For connected sockets, the first run of dev_pick_tx saves the
calculated txq in sk_tx_queue_mapping. This is not saved if
either the device has a queue select or the socket is not
connected. Next iterations of dev_pick_tx uses the cached value
of sk_tx_queue_mapping.
Signed-off-by: NKrishna Kumar <krkumar2@in.ibm.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Signed-off-by: NEric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Now that software UFO is supported, UFO can be enabled on master
devices like bridge, bond even though the attached device doesn't
support this feature in hardware.

This allows UFO to be used between KVM host and guest even when a
physical interface attached to the bridge doesn't support UFO.
Signed-off-by: NSridhar Samudrala <sri@us.ibm.com>
Acked-by: NHerbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

For various purposes including a wireless extensions
bugfix, we need to hook into the netdev creation before
before netdev_register_kobject(). This will also ease
doing the dev type assignment that Marcel was working
on for cfg80211 drivers w/o touching them all.
Signed-off-by: NJohannes Berg <johannes@sipsolutions.net>
Signed-off-by: NMarcel Holtmann <marcel@holtmann.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Since commit 9b22ea56
( net: fix packet socket delivery in rx irq handler )

We lost rx timestamping of packets received on accelerated vlans.

Effect is that tcpdump on real dev can show strange timings, since it gets rx timestamps
too late (ie at skb dequeueing time, not at skb queueing time)

14:47:26.986871 IP 192.168.20.110 > 192.168.20.141: icmp 64: echo request seq 1
14:47:26.986786 IP 192.168.20.141 > 192.168.20.110: icmp 64: echo reply seq 1

14:47:27.986888 IP 192.168.20.110 > 192.168.20.141: icmp 64: echo request seq 2
14:47:27.986781 IP 192.168.20.141 > 192.168.20.110: icmp 64: echo reply seq 2

14:47:28.986896 IP 192.168.20.110 > 192.168.20.141: icmp 64: echo request seq 3
14:47:28.986780 IP 192.168.20.141 > 192.168.20.110: icmp 64: echo reply seq 3
Signed-off-by: NEric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This patch fixes commit e36b9d16. The approach
there is to call dev_close()/dev_open() whenever the device type is changed in
order to remap the device IP multicast addresses to HW multicast addresses.
This approach suffers from 2 drawbacks:

*. It assumes tha the device is UP when calling dev_close(), or otherwise
   dev_close() has no affect. It is worth to mention that initscripts (Redhat)
   and sysconfig (Suse) doesn't act the same in this matter. 
*. dev_close() has other side affects, like deleting entries from the routing
   table, which might be unnecessary.

The fix here is to directly remap the IP multicast addresses to HW multicast
addresses for a bonding device that changes its type, and nothing else.
Reported-by: NJason Gunthorpe <jgunthorpe@obsidianresearch.com>
Signed-off-by: NMoni Shoua <monis@voltaire.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The only valid usage for the bridge frame hooks are by a
GPL components (such as the bridge module).
The kernel should not leave a crack in the door for proprietary
networking stacks to slip in.
Signed-off-by: NStephen Hemminger <shemminger@vyatta.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Remove a debugging aid I accidently left in previous 'cleanup' patch
Signed-off-by: NEric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Pure style cleanup patch before surgery :)
Signed-off-by: NEric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Patch compiled and 32 simultaneous netperf testing ran fine.
Signed-off-by: NKrishna Kumar <krkumar2@in.ibm.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

There are not maste devices in mac802154 anymore, so drop
ARPHRD_IEEE802154_PHY definition.
Signed-off-by: NDmitry Eremin-Solenikov <dbaryshkov@gmail.com>

The networking code checks CAP_SYS_MODULE before using request_module() to
try to load a kernel module.  While this seems reasonable it's actually
weakening system security since we have to allow CAP_SYS_MODULE for things
like /sbin/ip and bluetoothd which need to be able to trigger module loads.
CAP_SYS_MODULE actually grants those binaries the ability to directly load
any code into the kernel.  We should instead be protecting modprobe and the
modules on disk, rather than granting random programs the ability to load code
directly into the kernel.  Instead we are going to gate those networking checks
on CAP_NET_ADMIN which still limits them to root but which does not grant
those processes the ability to load arbitrary code into the kernel.
Signed-off-by: NEric Paris <eparis@redhat.com>
Acked-by: NSerge Hallyn <serue@us.ibm.com>
Acked-by: NPaul Moore <paul.moore@hp.com>
Acked-by: NDavid S. Miller <davem@davemloft.net>
Signed-off-by: NJames Morris <jmorris@namei.org>

dev_queue_xmit enqueue's a skb and calls qdisc_run which
dequeue's the skb and xmits it. In most cases, the skb that
is enqueue'd is the same one that is dequeue'd (unless the
queue gets stopped or multiple cpu's write to the same queue
and ends in a race with qdisc_run). For default qdiscs, we
can remove the redundant enqueue/dequeue and simply xmit the
skb since the default qdisc is work-conserving.

