/*
* This is a module which is used for queueing packets and communicating with
- * userspace via nfetlink.
+ * userspace via nfnetlink.
*
* (C) 2005 by Harald Welte <laforge@netfilter.org>
+ * (C) 2007 by Patrick McHardy <kaber@trash.net>
*
* Based on the old ipv4-only ip_queue.c:
* (C) 2000-2002 James Morris <jmorris@intercode.com.au>
static DEFINE_SPINLOCK(instances_lock);
#define INSTANCE_BUCKETS 16
-static struct hlist_head instance_table[INSTANCE_BUCKETS];
+static struct hlist_head instance_table[INSTANCE_BUCKETS] __read_mostly;
static inline u_int8_t instance_hashfn(u_int16_t queue_num)
{
static struct nfqnl_instance *
instance_create(u_int16_t queue_num, int pid)
{
- struct nfqnl_instance *inst = NULL;
+ struct nfqnl_instance *inst;
unsigned int h;
+ int err;
spin_lock(&instances_lock);
- if (instance_lookup(queue_num))
+ if (instance_lookup(queue_num)) {
+ err = -EEXIST;
goto out_unlock;
+ }
inst = kzalloc(sizeof(*inst), GFP_ATOMIC);
- if (!inst)
+ if (!inst) {
+ err = -ENOMEM;
goto out_unlock;
+ }
inst->queue_num = queue_num;
inst->peer_pid = pid;
inst->copy_mode = NFQNL_COPY_NONE;
spin_lock_init(&inst->lock);
INIT_LIST_HEAD(&inst->queue_list);
- INIT_RCU_HEAD(&inst->rcu);
- if (!try_module_get(THIS_MODULE))
+ if (!try_module_get(THIS_MODULE)) {
+ err = -EAGAIN;
goto out_free;
+ }
h = instance_hashfn(queue_num);
hlist_add_head_rcu(&inst->hlist, &instance_table[h]);
kfree(inst);
out_unlock:
spin_unlock(&instances_lock);
- return NULL;
+ return ERR_PTR(err);
}
static void nfqnl_flush(struct nfqnl_instance *queue, nfqnl_cmpfn cmpfn,
static struct sk_buff *
nfqnl_build_packet_message(struct nfqnl_instance *queue,
- struct nf_queue_entry *entry, int *errp)
+ struct nf_queue_entry *entry)
{
sk_buff_data_t old_tail;
size_t size;
struct sk_buff *entskb = entry->skb;
struct net_device *indev;
struct net_device *outdev;
- __be32 tmp_uint;
- size = NLMSG_ALIGN(sizeof(struct nfgenmsg))
+ size = NLMSG_SPACE(sizeof(struct nfgenmsg))
+ nla_total_size(sizeof(struct nfqnl_msg_packet_hdr))
+ nla_total_size(sizeof(u_int32_t)) /* ifindex */
+ nla_total_size(sizeof(u_int32_t)) /* ifindex */
spin_lock_bh(&queue->lock);
- switch (queue->copy_mode) {
+ switch ((enum nfqnl_config_mode)queue->copy_mode) {
case NFQNL_COPY_META:
case NFQNL_COPY_NONE:
- data_len = 0;
break;
case NFQNL_COPY_PACKET:
if ((entskb->ip_summed == CHECKSUM_PARTIAL ||
entskb->ip_summed == CHECKSUM_COMPLETE) &&
- (*errp = skb_checksum_help(entskb))) {
+ skb_checksum_help(entskb)) {
spin_unlock_bh(&queue->lock);
return NULL;
}
size += nla_total_size(data_len);
break;
-
- default:
- *errp = -EINVAL;
- spin_unlock_bh(&queue->lock);
- return NULL;
}
entry->id = queue->id_sequence++;
indev = entry->indev;
if (indev) {
- tmp_uint = htonl(indev->ifindex);
#ifndef CONFIG_BRIDGE_NETFILTER
- NLA_PUT(skb, NFQA_IFINDEX_INDEV, sizeof(tmp_uint), &tmp_uint);
