net_sched: remove the unused parameter of qdisc_create_dflt()

The first parameter dev isn't in use in qdisc_create_dflt().

Signed-off-by: Changli Gao <xiaosuo@gmail.com>
Acked-by: Jamal Hadi Salim <hadi@cyberus.ca>
Signed-off-by: David S. Miller <davem@davemloft.net>

bonding: make release_and_destroy static

Only used in main file.

Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Acked-by: Andy Gospodarek <andy@greyhouse.net>
Signed-off-by: Jay Vosburgh <fubar@us.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

xfrm: make xfrm_bundle_ok local

Only used in one place.

Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

rtnetlink: remove rtnl_kill_links

The function rtnl_kill_links is defined but never used.

Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

xfrm6: make xfrm6_tunnel_free_spi local

Function only defined and used in one file.

Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

pch_gbe: make local functions static

Make routines that are only used in one file static.

Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

vmxnet3: make bit twiddle routines inline

Gcc doesn't usually handle inline across compilation units, and the
functions don't have to be global in scope. Move the set/reset flag
functions int the existing vmxnet3 header.

Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Acked-by: Shreyas Bhatewara <sbhatewara@vmware.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

bonding: make bond_resend_igmp_join_requests_delayed static

Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Acked-by: Flavio Leitner <fleitner@redhat.com>
Signed-off-by: Jay Vosburgh <fubar@us.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

sfc: make functions static

Make local functions and variable static. Do some rearrangement
of the string table stuff to put it where it gets used.

Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Acked-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

bridge: make br_parse_ip_options static

Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

socket: localize functions

A couple of functions in socket.c are only used there and
should be localized.

Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

netxen: make local function static.

Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

fib: introduce fib_alias_accessed() helper

Perf tools session at NFWS 2010 pointed out a false sharing on struct
fib_alias that can be avoided pretty easily, if we set FA_S_ACCESSED bit
only if needed (ie : not already set)

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

net/sched: fix missing spinlock init

Under network load, doing :

tc qdisc del dev eth0 root

triggers :

[  167.193087] BUG: spinlock bad magic on CPU#3, udpflood/4928
[  167.193139]  lock: c15bc324, .magic: 00000000, .owner:
<none>/-1, .owner_cpu: -1
[  167.193193] Pid: 4928, comm: udpflood Not tainted
2.6.36-rc7-11417-g215340c-dirty #323
[  167.193245] Call Trace:
[  167.193292]  [<c13abaa0>] ? printk+0x18/0x20
[  167.193342]  [<c11afb53>] spin_bug+0xa3/0xf0
[  167.193389]  [<c11afcdd>] do_raw_spin_lock+0x7d/0x160
[  167.193440]  [<c1313d4e>] ? __dev_xmit_skb+0x27e/0x2b0
[  167.193496]  [<c107382b>] ? trace_hardirqs_on+0xb/0x10
[  167.193545]  [<c13ae99a>] _raw_spin_lock+0x3a/0x40
[  167.193593]  [<c1313d4e>] ? __dev_xmit_skb+0x27e/0x2b0
[  167.193641]  [<c1313d4e>] __dev_xmit_skb+0x27e/0x2b0

commit 79640a4ca695 (add additional lock to qdisc to increase
throughput) forgot to initialize  noop_qdisc and noqueue_qdisc busylock

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

r8169: print errors when dma mapping fail

If dma mapping fail we are dropping packages or fail to open device.
But exact reason of drop/fail stays unknow for a user, so print errors.

Signed-off-by: Stanislaw Gruszka <sgruszka@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

r8169: (re)init phy on resume

Fix switching device to low-speed mode after resume reported in:
https://bugzilla.redhat.com/show_bug.cgi?id=502974

Reported-and-tested-by: Laurentiu Badea <bugzilla-redhat@wotevah.com>
Signed-off-by: Stanislaw Gruszka <sgruszka@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

r8169: changing mtu clean up

Since we do not change rx buffer size any longer, we can
clean up rtl8169_change_mtu and in consequence rtl8169_down.

Signed-off-by: Stanislaw Gruszka <sgruszka@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

r8169: do not account fragments as packets

Only increase tx_{packets,dropped} statistics when transmit or drop
full skb, not just fragment.

Signed-off-by: Stanislaw Gruszka <sgruszka@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

r8169: use pointer to struct device as local variable

Signed-off-by: Stanislaw Gruszka <sgruszka@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

r8169: replace PCI_DMA_{TO,FROM}DEVICE to DMA_{TO,FROM}_DEVICE

Signed-off-by: Stanislaw Gruszka <sgruszka@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

r8169: init rx ring cleanup

Signed-off-by: Stanislaw Gruszka <sgruszka@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

r8169: check dma mapping failures

Check possible dma mapping errors and do clean up if it happens.

