br2684.c

Linux Kernel 2.6.9 for OMAP1710

/*
Experimental ethernet netdevice using ATM AAL5 as underlying carrier
(RFC1483 obsoleted by RFC2684) for Linux 2.4
Author: Marcell GAL, 2000, XDSL Ltd, Hungary
*/

#include <linux/module.h>
#include <linux/config.h>
#include <linux/init.h>
#include <linux/kernel.h>
#include <linux/list.h>
#include <linux/netdevice.h>
#include <linux/skbuff.h>
#include <linux/etherdevice.h>
#include <linux/rtnetlink.h>
#include <linux/ip.h>
#include <asm/uaccess.h>
#include <net/arp.h>
#include <linux/atm.h>
#include <linux/atmdev.h>
#include <linux/seq_file.h>

#include <linux/atmbr2684.h>

#include "common.h"
#include "ipcommon.h"

/*
 * Define this to use a version of the code which interacts with the higher
 * layers in a more intellegent way, by always reserving enough space for
 * our header at the begining of the packet.  However, there may still be
 * some problems with programs like tcpdump.  In 2.5 we'll sort out what
 * we need to do to get this perfect.  For now we just will copy the packet
 * if we need space for the header
 */
/* #define FASTER_VERSION */

#ifdef DEBUG
#define DPRINTK(format, args...) printk(KERN_DEBUG "br2684: " format, ##args)
#else
#define DPRINTK(format, args...)
#endif

#ifdef SKB_DEBUG
static void skb_debug(const struct sk_buff *skb)
{
#define NUM2PRINT 50
	char buf[NUM2PRINT * 3 + 1];	/* 3 chars per byte */
	int i = 0;
	for (i = 0; i < skb->len && i < NUM2PRINT; i++) {
		sprintf(buf + i * 3, "%2.2x ", 0xff & skb->data[i]);
	}
	printk(KERN_DEBUG "br2684: skb: %s\n", buf);
}
#else
#define skb_debug(skb)	do {} while (0)
#endif

static unsigned char llc_oui_pid_pad[] =
    { 0xAA, 0xAA, 0x03, 0x00, 0x80, 0xC2, 0x00, 0x07, 0x00, 0x00 };
#define PADLEN	(2)

enum br2684_encaps {
	e_vc  = BR2684_ENCAPS_VC,
	e_llc = BR2684_ENCAPS_LLC,
};

struct br2684_vcc {
	struct atm_vcc  *atmvcc;
	struct net_device *device;
	/* keep old push,pop functions for chaining */
	void (*old_push)(struct atm_vcc *vcc,struct sk_buff *skb);
	/* void (*old_pop)(struct atm_vcc *vcc,struct sk_buff *skb); */
	enum br2684_encaps encaps;
	struct list_head brvccs;
#ifdef CONFIG_ATM_BR2684_IPFILTER
	struct br2684_filter filter;
#endif /* CONFIG_ATM_BR2684_IPFILTER */
#ifndef FASTER_VERSION
	unsigned copies_needed, copies_failed;
#endif /* FASTER_VERSION */
};

struct br2684_dev {
	struct net_device *net_dev;
	struct list_head br2684_devs;
	int number;
	struct list_head brvccs; /* one device <=> one vcc (before xmas) */
	struct net_device_stats stats;
	int mac_was_set;
};

/*
 * This lock should be held for writing any time the list of devices or
 * their attached vcc's could be altered.  It should be held for reading
 * any time these are being queried.  Note that we sometimes need to
 * do read-locking under interrupt context, so write locking must block
 * the current CPU's interrupts
 */
static rwlock_t devs_lock = RW_LOCK_UNLOCKED;

static LIST_HEAD(br2684_devs);

static inline struct br2684_dev *BRPRIV(const struct net_device *net_dev)
{
	return (struct br2684_dev *) net_dev->priv;
}

static inline struct net_device *list_entry_brdev(const struct list_head *le)
{
	return list_entry(le, struct br2684_dev, br2684_devs)->net_dev;
}

static inline struct br2684_vcc *BR2684_VCC(const struct atm_vcc *atmvcc)
{
	return (struct br2684_vcc *) (atmvcc->user_back);
}

static inline struct br2684_vcc *list_entry_brvcc(const struct list_head *le)
{
	return list_entry(le, struct br2684_vcc, brvccs);
}

/* Caller should hold read_lock(&devs_lock) */
static struct net_device *br2684_find_dev(const struct br2684_if_spec *s)
{
	struct list_head *lh;
	struct net_device *net_dev;
	switch (s->method) {
	case BR2684_FIND_BYNUM:
		list_for_each(lh, &br2684_devs) {
			net_dev = list_entry_brdev(lh);
			if (BRPRIV(net_dev)->number == s->spec.devnum)
				return net_dev;
		}
		break;
	case BR2684_FIND_BYIFNAME:
		list_for_each(lh, &br2684_devs) {
			net_dev = list_entry_brdev(lh);
			if (!strncmp(net_dev->name, s->spec.ifname, IFNAMSIZ))
				return net_dev;
		}
		break;
	}
	return NULL;
}

