fedora core 6 1.2949 + vserver 2.2.0
[linux-2.6.git] / net / ipv4 / ip_gre.c
index 8848355..476cb60 100644 (file)
@@ -10,7 +10,7 @@
  *
  */
 
-#include <linux/config.h>
+#include <linux/capability.h>
 #include <linux/module.h>
 #include <linux/types.h>
 #include <linux/sched.h>
@@ -28,6 +28,7 @@
 #include <linux/inetdevice.h>
 #include <linux/igmp.h>
 #include <linux/netfilter_ipv4.h>
+#include <linux/if_ether.h>
 
 #include <net/sock.h>
 #include <net/ip.h>
@@ -143,7 +144,7 @@ static struct net_device *ipgre_fb_tunnel_dev;
  */
 
 #define HASH_SIZE  16
-#define HASH(addr) ((addr^(addr>>4))&0xF)
+#define HASH(addr) (((__force u32)addr^((__force u32)addr>>4))&0xF)
 
 static struct ip_tunnel *tunnels[4][HASH_SIZE];
 
@@ -156,7 +157,7 @@ static DEFINE_RWLOCK(ipgre_lock);
 
 /* Given src, dst and key, find appropriate for input tunnel. */
 
-static struct ip_tunnel * ipgre_tunnel_lookup(u32 remote, u32 local, u32 key)
+static struct ip_tunnel * ipgre_tunnel_lookup(__be32 remote, __be32 local, __be32 key)
 {
        unsigned h0 = HASH(remote);
        unsigned h1 = HASH(key);
@@ -187,15 +188,15 @@ static struct ip_tunnel * ipgre_tunnel_lookup(u32 remote, u32 local, u32 key)
        }
 
        if (ipgre_fb_tunnel_dev->flags&IFF_UP)
-               return ipgre_fb_tunnel_dev->priv;
+               return netdev_priv(ipgre_fb_tunnel_dev);
        return NULL;
 }
 
 static struct ip_tunnel **ipgre_bucket(struct ip_tunnel *t)
 {
-       u32 remote = t->parms.iph.daddr;
-       u32 local = t->parms.iph.saddr;
-       u32 key = t->parms.i_key;
+       __be32 remote = t->parms.iph.daddr;
+       __be32 local = t->parms.iph.saddr;
+       __be32 key = t->parms.i_key;
        unsigned h = HASH(key);
        int prio = 0;
 
@@ -235,9 +236,9 @@ static void ipgre_tunnel_unlink(struct ip_tunnel *t)
 
 static struct ip_tunnel * ipgre_tunnel_locate(struct ip_tunnel_parm *parms, int create)
 {
-       u32 remote = parms->iph.daddr;
-       u32 local = parms->iph.saddr;
-       u32 key = parms->i_key;
+       __be32 remote = parms->iph.daddr;
+       __be32 local = parms->iph.saddr;
+       __be32 key = parms->i_key;
        struct ip_tunnel *t, **tp, *nt;
        struct net_device *dev;
        unsigned h = HASH(key);
@@ -277,7 +278,7 @@ static struct ip_tunnel * ipgre_tunnel_locate(struct ip_tunnel_parm *parms, int
          return NULL;
 
        dev->init = ipgre_tunnel_init;
-       nt = dev->priv;
+       nt = netdev_priv(dev);
        nt->parms = *parms;
 
        if (register_netdevice(dev) < 0) {
@@ -285,12 +286,8 @@ static struct ip_tunnel * ipgre_tunnel_locate(struct ip_tunnel_parm *parms, int
                goto failed;
        }
 
-       nt = dev->priv;
-       nt->parms = *parms;
-
        dev_hold(dev);
        ipgre_tunnel_link(nt);
-       /* Do not decrement MOD_USE_COUNT here. */
        return nt;
 
 failed:
@@ -299,7 +296,7 @@ failed:
 
 static void ipgre_tunnel_uninit(struct net_device *dev)
 {
-       ipgre_tunnel_unlink((struct ip_tunnel*)dev->priv);
+       ipgre_tunnel_unlink(netdev_priv(dev));
        dev_put(dev);
 }
 
