linux/net/batman-adv/fragmentation.c

534 lines
16 KiB
C
Raw Normal View History

/* Copyright (C) 2013-2016 B.A.T.M.A.N. contributors:
*
* Martin Hundebøll <martin@hundeboll.net>
*
* This program is free software; you can redistribute it and/or
* modify it under the terms of version 2 of the GNU General Public
* License as published by the Free Software Foundation.
*
* This program is distributed in the hope that it will be useful, but
* WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
* General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with this program; if not, see <http://www.gnu.org/licenses/>.
*/
#include "fragmentation.h"
2015-04-18 01:40:28 +08:00
#include "main.h"
#include <linux/atomic.h>
#include <linux/byteorder/generic.h>
#include <linux/etherdevice.h>
#include <linux/fs.h>
#include <linux/if_ether.h>
#include <linux/jiffies.h>
#include <linux/kernel.h>
#include <linux/lockdep.h>
2015-04-18 01:40:28 +08:00
#include <linux/netdevice.h>
#include <linux/skbuff.h>
#include <linux/slab.h>
#include <linux/spinlock.h>
#include <linux/string.h>
#include "hard-interface.h"
#include "originator.h"
2015-04-18 01:40:28 +08:00
#include "packet.h"
#include "routing.h"
2015-04-18 01:40:28 +08:00
#include "send.h"
#include "soft-interface.h"
/**
* batadv_frag_clear_chain - delete entries in the fragment buffer chain
* @head: head of chain with entries.
*
* Free fragments in the passed hlist. Should be called with appropriate lock.
*/
static void batadv_frag_clear_chain(struct hlist_head *head)
{
struct batadv_frag_list_entry *entry;
struct hlist_node *node;
hlist_for_each_entry_safe(entry, node, head, list) {
hlist_del(&entry->list);
kfree_skb(entry->skb);
kfree(entry);
}
}
/**
* batadv_frag_purge_orig - free fragments associated to an orig
* @orig_node: originator to free fragments from
* @check_cb: optional function to tell if an entry should be purged
*/
void batadv_frag_purge_orig(struct batadv_orig_node *orig_node,
bool (*check_cb)(struct batadv_frag_table_entry *))
{
struct batadv_frag_table_entry *chain;
u8 i;
for (i = 0; i < BATADV_FRAG_BUFFER_COUNT; i++) {
chain = &orig_node->fragments[i];
spin_lock_bh(&chain->lock);
if (!check_cb || check_cb(chain)) {
batadv_frag_clear_chain(&chain->head);
chain->size = 0;
}
spin_unlock_bh(&chain->lock);
}
}
/**
* batadv_frag_size_limit - maximum possible size of packet to be fragmented
*
* Return: the maximum size of payload that can be fragmented.
*/
static int batadv_frag_size_limit(void)
{
int limit = BATADV_FRAG_MAX_FRAG_SIZE;
limit -= sizeof(struct batadv_frag_packet);
limit *= BATADV_FRAG_MAX_FRAGMENTS;
return limit;
}
/**
* batadv_frag_init_chain - check and prepare fragment chain for new fragment
* @chain: chain in fragments table to init
* @seqno: sequence number of the received fragment
*
* Make chain ready for a fragment with sequence number "seqno". Delete existing
* entries if they have an "old" sequence number.
*
* Caller must hold chain->lock.
*
* Return: true if chain is empty and caller can just insert the new fragment
* without searching for the right position.
*/
static bool batadv_frag_init_chain(struct batadv_frag_table_entry *chain,
u16 seqno)
{
lockdep_assert_held(&chain->lock);
if (chain->seqno == seqno)
return false;
if (!hlist_empty(&chain->head))
batadv_frag_clear_chain(&chain->head);
chain->size = 0;
chain->seqno = seqno;
return true;
}
/**
* batadv_frag_insert_packet - insert a fragment into a fragment chain
* @orig_node: originator that the fragment was received from
* @skb: skb to insert
* @chain_out: list head to attach complete chains of fragments to
*
* Insert a new fragment into the reverse ordered chain in the right table
* entry. The hash table entry is cleared if "old" fragments exist in it.
*
* Return: true if skb is buffered, false on error. If the chain has all the
* fragments needed to merge the packet, the chain is moved to the passed head
* to avoid locking the chain in the table.
