svcsock.c

Linux Kernel 2.6.9 for OMAP1710

/*
 * linux/net/sunrpc/svcsock.c
 *
 * These are the RPC server socket internals.
 *
 * The server scheduling algorithm does not always distribute the load
 * evenly when servicing a single client. May need to modify the
 * svc_sock_enqueue procedure...
 *
 * TCP support is largely untested and may be a little slow. The problem
 * is that we currently do two separate recvfrom's, one for the 4-byte
 * record length, and the second for the actual record. This could possibly
 * be improved by always reading a minimum size of around 100 bytes and
 * tucking any superfluous bytes away in a temporary store. Still, that
 * leaves write requests out in the rain. An alternative may be to peek at
 * the first skb in the queue, and if it matches the next TCP sequence
 * number, to extract the record marker. Yuck.
 *
 * Copyright (C) 1995, 1996 Olaf Kirch <[e-mail]>
 */

#include <linux/sched.h>
#include <linux/errno.h>
#include <linux/fcntl.h>
#include <linux/net.h>
#include <linux/in.h>
#include <linux/inet.h>
#include <linux/udp.h>
#include <linux/tcp.h>
#include <linux/unistd.h>
#include <linux/slab.h>
#include <linux/netdevice.h>
#include <linux/skbuff.h>
#include <linux/suspend.h>
#include <net/sock.h>
#include <net/checksum.h>
#include <net/ip.h>
#include <net/tcp.h>
#include <asm/uaccess.h>
#include <asm/ioctls.h>

#include <linux/sunrpc/types.h>
#include <linux/sunrpc/xdr.h>
#include <linux/sunrpc/svcsock.h>
#include <linux/sunrpc/stats.h>

/* SMP locking strategy:
 *
 * 	svc_serv->sv_lock protects most stuff for that service.
 *
 *	Some flags can be set to certain values at any time
 *	providing that certain rules are followed:
 *
 *	SK_BUSY  can be set to 0 at any time.  
 *		svc_sock_enqueue must be called afterwards
 *	SK_CONN, SK_DATA, can be set or cleared at any time.
 *		after a set, svc_sock_enqueue must be called.	
 *		after a clear, the socket must be read/accepted
 *		 if this succeeds, it must be set again.
 *	SK_CLOSE can set at any time. It is never cleared.
 *
 */

#define RPCDBG_FACILITY	RPCDBG_SVCSOCK


static struct svc_sock *svc_setup_socket(struct svc_serv *, struct socket *,
					 int *errp, int pmap_reg);
static void		svc_udp_data_ready(struct sock *, int);
static int		svc_udp_recvfrom(struct svc_rqst *);
static int		svc_udp_sendto(struct svc_rqst *);

static struct svc_deferred_req *svc_deferred_dequeue(struct svc_sock *svsk);
static int svc_deferred_recv(struct svc_rqst *rqstp);
static struct cache_deferred_req *svc_defer(struct cache_req *req);

/*
 * Queue up an idle server thread.  Must have serv->sv_lock held.
 * Note: this is really a stack rather than a queue, so that we only
 * use as many different threads as we need, and the rest don't polute
 * the cache.
 */
static inline void
svc_serv_enqueue(struct svc_serv *serv, struct svc_rqst *rqstp)
{
	list_add(&rqstp->rq_list, &serv->sv_threads);
}

/*
 * Dequeue an nfsd thread.  Must have serv->sv_lock held.
 */
static inline void
svc_serv_dequeue(struct svc_serv *serv, struct svc_rqst *rqstp)
{
	list_del(&rqstp->rq_list);
}

/*
 * Release an skbuff after use
 */
static inline void
svc_release_skb(struct svc_rqst *rqstp)
{
	struct sk_buff *skb = rqstp->rq_skbuff;
	struct svc_deferred_req *dr = rqstp->rq_deferred;

	if (skb) {
		rqstp->rq_skbuff = NULL;

		dprintk("svc: service %p, releasing skb %p\n", rqstp, skb);
		skb_free_datagram(rqstp->rq_sock->sk_sk, skb);
	}
	if (dr) {
		rqstp->rq_deferred = NULL;
		kfree(dr);
	}
}