@@ -322,12 +319,12 @@ static void ipgre_err(struct sk_buff *skb, u32 info)
  */
 
        struct iphdr *iph = (struct iphdr*)skb->data;
-       u16          *p = (u16*)(skb->data+(iph->ihl<<2));
+       __be16       *p = (__be16*)(skb->data+(iph->ihl<<2));
        int grehlen = (iph->ihl<<2) + 4;
        int type = skb->h.icmph->type;
        int code = skb->h.icmph->code;
        struct ip_tunnel *t;
-       u16 flags;
+       __be16 flags;
 
        flags = p[0];
        if (flags&(GRE_CSUM|GRE_KEY|GRE_SEQ|GRE_ROUTING|GRE_VERSION)) {
@@ -373,7 +370,7 @@ static void ipgre_err(struct sk_buff *skb, u32 info)
        }
 
        read_lock(&ipgre_lock);
-       t = ipgre_tunnel_lookup(iph->daddr, iph->saddr, (flags&GRE_KEY) ? *(((u32*)p) + (grehlen>>2) - 1) : 0);
+       t = ipgre_tunnel_lookup(iph->daddr, iph->saddr, (flags&GRE_KEY) ? *(((__be32*)p) + (grehlen>>2) - 1) : 0);
        if (t == NULL || t->parms.iph.daddr == 0 || MULTICAST(t->parms.iph.daddr))
                goto out;
 
@@ -391,13 +388,14 @@ out:
 #else
        struct iphdr *iph = (struct iphdr*)dp;
        struct iphdr *eiph;
-       u16          *p = (u16*)(dp+(iph->ihl<<2));
+       __be16       *p = (__be16*)(dp+(iph->ihl<<2));
        int type = skb->h.icmph->type;
        int code = skb->h.icmph->code;
        int rel_type = 0;
        int rel_code = 0;
-       int rel_info = 0;
-       u16 flags;
+       __be32 rel_info = 0;
+       __u32 n = 0;
+       __be16 flags;
        int grehlen = (iph->ihl<<2) + 4;
        struct sk_buff *skb2;
        struct flowi fl;
@@ -425,14 +423,16 @@ out:
        default:
                return;
        case ICMP_PARAMETERPROB:
-               if (skb->h.icmph->un.gateway < (iph->ihl<<2))
+               n = ntohl(skb->h.icmph->un.gateway) >> 24;
+               if (n < (iph->ihl<<2))
                        return;
 
                /* So... This guy found something strange INSIDE encapsulated
                   packet. Well, he is fool, but what can we do ?
                 */
                rel_type = ICMP_PARAMETERPROB;
-               rel_info = skb->h.icmph->un.gateway - grehlen;
+               n -= grehlen;
+               rel_info = htonl(n << 24);
                break;
 
        case ICMP_DEST_UNREACH:
@@ -443,13 +443,14 @@ out:
                        return;
                case ICMP_FRAG_NEEDED:
                        /* And it is the only really necessary thing :-) */
-                       rel_info = ntohs(skb->h.icmph->un.frag.mtu);
-                       if (rel_info < grehlen+68)
+                       n = ntohs(skb->h.icmph->un.frag.mtu);
+                       if (n < grehlen+68)
                                return;
-                       rel_info -= grehlen;
+                       n -= grehlen;
                        /* BSD 4.2 MORE DOES NOT EXIST IN NATURE. */
-                       if (rel_info > ntohs(eiph->tot_len))
+                       if (n > ntohs(eiph->tot_len))
                                return;
+                       rel_info = htonl(n);
                        break;
                default:
                        /* All others are translated to HOST_UNREACH.
@@ -511,14 +512,13 @@ out:
 
        /* change mtu on this route */
        if (type == ICMP_DEST_UNREACH && code == ICMP_FRAG_NEEDED) {
-               if (rel_info > dst_mtu(skb2->dst)) {
+               if (n > dst_mtu(skb2->dst)) {
                        kfree_skb(skb2);
                        return;
                }
-               skb2->dst->ops->update_pmtu(skb2->dst, rel_info);
-               rel_info = htonl(rel_info);
+               skb2->dst->ops->update_pmtu(skb2->dst, n);
        } else if (type == ICMP_TIME_EXCEEDED) {
-               struct ip_tunnel *t = (struct ip_tunnel*)skb2->dev->priv;
+               struct ip_tunnel *t = netdev_priv(skb2->dev);
                if (t->parms.iph.ttl) {
                        rel_type = ICMP_DEST_UNREACH;
                        rel_code = ICMP_HOST_UNREACH;
@@ -556,9 +556,9 @@ static int ipgre_rcv(struct sk_buff *skb)
 {
        struct iphdr *iph;
        u8     *h;
-       u16    flags;
-       u16    csum = 0;
-       u32    key = 0;
+       __be16    flags;
+       __sum16   csum = 0;
+       __be32 key = 0;
        u32    seqno = 0;
        struct ip_tunnel *tunnel;
        int    offset = 4;
@@ -568,7 +568,7 @@ static int ipgre_rcv(struct sk_buff *skb)
 
