tcp.c

讲述linux的初始化过程

		while (seglen > 0) {
			int copy, tmp, queue_it;

			if (err)
				goto do_fault2;

			/* Stop on errors. */
			if (sk->err)
				goto do_sock_err;

			/* Make sure that we are established. */
			if (sk->shutdown & SEND_SHUTDOWN)
				goto do_shutdown;
	
			/* Now we need to check if we have a half
			 * built packet we can tack some data onto.
			 */
			skb = sk->write_queue.prev;
			if (tp->send_head &&
			    (mss_now - skb->len) > 0) {
				copy = skb->len;
				if (skb_tailroom(skb) > 0) {
					int last_byte_was_odd = (copy % 4);

					copy = mss_now - copy;
					if(copy > skb_tailroom(skb))
						copy = skb_tailroom(skb);
					if(copy > seglen)
						copy = seglen;
					if(last_byte_was_odd) {
						if(copy_from_user(skb_put(skb, copy),
								  from, copy))
							err = -EFAULT;
						skb->csum = csum_partial(skb->data,
									 skb->len, 0);
					} else {
						skb->csum =
							csum_and_copy_from_user(
							from, skb_put(skb, copy),
							copy, skb->csum, &err);
					}
					/*
					 * FIXME: the *_user functions should
					 *	  return how much data was
					 *	  copied before the fault
					 *	  occurred and then a partial
					 *	  packet with this data should
					 *	  be sent.  Unfortunately
					 *	  csum_and_copy_from_user doesn't
					 *	  return this information.
					 *	  ATM it might send partly zeroed
					 *	  data in this case.
					 */
					tp->write_seq += copy;
					TCP_SKB_CB(skb)->end_seq += copy;
					from += copy;
					copied += copy;
					seglen -= copy;
					if (PSH_NEEDED ||
					    after(tp->write_seq, tp->pushed_seq+(tp->max_window>>1))) {
						TCP_SKB_CB(skb)->flags |= TCPCB_FLAG_PSH;
						tp->pushed_seq = tp->write_seq;
					}
					if (flags&MSG_OOB) {
						tp->urg_mode = 1;
						tp->snd_up = tp->write_seq;
						TCP_SKB_CB(skb)->sacked |= TCPCB_URG;
					}
					continue;
				} else {
					TCP_SKB_CB(skb)->flags |= TCPCB_FLAG_PSH;
					tp->pushed_seq = tp->write_seq;
				}
			}

			copy = min(seglen, mss_now);

			/* Determine how large of a buffer to allocate.  */
			tmp = MAX_TCP_HEADER + 15 + tp->mss_cache;
			if (copy < mss_now && !(flags & MSG_OOB)) {
				/* What is happening here is that we want to
				 * tack on later members of the users iovec
				 * if possible into a single frame.  When we
				 * leave this loop our we check to see if
				 * we can send queued frames onto the wire.
				 */
				queue_it = 1;
			} else {
				queue_it = 0;
			}

			skb = NULL;
			if (tcp_memory_free(sk))
				skb = tcp_alloc_skb(sk, tmp, sk->allocation);
			if (skb == NULL) {
				/* If we didn't get any memory, we need to sleep. */
				set_bit(SOCK_ASYNC_NOSPACE, &sk->socket->flags);
				set_bit(SOCK_NOSPACE, &sk->socket->flags);

				__tcp_push_pending_frames(sk, tp, mss_now, 1);

				if (!timeo) {
					err = -EAGAIN;
					goto do_interrupted;
				}
				if (signal_pending(current)) {
					err = sock_intr_errno(timeo);
					goto do_interrupted;
				}
				timeo = wait_for_tcp_memory(sk, timeo);

				/* If SACK's were formed or PMTU events happened,
				 * we must find out about it.
				 */
				mss_now = tcp_current_mss(sk);
				continue;
			}

			seglen -= copy;

			/* Prepare control bits for TCP header creation engine. */
			if (PSH_NEEDED ||
			    after(tp->write_seq+copy, tp->pushed_seq+(tp->max_window>>1))) {
				TCP_SKB_CB(skb)->flags = TCPCB_FLAG_ACK|TCPCB_FLAG_PSH;
				tp->pushed_seq = tp->write_seq + copy;
			} else {
				TCP_SKB_CB(skb)->flags = TCPCB_FLAG_ACK;
			}
			TCP_SKB_CB(skb)->sacked = 0;
			if (flags & MSG_OOB) {
				TCP_SKB_CB(skb)->sacked |= TCPCB_URG;
				tp->urg_mode = 1;
				tp->snd_up = tp->write_seq + copy;
			}

			/* TCP data bytes are SKB_PUT() on top, later
			 * TCP+IP+DEV headers are SKB_PUSH()'d beneath.
			 * Reserve header space and checksum the data.
			 */
			skb_reserve(skb, MAX_TCP_HEADER);
			skb->csum = csum_and_copy_from_user(from,
					skb_put(skb, copy), copy, 0, &err);

			if (err)
				goto do_fault;

			from += copy;
			copied += copy;