The patch uses a new flag - TCQ_F_CAN_BYPASS to identify the
default fast queue. The controversial part of the patch is
incrementing qlen when a skb is requeued - this is to avoid
checks like the second line below:

+  } else if ((q->flags & TCQ_F_CAN_BYPASS) && !qdisc_qlen(q) &&
>>         !q->gso_skb &&
+          !test_and_set_bit(__QDISC_STATE_RUNNING, &q->state)) {

Results of a 2 hour testing for multiple netperf sessions (1,
2, 4, 8, 12 sessions on a 4 cpu system-X). The BW numbers are
aggregate Mb/s across iterations tested with this version on
System-X boxes with Chelsio 10gbps cards:

----------------------------------
Size |  ORG BW          NEW BW   |
----------------------------------
128K |  156964          159381   |
256K |  158650          162042   |
----------------------------------

Changes from ver1:

1. Move sch_direct_xmit declaration from sch_generic.h to
   pkt_sched.h
2. Update qdisc basic statistics for direct xmit path.
3. Set qlen to zero in qdisc_reset.
4. Changed some function names to more meaningful ones.
Signed-off-by: NKrishna Kumar <krkumar2@in.ibm.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

String literals are constant, and usually, we can also tag the array
of pointers const too, moving it to the .rodata section.
Signed-off-by: NJan Engelhardt <jengelh@medozas.de>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

-tip testing found this lockdep warning:

[    2.272010] calling  net_dev_init+0x0/0x164 @ 1
[    2.276033] device class 'net': registering
[    2.280191] INFO: trying to register non-static key.
[    2.284005] the code is fine but needs lockdep annotation.
[    2.284005] turning off the locking correctness validator.
[    2.284005] Pid: 1, comm: swapper Not tainted 2.6.31-rc5-tip #1145
[    2.284005] Call Trace:
[    2.284005]  [<7958eb4e>] ? printk+0xf/0x11
[    2.284005]  [<7904f83c>] __lock_acquire+0x11b/0x622
[    2.284005]  [<7908c9b7>] ? alloc_debug_processing+0xf9/0x144
[    2.284005]  [<7904e2be>] ? mark_held_locks+0x3a/0x52
[    2.284005]  [<7908dbc4>] ? kmem_cache_alloc+0xa8/0x13f
[    2.284005]  [<7904e475>] ? trace_hardirqs_on_caller+0xa2/0xc3
[    2.284005]  [<7904fdf6>] lock_acquire+0xb3/0xd0
[    2.284005]  [<79489678>] ? alloc_netdev_mq+0xf5/0x1ad
[    2.284005]  [<79591514>] _spin_lock_bh+0x2d/0x5d
[    2.284005]  [<79489678>] ? alloc_netdev_mq+0xf5/0x1ad
[    2.284005]  [<79489678>] alloc_netdev_mq+0xf5/0x1ad
[    2.284005]  [<793a38f2>] ? loopback_setup+0x0/0x74
[    2.284005]  [<798eecd0>] loopback_net_init+0x20/0x5d
[    2.284005]  [<79483efb>] register_pernet_device+0x23/0x4b
[    2.284005]  [<798f5c9f>] net_dev_init+0x115/0x164
[    2.284005]  [<7900104f>] do_one_initcall+0x4a/0x11a
[    2.284005]  [<798f5b8a>] ? net_dev_init+0x0/0x164
[    2.284005]  [<79066f6d>] ? register_irq_proc+0x8c/0xa8
[    2.284005]  [<798cc29a>] do_basic_setup+0x42/0x52
[    2.284005]  [<798cc30a>] kernel_init+0x60/0xa1
[    2.284005]  [<798cc2aa>] ? kernel_init+0x0/0xa1
[    2.284005]  [<79003e03>] kernel_thread_helper+0x7/0x10
[    2.284078] device: 'lo': device_add
[    2.288248] initcall net_dev_init+0x0/0x164 returned 0 after 11718 usecs
[    2.292010] calling  neigh_init+0x0/0x66 @ 1
[    2.296010] initcall neigh_init+0x0/0x66 returned 0 after 0 usecs

it's using an zero-initialized spinlock. This is a side-effect of:

        dev_unicast_init(dev);

in alloc_netdev_mq() making use of dev->addr_list_lock.