+ NLA_PUT_BE32(skb, NFQA_IFINDEX_INDEV, htonl(indev->ifindex));
#else
if (entry->pf == PF_BRIDGE) {
/* Case 1: indev is physical input device, we need to
* look for bridge group (when called from
* netfilter_bridge) */
- NLA_PUT(skb, NFQA_IFINDEX_PHYSINDEV, sizeof(tmp_uint),
- &tmp_uint);
+ NLA_PUT_BE32(skb, NFQA_IFINDEX_PHYSINDEV,
+ htonl(indev->ifindex));
/* this is the bridge group "brX" */
- tmp_uint = htonl(indev->br_port->br->dev->ifindex);
- NLA_PUT(skb, NFQA_IFINDEX_INDEV, sizeof(tmp_uint),
- &tmp_uint);
+ NLA_PUT_BE32(skb, NFQA_IFINDEX_INDEV,
+ htonl(indev->br_port->br->dev->ifindex));
} else {
/* Case 2: indev is bridge group, we need to look for
* physical device (when called from ipv4) */
- NLA_PUT(skb, NFQA_IFINDEX_INDEV, sizeof(tmp_uint),
- &tmp_uint);
- if (entskb->nf_bridge
- && entskb->nf_bridge->physindev) {
- tmp_uint = htonl(entskb->nf_bridge->physindev->ifindex);
- NLA_PUT(skb, NFQA_IFINDEX_PHYSINDEV,
- sizeof(tmp_uint), &tmp_uint);
- }
+ NLA_PUT_BE32(skb, NFQA_IFINDEX_INDEV,
+ htonl(indev->ifindex));
+ if (entskb->nf_bridge && entskb->nf_bridge->physindev)
+ NLA_PUT_BE32(skb, NFQA_IFINDEX_PHYSINDEV,
+ htonl(entskb->nf_bridge->physindev->ifindex));
}
#endif
}
if (outdev) {
- tmp_uint = htonl(outdev->ifindex);
#ifndef CONFIG_BRIDGE_NETFILTER
- NLA_PUT(skb, NFQA_IFINDEX_OUTDEV, sizeof(tmp_uint), &tmp_uint);
+ NLA_PUT_BE32(skb, NFQA_IFINDEX_OUTDEV, htonl(outdev->ifindex));
#else
if (entry->pf == PF_BRIDGE) {
/* Case 1: outdev is physical output device, we need to
* look for bridge group (when called from
* netfilter_bridge) */
- NLA_PUT(skb, NFQA_IFINDEX_PHYSOUTDEV, sizeof(tmp_uint),
- &tmp_uint);
+ NLA_PUT_BE32(skb, NFQA_IFINDEX_PHYSOUTDEV,
+ htonl(outdev->ifindex));
/* this is the bridge group "brX" */
- tmp_uint = htonl(outdev->br_port->br->dev->ifindex);
- NLA_PUT(skb, NFQA_IFINDEX_OUTDEV, sizeof(tmp_uint),
- &tmp_uint);
+ NLA_PUT_BE32(skb, NFQA_IFINDEX_OUTDEV,
+ htonl(outdev->br_port->br->dev->ifindex));
} else {
/* Case 2: outdev is bridge group, we need to look for
* physical output device (when called from ipv4) */
- NLA_PUT(skb, NFQA_IFINDEX_OUTDEV, sizeof(tmp_uint),
- &tmp_uint);
- if (entskb->nf_bridge
- && entskb->nf_bridge->physoutdev) {
- tmp_uint = htonl(entskb->nf_bridge->physoutdev->ifindex);
- NLA_PUT(skb, NFQA_IFINDEX_PHYSOUTDEV,
- sizeof(tmp_uint), &tmp_uint);
- }
+ NLA_PUT_BE32(skb, NFQA_IFINDEX_OUTDEV,
+ htonl(outdev->ifindex));
+ if (entskb->nf_bridge && entskb->nf_bridge->physoutdev)
+ NLA_PUT_BE32(skb, NFQA_IFINDEX_PHYSOUTDEV,
+ htonl(entskb->nf_bridge->physoutdev->ifindex));
}
#endif
}
- if (entskb->mark) {
- tmp_uint = htonl(entskb->mark);
- NLA_PUT(skb, NFQA_MARK, sizeof(u_int32_t), &tmp_uint);
- }
+ if (entskb->mark)
+ NLA_PUT_BE32(skb, NFQA_MARK, htonl(entskb->mark));
if (indev && entskb->dev) {
struct nfqnl_msg_packet_hw phw;
if (data_len) {
struct nlattr *nla;