/*
 * Any space to write?
 */
static inline unsigned long
svc_sock_wspace(struct svc_sock *svsk)
{
	int wspace;

	if (svsk->sk_sock->type == SOCK_STREAM)
		wspace = sk_stream_wspace(svsk->sk_sk);
	else
		wspace = sock_wspace(svsk->sk_sk);

	return wspace;
}

/*
 * Queue up a socket with data pending. If there are idle nfsd
 * processes, wake 'em up.
 *
 */
static void
svc_sock_enqueue(struct svc_sock *svsk)
{
	struct svc_serv	*serv = svsk->sk_server;
	struct svc_rqst	*rqstp;

	if (!(svsk->sk_flags &
	      ( (1<<SK_CONN)|(1<<SK_DATA)|(1<<SK_CLOSE)|(1<<SK_DEFERRED)) ))
		return;
	if (test_bit(SK_DEAD, &svsk->sk_flags))
		return;

	spin_lock_bh(&serv->sv_lock);

	if (!list_empty(&serv->sv_threads) && 
	    !list_empty(&serv->sv_sockets))
		printk(KERN_ERR
			"svc_sock_enqueue: threads and sockets both waiting??\n");

	if (test_bit(SK_DEAD, &svsk->sk_flags)) {
		/* Don't enqueue dead sockets */
		dprintk("svc: socket %p is dead, not enqueued\n", svsk->sk_sk);
		goto out_unlock;
	}

	if (test_bit(SK_BUSY, &svsk->sk_flags)) {
		/* Don't enqueue socket while daemon is receiving */
		dprintk("svc: socket %p busy, not enqueued\n", svsk->sk_sk);
		goto out_unlock;
	}

	set_bit(SOCK_NOSPACE, &svsk->sk_sock->flags);
	if (((svsk->sk_reserved + serv->sv_bufsz)*2
	     > svc_sock_wspace(svsk))
	    && !test_bit(SK_CLOSE, &svsk->sk_flags)
	    && !test_bit(SK_CONN, &svsk->sk_flags)) {
		/* Don't enqueue while not enough space for reply */
		dprintk("svc: socket %p  no space, %d*2 > %ld, not enqueued\n",
			svsk->sk_sk, svsk->sk_reserved+serv->sv_bufsz,
			svc_sock_wspace(svsk));
		goto out_unlock;
	}
	clear_bit(SOCK_NOSPACE, &svsk->sk_sock->flags);

	/* Mark socket as busy. It will remain in this state until the
	 * server has processed all pending data and put the socket back
	 * on the idle list.
	 */
	set_bit(SK_BUSY, &svsk->sk_flags);

	if (!list_empty(&serv->sv_threads)) {
		rqstp = list_entry(serv->sv_threads.next,
				   struct svc_rqst,
				   rq_list);
		dprintk("svc: socket %p served by daemon %p\n",
			svsk->sk_sk, rqstp);
		svc_serv_dequeue(serv, rqstp);
		if (rqstp->rq_sock)
			printk(KERN_ERR 
				"svc_sock_enqueue: server %p, rq_sock=%p!\n",
				rqstp, rqstp->rq_sock);
		rqstp->rq_sock = svsk;
		svsk->sk_inuse++;
		rqstp->rq_reserved = serv->sv_bufsz;
		svsk->sk_reserved += rqstp->rq_reserved;
		wake_up(&rqstp->rq_wait);
	} else {
		dprintk("svc: socket %p put into queue\n", svsk->sk_sk);
		list_add_tail(&svsk->sk_ready, &serv->sv_sockets);
	}

out_unlock:
	spin_unlock_bh(&serv->sv_lock);
}

/*
 * Dequeue the first socket.  Must be called with the serv->sv_lock held.
 */
static inline struct svc_sock *
svc_sock_dequeue(struct svc_serv *serv)
{
	struct svc_sock	*svsk;

	if (list_empty(&serv->sv_sockets))
		return NULL;

	svsk = list_entry(serv->sv_sockets.next,
			  struct svc_sock, sk_ready);
	list_del_init(&svsk->sk_ready);

	dprintk("svc: socket %p dequeued, inuse=%d\n",
		svsk->sk_sk, svsk->sk_inuse);

	return svsk;
}

/*
 * Having read something from a socket, check whether it
 * needs to be re-enqueued.
 * Note: SK_DATA only gets cleared when a read-attempt finds
 * no (or insufficient) data.
 */
static inline void
svc_sock_received(struct svc_sock *svsk)
{
	clear_bit(SK_BUSY, &svsk->sk_flags);
	svc_sock_enqueue(svsk);
}