*/
static bool batadv_frag_insert_packet(struct batadv_orig_node *orig_node,
struct sk_buff *skb,
struct hlist_head *chain_out)
{
struct batadv_frag_table_entry *chain;
struct batadv_frag_list_entry *frag_entry_new = NULL, *frag_entry_curr;
batman-adv: Fix out-of-order fragmentation support batadv_frag_insert_packet was unable to handle out-of-order packets because it dropped them directly. This is caused by the way the fragmentation lists is checked for the correct place to insert a fragmentation entry. The fragmentation code keeps the fragments in lists. The fragmentation entries are kept in descending order of sequence number. The list is traversed and each entry is compared with the new fragment. If the current entry has a smaller sequence number than the new fragment then the new one has to be inserted before the current entry. This ensures that the list is still in descending order. An out-of-order packet with a smaller sequence number than all entries in the list still has to be added to the end of the list. The used hlist has no information about the last entry in the list inside hlist_head and thus the last entry has to be calculated differently. Currently the code assumes that the iterator variable of hlist_for_each_entry can be used for this purpose after the hlist_for_each_entry finished. This is obviously wrong because the iterator variable is always NULL when the list was completely traversed. Instead the information about the last entry has to be stored in a different variable. This problem was introduced in 610bfc6bc99bc83680d190ebc69359a05fc7f605 ("batman-adv: Receive fragmented packets and merge"). Signed-off-by: Sven Eckelmann <sven@narfation.org> Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch> Signed-off-by: Antonio Quartulli <antonio@meshcoding.com>
2014-05-26 23:21:39 +08:00
struct batadv_frag_list_entry *frag_entry_last = NULL;
struct batadv_frag_packet *frag_packet;
u8 bucket;
u16 seqno, hdr_size = sizeof(struct batadv_frag_packet);
bool ret = false;
/* Linearize packet to avoid linearizing 16 packets in a row when doing
* the later merge. Non-linear merge should be added to remove this
* linearization.
*/
if (skb_linearize(skb) < 0)
goto err;
frag_packet = (struct batadv_frag_packet *)skb->data;
seqno = ntohs(frag_packet->seqno);
bucket = seqno % BATADV_FRAG_BUFFER_COUNT;
frag_entry_new = kmalloc(sizeof(*frag_entry_new), GFP_ATOMIC);
if (!frag_entry_new)
goto err;
frag_entry_new->skb = skb;
frag_entry_new->no = frag_packet->no;
/* Select entry in the "chain table" and delete any prior fragments
* with another sequence number. batadv_frag_init_chain() returns true,
* if the list is empty at return.
*/
chain = &orig_node->fragments[bucket];
spin_lock_bh(&chain->lock);
if (batadv_frag_init_chain(chain, seqno)) {
hlist_add_head(&frag_entry_new->list, &chain->head);
chain->size = skb->len - hdr_size;
chain->timestamp = jiffies;
chain->total_size = ntohs(frag_packet->total_size);
ret = true;
goto out;
}
/* Find the position for the new fragment. */
hlist_for_each_entry(frag_entry_curr, &chain->head, list) {
/* Drop packet if fragment already exists. */
if (frag_entry_curr->no == frag_entry_new->no)
goto err_unlock;
/* Order fragments from highest to lowest. */
if (frag_entry_curr->no < frag_entry_new->no) {
hlist_add_before(&frag_entry_new->list,
&frag_entry_curr->list);
chain->size += skb->len - hdr_size;
chain->timestamp = jiffies;
ret = true;
goto out;
}
batman-adv: Fix out-of-order fragmentation support batadv_frag_insert_packet was unable to handle out-of-order packets because it dropped them directly. This is caused by the way the fragmentation lists is checked for the correct place to insert a fragmentation entry. The fragmentation code keeps the fragments in lists. The fragmentation entries are kept in descending order of sequence number. The list is traversed and each entry is compared with the new fragment. If the current entry has a smaller sequence number than the new fragment then the new one has to be inserted before the current entry. This ensures that the list is still in descending order. An out-of-order packet with a smaller sequence number than all entries in the list still has to be added to the end of the list. The used hlist has no information about the last entry in the list inside hlist_head and thus the last entry has to be calculated differently. Currently the code assumes that the iterator variable of hlist_for_each_entry can be used for this purpose after the hlist_for_each_entry finished. This is obviously wrong because the iterator variable is always NULL when the list was completely traversed. Instead the information about the last entry has to be stored in a different variable. This problem was introduced in 610bfc6bc99bc83680d190ebc69359a05fc7f605 ("batman-adv: Receive fragmented packets and merge"). Signed-off-by: Sven Eckelmann <sven@narfation.org> Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch> Signed-off-by: Antonio Quartulli <antonio@meshcoding.com>
2014-05-26 23:21:39 +08:00
/* store current entry because it could be the last in list */
frag_entry_last = frag_entry_curr;
}