The device has just been allocated freshly, it's not accessible
anywhere yet so no locking is needed at all - in fact it's wrong
to lock it here (the lock isnt initialized yet).

This bug was introduced via:

| commit a6ac65db
| Date:   Thu Jul 30 01:06:12 2009 +0000
|
|     net: restore the original spinlock to protect unicast list
Signed-off-by: NIngo Molnar <mingo@elte.hu>
Acked-by: NJiri Pirko <jpirko@redhat.com>
Tested-by: NMark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

There is a path when an assetion in dev_unicast_sync() appears.

igmp6_group_added -> dev_mc_add -> __dev_set_rx_mode ->
-> vlan_dev_set_rx_mode -> dev_unicast_sync

Therefore we cannot protect this list with rtnl. This patch restores the
original protecting this list with spinlock.
Signed-off-by: NJiri Pirko <jpirko@redhat.com>
Tested-by: NMeelis Roos <mroos@linux.ee>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

In order to make cfg80211/nl80211 aware of network namespaces,
we have to do the following things:

 * del_virtual_intf method takes an interface index rather
   than a netdev pointer - simply change this

 * nl80211 uses init_net a lot, it changes to use the sender's
   network namespace

 * scan requests use the interface index, hold a netdev pointer
   and reference instead

 * we want a wiphy and its associated virtual interfaces to be
   in one netns together, so
    - we need to be able to change ns for a given interface, so
      export dev_change_net_namespace()
    - for each virtual interface set the NETIF_F_NETNS_LOCAL
      flag, and clear that flag only when the wiphy changes ns,
      to disallow breaking this invariant

 * when a network namespace goes away, we need to reparent the
   wiphy to init_net

 * cfg80211 users that support creating virtual interfaces must
   create them in the wiphy's namespace, currently this affects
   only mac80211

The end result is that you can now switch an entire wiphy into
a different network namespace with the new command
	iw phy#<idx> set netns <pid>
and all virtual interfaces will follow (or the operation fails).
Signed-off-by: NJohannes Berg <johannes@sipsolutions.net>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

For mac80211, with the master netdev removal, we need to be
able to sync a multicast address list onto another list that
is not tracked within a netdev, so we need access to the
functions doing that.
Signed-off-by: NJohannes Berg <johannes@sipsolutions.net>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

This patch converts the remaining occurences of raw return values to their
symbolic counterparts in ndo_start_xmit() functions that were missed by the
previous automatic conversion.

Additionally code that assumed the symbolic value of NETDEV_TX_OK to be zero
is changed to explicitly use NETDEV_TX_OK.
Signed-off-by: NPatrick McHardy <kaber@trash.net>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

When NAPI is disabled while we're in net_rx_action, we end up
calling __napi_complete without flushing GRO packets.  This is
a bug as it would cause the GRO packets to linger, of course it
also literally BUGs to catch error like this :)

This patch changes it to napi_complete, with the obligatory IRQ
reenabling.  This should be safe because we've only just disabled
IRQs and it does not materially affect the test conditions in
between.
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

In order to get the tun driver to account packets, we need to be
able to receive packets with destructors set.  To be on the safe
side, I added an skb_orphan call for all protocols by default since
some of them (IP in particular) cannot handle receiving packets
destructors properly.

Now it seems that at least one protocol (CAN) expects to be able
to pass skb->sk through the rx path without getting clobbered.

So this patch attempts to fix this properly by moving the skb_orphan
call to where it's actually needed.  In particular, I've added it
to skb_set_owner_[rw] which is what most users of skb->destructor
call.

This is actually an improvement for tun too since it means that
we only give back the amount charged to the socket when the skb
is passed to another socket that will also be charged accordingly.
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>
Tested-by: NOliver Hartkopp <olver@hartkopp.net>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This patch is inspired by patch recently posted by Johannes Berg. Basically what
my patch does is to group list and a count of addresses into newly introduced
structure netdev_hw_addr_list. This brings us two benefits:
1) struct net_device becames a bit nicer.
2) in the future there will be a possibility to operate with lists independently
   on netdevices (with exporting right functions).
I wanted to introduce this patch before I'll post a multicast lists conversion.
Signed-off-by: NJiri Pirko <jpirko@redhat.com>

 drivers/net/bnx2.c              |    4 +-
 drivers/net/e1000/e1000_main.c  |    4 +-
 drivers/net/ixgbe/ixgbe_main.c  |    6 +-
 drivers/net/mv643xx_eth.c       |    2 +-
 drivers/net/niu.c               |    4 +-
 drivers/net/virtio_net.c        |   10 ++--
 drivers/s390/net/qeth_l2_main.c |    2 +-
 include/linux/netdevice.h       |   17 +++--
 net/core/dev.c                  |  130 ++++++++++++++++++--------------------
 9 files changed, 89 insertions(+), 90 deletions(-)
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This patch changes FDB entry check for ATM LANE bridge integration.
There's no point in holding a FDB entry around SKB building.

br_fdb_get()/br_fdb_put() pair are changed into single br_fdb_test_addr()
hook that checks if the addr has FDB entry pointing to other port
to the one the request arrived on.