/**
 * svc_reserve - change the space reserved for the reply to a request.
 * @rqstp:  The request in question
 * @space: new max space to reserve
 *
 * Each request reserves some space on the output queue of the socket
 * to make sure the reply fits.  This function reduces that reserved
 * space to be the amount of space used already, plus @space.
 *
 */
void svc_reserve(struct svc_rqst *rqstp, int space)
{
	space += rqstp->rq_res.head[0].iov_len;

	if (space < rqstp->rq_reserved) {
		struct svc_sock *svsk = rqstp->rq_sock;
		spin_lock_bh(&svsk->sk_server->sv_lock);
		svsk->sk_reserved -= (rqstp->rq_reserved - space);
		rqstp->rq_reserved = space;
		spin_unlock_bh(&svsk->sk_server->sv_lock);

		svc_sock_enqueue(svsk);
	}
}

/*
 * Release a socket after use.
 */
static inline void
svc_sock_put(struct svc_sock *svsk)
{
	struct svc_serv *serv = svsk->sk_server;

	spin_lock_bh(&serv->sv_lock);
	if (!--(svsk->sk_inuse) && test_bit(SK_DEAD, &svsk->sk_flags)) {
		spin_unlock_bh(&serv->sv_lock);
		dprintk("svc: releasing dead socket\n");
		sock_release(svsk->sk_sock);
		kfree(svsk);
	}
	else
		spin_unlock_bh(&serv->sv_lock);
}

static void
svc_sock_release(struct svc_rqst *rqstp)
{
	struct svc_sock	*svsk = rqstp->rq_sock;

	svc_release_skb(rqstp);

	svc_free_allpages(rqstp);
	rqstp->rq_res.page_len = 0;
	rqstp->rq_res.page_base = 0;


	/* Reset response buffer and release
	 * the reservation.
	 * But first, check that enough space was reserved
	 * for the reply, otherwise we have a bug!
	 */
	if ((rqstp->rq_res.len) >  rqstp->rq_reserved)
		printk(KERN_ERR "RPC request reserved %d but used %d\n",
		       rqstp->rq_reserved,
		       rqstp->rq_res.len);

	rqstp->rq_res.head[0].iov_len = 0;
	svc_reserve(rqstp, 0);
	rqstp->rq_sock = NULL;

	svc_sock_put(svsk);
}

/*
 * External function to wake up a server waiting for data
 */
void
svc_wake_up(struct svc_serv *serv)
{
	struct svc_rqst	*rqstp;

	spin_lock_bh(&serv->sv_lock);
	if (!list_empty(&serv->sv_threads)) {
		rqstp = list_entry(serv->sv_threads.next,
				   struct svc_rqst,
				   rq_list);
		dprintk("svc: daemon %p woken up.\n", rqstp);
		/*
		svc_serv_dequeue(serv, rqstp);
		rqstp->rq_sock = NULL;
		 */
		wake_up(&rqstp->rq_wait);
	}
	spin_unlock_bh(&serv->sv_lock);
}

/*
 * Generic sendto routine
 */
static int
svc_sendto(struct svc_rqst *rqstp, struct xdr_buf *xdr)
{
	struct svc_sock	*svsk = rqstp->rq_sock;
	struct socket	*sock = svsk->sk_sock;
	int		slen;
	char 		buffer[CMSG_SPACE(sizeof(struct in_pktinfo))];
	struct cmsghdr *cmh = (struct cmsghdr *)buffer;
	struct in_pktinfo *pki = (struct in_pktinfo *)CMSG_DATA(cmh);
	int		len = 0;
	int		result;
	int		size;
	struct page	**ppage = xdr->pages;
	size_t		base = xdr->page_base;
	unsigned int	pglen = xdr->page_len;
	unsigned int	flags = MSG_MORE;

	slen = xdr->len;

	if (rqstp->rq_prot == IPPROTO_UDP) {
		/* set the source and destination */
		struct msghdr	msg;
		msg.msg_name    = &rqstp->rq_addr;
		msg.msg_namelen = sizeof(rqstp->rq_addr);
		msg.msg_iov     = NULL;
		msg.msg_iovlen  = 0;
		msg.msg_flags	= MSG_MORE;

		msg.msg_control = cmh;
		msg.msg_controllen = sizeof(buffer);
		cmh->cmsg_len = CMSG_LEN(sizeof(*pki));
		cmh->cmsg_level = SOL_IP;
		cmh->cmsg_type = IP_PKTINFO;
		pki->ipi_ifindex = 0;
		pki->ipi_spec_dst.s_addr = rqstp->rq_daddr;

		if (sock_sendmsg(sock, &msg, 0) < 0)
			goto out;
	}

	/* send head */
	if (slen == xdr->head[0].iov_len)
		flags = 0;
	len = sock->ops->sendpage(sock, rqstp->rq_respages[0], 0, xdr->head[0].iov_len, flags);
	if (len != xdr->head[0].iov_len)
		goto out;
	slen -= xdr->head[0].iov_len;
	if (slen == 0)
		goto out;