        iph = skb->nh.iph;
        h = skb->data;
-       flags = *(u16*)h;
+       flags = *(__be16*)h;
 
        if (flags&(GRE_CSUM|GRE_KEY|GRE_ROUTING|GRE_SEQ|GRE_VERSION)) {
                /* - Version must be 0.
@@ -578,24 +578,25 @@ static int ipgre_rcv(struct sk_buff *skb)
                        goto drop_nolock;
 
                if (flags&GRE_CSUM) {
-                       if (skb->ip_summed == CHECKSUM_HW) {
-                               csum = (u16)csum_fold(skb->csum);
-                               if (csum)
-                                       skb->ip_summed = CHECKSUM_NONE;
-                       }
-                       if (skb->ip_summed == CHECKSUM_NONE) {
-                               skb->csum = skb_checksum(skb, 0, skb->len, 0);
-                               skb->ip_summed = CHECKSUM_HW;
-                               csum = (u16)csum_fold(skb->csum);
+                       switch (skb->ip_summed) {
+                       case CHECKSUM_COMPLETE:
+                               csum = csum_fold(skb->csum);
+                               if (!csum)
+                                       break;
+                               /* fall through */
+                       case CHECKSUM_NONE:
+                               skb->csum = 0;
+                               csum = __skb_checksum_complete(skb);
+                               skb->ip_summed = CHECKSUM_COMPLETE;
                        }
                        offset += 4;
                }
                if (flags&GRE_KEY) {
-                       key = *(u32*)(h + offset);
+                       key = *(__be32*)(h + offset);
                        offset += 4;
                }
                if (flags&GRE_SEQ) {
-                       seqno = ntohl(*(u32*)(h + offset));
+                       seqno = ntohl(*(__be32*)(h + offset));
                        offset += 4;
                }
        }
@@ -604,22 +605,21 @@ static int ipgre_rcv(struct sk_buff *skb)
        if ((tunnel = ipgre_tunnel_lookup(iph->saddr, iph->daddr, key)) != NULL) {
                secpath_reset(skb);
 
-               skb->protocol = *(u16*)(h + 2);
+               skb->protocol = *(__be16*)(h + 2);
                /* WCCP version 1 and 2 protocol decoding.
                 * - Change protocol to IP
                 * - When dealing with WCCPv2, Skip extra 4 bytes in GRE header
                 */
                if (flags == 0 &&
-                   skb->protocol == __constant_htons(ETH_P_WCCP)) {
-                       skb->protocol = __constant_htons(ETH_P_IP);
+                   skb->protocol == htons(ETH_P_WCCP)) {
+                       skb->protocol = htons(ETH_P_IP);
                        if ((*(h + offset) & 0xF0) != 0x40) 
                                offset += 4;
                }
 
                skb->mac.raw = skb->nh.raw;
                skb->nh.raw = __pskb_pull(skb, offset);
-               skb_postpull_rcsum(skb, skb->mac.raw, offset);
-               memset(&(IPCB(skb)->opt), 0, sizeof(struct ip_options));
+               skb_postpull_rcsum(skb, skb->h.raw, offset);
                skb->pkt_type = PACKET_HOST;
 #ifdef CONFIG_NET_IPGRE_BROADCAST
                if (MULTICAST(iph->daddr)) {
@@ -657,7 +657,7 @@ static int ipgre_rcv(struct sk_buff *skb)
                read_unlock(&ipgre_lock);
                return(0);
        }
-       icmp_send(skb, ICMP_DEST_UNREACH, ICMP_PROT_UNREACH, 0);
+       icmp_send(skb, ICMP_DEST_UNREACH, ICMP_PORT_UNREACH, 0);
 
 drop:
        read_unlock(&ipgre_lock);
@@ -668,18 +668,18 @@ drop_nolock:
 
 static int ipgre_tunnel_xmit(struct sk_buff *skb, struct net_device *dev)
 {
-       struct ip_tunnel *tunnel = (struct ip_tunnel*)dev->priv;
+       struct ip_tunnel *tunnel = netdev_priv(dev);
        struct net_device_stats *stats = &tunnel->stat;
        struct iphdr  *old_iph = skb->nh.iph;
        struct iphdr  *tiph;
        u8     tos;
-       u16    df;
+       __be16 df;
        struct rtable *rt;                      /* Route to the other host */
        struct net_device *tdev;                        /* Device to other host */
        struct iphdr  *iph;                     /* Our new IP header */
        int    max_headroom;                    /* The extra header space needed */
        int    gre_hlen;
-       u32    dst;
+       __be32 dst;
        int    mtu;
 
        if (tunnel->recursion++) {
@@ -831,6 +831,8 @@ static int ipgre_tunnel_xmit(struct sk_buff *skb, struct net_device *dev)
        skb->h.raw = skb->nh.raw;
        skb->nh.raw = skb_push(skb, gre_hlen);
        memset(&(IPCB(skb)->opt), 0, sizeof(IPCB(skb)->opt));
+       IPCB(skb)->flags &= ~(IPSKB_XFRM_TUNNEL_SIZE | IPSKB_XFRM_TRANSFORMED |
+                             IPSKB_REROUTED);
        dst_release(skb->dst);
        skb->dst = &rt->u.dst;
 