Fix overwrap bug in rtl8169_tx_clear on the way.

Signed-off-by: Stanislaw Gruszka <sgruszka@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

bnx2x: Update bnx2x to use new vlan accleration.

Make the bnx2x driver use the new vlan accleration model.

Signed-off-by: Hao Zheng <hzheng@nicira.com>
Signed-off-by: Jesse Gross <jesse@nicira.com>
CC: Eilon Greenstein <eilong@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

ixgbe: Update ixgbe to use new vlan accleration.

Make the ixgbe driver use the new vlan accleration model.

Signed-off-by: Jesse Gross <jesse@nicira.com>
CC: Peter Waskiewicz <peter.p.waskiewicz.jr@intel.com>
CC: Emil Tantilov <emil.s.tantilov@intel.com>
CC: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

bnx2: Update bnx2 to use new vlan accleration.

Make the bnx2 driver use the new vlan accleration model.

Signed-off-by: Jesse Gross <jesse@nicira.com>
CC: Michael Chan <mchan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

bridge: Add support for TX vlan offload.

If some of the underlying devices support it, enable vlan offload on
transmit for bridge devices.  This allows senders to take advantage of the
hardware support, similar to other forms of acceleration.

Signed-off-by: Jesse Gross <jesse@nicira.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

ethtool: Add support for vlan accleration.

Now that vlan acceleration is handled consistently regardless of usage,
it is possible to enable and disable it at will.  This adds support for
Ethtool operations that change the offloading status for debugging
purposes, similar to other forms of hardware acceleration.

Signed-off-by: Jesse Gross <jesse@nicira.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

vlan: Centralize handling of hardware acceleration.

Currently each driver that is capable of vlan hardware acceleration
must be aware of the vlan groups that are configured and then pass
the stripped tag to a specialized receive function.  This is

different from other types of hardware offload in that it places a
significant amount of knowledge in the driver itself rather keeping
it in the networking core.

This makes vlan offloading function more similarly to other forms
of offloading (such as checksum offloading or TSO) by doing the
following:
* On receive, stripped vlans are passed directly to the network
core, without attempting to check for vlan groups or reconstructing
the header if no group
* vlans are made less special by folding the logic into the main
receive routines
* On transmit, the device layer will add the vlan header in software
if the hardware doesn't support it, instead of spreading that logic
out in upper layers, such as bonding.

There are a number of advantages to this:
* Fixes all bugs with drivers incorrectly dropping vlan headers at once.
* Avoids having to disable VLAN acceleration when in promiscuous mode
(good for bridging since it always puts devices in promiscuous mode).
* Keeps VLAN tag separate until given to ultimate consumer, which
avoids needing to do header reconstruction as in tg3 unless absolutely
necessary.
* Consolidates common code in core networking.

Signed-off-by: Jesse Gross <jesse@nicira.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

vlan: Avoid hash table lookup to find group.

A struct net_device always maps to zero or one vlan groups and we
always know the device when we are looking up a group.  We currently
do a hash table lookup on the device to find the group but it is
much simpler to just store a pointer.

Signed-off-by: Jesse Gross <jesse@nicira.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

vlan: Enable software emulation for vlan accleration.

Currently users of hardware vlan accleration need to know whether
the device supports it before generating packets.  However, vlan
acceleration will soon be available in a more flexible manner so
knowing ahead of time becomes much more difficult.  This adds
a software fallback path for vlan packets on devices without the
necessary offloading support, similar to other types of hardware
accleration.

Signed-off-by: Jesse Gross <jesse@nicira.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

vlan: Don't check for vlan group before vlan_tx_tag_present.

Many (but not all) drivers check to see whether there is a vlan
group configured before using a tag stored in the skb.  There's
not much point in this check since it just throws away data that
should only be present in the expected circumstances.  However,
it will soon be legal and expected to get a vlan tag when no
vlan group is configured, so remove this check from all drivers
to avoid dropping the tags.

Signed-off-by: Jesse Gross <jesse@nicira.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

vlan: Rename VLAN_GROUP_ARRAY_LEN to VLAN_N_VID.

VLAN_GROUP_ARRAY_LEN is simply the number of possible vlan VIDs.
Since vlan groups will soon be more of an implementation detail
for vlan devices, rename the constant to be descriptive of its
actual purpose.

Signed-off-by: Jesse Gross <jesse@nicira.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

ebtables: Allow filtering of hardware accelerated vlan frames.