- int size = nla_attr_size(data_len);
+ int sz = nla_attr_size(data_len);
if (skb_tailroom(skb) < nla_total_size(data_len)) {
printk(KERN_WARNING "nf_queue: no tailroom!\n");
nla = (struct nlattr *)skb_put(skb, nla_total_size(data_len));
nla->nla_type = NFQA_PAYLOAD;
- nla->nla_len = size;
+ nla->nla_len = sz;
if (skb_copy_bits(entskb, 0, nla_data(nla), data_len))
BUG();
nla_put_failure:
if (skb)
kfree_skb(skb);
- *errp = -EINVAL;
if (net_ratelimit())
printk(KERN_ERR "nf_queue: error creating packet message\n");
return NULL;
static int
nfqnl_enqueue_packet(struct nf_queue_entry *entry, unsigned int queuenum)
{
- int status = -EINVAL;
struct sk_buff *nskb;
struct nfqnl_instance *queue;
+ int err;
/* rcu_read_lock()ed by nf_hook_slow() */
queue = instance_lookup(queuenum);
if (!queue)
- return -EINVAL;
+ goto err_out;
if (queue->copy_mode == NFQNL_COPY_NONE)
- return -EAGAIN;
+ goto err_out;
- nskb = nfqnl_build_packet_message(queue, entry, &status);
+ nskb = nfqnl_build_packet_message(queue, entry);
if (nskb == NULL)
- return status;
+ goto err_out;
spin_lock_bh(&queue->lock);
if (queue->queue_total >= queue->queue_maxlen) {
queue->queue_dropped++;
- status = -ENOSPC;
if (net_ratelimit())
printk(KERN_WARNING "nf_queue: full at %d entries, "
- "dropping packets(s). Dropped: %d\n",
- queue->queue_total, queue->queue_dropped);
+ "dropping packets(s).\n",
+ queue->queue_total);
goto err_out_free_nskb;
}
/* nfnetlink_unicast will either free the nskb or add it to a socket */
- status = nfnetlink_unicast(nskb, queue->peer_pid, MSG_DONTWAIT);
- if (status < 0) {
+ err = nfnetlink_unicast(nskb, &init_net, queue->peer_pid, MSG_DONTWAIT);
+ if (err < 0) {
queue->queue_user_dropped++;
goto err_out_unlock;
}
__enqueue_entry(queue, entry);
spin_unlock_bh(&queue->lock);
- return status;
+ return 0;
err_out_free_nskb:
kfree_skb(nskb);
-
err_out_unlock:
spin_unlock_bh(&queue->lock);
- return status;
+err_out:
+ return -1;
}
static int
nfqnl_mangle(void *data, int data_len, struct nf_queue_entry *e)
{
+ struct sk_buff *nskb;
int diff;
- int err;
diff = data_len - e->skb->len;
if (diff < 0) {
if (data_len > 0xFFFF)
return -EINVAL;
if (diff > skb_tailroom(e->skb)) {
- err = pskb_expand_head(e->skb, 0,
- diff - skb_tailroom(e->skb),
- GFP_ATOMIC);
- if (err) {
+ nskb = skb_copy_expand(e->skb, skb_headroom(e->skb),
+ diff, GFP_ATOMIC);
+ if (!nskb) {
printk(KERN_WARNING "nf_queue: OOM "
"in mangle, dropping packet\n");
- return err;
+ return -ENOMEM;
}
+ kfree_skb(e->skb);
+ e->skb = nskb;
}
skb_put(e->skb, diff);
}
{
struct net_device *dev = ptr;
- if (dev->nd_net != &init_net)
+ if (!net_eq(dev_net(dev), &init_net))
return NOTIFY_DONE;
/* Drop any packets associated with the downed device */
{
struct netlink_notify *n = ptr;
- if (event == NETLINK_URELEASE &&
- n->protocol == NETLINK_NETFILTER && n->pid) {
+ if (event == NETLINK_URELEASE && n->protocol == NETLINK_NETFILTER) {