/*
 * Send a packet out a particular vcc.  Not to useful right now, but paves
 * the way for multiple vcc's per itf.  Returns true if we can send,
 * otherwise false
 */
static int br2684_xmit_vcc(struct sk_buff *skb, struct br2684_dev *brdev,
	struct br2684_vcc *brvcc)
{
	struct atm_vcc *atmvcc;
#ifdef FASTER_VERSION
	if (brvcc->encaps == e_llc)
		memcpy(skb_push(skb, 8), llc_oui_pid_pad, 8);
	/* last 2 bytes of llc_oui_pid_pad are managed by header routines;
	   yes, you got it: 8 + 2 = sizeof(llc_oui_pid_pad)
	 */
#else
	int minheadroom = (brvcc->encaps == e_llc) ? 10 : 2;
	if (skb_headroom(skb) < minheadroom) {
		struct sk_buff *skb2 = skb_realloc_headroom(skb, minheadroom);
		brvcc->copies_needed++;
		dev_kfree_skb(skb);
		if (skb2 == NULL) {
			brvcc->copies_failed++;
			return 0;
		}
		skb = skb2;
	}
	skb_push(skb, minheadroom);
	if (brvcc->encaps == e_llc)
		memcpy(skb->data, llc_oui_pid_pad, 10);
	else
		memset(skb->data, 0, 2);
#endif /* FASTER_VERSION */
	skb_debug(skb);

	ATM_SKB(skb)->vcc = atmvcc = brvcc->atmvcc;
	DPRINTK("atm_skb(%p)->vcc(%p)->dev(%p)\n", skb, atmvcc, atmvcc->dev);
	if (!atm_may_send(atmvcc, skb->truesize)) {
		/* we free this here for now, because we cannot know in a higher 
			layer whether the skb point it supplied wasn't freed yet.
			now, it always is.
		*/
		dev_kfree_skb(skb);
		return 0;
		}
	atomic_add(skb->truesize, &atmvcc->sk->sk_wmem_alloc);
	ATM_SKB(skb)->atm_options = atmvcc->atm_options;
	brdev->stats.tx_packets++;
	brdev->stats.tx_bytes += skb->len;
	atmvcc->send(atmvcc, skb);
	return 1;
}

static inline struct br2684_vcc *pick_outgoing_vcc(struct sk_buff *skb,
	struct br2684_dev *brdev)
{
	return list_empty(&brdev->brvccs) ? NULL :
	    list_entry_brvcc(brdev->brvccs.next); /* 1 vcc/dev right now */
}

static int br2684_start_xmit(struct sk_buff *skb, struct net_device *dev)
{
	struct br2684_dev *brdev = BRPRIV(dev);
	struct br2684_vcc *brvcc;

	DPRINTK("br2684_start_xmit, skb->dst=%p\n", skb->dst);
	read_lock(&devs_lock);
	brvcc = pick_outgoing_vcc(skb, brdev);
	if (brvcc == NULL) {
		DPRINTK("no vcc attached to dev %s\n", dev->name);
		brdev->stats.tx_errors++;
		brdev->stats.tx_carrier_errors++;
		/* netif_stop_queue(dev); */
		dev_kfree_skb(skb);
		read_unlock(&devs_lock);
		return -EUNATCH;
	}
	if (!br2684_xmit_vcc(skb, brdev, brvcc)) {
		/*
		 * We should probably use netif_*_queue() here, but that
		 * involves added complication.  We need to walk before
		 * we can run
		 */
		/* don't free here! this pointer might be no longer valid!
		dev_kfree_skb(skb);
		*/
		brdev->stats.tx_errors++;
		brdev->stats.tx_fifo_errors++;
	}
	read_unlock(&devs_lock);
	return 0;
}

static struct net_device_stats *br2684_get_stats(struct net_device *dev)
{
	DPRINTK("br2684_get_stats\n");
	return &BRPRIV(dev)->stats;
}

#ifdef FASTER_VERSION
/*
 * These mirror eth_header and eth_header_cache.  They are not usually
 * exported for use in modules, so we grab them from net_device
 * after ether_setup() is done with it.  Bit of a hack.
 */
static int (*my_eth_header)(struct sk_buff *, struct net_device *,
	unsigned short, void *, void *, unsigned);
static int (*my_eth_header_cache)(struct neighbour *, struct hh_cache *);

static int
br2684_header(struct sk_buff *skb, struct net_device *dev,
	      unsigned short type, void *daddr, void *saddr, unsigned len)
{
	u16 *pad_before_eth;
	int t = my_eth_header(skb, dev, type, daddr, saddr, len);
	if (t > 0) {
		pad_before_eth = (u16 *) skb_push(skb, 2);
		*pad_before_eth = 0;
		return dev->hard_header_len;	/* or return 16; ? */
	} else
		return t;
}

static int
br2684_header_cache(struct neighbour *neigh, struct hh_cache *hh)
{
/* hh_data is 16 bytes long. if encaps is ether-llc we need 24, so
xmit will add the additional header part in that case */
	u16 *pad_before_eth = (u16 *)(hh->hh_data);
	int t = my_eth_header_cache(neigh, hh);
	DPRINTK("br2684_header_cache, neigh=%p, hh_cache=%p\n", neigh, hh);
	if (t < 0)
		return t;
	else {
		*pad_before_eth = 0;
		hh->hh_len = PADLEN + ETH_HLEN;
	}
	return 0;
}

