Idea is from Patrick McHardy.

Instead of managing the list of qdiscs on the device level, manage it
in the root qdisc of a netdev_queue.  This solves all kinds of
visibility issues during qdisc destruction.

The way to iterate over all qdiscs of a netdev_queue is to visit
the netdev_queue->qdisc, and then traverse it's list.

The only special case is to ignore builting qdiscs at the root when
dumping or doing a qdisc_lookup().  That was not needed previously
because builtin qdiscs were not added to the device's qdisc_list.
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

We can simply use the qdisc->q.lock for all of the
qdisc tree synchronization.
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

And give it it's own lock.
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

When we have shared qdiscs, packets come out of the qdiscs
for multiple transmit queues.

Therefore it doesn't make any sense to schedule the transmit
queue when logically we cannot know ahead of time the TX
queue of the SKB that the qdisc->dequeue() will give us.

Just for sanity I added a BUG check to make sure we never
get into a state where the noop_qdisc is scheduled.
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

It just xor hashes over IPv4/IPv6 addresses and ports of transport.

The only assumption it makes is that skb_network_header() is set
correctly.

With bug fixes from Eric Dumazet.
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Devices or device layers can set this to control the queue selection
performed by dev_pick_tx().

This function runs under RCU protection, which allows overriding
functions to have some way of synchronizing with things like dynamic
->real_num_tx_queues adjustments.

This makes the spinlock prefetch in dev_queue_xmit() a little bit
less effective, but that's the price right now for correctness.
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This effectively "flips the switch" by making the core networking
and multiqueue-aware drivers use the new TX multiqueue structures.

Non-multiqueue drivers need no changes.  The interfaces they use such
as netif_stop_queue() degenerate into an operation on TX queue zero.
So everything "just works" for them.

Code that really wants to do "X" to all TX queues now invokes a
routine that does so, such as netif_tx_wake_all_queues(),
netif_tx_stop_all_queues(), etc.

pktgen and netpoll required a little bit more surgery than the others.

In particular the pktgen changes, whilst functional, could be largely
improved.  The initial check in pktgen_xmit() will sometimes check the
wrong queue, which is mostly harmless.  The thing to do is probably to
invoke fill_packet() earlier.

The bulk of the netpoll changes is to make the code operate solely on
the TX queue indicated by by the SKB queue mapping.

Setting of the SKB queue mapping is entirely confined inside of
net/core/dev.c:dev_pick_tx().  If we end up needing any kind of
special semantics (drops, for example) it will be implemented here.

Finally, we now have a "real_num_tx_queues" which is where the driver
indicates how many TX queues are actually active.

With IGB changes from Jeff Kirsher.
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

alloc_netdev_mq() now allocates an array of netdev_queue
structures for TX, based upon the queue_count argument.

Furthermore, all accesses to the TX queues are now vectored
through the netdev_get_tx_queue() and netdev_for_each_tx_queue()
interfaces.  This makes it easy to grep the tree for all
things that want to get to a TX queue of a net device.

Problem spots which are not really multiqueue aware yet, and
only work with one queue, can easily be spotted by grepping
for all netdev_get_tx_queue() calls that pass in a zero index.
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Now that we have a specific lock to protect the network
device unicast and multicast lists, remove extraneous
grabs of the TX lock in cases where the code only needs
address list protection.
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Add netif_addr_{lock,unlock}{,_bh}() helpers.

Use them to protect operations that operate on or read
the network device unicast and multicast address lists.

Also use them in cases where the code simply wants to
block calls into the driver's ->set_rx_mode() and
->set_multicast_list() methods.
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This will be used to protect the per-device unicast and multicast
address lists, as well as the callbacks into the drivers which
configure such state such as ->set_rx_mode() and ->set_multicast_list().
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

When VLAN header stripping is used, packets currently bypass packet
sockets (and other network taps) completely. For locally existing
VLANs, they appear directly on the VLAN device, for unknown VLANs
they are silently dropped.

Add a new function netif_nit_deliver() to deliver incoming packets
to all network interface taps and use it in __vlan_hwaccel_rx() to
make VLAN packets visible on the underlying device.
Signed-off-by: NPatrick McHardy <kaber@trash.net>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Accesses are mostly structured such that when there are multiple TX
queues the code transformations will be a little bit simpler.
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This allows us to use this calling convention all the way down into
qdisc_restart().
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Only plain netif_schedule() remains taking a net_device, mostly as a
compatability item while we transition the rest of these interfaces.

