, struct sockaddr __user *upeer_sockaddr,
			      int __user *upeer_addrlen, int flags)
{
	struct file *newfile;
	int newfd;

	if (flags & ~(SOCK_CLOEXEC | SOCK_NONBLOCK))
		return -EINVAL;

	if (SOCK_NONBLOCK != O_NONBLOCK && (flags & SOCK_NONBLOCK))
		flags = (flags & ~SOCK_NONBLOCK) | O_NONBLOCK;

	newfd = get_unused_fd_flags(flags);
	if (unlikely(newfd < 0))
		return newfd;

   /* 实际处理函数 */
	newfile = do_accept(file, 0, upeer_sockaddr, upeer_addrlen,
			    flags);
	if (IS_ERR(newfile)) {
		put_unused_fd(newfd);
		return PTR_ERR(newfile);
	}
	fd_install(newfd, newfile);
	return newfd;
}

do_accept()

< net/socket.c >

struct file *do_accept(struct file *file, unsigned file_flags,
		       struct sockaddr __user *upeer_sockaddr,
		       int __user *upeer_addrlen, int flags)
{
	struct socket *sock, *newsock;
	struct file *newfile;
	int err, len;
	struct sockaddr_storage address;

	sock = sock_from_file(file);
	if (!sock)
		return ERR_PTR(-ENOTSOCK);

	newsock = sock_alloc();
	if (!newsock)
		return ERR_PTR(-ENFILE);

	newsock->type = sock->type;
	newsock->ops = sock->ops;

	/*
	 * We don't need try_module_get here, as the listening socket (sock)
	 * has the protocol module (sock->ops->owner) held.
	 */
	__module_get(newsock->ops->owner);

	newfile = sock_alloc_file(newsock, flags, sock->sk->sk_prot_creator->name);
	if (IS_ERR(newfile))
		return newfile;

	err = security_socket_accept(sock, newsock);
	if (err)
		goto out_fd;

	err = sock->ops->accept(sock, newsock, sock->file->f_flags | file_flags,
					false);
	if (err < 0)
		goto out_fd;

	if (upeer_sockaddr) {
		len = newsock->ops->getname(newsock,
					(struct sockaddr *)&address, 2);
		if (len < 0) {
			err = -ECONNABORTED;
			goto out_fd;
		}
		err = move_addr_to_user(&address,
					len, upeer_sockaddr, upeer_addrlen);
		if (err < 0)
			goto out_fd;
	}

	/* File flags are not inherited via accept() unlike another OSes. */
	return newfile;
out_fd:
	fput(newfile);
	return ERR_PTR(err);
}

accept()注册

.accept = inet_accept

< net/dccp/ipv4.c >

static const struct proto_ops inet_dccp_ops = {
	.family		   = PF_INET,
	.owner		   = THIS_MODULE,
	.release	   = inet_release,
	.bind		   = inet_bind,
	.connect	   = inet_stream_connect,
	.socketpair	   = sock_no_socketpair,
	.accept		   = inet_accept,
	.getname	   = inet_getname,
	/* FIXME: work on tcp_poll to rename it to inet_csk_poll */
	.poll		   = dccp_poll,
	.ioctl		   = inet_ioctl,
	.gettstamp	   = sock_gettstamp,
	/* FIXME: work on inet_listen to rename it to sock_common_listen */
	.listen		   = inet_dccp_listen,
	.shutdown	   = inet_shutdown,
	.setsockopt	   = sock_common_setsockopt,
	.getsockopt	   = sock_common_getsockopt,
	.sendmsg	   = inet_sendmsg,
	.recvmsg	   = sock_common_recvmsg,
	.mmap		   = sock_no_mmap,
	.sendpage	   = sock_no_sendpage,
};

inet_accept()

< net/ipv4/af_inet.c >

int inet_accept(struct socket *sock, struct socket *newsock, int flags,
		bool kern)
{
	struct sock *sk1 = sock->sk, *sk2;
	int err = -EINVAL;

	/* IPV6_ADDRFORM can change sk->sk_prot under us. */
	sk2 = READ_ONCE(sk1->sk_prot)->accept(sk1, flags, &err, kern);          /* 从建立连接成功的sock队列中拿出一个sock，赋值给sk2 */
	if (!sk2)
		goto do_err;

	lock_sock(sk2);

	sock_rps_record_flow(sk2);                                                               /* 处理流控表相关 */
	WARN_ON(!((1 << sk2->sk_state) &
		  (TCPF_ESTABLISHED | TCPF_SYN_RECV |
		  TCPF_CLOSE_WAIT | TCPF_CLOSE)));

	if (test_bit(SOCK_SUPPORT_ZC, &sock->flags))
		set_bit(SOCK_SUPPORT_ZC, &newsock->flags);
	sock_graft(sk2, newsock);

	newsock->state = SS_CONNECTED;
	err = 0;
	release_sock(sk2);
do_err:
	return err;
}