We cannot call module_put(THIS_MODULE) if this is our last reference. Otherwise,
this call may cleanup our module before it returns.

Gladly, the kthread API provides a simple wrapper for us. So lets use
module_put_and_exit() to avoid a race condition with the module cleanup code.
Signed-off-by: NDavid Herrmann <dh.herrmann@googlemail.com>
Acked-by: NMarcel Holtmann <marcel@holtmann.org>
Signed-off-by: NGustavo F. Padovan <padovan@profusion.mobi>

We cannot call module_put(THIS_MODULE) if this is our last reference. Otherwise,
this call may cleanup our module before it returns.

Gladly, the kthread API provides a simple wrapper for us. So lets use
module_put_and_exit() to avoid a race condition with the module cleanup code.
Signed-off-by: NDavid Herrmann <dh.herrmann@googlemail.com>
Acked-by: NMarcel Holtmann <marcel@holtmann.org>
Signed-off-by: NGustavo F. Padovan <padovan@profusion.mobi>

Timers set by __set_chan_timer() should use miliseconds instead of
jiffies. Commit 942ecc9c updated
l2cap_set_timer() so it expects timeout to be specified in msecs
instead of jiffies. This makes timeouts unreliable when CONFIG_HZ
is not set to 1000.
Signed-off-by: NAndrzej Kaczmarek <andrzej.kaczmarek@tieto.com>
Signed-off-by: NGustavo F. Padovan <padovan@profusion.mobi>

This reverts commit 33060542.
The commit introduces regression when two 2.1 devices attempt
establish rfcomm channel. Such connection is refused since there's
a security block issue on l2cap. It means the link is unencrypted.

2011-09-16 18:08:46.567616 < ACL data: handle 1 flags 0x00 dlen 24
    0000: 14 00 40 00 06 00 02 00  0f 35 03 19 12 00 ff ff
..@......5....˙˙
    0010: 35 05 0a 00 00 ff ff 00                           5....˙˙.
2011-09-16 18:08:46.572377 > HCI Event: Number of Completed Packets
(0x13) plen 5
    handle 1 packets 1
2011-09-16 18:08:46.577931 > ACL data: handle 1 flags 0x02 dlen 88
    L2CAP(d): cid 0x0040 len 84 [psm 0]
      0000: 07 00 02 00 4f 00 4c 35  4a 35 48 09 00 00 0a 00
....O.L5J5H.....
      0010: 01 00 00 09 00 01 35 03  19 12 00 09 00 05 35 03
......5.......5.
      0020: 19 10 02 09 00 09 35 08  35 06 19 12 00 09 01 02
......5.5.......
      0030: 09 02 00 09 01 02 09 02  01 09 00 0a 09 02 02 09
................
      0040: 00 00 09 02 03 09 00 00  09 02 04 28 01 09 02 05
...........(....
      0050: 09 00 02 00                                       ....
2011-09-16 18:08:46.626057 < HCI Command: Authentication Requested
(0x01|0x0011) plen 2
    handle 1
2011-09-16 18:08:46.627614 > HCI Event: Command Status (0x0f) plen 4
    Authentication Requested (0x01|0x0011) status 0x00 ncmd 1
2011-09-16 18:08:46.627675 > HCI Event: Link Key Request (0x17) plen 6
    bdaddr 00:00:F2:6A:29:69
2011-09-16 18:08:46.634999 < HCI Command: Link Key Request Reply
(0x01|0x000b) plen 22
    bdaddr 00:00:F2:6A:29:69 key 58CD393179FC902E5E8F512A855EE532
2011-09-16 18:08:46.683278 > HCI Event: Command Complete (0x0e) plen 10
    Link Key Request Reply (0x01|0x000b) ncmd 1
    status 0x00 bdaddr 00:00:F2:6A:29:69
2011-09-16 18:08:46.764729 > HCI Event: Auth Complete (0x06) plen 3
    status 0x00 handle 1
2011-09-16 18:08:46.764821 < ACL data: handle 1 flags 0x00 dlen 12
    0000: 08 00 01 00 02 05 04 00  03 00 41 00              ..........A.
2011-09-16 18:08:46.764851 > HCI Event: Command Status (0x0f) plen 4
    Unknown (0x00|0x0000) status 0x00 ncmd 2
2011-09-16 18:08:46.768117 > HCI Event: Number of Completed Packets
(0x13) plen 5
    handle 1 packets 1
2011-09-16 18:08:46.770894 > ACL data: handle 1 flags 0x02 dlen 16
    L2CAP(s): Connect rsp: dcid 0x0000 scid 0x0041 result 3 status 0
      Connection refused - security block
2011-09-16 18:08:49.000691 < ACL data: handle 1 flags 0x00 dlen 12
    0000: 08 00 01 00 06 06 04 00  40 00 40 00              ........@.@.
2011-09-16 18:08:49.015675 > HCI Event: Number of Completed Packets
(0x13) plen 5
    handle 1 packets 1
2011-09-16 18:08:49.016927 > ACL data: handle 1 flags 0x02 dlen 12
    L2CAP(s): Disconn rsp: dcid 0x0040 scid 0x0040
2011-09-16 18:08:51.009480 < HCI Command: Disconnect (0x01|0x0006) plen
3
    handle 1 reason 0x13
    Reason: Remote User Terminated Connection
2011-09-16 18:08:51.011525 > HCI Event: Command Status (0x0f) plen 4
    Disconnect (0x01|0x0006) status 0x00 ncmd 1
2011-09-16 18:08:51.123494 > HCI Event: Disconn Complete (0x05) plen 4
    status 0x00 handle 1 reason 0x16
    Reason: Connection Terminated by Local Host
Signed-off-by: NArek Lichwa <arkadiusz.lichwa@tieto.com>
Signed-off-by: NGustavo F. Padovan <padovan@profusion.mobi>