Everything else calls netif_schedule_queue() or __netif_schedule(),
both of which take a netdev_queue pointer.
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

We schedule queues, not the device, for output queue processing in BH.
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Now that our qdisc management is bi-directional, per-queue, and fully
orthogonal, there is no reason to have a special ingress qdisc pointer
in struct net_device.
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Now qdisc, qdisc_sleeping, and qdisc_list also live there.
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Every qdisc is assosciated with a queue, and in the case of ingress
qdiscs that will now be netdev->rx_queue so using that queue's lock is
the thing to do.
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The lock is now an attribute of the device queue.

One thing to notice is that "suspicious" places
emerge which will need specific training about
multiple queue handling.  They are so marked with
explicit "netdev->rx_queue" and "netdev->tx_queue"
references.
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

A netdev_queue is an entity managed by a qdisc.

Currently there is one RX and one TX queue, and a netdev_queue merely
contains a backpointer to the net_device.

The Qdisc struct is augmented with a netdev_queue pointer as well.

Eventually the 'dev' Qdisc member will go away and we will have the
resulting hierarchy:

	net_device --> netdev_queue --> Qdisc

Also, qdisc_alloc() and qdisc_create_dflt() now take a netdev_queue
pointer argument.
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Commit dad9b335 (netdevice: Fix promiscuity and allmulti overflow) broke
dev_set_promiscuity() by returning on success without reprogramming the
device.
Signed-off-by: NPatrick McHardy <kaber@trash.net>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

v1->v2: Use strlcpy() to ensure s[i].name be null-termination.

1. In netdev_boot_setup_add(), a long name will leak.
   ex. : dev=21,0x1234,0x1234,0x2345,eth123456789verylongname.........
2. In netdev_boot_setup_check(), mismatch will happen if s[i].name
   is a substring of dev->name.
   ex. : dev=...eth1 dev=...eth11

[ With feedback from Ben Hutchings. ]
Signed-off-by: NWang Chen <wangchen@cn.fujitsu.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Signed-off-by: NWang Chen <wangchen@cn.fujitsu.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Alexey Dobriyan <adobriyan@gmail.com> writes:
> Subject: ICMP sockets destruction vs ICMP packets oops

