/*
* IPv6 fragment reassembly
- * Linux INET6 implementation
+ * Linux INET6 implementation
*
* Authors:
- * Pedro Roque <roque@di.fc.ul.pt>
+ * Pedro Roque <roque@di.fc.ul.pt>
*
* $Id: reassembly.c,v 1.26 2001/03/07 22:00:57 davem Exp $
*
* 2 of the License, or (at your option) any later version.
*/
-/*
- * Fixes:
+/*
+ * Fixes:
* Andi Kleen Make it work with multiple hosts.
* More RFC compliance.
*
#include <net/snmp.h>
#include <net/ipv6.h>
+#include <net/ip6_route.h>
#include <net/protocol.h>
#include <net/transp_v6.h>
#include <net/rawv6.h>
#include <net/ndisc.h>
#include <net/addrconf.h>
-int sysctl_ip6frag_high_thresh = 256*1024;
-int sysctl_ip6frag_low_thresh = 192*1024;
+int sysctl_ip6frag_high_thresh __read_mostly = 256*1024;
+int sysctl_ip6frag_low_thresh __read_mostly = 192*1024;
-int sysctl_ip6frag_time = IPV6_FRAG_TIMEOUT;
+int sysctl_ip6frag_time __read_mostly = IPV6_FRAG_TIMEOUT;
struct ip6frag_skb_cb
{
struct hlist_node list;
struct list_head lru_list; /* lru list member */
- __u32 id; /* fragment id */
+ __be32 id; /* fragment id */
struct in6_addr saddr;
struct in6_addr daddr;
int len;
int meat;
int iif;
- struct timeval stamp;
+ ktime_t stamp;
unsigned int csum;
__u8 last_in; /* has first/last segment arrived? */
#define COMPLETE 4
* callers should be careful not to use the hash value outside the ipfrag_lock
* as doing so could race with ipfrag_hash_rnd being recalculated.
*/
-static unsigned int ip6qhashfn(u32 id, struct in6_addr *saddr,
+static unsigned int ip6qhashfn(__be32 id, struct in6_addr *saddr,
struct in6_addr *daddr)
{
u32 a, b, c;
- a = saddr->s6_addr32[0];
- b = saddr->s6_addr32[1];
- c = saddr->s6_addr32[2];
+ a = (__force u32)saddr->s6_addr32[0];
+ b = (__force u32)saddr->s6_addr32[1];
+ c = (__force u32)saddr->s6_addr32[2];
a += JHASH_GOLDEN_RATIO;
b += JHASH_GOLDEN_RATIO;
c += ip6_frag_hash_rnd;
__jhash_mix(a, b, c);
- a += saddr->s6_addr32[3];
- b += daddr->s6_addr32[0];
- c += daddr->s6_addr32[1];
+ a += (__force u32)saddr->s6_addr32[3];
+ b += (__force u32)daddr->s6_addr32[0];
+ c += (__force u32)daddr->s6_addr32[1];
__jhash_mix(a, b, c);
- a += daddr->s6_addr32[2];
- b += daddr->s6_addr32[3];
- c += id;
+ a += (__force u32)daddr->s6_addr32[2];
+ b += (__force u32)daddr->s6_addr32[3];
+ c += (__force u32)id;
__jhash_mix(a, b, c);
return c & (IP6Q_HASHSZ - 1);
}
static struct timer_list ip6_frag_secret_timer;
-int sysctl_ip6frag_secret_interval = 10 * 60 * HZ;
+int sysctl_ip6frag_secret_interval __read_mostly = 10 * 60 * HZ;
static void ip6_frag_secret_rebuild(unsigned long dummy)
{
}
}
-static void ip6_evictor(void)
+static void ip6_evictor(struct inet6_dev *idev)
{
struct frag_queue *fq;
struct list_head *tmp;
spin_unlock(&fq->lock);
fq_put(fq, &work);
- IP6_INC_STATS_BH(IPSTATS_MIB_REASMFAILS);
+ IP6_INC_STATS_BH(idev, IPSTATS_MIB_REASMFAILS);
}
}
static void ip6_frag_expire(unsigned long data)
{