FDB entry refcounting is removed as it's not used anywhere else.
Signed-off-by: NMichał Mirosław <mirq-linux@rere.qmqm.pl>
Acked-by: NStephen Hemminger <shemminger@vyatta.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Commit b00055aa " [NET] core: add
RFC2863 operstate" defined new interface flag values.  Its
documentation specified that these flags could be accessed from user
space via SIOCGIFFLAGS.  However, this does not work because the new
flags do not fit in that ioctl's argument width.

Change the documentation to match the code's behavior.  Also change
the source to explicitly show the truncation.  This _should_ have no
effect on executable code, and did not with gcc 4.2.4 generating x86
code.

A new ioctl could be defined to return all interface flags to user
space.  However, since this has been broken for three years with no
one complaining, there doesn't seem much need.  They are still
accessible via netlink.
Reported-by: N"Fredrik Arnerup" <fredrik.arnerup@edgeware.tv>
Signed-off-by: NJohn Dykstra <john.dykstra1@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

IEEE 802.15.4 stack requires several constants to be defined/adjusted.
Signed-off-by: NDmitry Eremin-Solenikov <dbaryshkov@gmail.com>
Signed-off-by: NSergey Lapin <slapin@ossfans.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

commit f001fde5
(net: introduce a list of device addresses dev_addr_list (v6))
added one regression Vegard Nossum found in its testings.

With kmemcheck help, Vegard found some uninitialized memory
was read and reported to user, potentialy leaking kernel data.
( thread can be found on http://lkml.org/lkml/2009/5/30/177 )

dev_addr_init() incorrectly uses sizeof() operator. We were
initializing one byte instead of MAX_ADDR_LEN bytes.
Reported-by: NVegard Nossum <vegard.nossum@gmail.com>
Signed-off-by: NEric Dumazet <eric.dumazet@gmail.com>
Acked-by: NJiri Pirko <jpirko@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Signed-off-by: NDavid S. Miller <davem@davemloft.net>

NETDEV_UP is called after the device is set UP, but sometimes
it is useful to be able to veto the device UP. Introduce a
new NETDEV_PRE_UP notifier that can be used for exactly this.
The first use case will be cfg80211 denying interfaces to be
set UP if the device is known to be rfkill'ed.
Signed-off-by: NJohannes Berg <johannes@sipsolutions.net>
Acked-by: NDavid S. Miller <davem@davemloft.net>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

Define three accessors to get/set dst attached to a skb

struct dst_entry *skb_dst(const struct sk_buff *skb)

void skb_dst_set(struct sk_buff *skb, struct dst_entry *dst)

void skb_dst_drop(struct sk_buff *skb)
This one should replace occurrences of :
dst_release(skb->dst)
skb->dst = NULL;

Delete skb->dst field
Signed-off-by: NEric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This patch converts unicast address list to standard list_head using
previously introduced struct netdev_hw_addr. It also relaxes the
locking. Original spinlock (still used for multicast addresses) is not
needed and is no longer used for a protection of this list. All
reading and writing takes place under rtnl (with no changes).

I also removed a possibility to specify the length of the address
while adding or deleting unicast address. It's always dev->addr_len.

The convertion touched especially e1000 and ixgbe codes when the
change is not so trivial.
Signed-off-by: NJiri Pirko <jpirko@redhat.com>

 drivers/net/bnx2.c               |   13 +--
 drivers/net/e1000/e1000_main.c   |   24 +++--
 drivers/net/ixgbe/ixgbe_common.c |   14 ++--
 drivers/net/ixgbe/ixgbe_common.h |    4 +-
 drivers/net/ixgbe/ixgbe_main.c   |    6 +-
 drivers/net/ixgbe/ixgbe_type.h   |    4 +-
 drivers/net/macvlan.c            |   11 +-
 drivers/net/mv643xx_eth.c        |   11 +-
 drivers/net/niu.c                |    7 +-
 drivers/net/virtio_net.c         |    7 +-
 drivers/s390/net/qeth_l2_main.c  |    6 +-
 drivers/scsi/fcoe/fcoe.c         |   16 ++--
 include/linux/netdevice.h        |   18 ++--
 net/8021q/vlan.c                 |    4 +-
 net/8021q/vlan_dev.c             |   10 +-
 net/core/dev.c                   |  195 +++++++++++++++++++++++++++-----------
 net/dsa/slave.c                  |   10 +-
 net/packet/af_packet.c           |    4 +-
 18 files changed, 227 insertions(+), 137 deletions(-)
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Use ALIGN() and PTR_ALIGN() macros instead of handcoding them.