			TCP_SKB_CB(skb)->seq = tp->write_seq;
			TCP_SKB_CB(skb)->end_seq = TCP_SKB_CB(skb)->seq + copy;

			/* This advances tp->write_seq for us. */
			tcp_send_skb(sk, skb, queue_it, mss_now);
		}
	}
	err = copied;
out:
	__tcp_push_pending_frames(sk, tp, mss_now, tp->nonagle);
out_unlock:
	TCP_CHECK_TIMER(sk);
	release_sock(sk);
	return err;

do_sock_err:
	if (copied)
		err = copied;
	else
		err = sock_error(sk);
	goto out;
do_shutdown:
	if (copied)
		err = copied;
	else {
		if (!(flags&MSG_NOSIGNAL))
			send_sig(SIGPIPE, current, 0);
		err = -EPIPE;
	}
	goto out;
do_interrupted:
	if (copied)
		err = copied;
	goto out_unlock;
do_fault:
	__kfree_skb(skb);
do_fault2:
	if (copied)
		err = copied;
	else
		err = -EFAULT;
	goto out;
}

#undef PSH_NEEDED

/*
 *	Handle reading urgent data. BSD has very simple semantics for
 *	this, no blocking and very strange errors 8)
 */

static int tcp_recv_urg(struct sock * sk, long timeo,
			struct msghdr *msg, int len, int flags, 
			int *addr_len)
{
	struct tcp_opt *tp = &(sk->tp_pinfo.af_tcp);

	/* No URG data to read. */
	if (sk->urginline || !tp->urg_data || tp->urg_data == TCP_URG_READ)
		return -EINVAL;	/* Yes this is right ! */

	if (sk->state==TCP_CLOSE && !sk->done)
		return -ENOTCONN;

	if (tp->urg_data & TCP_URG_VALID) {
		int err = 0; 
		char c = tp->urg_data;

		if (!(flags & MSG_PEEK))
			tp->urg_data = TCP_URG_READ;

		/* Read urgent data. */
		msg->msg_flags|=MSG_OOB;

		if(len>0) {
			if (!(flags & MSG_PEEK))
				err = memcpy_toiovec(msg->msg_iov, &c, 1);
			len = 1;
		} else
			msg->msg_flags|=MSG_TRUNC;

		return err ? -EFAULT : len;
	}

	if (sk->state == TCP_CLOSE || (sk->shutdown & RCV_SHUTDOWN))
		return 0;

	/* Fixed the recv(..., MSG_OOB) behaviour.  BSD docs and
	 * the available implementations agree in this case:
	 * this call should never block, independent of the
	 * blocking state of the socket.
	 * Mike <[e-mail]>
	 */
	return -EAGAIN;
}

/*
 *	Release a skb if it is no longer needed. This routine
 *	must be called with interrupts disabled or with the
 *	socket locked so that the sk_buff queue operation is ok.
 */

static inline void tcp_eat_skb(struct sock *sk, struct sk_buff * skb)
{
	__skb_unlink(skb, &sk->receive_queue);
	__kfree_skb(skb);
}

/* Clean up the receive buffer for full frames taken by the user,
 * then send an ACK if necessary.  COPIED is the number of bytes
 * tcp_recvmsg has given to the user so far, it speeds up the
 * calculation of whether or not we must ACK for the sake of
 * a window update.
 */
static void cleanup_rbuf(struct sock *sk, int copied)
{
	struct tcp_opt *tp = &(sk->tp_pinfo.af_tcp);
	struct sk_buff *skb;
	int time_to_ack = 0;

	/* NOTE! The socket must be locked, so that we don't get
	 * a messed-up receive queue.
	 */
	while ((skb=skb_peek(&sk->receive_queue)) != NULL) {
		if (!skb->used)
			break;
		tcp_eat_skb(sk, skb);
	}

	if (tcp_ack_scheduled(tp)) {
		   /* Delayed ACKs frequently hit locked sockets during bulk receive. */
		if (tp->ack.blocked
		    /* Once-per-two-segments ACK was not sent by tcp_input.c */
		    || tp->rcv_nxt - tp->rcv_wup > tp->ack.rcv_mss
		    /*
		     * If this read emptied read buffer, we send ACK, if
		     * connection is not bidirectional, user drained
		     * receive buffer and there was a small segment
		     * in queue.
		     */
		    || (copied > 0 &&
			(tp->ack.pending&TCP_ACK_PUSHED) &&
			!tp->ack.pingpong &&
			atomic_read(&sk->rmem_alloc) == 0)) {
			time_to_ack = 1;
		}
	}

  	/* We send an ACK if we can now advertise a non-zero window
	 * which has been raised "significantly".
	 *
	 * Even if window raised up to infinity, do not send window open ACK
	 * in states, where we will not receive more. It is useless.
  	 */
	if(copied > 0 && !time_to_ack && !(sk->shutdown&RCV_SHUTDOWN)) {
		__u32 rcv_window_now = tcp_receive_window(tp);