struct frag_queue *fq = (struct frag_queue *) data;
- struct net_device *dev;
+ struct net_device *dev = NULL;
spin_lock(&fq->lock);
fq_kill(fq);
- IP6_INC_STATS_BH(IPSTATS_MIB_REASMTIMEOUT);
- IP6_INC_STATS_BH(IPSTATS_MIB_REASMFAILS);
+ dev = dev_get_by_index(fq->iif);
+ if (!dev)
+ goto out;
+
+ rcu_read_lock();
+ IP6_INC_STATS_BH(__in6_dev_get(dev), IPSTATS_MIB_REASMTIMEOUT);
+ IP6_INC_STATS_BH(__in6_dev_get(dev), IPSTATS_MIB_REASMFAILS);
+ rcu_read_unlock();
/* Don't send error if the first segment did not arrive. */
if (!(fq->last_in&FIRST_IN) || !fq->fragments)
goto out;
- dev = dev_get_by_index(fq->iif);
- if (!dev)
- goto out;
-
/*
But use as source device on which LAST ARRIVED
segment was received. And do not use fq->dev
*/
fq->fragments->dev = dev;
icmpv6_send(fq->fragments, ICMPV6_TIME_EXCEED, ICMPV6_EXC_FRAGTIME, 0, dev);
- dev_put(dev);
out:
+ if (dev)
+ dev_put(dev);
spin_unlock(&fq->lock);
fq_put(fq, NULL);
}
hash = ip6qhashfn(fq_in->id, &fq_in->saddr, &fq_in->daddr);
#ifdef CONFIG_SMP
hlist_for_each_entry(fq, n, &ip6_frag_hash[hash], list) {
- if (fq->id == fq_in->id &&
+ if (fq->id == fq_in->id &&
ipv6_addr_equal(&fq_in->saddr, &fq->saddr) &&
ipv6_addr_equal(&fq_in->daddr, &fq->daddr)) {
atomic_inc(&fq->refcnt);
static struct frag_queue *
-ip6_frag_create(u32 id, struct in6_addr *src, struct in6_addr *dst)
+ip6_frag_create(__be32 id, struct in6_addr *src, struct in6_addr *dst,
+ struct inet6_dev *idev)
{
struct frag_queue *fq;
return ip6_frag_intern(fq);
oom:
- IP6_INC_STATS_BH(IPSTATS_MIB_REASMFAILS);
+ IP6_INC_STATS_BH(idev, IPSTATS_MIB_REASMFAILS);
return NULL;
}
static __inline__ struct frag_queue *
-fq_find(u32 id, struct in6_addr *src, struct in6_addr *dst)
+fq_find(__be32 id, struct in6_addr *src, struct in6_addr *dst,
+ struct inet6_dev *idev)
{
struct frag_queue *fq;
struct hlist_node *n;
read_lock(&ip6_frag_lock);
hash = ip6qhashfn(id, src, dst);
hlist_for_each_entry(fq, n, &ip6_frag_hash[hash], list) {
- if (fq->id == id &&
+ if (fq->id == id &&
ipv6_addr_equal(src, &fq->saddr) &&
ipv6_addr_equal(dst, &fq->daddr)) {
atomic_inc(&fq->refcnt);
}
read_unlock(&ip6_frag_lock);
- return ip6_frag_create(id, src, dst);
+ return ip6_frag_create(id, src, dst, idev);
}
-static void ip6_frag_queue(struct frag_queue *fq, struct sk_buff *skb,
+static void ip6_frag_queue(struct frag_queue *fq, struct sk_buff *skb,
struct frag_hdr *fhdr, int nhoff)
{
struct sk_buff *prev, *next;
goto err;
offset = ntohs(fhdr->frag_off) & ~0x7;
- end = offset + (ntohs(skb->nh.ipv6h->payload_len) -
- ((u8 *) (fhdr + 1) - (u8 *) (skb->nh.ipv6h + 1)));
+ end = offset + (ntohs(ipv6_hdr(skb)->payload_len) -
+ ((u8 *)(fhdr + 1) - (u8 *)(ipv6_hdr(skb) + 1)));
if ((unsigned int)end > IPV6_MAXPLEN) {
- IP6_INC_STATS_BH(IPSTATS_MIB_INHDRERRORS);
- icmpv6_param_prob(skb,ICMPV6_HDR_FIELD, (u8*)&fhdr->frag_off - skb->nh.raw);
- return;
+ IP6_INC_STATS_BH(ip6_dst_idev(skb->dst),