Get rid of NETDEV_ALIGN_CONST ugly define
Signed-off-by: NEric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

As we know the only packets which need the final pskb_may_pull
are completely non-linear, and have all the required bits in
frag0, we can perform a straight memcpy instead of going through
pskb_may_pull and doing skb_copy_bits.
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

For the overwhelming majority of cases, skb_gro_header's return
value cannot be NULL.  Yet we must check it because of its current
form.  This patch splits it up into multiple functions in order
to avoid this.
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

By caching frag0_len, we can avoid checking both frag0 and the
length separately in skb_gro_header.  This helps as skb_gro_header
is called four times per packet which amounts to a few million
times at 10Gb/s.
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Currently skb_gro_header is used for packets which put the hardware
header in skb->data with the rest in frags.  Since the drivers that
need this optimisation all provide completely non-linear packets,
we can gain extra optimisations by only performing the frag0
optimisation for completely non-linear packets.

In particular, we can simply test frag0 (instead of skb_headlen)
to see whether the optimisation is in force.
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The function skb_gro_header is called four times per packet which
quickly adds up at 10Gb/s.  This patch inlines it to allow better
optimisations.

Some architectures perform multiplication for page_address, which
is done by each skb_gro_header invocation.  This patch caches that
value in skb->cb to avoid the unnecessary multiplications.
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

We would like to get rid of netdev->trans_start = jiffies; that about all net
drivers have to use in their start_xmit() function, and use txq->trans_start
instead.

This can be done generically in core network, as suggested by David.

Some devices, (particularly loopback) dont need trans_start update, because
they dont have transmit watchdog. We could add a new device flag, or rely
on fact that txq->tran_start can be updated is txq->xmit_lock_owner is
different than -1. Use a helper function to hide our choice.
Signed-off-by: NEric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

All drivers are already converted to new net_device_ops API
and nobody uses old API anymore.
Signed-off-by: NAlexander Beregalov <a.beregalov@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Patch to add the ability to detect drops in hardware interfaces via dropwatch.
Adds a tracepoint to net_rx_action to signal everytime a napi instance is
polled.  The dropmon code then periodically checks to see if the rx_frames
counter has changed, and if so, adds a drop notification to the netlink
protocol, using the reserved all-0's vector to indicate the drop location was in
hardware, rather than somewhere in the code.
Signed-off-by: NNeil Horman <nhorman@tuxdriver.com>

 include/linux/net_dropmon.h |    8 ++
 include/trace/napi.h        |   11 +++
 net/core/dev.c              |    5 +
 net/core/drop_monitor.c     |  124 ++++++++++++++++++++++++++++++++++++++++++--
 net/core/net-traces.c       |    4 +
 net/core/netpoll.c          |    2
 6 files changed, 149 insertions(+), 5 deletions(-)
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

One point of contention in high network loads is the dst_release() performed
when a transmited skb is freed. This is because NIC tx completion calls
dev_kree_skb() long after original call to dev_queue_xmit(skb).

CPU cache is cold and the atomic op in dst_release() stalls. On SMP, this is
quite visible if one CPU is 100% handling softirqs for a network device,
since dst_clone() is done by other cpus, involving cache line ping pongs.

It seems right place to release dst is in dev_hard_start_xmit(), for most
devices but ones that are virtual, and some exceptions.

David Miller suggested to define a new device flag, set in alloc_netdev_mq()
(so that most devices set it at init time), and carefuly unset in devices
which dont want a NULL skb->dst in their ndo_start_xmit().

List of devices that must clear this flag is :

- loopback device, because it calls netif_rx() and quoting Patrick :
    "ip_route_input() doesn't accept loopback addresses, so loopback packets
     already need to have a dst_entry attached."
- appletalk/ipddp.c : needs skb->dst in its xmit function

- And all devices that call again dev_queue_xmit() from their xmit function
(as some classifiers need skb->dst) : bonding, vlan, macvlan, eql, ifb, hdlc_fr
Signed-off-by: NEric Dumazet <dada1@cosmosbay.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>