batman-adv: Fix out-of-order fragmentation support batadv_frag_insert_packet was unable to handle out-of-order packets because it dropped them directly. This is caused by the way the fragmentation lists is checked for the correct place to insert a fragmentation entry. The fragmentation code keeps the fragments in lists. The fragmentation entries are kept in descending order of sequence number. The list is traversed and each entry is compared with the new fragment. If the current entry has a smaller sequence number than the new fragment then the new one has to be inserted before the current entry. This ensures that the list is still in descending order. An out-of-order packet with a smaller sequence number than all entries in the list still has to be added to the end of the list. The used hlist has no information about the last entry in the list inside hlist_head and thus the last entry has to be calculated differently. Currently the code assumes that the iterator variable of hlist_for_each_entry can be used for this purpose after the hlist_for_each_entry finished. This is obviously wrong because the iterator variable is always NULL when the list was completely traversed. Instead the information about the last entry has to be stored in a different variable. This problem was introduced in 610bfc6bc99bc83680d190ebc69359a05fc7f605 ("batman-adv: Receive fragmented packets and merge"). Signed-off-by: Sven Eckelmann <sven@narfation.org> Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch> Signed-off-by: Antonio Quartulli <antonio@meshcoding.com>
2014-05-26 23:21:39 +08:00
/* Reached the end of the list, so insert after 'frag_entry_last'. */
if (likely(frag_entry_last)) {
hlist_add_behind(&frag_entry_new->list, &frag_entry_last->list);
chain->size += skb->len - hdr_size;
chain->timestamp = jiffies;
ret = true;
}
out:
if (chain->size > batadv_frag_size_limit() ||
chain->total_size != ntohs(frag_packet->total_size) ||
chain->total_size > batadv_frag_size_limit()) {
/* Clear chain if total size of either the list or the packet
* exceeds the maximum size of one merged packet. Don't allow
* packets to have different total_size.
*/
batadv_frag_clear_chain(&chain->head);
chain->size = 0;
} else if (ntohs(frag_packet->total_size) == chain->size) {
/* All fragments received. Hand over chain to caller. */
hlist_move_list(&chain->head, chain_out);
chain->size = 0;
}
err_unlock:
spin_unlock_bh(&chain->lock);
err:
if (!ret)
kfree(frag_entry_new);
return ret;
}
/**
* batadv_frag_merge_packets - merge a chain of fragments
* @chain: head of chain with fragments
*
* Expand the first skb in the chain and copy the content of the remaining
* skb's into the expanded one. After doing so, clear the chain.
*
* Return: the merged skb or NULL on error.
*/
static struct sk_buff *
batadv_frag_merge_packets(struct hlist_head *chain)
{
struct batadv_frag_packet *packet;
struct batadv_frag_list_entry *entry;
struct sk_buff *skb_out;
int size, hdr_size = sizeof(struct batadv_frag_packet);
/* Remove first entry, as this is the destination for the rest of the
* fragments.
*/
entry = hlist_entry(chain->first, struct batadv_frag_list_entry, list);
hlist_del(&entry->list);
skb_out = entry->skb;
kfree(entry);
packet = (struct batadv_frag_packet *)skb_out->data;
size = ntohs(packet->total_size);
/* Make room for the rest of the fragments. */
batman-adv: Calculate extra tail size based on queued fragments The fragmentation code was replaced in 610bfc6bc99bc83680d190ebc69359a05fc7f605 ("batman-adv: Receive fragmented packets and merge"). The new code provided a mostly unused parameter skb for the merging function. It is used inside the function to calculate the additionally needed skb tailroom. But instead of increasing its own tailroom, it is only increasing the tailroom of the first queued skb. This is not correct in some situations because the first queued entry can be a different one than the parameter. An observed problem was: 1. packet with size 104, total_size 1464, fragno 1 was received - packet is queued 2. packet with size 1400, total_size 1464, fragno 0 was received - packet is queued at the end of the list 3. enough data was received and can be given to the merge function (1464 == (1400 - 20) + (104 - 20)) - merge functions gets 1400 byte large packet as skb argument 4. merge function gets first entry in queue (104 byte) - stored as skb_out 5. merge function calculates the required extra tail as total_size - skb->len - pskb_expand_head tail of skb_out with 64 bytes 6. merge function tries to squeeze the extra 1380 bytes from the second queued skb (1400 byte aka skb parameter) in the 64 extra tail bytes of skb_out Instead calculate the extra required tail bytes for skb_out also using skb_out instead of using the parameter skb. The skb parameter is only used to get the total_size from the last received packet. This is also the total_size used to decide that all fragments were received. Reported-by: Philipp Psurek <philipp.psurek@gmail.com> Signed-off-by: Sven Eckelmann <sven@narfation.org> Acked-by: Martin Hundebøll <martin@hundeboll.net> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-12-20 20:48:55 +08:00
if (pskb_expand_head(skb_out, 0, size - skb_out->len, GFP_ATOMIC) < 0) {
kfree_skb(skb_out);
skb_out = NULL;
goto free;
}
/* Move the existing MAC header to just before the payload. (Override
* the fragment header.)