+ IPSTATS_MIB_INHDRERRORS);
+ icmpv6_param_prob(skb, ICMPV6_HDR_FIELD,
+ ((u8 *)&fhdr->frag_off -
+ skb_network_header(skb)));
+ return;
}
- if (skb->ip_summed == CHECKSUM_COMPLETE)
- skb->csum = csum_sub(skb->csum,
- csum_partial(skb->nh.raw, (u8*)(fhdr+1)-skb->nh.raw, 0));
+ if (skb->ip_summed == CHECKSUM_COMPLETE) {
+ const unsigned char *nh = skb_network_header(skb);
+ skb->csum = csum_sub(skb->csum,
+ csum_partial(nh, (u8 *)(fhdr + 1) - nh,
+ 0));
+ }
/* Is this the final fragment? */
if (!(fhdr->frag_off & htons(IP6_MF))) {
/* RFC2460 says always send parameter problem in
* this case. -DaveM
*/
- IP6_INC_STATS_BH(IPSTATS_MIB_INHDRERRORS);
- icmpv6_param_prob(skb, ICMPV6_HDR_FIELD,
+ IP6_INC_STATS_BH(ip6_dst_idev(skb->dst),
+ IPSTATS_MIB_INHDRERRORS);
+ icmpv6_param_prob(skb, ICMPV6_HDR_FIELD,
offsetof(struct ipv6hdr, payload_len));
return;
}
/* Point into the IP datagram 'data' part. */
if (!pskb_pull(skb, (u8 *) (fhdr + 1) - skb->data))
goto err;
-
+
if (pskb_trim_rcsum(skb, end - offset))
goto err;
if (skb->dev)
fq->iif = skb->dev->ifindex;
skb->dev = NULL;
- skb_get_timestamp(skb, &fq->stamp);
+ fq->stamp = skb->tstamp;
fq->meat += skb->len;
atomic_add(skb->truesize, &ip6_frag_mem);
return;
err:
- IP6_INC_STATS(IPSTATS_MIB_REASMFAILS);
+ IP6_INC_STATS(ip6_dst_idev(skb->dst), IPSTATS_MIB_REASMFAILS);
kfree_skb(skb);
}
BUG_TRAP(FRAG6_CB(head)->offset == 0);
/* Unfragmented part is taken from the first segment. */
- payload_len = (head->data - head->nh.raw) - sizeof(struct ipv6hdr) + fq->len - sizeof(struct frag_hdr);
+ payload_len = ((head->data - skb_network_header(head)) -
+ sizeof(struct ipv6hdr) + fq->len -
+ sizeof(struct frag_hdr));
if (payload_len > IPV6_MAXPLEN)
goto out_oversize;
/* We have to remove fragment header from datagram and to relocate
* header in order to calculate ICV correctly. */
nhoff = fq->nhoffset;
- head->nh.raw[nhoff] = head->h.raw[0];
- memmove(head->head + sizeof(struct frag_hdr), head->head,
+ skb_network_header(head)[nhoff] = skb_transport_header(head)[0];
+ memmove(head->head + sizeof(struct frag_hdr), head->head,
(head->data - head->head) - sizeof(struct frag_hdr));
- head->mac.raw += sizeof(struct frag_hdr);
- head->nh.raw += sizeof(struct frag_hdr);
+ head->mac_header += sizeof(struct frag_hdr);
+ head->network_header += sizeof(struct frag_hdr);
skb_shinfo(head)->frag_list = head->next;
- head->h.raw = head->data;
- skb_push(head, head->data - head->nh.raw);
+ skb_reset_transport_header(head);
+ skb_push(head, head->data - skb_network_header(head));
atomic_sub(head->truesize, &ip6_frag_mem);
for (fp=head->next; fp; fp = fp->next) {
head->next = NULL;
head->dev = dev;
- skb_set_timestamp(head, &fq->stamp);
- head->nh.ipv6h->payload_len = htons(payload_len);
+ head->tstamp = fq->stamp;
+ ipv6_hdr(head)->payload_len = htons(payload_len);
IP6CB(head)->nhoff = nhoff;
*skb_in = head;