> After icmp_sk_exit() nuked ICMP sockets, we get an interrupt.
> icmp_reply() wants ICMP socket.
>
> Steps to reproduce:
>
> 	launch shell in new netns
> 	move real NIC to netns
> 	setup routing
> 	ping -i 0
> 	exit from shell
>
> BUG: unable to handle kernel NULL pointer dereference at 0000000000000000
> IP: [<ffffffff803fce17>] icmp_sk+0x17/0x30
> PGD 17f3cd067 PUD 17f3ce067 PMD 0 
> Oops: 0000 [1] PREEMPT SMP DEBUG_PAGEALLOC
> CPU 0 
> Modules linked in: usblp usbcore
> Pid: 0, comm: swapper Not tainted 2.6.26-rc6-netns-ct #4
> RIP: 0010:[<ffffffff803fce17>]  [<ffffffff803fce17>] icmp_sk+0x17/0x30
> RSP: 0018:ffffffff8057fc30  EFLAGS: 00010286
> RAX: 0000000000000000 RBX: 0000000000000000 RCX: ffff81017c7db900
> RDX: 0000000000000034 RSI: ffff81017c7db900 RDI: ffff81017dc41800
> RBP: ffffffff8057fc40 R08: 0000000000000001 R09: 000000000000a815
> R10: 0000000000000000 R11: 0000000000000001 R12: ffffffff8057fd28
> R13: ffffffff8057fd00 R14: ffff81017c7db938 R15: ffff81017dc41800
> FS:  0000000000000000(0000) GS:ffffffff80525000(0000) knlGS:0000000000000000
> CS:  0010 DS: 0018 ES: 0018 CR0: 000000008005003b
> CR2: 0000000000000000 CR3: 000000017fcda000 CR4: 00000000000006e0
> DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
> DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400
> Process swapper (pid: 0, threadinfo ffffffff8053a000, task ffffffff804fa4a0)
> Stack:  0000000000000000 ffff81017c7db900 ffffffff8057fcf0 ffffffff803fcfe4
>  ffffffff804faa38 0000000000000246 0000000000005a40 0000000000000246
>  000000000001ffff ffff81017dd68dc0 0000000000005a40 0000000055342436
> Call Trace:
>  <IRQ>  [<ffffffff803fcfe4>] icmp_reply+0x44/0x1e0
>  [<ffffffff803d3a0a>] ? ip_route_input+0x23a/0x1360
>  [<ffffffff803fd645>] icmp_echo+0x65/0x70
>  [<ffffffff803fd300>] icmp_rcv+0x180/0x1b0
>  [<ffffffff803d6d84>] ip_local_deliver+0xf4/0x1f0
>  [<ffffffff803d71bb>] ip_rcv+0x33b/0x650
>  [<ffffffff803bb16a>] netif_receive_skb+0x27a/0x340
>  [<ffffffff803be57d>] process_backlog+0x9d/0x100
>  [<ffffffff803bdd4d>] net_rx_action+0x18d/0x250
>  [<ffffffff80237be5>] __do_softirq+0x75/0x100
>  [<ffffffff8020c97c>] call_softirq+0x1c/0x30
>  [<ffffffff8020f085>] do_softirq+0x65/0xa0
>  [<ffffffff80237af7>] irq_exit+0x97/0xa0
>  [<ffffffff8020f198>] do_IRQ+0xa8/0x130
>  [<ffffffff80212ee0>] ? mwait_idle+0x0/0x60
>  [<ffffffff8020bc46>] ret_from_intr+0x0/0xf
>  <EOI>  [<ffffffff80212f2c>] ? mwait_idle+0x4c/0x60
>  [<ffffffff80212f23>] ? mwait_idle+0x43/0x60
>  [<ffffffff8020a217>] ? cpu_idle+0x57/0xa0
>  [<ffffffff8040f380>] ? rest_init+0x70/0x80
> Code: 10 5b 41 5c 41 5d 41 5e c9 c3 66 2e 0f 1f 84 00 00 00 00 00 55 48 89 e5 53
> 48 83 ec 08 48 8b 9f 78 01 00 00 e8 2b c7 f1 ff 89 c0 <48> 8b 04 c3 48 83 c4 08
> 5b c9 c3 66 66 66 66 66 2e 0f 1f 84 00
> RIP  [<ffffffff803fce17>] icmp_sk+0x17/0x30
>  RSP <ffffffff8057fc30>
> CR2: 0000000000000000
> ---[ end trace ea161157b76b33e8 ]---
> Kernel panic - not syncing: Aiee, killing interrupt handler!

Receiving packets while we are cleaning up a network namespace is a
racy proposition. It is possible when the packet arrives that we have
removed some but not all of the state we need to fully process it.  We
have the choice of either playing wack-a-mole with the cleanup routines
or simply dropping packets when we don't have a network namespace to
handle them.

Since the check looks inexpensive in netif_receive_skb let's just
drop the incoming packets.
Signed-off-by: NEric W. Biederman <ebiederm@xmission.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Large Receive Offload (LRO) is only appropriate for packets that are
destined for the host, and should be disabled if received packets may be
forwarded.  It can also confuse the GSO on output.

Add dev_disable_lro() function which uses the appropriate ethtool ops to
disable LRO if enabled.

Add calls to dev_disable_lro() in br_add_if() and functions that enable
IPv4 and IPv6 forwarding.
Signed-off-by: NBen Hutchings <bhutchings@solarflare.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Max of promiscuity and allmulti plus positive @inc can cause overflow.
Fox example: when allmulti=0xFFFFFFFF, any caller give dev_set_allmulti() a
positive @inc will cause allmulti be off.
This is not what we want, though it's rare case.
The fix is that only negative @inc will cause allmulti or promiscuity be off
and when any caller makes the counters touch the roof, we return error.

Change of v2:
Change void function dev_set_promiscuity/allmulti to return int.
So callers can get the overflow error.
Caller's fix will be done later.

Change of v3:
1. Since we return error to caller, we don't need to print KERN_ERROR,
KERN_WARNING is enough.
2. In dev_set_promiscuity(), if __dev_set_promiscuity() failed, we
return at once.
Signed-off-by: NWang Chen <wangchen@cn.fujitsu.com>
Acked-by: NPatrick McHardy <kaber@trash.net>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Add NETDEV_BONDING_FAILOVER event to be used in a successive patch
by bonding to announce fail-over for the active-backup mode through the
netdev events notifier chain mechanism. Such an event can be of use for the
RDMA CM (communication manager) to let native RDMA ULPs (eg NFS-RDMA, iSER)
always be aligned with the IP stack, in the sense that they use the same
ports/links as the stack does. More usages can be done to allow monitoring
tools based on netlink events being aware to bonding fail-over.
Signed-off-by: NOr Gerlitz <ogerlitz@voltaire.com>
Signed-off-by: NJay Vosburgh <fubar@us.ibm.com>
Signed-off-by: NJeff Garzik <jgarzik@redhat.com>

Selected device feature bits can be propagated to VLAN devices, so we
can make use of TX checksum offload and TSO on VLAN-tagged packets.
However, if the physical device does not do VLAN tag insertion or
generic checksum offload then the test for TX checksum offload in
dev_queue_xmit() will see a protocol of htons(ETH_P_8021Q) and yield
false.