*/
skb_pull_rcsum(skb_out, hdr_size);
memmove(skb_out->data - ETH_HLEN, skb_mac_header(skb_out), ETH_HLEN);
skb_set_mac_header(skb_out, -ETH_HLEN);
skb_reset_network_header(skb_out);
skb_reset_transport_header(skb_out);
/* Copy the payload of the each fragment into the last skb */
hlist_for_each_entry(entry, chain, list) {
size = entry->skb->len - hdr_size;
memcpy(skb_put(skb_out, size), entry->skb->data + hdr_size,
size);
}
free:
/* Locking is not needed, because 'chain' is not part of any orig. */
batadv_frag_clear_chain(chain);
return skb_out;
}
/**
* batadv_frag_skb_buffer - buffer fragment for later merge
* @skb: skb to buffer
* @orig_node_src: originator that the skb is received from
*
* Add fragment to buffer and merge fragments if possible.
*
* There are three possible outcomes: 1) Packet is merged: Return true and
* set *skb to merged packet; 2) Packet is buffered: Return true and set *skb
* to NULL; 3) Error: Return false and leave skb as is.
*
* Return: true when packet is merged or buffered, false when skb is not not
* used.
*/
bool batadv_frag_skb_buffer(struct sk_buff **skb,
struct batadv_orig_node *orig_node_src)
{
struct sk_buff *skb_out = NULL;
struct hlist_head head = HLIST_HEAD_INIT;
bool ret = false;
/* Add packet to buffer and table entry if merge is possible. */
if (!batadv_frag_insert_packet(orig_node_src, *skb, &head))
goto out_err;
/* Leave if more fragments are needed to merge. */
if (hlist_empty(&head))
goto out;
skb_out = batadv_frag_merge_packets(&head);
if (!skb_out)
goto out_err;
out:
*skb = skb_out;
ret = true;
out_err:
return ret;
}
/**
* batadv_frag_skb_fwd - forward fragments that would exceed MTU when merged
* @skb: skb to forward
* @recv_if: interface that the skb is received on
* @orig_node_src: originator that the skb is received from
*
* Look up the next-hop of the fragments payload and check if the merged packet
* will exceed the MTU towards the next-hop. If so, the fragment is forwarded
* without merging it.
*
* Return: true if the fragment is consumed/forwarded, false otherwise.
*/
bool batadv_frag_skb_fwd(struct sk_buff *skb,
struct batadv_hard_iface *recv_if,
struct batadv_orig_node *orig_node_src)
{
struct batadv_priv *bat_priv = netdev_priv(recv_if->soft_iface);
struct batadv_orig_node *orig_node_dst;
struct batadv_neigh_node *neigh_node = NULL;
struct batadv_frag_packet *packet;
u16 total_size;
bool ret = false;
packet = (struct batadv_frag_packet *)skb->data;
orig_node_dst = batadv_orig_hash_find(bat_priv, packet->dest);
if (!orig_node_dst)
goto out;
neigh_node = batadv_find_router(bat_priv, orig_node_dst, recv_if);
if (!neigh_node)
goto out;
/* Forward the fragment, if the merged packet would be too big to
* be assembled.
*/
total_size = ntohs(packet->total_size);
if (total_size > neigh_node->if_incoming->net_dev->mtu) {
batadv_inc_counter(bat_priv, BATADV_CNT_FRAG_FWD);
batadv_add_counter(bat_priv, BATADV_CNT_FRAG_FWD_BYTES,
skb->len + ETH_HLEN);
packet->ttl--;
batadv_send_unicast_skb(skb, neigh_node);
ret = true;
}
out:
if (orig_node_dst)
batadv_orig_node_put(orig_node_dst);
if (neigh_node)
batadv_neigh_node_put(neigh_node);
return ret;
}
/**
* batadv_frag_create - create a fragment from skb
* @skb: skb to create fragment from
* @frag_head: header to use in new fragment
* @mtu: size of new fragment
*
* Split the passed skb into two fragments: A new one with size matching the
* passed mtu and the old one with the rest. The new skb contains data from the
* tail of the old skb.
*
* Return: the new fragment, NULL on error.