@@ -858,11 +860,11 @@ static int ipgre_tunnel_xmit(struct sk_buff *skb, struct net_device *dev)
                        iph->ttl = dst_metric(&rt->u.dst, RTAX_HOPLIMIT);
        }
 
-       ((u16*)(iph+1))[0] = tunnel->parms.o_flags;
-       ((u16*)(iph+1))[1] = skb->protocol;
+       ((__be16*)(iph+1))[0] = tunnel->parms.o_flags;
+       ((__be16*)(iph+1))[1] = skb->protocol;
 
        if (tunnel->parms.o_flags&(GRE_KEY|GRE_CSUM|GRE_SEQ)) {
-               u32 *ptr = (u32*)(((u8*)iph) + tunnel->hlen - 4);
+               __be32 *ptr = (__be32*)(((u8*)iph) + tunnel->hlen - 4);
 
                if (tunnel->parms.o_flags&GRE_SEQ) {
                        ++tunnel->o_seqno;
@@ -875,7 +877,7 @@ static int ipgre_tunnel_xmit(struct sk_buff *skb, struct net_device *dev)
                }
                if (tunnel->parms.o_flags&GRE_CSUM) {
                        *ptr = 0;
-                       *(__u16*)ptr = ip_compute_csum((void*)(iph+1), skb->len - sizeof(struct iphdr));
+                       *(__sum16*)ptr = ip_compute_csum((void*)(iph+1), skb->len - sizeof(struct iphdr));
                }
        }
 
@@ -913,7 +915,7 @@ ipgre_tunnel_ioctl (struct net_device *dev, struct ifreq *ifr, int cmd)
                        t = ipgre_tunnel_locate(&p, 0);
                }
                if (t == NULL)
-                       t = (struct ip_tunnel*)dev->priv;
+                       t = netdev_priv(dev);
                memcpy(&p, &t->parms, sizeof(p));
                if (copy_to_user(ifr->ifr_ifru.ifru_data, &p, sizeof(p)))
                        err = -EFAULT;
@@ -953,7 +955,7 @@ ipgre_tunnel_ioctl (struct net_device *dev, struct ifreq *ifr, int cmd)
                        } else {
                                unsigned nflags=0;
 
-                               t = (struct ip_tunnel*)dev->priv;
+                               t = netdev_priv(dev);
 
                                if (MULTICAST(p.iph.daddr))
                                        nflags = IFF_BROADCAST;
@@ -1002,7 +1004,7 @@ ipgre_tunnel_ioctl (struct net_device *dev, struct ifreq *ifr, int cmd)
                        if ((t = ipgre_tunnel_locate(&p, 0)) == NULL)
                                goto done;
                        err = -EPERM;
-                       if (t == ipgre_fb_tunnel_dev->priv)
+                       if (t == netdev_priv(ipgre_fb_tunnel_dev))
                                goto done;
                        dev = t->dev;
                }
@@ -1019,12 +1021,12 @@ done:
 
 static struct net_device_stats *ipgre_tunnel_get_stats(struct net_device *dev)
 {
-       return &(((struct ip_tunnel*)dev->priv)->stat);
+       return &(((struct ip_tunnel*)netdev_priv(dev))->stat);
 }
 
 static int ipgre_tunnel_change_mtu(struct net_device *dev, int new_mtu)
 {
-       struct ip_tunnel *tunnel = (struct ip_tunnel*)dev->priv;
+       struct ip_tunnel *tunnel = netdev_priv(dev);
        if (new_mtu < 68 || new_mtu > 0xFFF8 - tunnel->hlen)
                return -EINVAL;
        dev->mtu = new_mtu;
@@ -1064,9 +1066,9 @@ static int ipgre_tunnel_change_mtu(struct net_device *dev, int new_mtu)
 static int ipgre_header(struct sk_buff *skb, struct net_device *dev, unsigned short type,
                        void *daddr, void *saddr, unsigned len)
 {
-       struct ip_tunnel *t = (struct ip_tunnel*)dev->priv;
+       struct ip_tunnel *t = netdev_priv(dev);
        struct iphdr *iph = (struct iphdr *)skb_push(skb, t->hlen);
-       u16 *p = (u16*)(iph+1);
+       __be16 *p = (__be16*)(iph+1);
 