This splits the checksum offload test into two functions:

- can_checksum_protocol() tests a given protocol against a feature bitmask

- dev_can_checksum() first tests the skb protocol against the device
  features; if that fails and the protocol is htons(ETH_P_8021Q) then
  it tests the encapsulated protocol against the effective device
  features for VLANs
Signed-off-by: NBen Hutchings <bhutchings@solarflare.com>
Acked-by: NPatrick McHardy <kaber@trash.net>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

As git-grep shows, open_softirq() is always called with the last argument
being NULL

block/blk-core.c:       open_softirq(BLOCK_SOFTIRQ, blk_done_softirq, NULL);
kernel/hrtimer.c:       open_softirq(HRTIMER_SOFTIRQ, run_hrtimer_softirq, NULL);
kernel/rcuclassic.c:    open_softirq(RCU_SOFTIRQ, rcu_process_callbacks, NULL);
kernel/rcupreempt.c:    open_softirq(RCU_SOFTIRQ, rcu_process_callbacks, NULL);
kernel/sched.c: open_softirq(SCHED_SOFTIRQ, run_rebalance_domains, NULL);
kernel/softirq.c:       open_softirq(TASKLET_SOFTIRQ, tasklet_action, NULL);
kernel/softirq.c:       open_softirq(HI_SOFTIRQ, tasklet_hi_action, NULL);
kernel/timer.c: open_softirq(TIMER_SOFTIRQ, run_timer_softirq, NULL);
net/core/dev.c: open_softirq(NET_TX_SOFTIRQ, net_tx_action, NULL);
net/core/dev.c: open_softirq(NET_RX_SOFTIRQ, net_rx_action, NULL);

This observation has already been made by Matthew Wilcox in June 2002
(http://www.cs.helsinki.fi/linux/linux-kernel/2002-25/0687.html)

"I notice that none of the current softirq routines use the data element
passed to them."

and the situation hasn't changed since them. So it appears we can safely
remove that extra argument to save 128 (54) bytes of kernel data (text).
Signed-off-by: NCarlos R. Mafra <crmafra@ift.unesp.br>
Signed-off-by: NIngo Molnar <mingo@elte.hu>
Signed-off-by: NThomas Gleixner <tglx@linutronix.de>

Am I just being particularly dim today, or can the call to
dev->change_rx_flags(dev, IFF_MULTICAST) in dev_change_flags() never
happen?

We've just set dev->flags = flags & IFF_MULTICAST, effectively. So the
condition '(dev->flags ^ flags) & IFF_MULTICAST' is _never_ going to be
true.
Signed-off-by: NDavid Woodhouse <dwmw2@infradead.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

device_rename can fail with -EEXIST or -ENOMEM, so handle any
problems.
Signed-off-by: NStephen Hemminger <shemminger@vyatta.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

dev_open() and dev_close() must be called holding the RTNL, since they
call device functions and netdevice notifiers that are promised the RTNL.
Signed-off-by: NBen Hutchings <bhutchings@solarflare.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

When a net namespace is destroyed, some devices (those, not killed
on ns stop explicitly) are moved back to init_net.

The problem, is that this net_ns change has one point of failure -
the __dev_alloc_name() may be called if a name collision occurs (and
this is easy to trigger). This allocator performs a likely-to-fail
GFP_ATOMIC allocation to find a suitable number. Other possible 
conditions that may cause error (for device being ns local or not
registered) are always false in this case.

So, when this call fails, the device is unregistered. But this is
*not* the right thing to do, since after this the device may be
released (and kfree-ed) improperly. E. g. bridges require more
actions (sysfs update, timer disarming, etc.), some other devices 
want to remove their private areas from lists, etc.

I. e. arbitrary use-after-free cases may occur.

The proposed fix is the following: since the only reason for the
dev_change_net_namespace to fail is the name generation, we may
give it a unique fall-back name w/o %d-s in it - the dev<ifindex>
one, since ifindexes are still unique.