*/
static struct sk_buff *batadv_frag_create(struct sk_buff *skb,
struct batadv_frag_packet *frag_head,
unsigned int mtu)
{
struct sk_buff *skb_fragment;
unsigned int header_size = sizeof(*frag_head);
unsigned int fragment_size = mtu - header_size;
skb_fragment = netdev_alloc_skb(NULL, mtu + ETH_HLEN);
if (!skb_fragment)
goto err;
skb_fragment->priority = skb->priority;
/* Eat the last mtu-bytes of the skb */
skb_reserve(skb_fragment, header_size + ETH_HLEN);
skb_split(skb, skb_fragment, skb->len - fragment_size);
/* Add the header */
skb_push(skb_fragment, header_size);
memcpy(skb_fragment->data, frag_head, header_size);
err:
return skb_fragment;
}
/**
* batadv_frag_send_packet - create up to 16 fragments from the passed skb
* @skb: skb to create fragments from
* @orig_node: final destination of the created fragments
* @neigh_node: next-hop of the created fragments
*
* Return: the netdev tx status or -1 in case of error.
* When -1 is returned the skb is not consumed.
*/
int batadv_frag_send_packet(struct sk_buff *skb,
struct batadv_orig_node *orig_node,
struct batadv_neigh_node *neigh_node)
{
struct batadv_priv *bat_priv;
struct batadv_hard_iface *primary_if = NULL;
struct batadv_frag_packet frag_header;
struct sk_buff *skb_fragment;
unsigned int mtu = neigh_node->if_incoming->net_dev->mtu;
unsigned int header_size = sizeof(frag_header);
unsigned int max_fragment_size, max_packet_size;
int ret = -1;
/* To avoid merge and refragmentation at next-hops we never send
* fragments larger than BATADV_FRAG_MAX_FRAG_SIZE
*/
mtu = min_t(unsigned int, mtu, BATADV_FRAG_MAX_FRAG_SIZE);
max_fragment_size = mtu - header_size;
max_packet_size = max_fragment_size * BATADV_FRAG_MAX_FRAGMENTS;
/* Don't even try to fragment, if we need more than 16 fragments */
if (skb->len > max_packet_size)
goto out;
bat_priv = orig_node->bat_priv;
primary_if = batadv_primary_if_get_selected(bat_priv);
if (!primary_if)
goto out;
/* Create one header to be copied to all fragments */
frag_header.packet_type = BATADV_UNICAST_FRAG;
frag_header.version = BATADV_COMPAT_VERSION;
frag_header.ttl = BATADV_TTL;
frag_header.seqno = htons(atomic_inc_return(&bat_priv->frag_seqno));
frag_header.reserved = 0;
frag_header.no = 0;
frag_header.total_size = htons(skb->len);
/* skb->priority values from 256->263 are magic values to
* directly indicate a specific 802.1d priority. This is used
* to allow 802.1d priority to be passed directly in from VLAN
* tags, etc.
*/
if (skb->priority >= 256 && skb->priority <= 263)
frag_header.priority = skb->priority - 256;
ether_addr_copy(frag_header.orig, primary_if->net_dev->dev_addr);
ether_addr_copy(frag_header.dest, orig_node->orig);
/* Eat and send fragments from the tail of skb */
while (skb->len > max_fragment_size) {
skb_fragment = batadv_frag_create(skb, &frag_header, mtu);
if (!skb_fragment)
goto out;
batadv_inc_counter(bat_priv, BATADV_CNT_FRAG_TX);
batadv_add_counter(bat_priv, BATADV_CNT_FRAG_TX_BYTES,
skb_fragment->len + ETH_HLEN);
ret = batadv_send_unicast_skb(skb_fragment, neigh_node);
if (ret != NET_XMIT_SUCCESS) {
/* return -1 so that the caller can free the original
* skb
*/
ret = -1;
goto out;
}
frag_header.no++;
/* The initial check in this function should cover this case */
if (frag_header.no == BATADV_FRAG_MAX_FRAGMENTS - 1) {
ret = -1;
goto out;
}
}
/* Make room for the fragment header. */
if (batadv_skb_head_push(skb, header_size) < 0 ||
pskb_expand_head(skb, header_size + ETH_HLEN, 0, GFP_ATOMIC) < 0)
goto out;
memcpy(skb->data, &frag_header, header_size);
/* Send the last fragment */
batadv_inc_counter(bat_priv, BATADV_CNT_FRAG_TX);
batadv_add_counter(bat_priv, BATADV_CNT_FRAG_TX_BYTES,
skb->len + ETH_HLEN);
ret = batadv_send_unicast_skb(skb, neigh_node);
out:
if (primary_if)
batadv_hardif_put(primary_if);
return ret;
}