        memcpy(iph, &t->parms.iph, sizeof(struct iphdr));
        p[0]            = t->parms.o_flags;
@@ -1091,7 +1093,7 @@ static int ipgre_header(struct sk_buff *skb, struct net_device *dev, unsigned sh
 
 static int ipgre_open(struct net_device *dev)
 {
-       struct ip_tunnel *t = (struct ip_tunnel*)dev->priv;
+       struct ip_tunnel *t = netdev_priv(dev);
 
        if (MULTICAST(t->parms.iph.daddr)) {
                struct flowi fl = { .oif = t->parms.link,
@@ -1105,17 +1107,17 @@ static int ipgre_open(struct net_device *dev)
                        return -EADDRNOTAVAIL;
                dev = rt->u.dst.dev;
                ip_rt_put(rt);
-               if (__in_dev_get(dev) == NULL)
+               if (__in_dev_get_rtnl(dev) == NULL)
                        return -EADDRNOTAVAIL;
                t->mlink = dev->ifindex;
-               ip_mc_inc_group(__in_dev_get(dev), t->parms.iph.daddr);
+               ip_mc_inc_group(__in_dev_get_rtnl(dev), t->parms.iph.daddr);
        }
        return 0;
 }
 
 static int ipgre_close(struct net_device *dev)
 {
-       struct ip_tunnel *t = (struct ip_tunnel*)dev->priv;
+       struct ip_tunnel *t = netdev_priv(dev);
        if (MULTICAST(t->parms.iph.daddr) && t->mlink) {
                struct in_device *in_dev = inetdev_by_index(t->mlink);
                if (in_dev) {
@@ -1140,7 +1142,7 @@ static void ipgre_tunnel_setup(struct net_device *dev)
 
        dev->type               = ARPHRD_IPGRE;
        dev->hard_header_len    = LL_MAX_HEADER + sizeof(struct iphdr) + 4;
-       dev->mtu                = 1500 - sizeof(struct iphdr) - 4;
+       dev->mtu                = ETH_DATA_LEN - sizeof(struct iphdr) - 4;
        dev->flags              = IFF_NOARP;
        dev->iflink             = 0;
        dev->addr_len           = 4;
@@ -1152,10 +1154,10 @@ static int ipgre_tunnel_init(struct net_device *dev)
        struct ip_tunnel *tunnel;
        struct iphdr *iph;
        int hlen = LL_MAX_HEADER;
-       int mtu = 1500;
+       int mtu = ETH_DATA_LEN;
        int addend = sizeof(struct iphdr) + 4;
 
-       tunnel = (struct ip_tunnel*)dev->priv;
+       tunnel = netdev_priv(dev);
        iph = &tunnel->parms.iph;
 
        tunnel->dev = dev;
@@ -1217,9 +1219,9 @@ static int ipgre_tunnel_init(struct net_device *dev)
        return 0;
 }
 
-int __init ipgre_fb_tunnel_init(struct net_device *dev)
+static int __init ipgre_fb_tunnel_init(struct net_device *dev)
 {
-       struct ip_tunnel *tunnel = (struct ip_tunnel*)dev->priv;
+       struct ip_tunnel *tunnel = netdev_priv(dev);
        struct iphdr *iph = &tunnel->parms.iph;
 
        tunnel->dev = dev;
@@ -1277,12 +1279,28 @@ err1:
        goto out;
 }
 
-static void ipgre_fini(void)
+static void __exit ipgre_destroy_tunnels(void)
+{
+       int prio;
+
+       for (prio = 0; prio < 4; prio++) {
+               int h;
+               for (h = 0; h < HASH_SIZE; h++) {
+                       struct ip_tunnel *t;
+                       while ((t = tunnels[prio][h]) != NULL)
+                               unregister_netdevice(t->dev);
+               }
+       }
+}
+
+static void __exit ipgre_fini(void)
 {
        if (inet_del_protocol(&ipgre_protocol, IPPROTO_GRE) < 0)
                printk(KERN_INFO "ipgre close: can't remove protocol\n");
 
-       unregister_netdev(ipgre_fb_tunnel_dev);
+       rtnl_lock();
+       ipgre_destroy_tunnels();
+       rtnl_unlock();
 }
 
 module_init(ipgre_init);