Only AID values 1-2007 are valid, but some APs have been
found to send random bogus values, in the reported case an
AP that was sending the AID field value 0xffff, an AID of
0x3fff (16383).

There isn't much we can do but disable powersave since
there's no way it can work properly in this case.

Cc: stable@vger.kernel.org
Reported-by: NBill C Riemers <briemers@redhat.com>
Signed-off-by: NJohannes Berg <johannes.berg@intel.com>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

We need to verify whether the command is successful before allocating
the station entry to avoid extra processing. This also fixes a memory
leak on the error path.
Signed-off-by: NJouni Malinen <j@w1.fi>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

When going back on-channel, we should reconfigure
the hw iff the hardware is not already configured
to the operational channel.
Signed-off-by: NEliad Peller <eliad@wizery.com>
Cc: stable@kernel.org # 2.6.39+
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

The offchannel code is currently broken - we should
remain_off_channel if the work was started, and
the work's channel and channel_type are the same
as local->tmp_channel and local->tmp_channel_type.

However, if wk->chan_type and local->tmp_channel_type
coexist (e.g. have the same channel type), we won't
remain_off_channel.

This behavior was introduced by commit da2fd1f0
("mac80211: Allow work items to use existing
channel type.")
Tested-by: NBen Greear <greearb@candelatech.com>
Signed-off-by: NEliad Peller <eliad@wizery.com>
Cc: stable@kernel.org # 2.6.39+
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

The HCI_MGMT flag should only be set when user space requests the full
controller information. This way we avoid potential issues with setting
change events ariving before the actual read_controller_info command
finishes.
Signed-off-by: NJohan Hedberg <johan.hedberg@intel.com>
Signed-off-by: NGustavo F. Padovan <padovan@profusion.mobi>

I've noticed that my CSR usb dongle was not working if it was plugged in when
PC was booting. It looks like I get two HCI reset command complete events (see
hcidump logs below).
The root cause is reset called from off_timer. Timeout for this reset to
complete is set to 250ms and my bt dongle requires more time for replying with
command complete event. After that, chip seems to reply with reset command
complete event for next non-reset command.