	/* send page data */
	size = PAGE_SIZE - base < pglen ? PAGE_SIZE - base : pglen;
	while (pglen > 0) {
		if (slen == size)
			flags = 0;
		result = sock->ops->sendpage(sock, *ppage, base, size, flags);
		if (result > 0)
			len += result;
		if (result != size)
			goto out;
		slen -= size;
		pglen -= size;
		size = PAGE_SIZE < pglen ? PAGE_SIZE : pglen;
		base = 0;
		ppage++;
	}
	/* send tail */
	if (xdr->tail[0].iov_len) {
		result = sock->ops->sendpage(sock, rqstp->rq_respages[rqstp->rq_restailpage], 
					     ((unsigned long)xdr->tail[0].iov_base)& (PAGE_SIZE-1),
					     xdr->tail[0].iov_len, 0);

		if (result > 0)
			len += result;
	}
out:
	dprintk("svc: socket %p sendto([%p %Zu... ], %d) = %d (addr %x)\n",
			rqstp->rq_sock, xdr->head[0].iov_base, xdr->head[0].iov_len, xdr->len, len,
		rqstp->rq_addr.sin_addr.s_addr);

	return len;
}

/*
 * Check input queue length
 */
static int
svc_recv_available(struct svc_sock *svsk)
{
	mm_segment_t	oldfs;
	struct socket	*sock = svsk->sk_sock;
	int		avail, err;

	oldfs = get_fs(); set_fs(KERNEL_DS);
	err = sock->ops->ioctl(sock, TIOCINQ, (unsigned long) &avail);
	set_fs(oldfs);

	return (err >= 0)? avail : err;
}

/*
 * Generic recvfrom routine.
 */
static int
svc_recvfrom(struct svc_rqst *rqstp, struct kvec *iov, int nr, int buflen)
{
	struct msghdr	msg;
	struct socket	*sock;
	int		len, alen;

	rqstp->rq_addrlen = sizeof(rqstp->rq_addr);
	sock = rqstp->rq_sock->sk_sock;

	msg.msg_name    = &rqstp->rq_addr;
	msg.msg_namelen = sizeof(rqstp->rq_addr);
	msg.msg_control = NULL;
	msg.msg_controllen = 0;

	msg.msg_flags	= MSG_DONTWAIT;

	len = kernel_recvmsg(sock, &msg, iov, nr, buflen, MSG_DONTWAIT);

	/* sock_recvmsg doesn't fill in the name/namelen, so we must..
	 * possibly we should cache this in the svc_sock structure
	 * at accept time. FIXME
	 */
	alen = sizeof(rqstp->rq_addr);
	sock->ops->getname(sock, (struct sockaddr *)&rqstp->rq_addr, &alen, 1);

	dprintk("svc: socket %p recvfrom(%p, %Zu) = %d\n",
		rqstp->rq_sock, iov[0].iov_base, iov[0].iov_len, len);

	return len;
}

/*
 * Set socket snd and rcv buffer lengths
 */
static inline void
svc_sock_setbufsize(struct socket *sock, unsigned int snd, unsigned int rcv)
{
#if 0
	mm_segment_t	oldfs;
	oldfs = get_fs(); set_fs(KERNEL_DS);
	sock_setsockopt(sock, SOL_SOCKET, SO_SNDBUF,
			(char*)&snd, sizeof(snd));
	sock_setsockopt(sock, SOL_SOCKET, SO_RCVBUF,
			(char*)&rcv, sizeof(rcv));
#else
	/* sock_setsockopt limits use to sysctl_?mem_max,
	 * which isn't acceptable.  Until that is made conditional
	 * on not having CAP_SYS_RESOURCE or similar, we go direct...
	 * DaveM said I could!
	 */
	lock_sock(sock->sk);
	sock->sk->sk_sndbuf = snd * 2;
	sock->sk->sk_rcvbuf = rcv * 2;
	sock->sk->sk_userlocks |= SOCK_SNDBUF_LOCK|SOCK_RCVBUF_LOCK;
	release_sock(sock->sk);
#endif
}
/*
 * INET callback when data has been received on the socket.
 */
static void
svc_udp_data_ready(struct sock *sk, int count)
{
	struct svc_sock	*svsk = (struct svc_sock *)(sk->sk_user_data);

	if (!svsk)
		goto out;
	dprintk("svc: socket %p(inet %p), count=%d, busy=%d\n",
		svsk, sk, count, test_bit(SK_BUSY, &svsk->sk_flags));
	set_bit(SK_DATA, &svsk->sk_flags);
	svc_sock_enqueue(svsk);
 out:
	if (sk->sk_sleep && waitqueue_active(sk->sk_sleep))
		wake_up_interruptible(sk->sk_sleep);
}

/*