/* Yes, and fold redundant checksum back. 8) */
if (head->ip_summed == CHECKSUM_COMPLETE)
- head->csum = csum_partial(head->nh.raw, head->h.raw-head->nh.raw, head->csum);
+ head->csum = csum_partial(skb_network_header(head),
+ skb_network_header_len(head),
+ head->csum);
- IP6_INC_STATS_BH(IPSTATS_MIB_REASMOKS);
+ rcu_read_lock();
+ IP6_INC_STATS_BH(__in6_dev_get(dev), IPSTATS_MIB_REASMOKS);
+ rcu_read_unlock();
fq->fragments = NULL;
return 1;
if (net_ratelimit())
printk(KERN_DEBUG "ip6_frag_reasm: no memory for reassembly\n");
out_fail:
- IP6_INC_STATS_BH(IPSTATS_MIB_REASMFAILS);
+ rcu_read_lock();
+ IP6_INC_STATS_BH(__in6_dev_get(dev), IPSTATS_MIB_REASMFAILS);
+ rcu_read_unlock();
return -1;
}
static int ipv6_frag_rcv(struct sk_buff **skbp)
{
- struct sk_buff *skb = *skbp;
+ struct sk_buff *skb = *skbp;
struct net_device *dev = skb->dev;
struct frag_hdr *fhdr;
struct frag_queue *fq;
- struct ipv6hdr *hdr;
-
- hdr = skb->nh.ipv6h;
+ struct ipv6hdr *hdr = ipv6_hdr(skb);
- IP6_INC_STATS_BH(IPSTATS_MIB_REASMREQDS);
+ IP6_INC_STATS_BH(ip6_dst_idev(skb->dst), IPSTATS_MIB_REASMREQDS);
/* Jumbo payload inhibits frag. header */
if (hdr->payload_len==0) {
- IP6_INC_STATS(IPSTATS_MIB_INHDRERRORS);
- icmpv6_param_prob(skb, ICMPV6_HDR_FIELD, skb->h.raw-skb->nh.raw);
+ IP6_INC_STATS(ip6_dst_idev(skb->dst), IPSTATS_MIB_INHDRERRORS);
+ icmpv6_param_prob(skb, ICMPV6_HDR_FIELD,
+ skb_network_header_len(skb));
return -1;
}
- if (!pskb_may_pull(skb, (skb->h.raw-skb->data)+sizeof(struct frag_hdr))) {
- IP6_INC_STATS(IPSTATS_MIB_INHDRERRORS);
- icmpv6_param_prob(skb, ICMPV6_HDR_FIELD, skb->h.raw-skb->nh.raw);
+ if (!pskb_may_pull(skb, (skb_transport_offset(skb) +
+ sizeof(struct frag_hdr)))) {
+ IP6_INC_STATS(ip6_dst_idev(skb->dst), IPSTATS_MIB_INHDRERRORS);
+ icmpv6_param_prob(skb, ICMPV6_HDR_FIELD,
+ skb_network_header_len(skb));
return -1;
}
- hdr = skb->nh.ipv6h;
- fhdr = (struct frag_hdr *)skb->h.raw;
+ hdr = ipv6_hdr(skb);
+ fhdr = (struct frag_hdr *)skb_transport_header(skb);
if (!(fhdr->frag_off & htons(0xFFF9))) {
/* It is not a fragmented frame */
- skb->h.raw += sizeof(struct frag_hdr);
- IP6_INC_STATS_BH(IPSTATS_MIB_REASMOKS);
+ skb->transport_header += sizeof(struct frag_hdr);
+ IP6_INC_STATS_BH(ip6_dst_idev(skb->dst), IPSTATS_MIB_REASMOKS);
- IP6CB(skb)->nhoff = (u8*)fhdr - skb->nh.raw;
+ IP6CB(skb)->nhoff = (u8 *)fhdr - skb_network_header(skb);
return 1;
}
if (atomic_read(&ip6_frag_mem) > sysctl_ip6frag_high_thresh)
- ip6_evictor();
+ ip6_evictor(ip6_dst_idev(skb->dst));
- if ((fq = fq_find(fhdr->identification, &hdr->saddr, &hdr->daddr)) != NULL) {
+ if ((fq = fq_find(fhdr->identification, &hdr->saddr, &hdr->daddr,
+ ip6_dst_idev(skb->dst))) != NULL) {
int ret = -1;
spin_lock(&fq->lock);
return ret;
}
- IP6_INC_STATS_BH(IPSTATS_MIB_REASMFAILS);
+ IP6_INC_STATS_BH(ip6_dst_idev(skb->dst), IPSTATS_MIB_REASMFAILS);
kfree_skb(skb);
return -1;
}