Attached patch increase mentioned timeout to HCI_INIT_TIMEOUT, this value is
already used for timeouting hci_reset_req in hci_dev_reset().

This might also be related to BT not working after suspend that was reported
here some time ago.

Hcidump log:

2011-09-12 23:13:27.379465 < HCI Command: Reset (0x03|0x0003) plen 0
2011-09-12 23:13:27.380797 > HCI Event: Command Complete (0x0e) plen 4
    Reset (0x03|0x0003) ncmd 1
    status 0x00
2011-09-12 23:13:27.380859 < HCI Command: Read Local Supported Features (0x04|0x000
3) plen 0
2011-09-12 23:13:27.760789 > HCI Event: Command Complete (0x0e) plen 4
    Reset (0x03|0x0003) ncmd 1
    status 0x00
2011-09-12 23:13:27.760831 < HCI Command: Read Local Version Information (0x04|0x00
01) plen 0
2011-09-12 23:13:27.764780 > HCI Event: Command Complete (0x0e) plen 12
    Read Local Version Information (0x04|0x0001) ncmd 1
    status 0x00
    HCI Version: 1.1 (0x1) HCI Revision: 0x36f
    LMP Version: 1.1 (0x1) LMP Subversion: 0x36f
    Manufacturer: Cambridge Silicon Radio (10)
Signed-off-by: NSzymon Janc <szymon@janc.net.pl>
Signed-off-by: NGustavo F. Padovan <padovan@profusion.mobi>

commit 2425717b (net: allow vlan traffic to be received under bond)
broke ARP processing on vlan on top of bonding.

       +-------+
eth0 --| bond0 |---bond0.103
eth1 --|       |
       +-------+

52870.115435: skb_gro_reset_offset <-napi_gro_receive
52870.115435: dev_gro_receive <-napi_gro_receive
52870.115435: napi_skb_finish <-napi_gro_receive
52870.115435: netif_receive_skb <-napi_skb_finish
52870.115435: get_rps_cpu <-netif_receive_skb
52870.115435: __netif_receive_skb <-netif_receive_skb
52870.115436: vlan_do_receive <-__netif_receive_skb
52870.115436: bond_handle_frame <-__netif_receive_skb
52870.115436: vlan_do_receive <-__netif_receive_skb
52870.115436: arp_rcv <-__netif_receive_skb
52870.115436: kfree_skb <-arp_rcv

Packet is dropped in arp_rcv() because its pkt_type was set to
PACKET_OTHERHOST in the first vlan_do_receive() call, since no eth0.103
exists.

We really need to change pkt_type only if no more rx_handler is about to
be called for the packet.
Signed-off-by: NEric Dumazet <eric.dumazet@gmail.com>
Reviewed-by: NJiri Pirko <jpirko@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The route lookup to find a previously auto-configured route for a prefixes used
to use rt6_lookup(), with the prefix from the RA used as an address. However,
that kind of lookup ignores routing tables, the prefix length and route flags,
so when there were other matching routes, even in different tables and/or with
a different prefix length, the wrong route would be manipulated.

Now, a new function "addrconf_get_prefix_route()" is used for the route lookup,
which searches in RT6_TABLE_PREFIX and takes the prefix-length and route flags
into account.
Signed-off-by: NAndreas Hofmeister <andi@collax.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

in func icmp6_dst_alloc,dst_metric_set call ipv6_cow_metrics to set metric.
ipv6_cow_metrics may will call rt6_bind_peer to set rt6_info->rt6i_peer.
So,we should move ipv6_addr_copy before dst_metric_set to make sure rt6_bind_peer success.
Signed-off-by: NGao feng <gaofeng@cn.fujitsu.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

We should return errcode from sock_alloc_send_skb()
Signed-off-by: NZheng Yan <zheng.z.yan@intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

commit 66b13d99 (ipv4: tcp: fix TOS value in ACK messages sent from
TIME_WAIT) fixed IPv4 only.