So make this change, raise the failure-case printk loglevel to 
EMERG and replace the unregister_netdevice call with BUG().

[ Use snprintf() -DaveM ]
Signed-off-by: NPavel Emelyanov <xemul@openvz.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

When a netdev is moved across namespaces with the
'dev_change_net_namespace' function, the 'device_rename' function is
used to fixup kobject and refresh the sysfs tree. The device_rename
function will call kobject_rename and this one will check if there is
an object with the same name and this is the case because we are
renaming the object with the same name.

The use of 'device_rename' seems for me wrong because we usually don't
rename it but just move it across namespaces. As we just want to do a
mini "netdev_[un]register", IMO the functions
'netdev_[un]register_kobject' should be used instead, like an usual
network device [un]registering.

This patch replace device_rename by netdev_unregister_kobject,
followed by netdev_register_kobject.

The netdev_register_kobject will call device_initialize and will raise
a warning indicating the device was already initialized. In order to
fix that, I split the device initialization into a separate function
and use it together with 'netdev_register_kobject' into
register_netdevice. So we can safely call 'netdev_register_kobject' in
'dev_change_net_namespace'.

This fix will allow to properly use the sysfs per namespace which is
coming from -mm tree.
Signed-off-by: NDaniel Lezcano <dlezcano@fr.ibm.com>
Acked-by: NBenjamin Thery <benjamin.thery@bull.net>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Remove the fixed size channels[NR_CPUS] array in net/core/dev.c and
dynamically allocate array based on nr_cpu_ids.
Signed-off-by: NMike Travis <travis@sgi.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Some drivers have duplicated unlikely() macros.  IS_ERR() already has
unlikely() in itself.

This patch cleans up such pointless code.
Signed-off-by: NHirofumi Nakagawa <hnakagawa@miraclelinux.com>
Acked-by: NDavid S. Miller <davem@davemloft.net>
Acked-by: NJeff Garzik <jeff@garzik.org>
Cc: Paul Clements <paul.clements@steeleye.com>
Cc: Richard Purdie <rpurdie@rpsys.net>
Cc: Alessandro Zummo <a.zummo@towertech.it>
Cc: David Brownell <david-b@pacbell.net>
Cc: James Bottomley <James.Bottomley@HansenPartnership.com>
Cc: Michael Halcrow <mhalcrow@us.ibm.com>
Cc: Anton Altaparmakov <aia21@cantab.net>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Carsten Otte <cotte@de.ibm.com>
Cc: Patrick McHardy <kaber@trash.net>
Cc: Paul Mundt <lethal@linux-sh.org>
Cc: Jaroslav Kysela <perex@perex.cz>
Cc: Takashi Iwai <tiwai@suse.de>
Acked-by: NMike Frysinger <vapier@gentoo.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

This patch effectively reverts commit d0498d9a
aka "[NET]: Do not allocate unneeded memory for dev->priv alignment."
It was found to be buggy because of final unconditional += NETDEV_ALIGN_CONST
removal.

For example, for sizeof(struct net_device) being 2048 bytes, "alloc_size"
was also 2048 bytes, but allocator with debugging options turned on started
giving out !32-byte aligned memory resulting in redzones overwrites.

Patch does small optimization in ->priv'less case: bumping size to next
32-byte boundary was always done to ensure ->priv will also be aligned.
But, no ->priv, no need to do that.
Signed-off-by: NAlexey Dobriyan <adobriyan@sw.ru>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The alloc_netdev_mq() tries to produce 32-bytes alignment for both
the net_device itself and its private data. The second alignment is
achieved by adding the NETDEV_ALIGN_CONST to the whole size of
the memory to be allocated.

However, for those devices that do not need the private area, this
addition just makes the net_device weight 1024 + 32 = 1068 bytes,
i.e. consume twice as much memory.

Since loopback device is such (sizeof_priv == 0 for it), and each
net namespace creates one, this can save a noticeable amount of
memory for kernel with net namespaces turned on.

After this set the lo device is actually allocated from a size-1024
kmem cache on i386 box even with NETPOLL and WIRELESS_EXT turned on.
Signed-off-by: NPavel Emelyanov <xemul@openvz.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

dev_set_net is called for
- just allocated devices
- devices moving from one namespace to another
release_net has proper check inside to distinguish these cases.
Signed-off-by: NDenis V. Lunev <den@openvz.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>