int i;
/* destroy all instances for this pid */
static int
nfqnl_recv_verdict(struct sock *ctnl, struct sk_buff *skb,
- struct nlmsghdr *nlh, struct nlattr *nfqa[])
+ const struct nlmsghdr *nlh,
+ const struct nlattr * const nfqa[])
{
struct nfgenmsg *nfmsg = NLMSG_DATA(nlh);
u_int16_t queue_num = ntohs(nfmsg->res_id);
}
if (nfqa[NFQA_MARK])
- entry->skb->mark = ntohl(*(__be32 *)
- nla_data(nfqa[NFQA_MARK]));
+ entry->skb->mark = ntohl(nla_get_be32(nfqa[NFQA_MARK]));
nf_reinject(entry, verdict);
return 0;
static int
nfqnl_recv_unsupp(struct sock *ctnl, struct sk_buff *skb,
- struct nlmsghdr *nlh, struct nlattr *nfqa[])
+ const struct nlmsghdr *nlh,
+ const struct nlattr * const nfqa[])
{
return -ENOTSUPP;
}
static int
nfqnl_recv_config(struct sock *ctnl, struct sk_buff *skb,
- struct nlmsghdr *nlh, struct nlattr *nfqa[])
+ const struct nlmsghdr *nlh,
+ const struct nlattr * const nfqa[])
{
struct nfgenmsg *nfmsg = NLMSG_DATA(nlh);
u_int16_t queue_num = ntohs(nfmsg->res_id);
/* Commands without queue context - might sleep */
switch (cmd->command) {
case NFQNL_CFG_CMD_PF_BIND:
- ret = nf_register_queue_handler(ntohs(cmd->pf),
- &nfqh);
- break;
+ return nf_register_queue_handler(ntohs(cmd->pf),
+ &nfqh);
case NFQNL_CFG_CMD_PF_UNBIND:
- ret = nf_unregister_queue_handler(ntohs(cmd->pf),
- &nfqh);
- break;
- default:
- break;
+ return nf_unregister_queue_handler(ntohs(cmd->pf),
+ &nfqh);
}
-
- if (ret < 0)
- return ret;
}
rcu_read_lock();
goto err_out_unlock;
}
queue = instance_create(queue_num, NETLINK_CB(skb).pid);
- if (!queue) {
- ret = -EINVAL;
+ if (IS_ERR(queue)) {
+ ret = PTR_ERR(queue);
goto err_out_unlock;
}
break;
case NFQNL_CFG_CMD_PF_UNBIND:
break;
default:
- ret = -EINVAL;
+ ret = -ENOTSUPP;
break;
}
}
}
static void *seq_start(struct seq_file *seq, loff_t *pos)
+ __acquires(instances_lock)
{
spin_lock(&instances_lock);
return get_idx(seq, *pos);
}
static void seq_stop(struct seq_file *s, void *v)
+ __releases(instances_lock)
{
spin_unlock(&instances_lock);
}
static int __init nfnetlink_queue_init(void)
{
int i, status = -ENOMEM;
-#ifdef CONFIG_PROC_FS
- struct proc_dir_entry *proc_nfqueue;
-#endif
for (i = 0; i < INSTANCE_BUCKETS; i++)
INIT_HLIST_HEAD(&instance_table[i]);
}
#ifdef CONFIG_PROC_FS
- proc_nfqueue = create_proc_entry("nfnetlink_queue", 0440,
- proc_net_netfilter);
- if (!proc_nfqueue)
+ if (!proc_create("nfnetlink_queue", 0440,
+ proc_net_netfilter, &nfqnl_file_ops))
goto cleanup_subsys;
- proc_nfqueue->proc_fops = &nfqnl_file_ops;
#endif
register_netdevice_notifier(&nfqnl_dev_notifier);
#endif
nfnetlink_subsys_unregister(&nfqnl_subsys);
netlink_unregister_notifier(&nfqnl_rtnl_notifier);
+
+ rcu_barrier(); /* Wait for completion of call_rcu()'s */
}
MODULE_DESCRIPTION("netfilter packet queue handler");
git://ftp.safe.ca / safe / jmp / linux-2.6 / blobdiff