This part is for the IPv6 side, adding a tclass param to ip6_xmit()

We alias tw_tclass and tw_tos, if socket family is INET6.

[ if sockets is ipv4-mapped, only IP_TOS socket option is used to fill
TOS field, TCLASS is not taken into account ]
Signed-off-by: NEric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The caif code will register its own pernet_operations, and then register
a netdevice_notifier. Each time the netdevice_notifier is triggered,
it'll do some stuff... including a lookup of its own pernet stuff with
net_generic().

If the net_generic() call ever returns NULL, the caif code will BUG().
That doesn't seem *so* unreasonable, I suppose — it does seem like it
should never happen.

However, it *does* happen. When we clone a network namespace,
setup_net() runs through all the pernet_operations one at a time. It
gets to loopback before it gets to caif. And loopback_net_init()
registers a netdevice... while caif hasn't been initialised. So the caif
netdevice notifier triggers, and immediately goes BUG().

We could imagine a complex and overengineered solution to this generic
class of problems, but this patch takes the simple approach. It just
makes caif_device_notify() *not* go looking for its pernet data
structures if the device it's being notified about isn't a caif device
in the first place.

Cc: stable@kernel.org
Signed-off-by: NDavid Woodhouse <David.Woodhouse@intel.com>
Acked-by: NSjur Brændeland <sjur.brandeland@stericsson.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Signed-off-by: NThomas Gleixner <tglx@linutronix.de>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Function tt_response_fill_table() actually uses a tt_local_entry pointer to
iterate either over the local or the global table entries (it depends on the
what hash table is passed as argument). To iterate over such entries the
hlist_for_each_entry_rcu() macro has to access their "hash_entry" field which
MUST be at the same position in both the tt_global/local_entry structures.
Reported-by: NSimon Wunderlich <siwu@hrz.tu-chemnitz.de>
Signed-off-by: NAntonio Quartulli <ordex@autistici.org>
Signed-off-by: NMarek Lindner <lindner_marek@yahoo.de>

After removing the batman-adv module, the hash may be already gone
when tt_global_del_orig() tries to clean the hash. This patch adds
a sanity check to avoid this.
Signed-off-by: NSimon Wunderlich <siwu@hrz.tu-chemnitz.de>
Tested-by: NAlexey Fisher <bug-track@fisher-privat.net>
Signed-off-by: NMarek Lindner <lindner_marek@yahoo.de>

struct tt_global_entry holds a reference to an orig_node which must be
decremented before deallocating the structure.
Signed-off-by: NSimon Wunderlich <siwu@hrz.tu-chemnitz.de>
Tested-by: NAlexey Fisher <bug-track@fisher-privat.net>
Signed-off-by: NMarek Lindner <lindner_marek@yahoo.de>

When hybrid mode is enabled (accept_ra == 2), the kernel also sees RAs
generated locally. This is useful since it allows the kernel to auto-configure
its own interface addresses.

However, if 'accept_ra_defrtr' and/or 'accept_ra_rtr_pref' are set and the
locally generated RAs announce the default route and/or other route information,
the kernel happily inserts bogus routes with its own address as gateway.

With this patch, adding routes from an RA will be skiped when the RAs source
address matches any local address, just as if 'accept_ra_defrtr' and
'accept_ra_rtr_pref' were set to 0.
Signed-off-by: NAndreas Hofmeister <andi@collax.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Dan Siemon would like to add tunnelling support to cls_flow

This preliminary patch introduces use of skb_header_pointer() to help
this task, while avoiding skb head reallocation because of deep packet
inspection.
Signed-off-by: NEric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

In func ipv4_dst_check,check_peer_pmtu should be called only when peer is updated.
So,if the peer is not updated in ip_rt_frag_needed,we can not inc __rt_peer_genid.
Signed-off-by: NGao feng <gaofeng@cn.fujitsu.com>
Acked-by: NEric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