		/* Optimize, __tcp_select_window() is not cheap. */
		if (2*rcv_window_now <= tp->window_clamp) {
			__u32 new_window = __tcp_select_window(sk);

			/* Send ACK now, if this read freed lots of space
			 * in our buffer. Certainly, new_window is new window.
			 * We can advertise it now, if it is not less than current one.
			 * "Lots" means "at least twice" here.
			 */
			if(new_window && new_window >= 2*rcv_window_now)
				time_to_ack = 1;
		}
	}
	if (time_to_ack)
		tcp_send_ack(sk);
}

/* Now socket state including sk->err is changed only under lock,
 * hence we may omit checks after joining wait queue.
 * We check receive queue before schedule() only as optimization;
 * it is very likely that release_sock() added new data.
 */

static long tcp_data_wait(struct sock *sk, long timeo)
{
	DECLARE_WAITQUEUE(wait, current);

	add_wait_queue(sk->sleep, &wait);

	__set_current_state(TASK_INTERRUPTIBLE);

	set_bit(SOCK_ASYNC_WAITDATA, &sk->socket->flags);
	release_sock(sk);

	if (skb_queue_empty(&sk->receive_queue))
		timeo = schedule_timeout(timeo);

	lock_sock(sk);
	clear_bit(SOCK_ASYNC_WAITDATA, &sk->socket->flags);

	remove_wait_queue(sk->sleep, &wait);
	__set_current_state(TASK_RUNNING);
	return timeo;
}

static void tcp_prequeue_process(struct sock *sk)
{
	struct sk_buff *skb;
	struct tcp_opt *tp = &(sk->tp_pinfo.af_tcp);

	net_statistics[smp_processor_id()*2+1].TCPPrequeued += skb_queue_len(&tp->ucopy.prequeue);

	/* RX process wants to run with disabled BHs, though it is not necessary */
	local_bh_disable();
	while ((skb = __skb_dequeue(&tp->ucopy.prequeue)) != NULL)
		sk->backlog_rcv(sk, skb);
	local_bh_enable();

	/* Clear memory counter. */
	tp->ucopy.memory = 0;
}

/*
 *	This routine copies from a sock struct into the user buffer. 
 *
 *	Technical note: in 2.3 we work on _locked_ socket, so that
 *	tricks with *seq access order and skb->users are not required.
 *	Probably, code can be easily improved even more.
 */
 
int tcp_recvmsg(struct sock *sk, struct msghdr *msg,
		int len, int nonblock, int flags, int *addr_len)
{
	struct tcp_opt *tp = &(sk->tp_pinfo.af_tcp);
	int copied = 0;
	u32 peek_seq;
	u32 *seq;
	unsigned long used;
	int err;
	int target;		/* Read at least this many bytes */
	long timeo;
	struct task_struct *user_recv = NULL;

	lock_sock(sk);

	TCP_CHECK_TIMER(sk);

	err = -ENOTCONN;
	if (sk->state == TCP_LISTEN)
		goto out;

	timeo = sock_rcvtimeo(sk, nonblock);

	/* Urgent data needs to be handled specially. */
	if (flags & MSG_OOB)
		goto recv_urg;

	seq = &tp->copied_seq;
	if (flags & MSG_PEEK) {
		peek_seq = tp->copied_seq;
		seq = &peek_seq;
	}

	target = sock_rcvlowat(sk, flags & MSG_WAITALL, len);

	do {
		struct sk_buff * skb;
		u32 offset;

		/* Are we at urgent data? Stop if we have read anything. */
		if (copied && tp->urg_data && tp->urg_seq == *seq)
			break;

		/* We need to check signals first, to get correct SIGURG
		 * handling. FIXME: Need to check this doesnt impact 1003.1g
		 * and move it down to the bottom of the loop
		 */
		if (signal_pending(current)) {
			if (copied)
				break;
			copied = timeo ? sock_intr_errno(timeo) : -EAGAIN;
			break;
		}

		/* Next get a buffer. */

		skb = skb_peek(&sk->receive_queue);
		do {
			if (!skb)
				break;

			/* Now that we have two receive queues this 
			 * shouldn't happen.
			 */
			if (before(*seq, TCP_SKB_CB(skb)->seq)) {
				printk(KERN_INFO "recvmsg bug: copied %X seq %X\n",
				       *seq, TCP_SKB_CB(skb)->seq);
				break;
			}
			offset = *seq - TCP_SKB_CB(skb)->seq;
			if (skb->h.th->syn)
				offset--;
			if (offset < skb->len)
				goto found_ok_skb;
			if (skb->h.th->fin)
				goto found_fin_ok;
			if (!(flags & MSG_PEEK))
				skb->used = 1;
			skb = skb->next;
		} while (skb != (struct sk_buff *)&sk->receive_queue);

		/* Well, if we have backlog, try to process it now yet. */

		if (copied >= target && sk->backlog.tail == NULL)
			break;

		if (copied) {
			if (sk->err ||
			    sk->state == TCP_CLOSE ||
			    (sk->shutdown & RCV_SHUTDOWN) ||
			    !timeo)