/*
 * This is similar to eth_type_trans, which cannot be used because of
 * our dev->hard_header_len
 */
static inline unsigned short br_type_trans(struct sk_buff *skb,
					       struct net_device *dev)
{
	struct ethhdr *eth;
	unsigned char *rawp;
	eth = eth_hdr(skb);

	if (*eth->h_dest & 1) {
		if (memcmp(eth->h_dest, dev->broadcast, ETH_ALEN) == 0)
			skb->pkt_type = PACKET_BROADCAST;
		else
			skb->pkt_type = PACKET_MULTICAST;
	}

	else if (memcmp(eth->h_dest, dev->dev_addr, ETH_ALEN))
		skb->pkt_type = PACKET_OTHERHOST;

	if (ntohs(eth->h_proto) >= 1536)
		return eth->h_proto;

	rawp = skb->data;

	/*
	 * This is a magic hack to spot IPX packets. Older Novell breaks
	 * the protocol design and runs IPX over 802.3 without an 802.2 LLC
	 * layer. We look for FFFF which isn't a used 802.2 SSAP/DSAP. This
	 * won't work for fault tolerant netware but does for the rest.
	 */
	if (*(unsigned short *) rawp == 0xFFFF)
		return htons(ETH_P_802_3);

	/*
	 * Real 802.2 LLC
	 */
	return htons(ETH_P_802_2);
}
#endif /* FASTER_VERSION */

/*
 * We remember when the MAC gets set, so we don't override it later with
 * the ESI of the ATM card of the first VC
 */
static int (*my_eth_mac_addr)(struct net_device *, void *);
static int br2684_mac_addr(struct net_device *dev, void *p)
{
	int err = my_eth_mac_addr(dev, p);
	if (!err)
		BRPRIV(dev)->mac_was_set = 1;
	return err;
}

#ifdef CONFIG_ATM_BR2684_IPFILTER
/* this IOCTL is experimental. */
static int br2684_setfilt(struct atm_vcc *atmvcc, void __user *arg)
{
	struct br2684_vcc *brvcc;
	struct br2684_filter_set fs;

	if (copy_from_user(&fs, arg, sizeof fs))
		return -EFAULT;
	if (fs.ifspec.method != BR2684_FIND_BYNOTHING) {
		/*
		 * This is really a per-vcc thing, but we can also search
		 * by device
		 */
		struct br2684_dev *brdev;
		read_lock(&devs_lock);
		brdev = BRPRIV(br2684_find_dev(&fs.ifspec));
		if (brdev == NULL || list_empty(&brdev->brvccs) ||
		    brdev->brvccs.next != brdev->brvccs.prev)  /* >1 VCC */
			brvcc = NULL;
		else
			brvcc = list_entry_brvcc(brdev->brvccs.next);
		read_unlock(&devs_lock);
		if (brvcc == NULL)
			return -ESRCH;
	} else
		brvcc = BR2684_VCC(atmvcc);
	memcpy(&brvcc->filter, &fs.filter, sizeof(brvcc->filter));
	return 0;
}

/* Returns 1 if packet should be dropped */
static inline int
packet_fails_filter(u16 type, struct br2684_vcc *brvcc, struct sk_buff *skb)
{
	if (brvcc->filter.netmask == 0)
		return 0;			/* no filter in place */
	if (type == __constant_htons(ETH_P_IP) &&
	    (((struct iphdr *) (skb->data))->daddr & brvcc->filter.
	     netmask) == brvcc->filter.prefix)
		return 0;
	if (type == __constant_htons(ETH_P_ARP))
		return 0;
	/* TODO: we should probably filter ARPs too.. don't want to have
	 *   them returning values that don't make sense, or is that ok?
	 */
	return 1;		/* drop */
}
#endif /* CONFIG_ATM_BR2684_IPFILTER */

static void br2684_close_vcc(struct br2684_vcc *brvcc)
{
	DPRINTK("removing VCC %p from dev %p\n", brvcc, brvcc->device);
	write_lock_irq(&devs_lock);
	list_del(&brvcc->brvccs);
	write_unlock_irq(&devs_lock);
	brvcc->atmvcc->user_back = NULL;	/* what about vcc->recvq ??? */
	brvcc->old_push(brvcc->atmvcc, NULL);	/* pass on the bad news */
	kfree(brvcc);
	module_put(THIS_MODULE);
}

/* when AAL5 PDU comes in: */
static void br2684_push(struct atm_vcc *atmvcc, struct sk_buff *skb)
{
	struct br2684_vcc *brvcc = BR2684_VCC(atmvcc);
	struct net_device *net_dev = brvcc->device;
	struct br2684_dev *brdev = BRPRIV(net_dev);
	int plen = sizeof(llc_oui_pid_pad) + ETH_HLEN;

	DPRINTK("br2684_push\n");