It was enabled by default and the messages guarded
by the define are useful.
Signed-off-by: NFlavio Leitner <fbl@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This helps in more control over debugging.
root@qemu-img-64:~# ls /pass/123
ls: cannot access /pass/123: No such file or directory
root@qemu-img-64:~# cat /sys/kernel/debug/tracing/trace
# tracer: nop
#
#           TASK-PID    CPU#    TIMESTAMP  FUNCTION
#              | |       |          |         |
              ls-1536  [001]    70.928584: 9p_protocol_dump: clnt 18446612132784021504 P9_TWALK(tag = 1)
000: 16 00 00 00 6e 01 00 01 00 00 00 02 00 00 00 01
010: 00 03 00 31 32 33 00 00 00 ff ff ff ff 00 00 00

              ls-1536  [001]    70.928587: <stack trace>
 => trace_9p_protocol_dump
 => p9pdu_finalize
 => p9_client_rpc
 => p9_client_walk
 => v9fs_vfs_lookup
 => d_alloc_and_lookup
 => walk_component
 => path_lookupat
              ls-1536  [000]    70.929696: 9p_protocol_dump: clnt 18446612132784021504 P9_RLERROR(tag = 1)
000: 0b 00 00 00 07 01 00 02 00 00 00 4e 03 00 02 00
010: 00 00 00 00 03 00 02 00 00 00 00 00 ff 43 00 00

              ls-1536  [000]    70.929697: <stack trace>
 => trace_9p_protocol_dump
 => p9_client_rpc
 => p9_client_walk
 => v9fs_vfs_lookup
 => d_alloc_and_lookup
 => walk_component
 => path_lookupat
 => do_path_lookup
Signed-off-by: NAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: NEric Van Hensbergen <ericvh@gmail.com>

Without this msize=4294967295 will result in a crash
Signed-off-by: NDan Carpenter <error27@gmail.com>
Signed-off-by: NAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: NEric Van Hensbergen <ericvh@gmail.com>

Instead of saying all integer argument option should be listed in the beginning
move integer parsing to each option type.
Signed-off-by: NAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: NEric Van Hensbergen <ericvh@gmail.com>

We dereferenced "req->tc" and "req->rc" before checking for NULL.
Signed-off-by: NDan Carpenter <error27@gmail.com>
Signed-off-by: NAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: NEric Van Hensbergen <ericvh@gmail.com>

* remove lot of update to different data structure
* add a seperate callback for zero copy request.
* above makes non zero copy code path simpler
* remove conditionalizing TREAD/TREADDIR/TWRITE in the zero copy path
* Fix the dotu p9_check_errors with zero copy. Add sufficient doc around
* Add support for both in and output buffers in zero copy callback
* pin and unpin pages in the same context
* use helpers instead of defining page offset and rest of page ourself
* Fix mem leak in p9_check_errors
* Remove 'E' and 'F' in p9pdu_vwritef
Signed-off-by: NAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: NEric Van Hensbergen <ericvh@gmail.com>

There is a long standing bug in linux tcp stack, about ACK messages sent
on behalf of TIME_WAIT sockets.

In the IP header of the ACK message, we choose to reflect TOS field of
incoming message, and this might break some setups.

Example of things that were broken :
  - Routing using TOS as a selector
  - Firewalls
  - Trafic classification / shaping

We now remember in timewait structure the inet tos field and use it in
ACK generation, and route lookup.

Notes :
 - We still reflect incoming TOS in RST messages.
 - We could extend MuraliRaja Muniraju patch to report TOS value in
netlink messages for TIME_WAIT sockets.
 - A patch is needed for IPv6
Signed-off-by: NEric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Renato Westphal noticed that since commit a2835763
"rtnetlink: handle rtnl_link netlink notifications manually" was merged
we no longer send a netlink message when a networking device is moved
from one network namespace to another.

Fix this by adding the missing manual notification in dev_change_net_namespaces.

Since all network devices that are processed by dev_change_net_namspaces are
in the initialized state the complicated tests that guard the manual
rtmsg_ifinfo calls in rollback_registered and register_netdevice are
unnecessary and we can just perform a plain notification.

Cc: stable@kernel.org
Tested-by: NRenato Westphal <renatowestphal@gmail.com>
Signed-off-by: NEric W. Biederman <ebiederm@xmission.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

There is bug in commit 5e2b61f7(ipv4: Remove flowi from struct rtable).
It makes xfrm4_fill_dst() modify wrong data structure.
Signed-off-by: NZheng Yan <zheng.z.yan@intel.com>
Reported-by: NKim Phillips <kim.phillips@freescale.com>
Acked-by: NEric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The commit f39925db
(ipv4: Cache learned redirect information in inetpeer.)
removed some ICMP packet validations which are required by
RFC 1122, section 3.2.2.2:
...
  A Redirect message SHOULD be silently discarded if the new
  gateway address it specifies is not on the same connected
  (sub-) net through which the Redirect arrived [INTRO:2,
  Appendix A], or if the source of the Redirect is not the
  current first-hop gateway for the specified destination (see
  Section 3.3.1).
Signed-off-by: NFlavio Leitner <fbl@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The pair of functions,

 * skb_clone_tx_timestamp()
 * skb_complete_tx_timestamp()

were designed to allow timestamping in PHY devices. The first
function, called during the MAC driver's hard_xmit method, identifies
PTP protocol packets, clones them, and gives them to the PHY device
driver. The PHY driver may hold onto the packet and deliver it at a
later time using the second function, which adds the packet to the
socket's error queue.

As pointed out by Johannes, nothing prevents the socket from
disappearing while the cloned packet is sitting in the PHY driver
awaiting a timestamp. This patch fixes the issue by taking a reference
on the socket for each such packet. In addition, the comments
regarding the usage of these function are expanded to highlight the
rule that PHY drivers must use skb_complete_tx_timestamp() to release
the packet, in order to release the socket reference, too.

These functions first appeared in v2.6.36.
Reported-by: NJohannes Berg <johannes@sipsolutions.net>
Signed-off-by: NRichard Cochran <richard.cochran@omicron.at>
Cc: <stable@vger.kernel.org>
Signed-off-by: NEric Dumazet <eric.dumazet@gmail.com>
Reviewed-by: NJohannes Berg <johannes@sipsolutions.net>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Now tcp_md5_hash_header() has a const tcphdr argument, we can add more
const attributes to callers.
Signed-off-by: NEric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

tcp_md5_hash_header() writes into skb header a temporary zero value,
this might confuse other users of this area.

Since tcphdr is small (20 bytes), copy it in a temporary variable and
make the change in the copy.
Signed-off-by: NEric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Signed-off-by: NMaciej Żenczykowski <maze@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Adding const qualifiers to pointers can ease code review, and spot some
bugs. It might allow compiler to optimize code further.

For example, is it legal to temporary write a null cksum into tcphdr
in tcp_md5_hash_header() ? I am afraid a sniffer could catch the
temporary null value...
Signed-off-by: NEric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Instead of using the dev->next chain and trying to resync at each call to
dev_seq_start, use the name hash, keeping the bucket and the offset in
seq->private field.

Tests revealed the following results for ifconfig > /dev/null
	* 1000 interfaces:
		* 0.114s without patch
		* 0.089s with patch
	* 3000 interfaces:
		* 0.489s without patch
		* 0.110s with patch
	* 5000 interfaces:
		* 1.363s without patch
		* 0.250s with patch
	* 128000 interfaces (other setup):
		* ~100s without patch
		* ~30s with patch
Signed-off-by: NMihai Maruseac <mmaruseac@ixiacom.com>
Signed-off-by: NEric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

I've split this bit out of the skb frag destructor patch since it helps enforce
the use of the fragment API.
Signed-off-by: NIan Campbell <ian.campbell@citrix.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>