An upcoming commit will allow packets with hardware vlan acceleration
information to be passed though more parts of the network stack, including
packets trunked through the bridge.  This adds support for matching and
filtering those packets through ebtables.

Signed-off-by: Jesse Gross <jesse@nicira.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

enic: Fix log message

Fix a log message

Signed-off-by: Vasanthy Kolluri <vkolluri@cisco.com>
Signed-off-by: Roopa Prabhu <roprabhu@cisco.com>
Signed-off-by: David Wang <dwang2@cisco.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

enic: Change min MTU

Change min MTU to 68.

Signed-off-by: Vasanthy Kolluri <vkolluri@cisco.com>
Signed-off-by: Roopa Prabhu <roprabhu@cisco.com>
Signed-off-by: David Wang <dwang2@cisco.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

enic: Replace firmware devcmd CMD_ENABLE with CMD_ENABLE_WAIT

Replace no wait CMD_ENABLE firmware devcmd with CMD_ENABLE_WAIT

Signed-off-by: Vasanthy Kolluri <vkolluri@cisco.com>
Signed-off-by: Roopa Prabhu <roprabhu@cisco.com>
Signed-off-by: David Wang <dwang2@cisco.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

enic: Make firmware cognizant of the user set mac address

Let the firmware know about the mac address set by the user using ndo_set_mac_address

Signed-off-by: Vasanthy Kolluri <vkolluri@cisco.com>
Signed-off-by: Roopa Prabhu <roprabhu@cisco.com>
Signed-off-by: David Wang <dwang2@cisco.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

enic: Add support for multiple hardware receive queues

Add support for multiple hardware receive queues. The ingress traffic is hashed into one of the receive queues based on IP or TCP or both headers. The max no. of receive queues supported is 8.

Signed-off-by: Vasanthy Kolluri <vkolluri@cisco.com>
Signed-off-by: Roopa Prabhu <roprabhu@cisco.com>
Signed-off-by: David Wang <dwang2@cisco.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

ibmveth: Free irq on error path

Free irq on error path.

Signed-off-by: Denis Kirjanov <dkirjanov@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

ibmveth: Cleanup error handling inside ibmveth_open

Remove duplicated code in one place.

Signed-off-by: Denis Kirjanov <dkirjanov@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

net: avoid RCU for NOCACHE dst

There is no point using RCU for dst we allocate for a very short time
(used once).

Change dst_release() to take DST_NOCACHE into account, but also change
skb_dst_set_noref() to force a refcount increment for such dst.

This is a _huge_ gain, because we dont waste memory to store xx thousand
of dsts. Instead of queueing them to RCU, we can free them instantly.

CPU caches can stay hot, re-using same memory blocks to hold temporary
dsts.

Note : remove unneeded smp_mb__before_atomic_dec(); in dst_release(),
since atomic_dec_return() implies a full memory barrier.

Stress test, 160.000.000 udp frames sent, IP route cache disabled
(DDOS).

Before:

real    0m38.091s
user    0m13.189s
sys     7m53.018s

After:

real 0m29.946s
user 0m12.157s
sys 7m40.605s

For reference, if IP route cache was enabled :

real 0m32.030s
user 0m10.521s
sys 8m15.243s

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

net: allocate tx queues in register_netdevice

This patch introduces netif_alloc_netdev_queues which is called from
register_device instead of alloc_netdev_mq.  This makes TX queue
allocation symmetric with RX allocation.  Also, queue locks allocation
is done in netdev_init_one_queue.  Change set_real_num_tx_queues to
fail if requested number < 1 or greater than number of allocated
queues.

Signed-off-by: Tom Herbert <therbert@google.com>
Acked-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

net: cleanups in RX queue allocation

Clean up in RX queue allocation.  In netif_set_real_num_rx_queues
return error on attempt to set zero queues, or requested number is
greater than number of allocated queues.  In netif_alloc_rx_queues,
do BUG_ON if queue_count is zero.

Signed-off-by: Tom Herbert <therbert@google.com>
Acked-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

net: fail alloc_netdev_mq if queue count < 1

In alloc_netdev_mq fail if requested queue_count < 1.

Signed-off-by: Tom Herbert <therbert@google.com>
Acked-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Merge branch 'for-davem' of git://git./linux/kernel/git/linville/wireless-next-2.6

phonet: remove the unused variable pn

Signed-off-by: Changli Gao <xiaosuo@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

netpoll: Revert napi_poll fix for bonding driver

In an erlier patch I modified napi_poll so that devices with IFF_MASTER polled
the per_cpu list instead of the device list for napi.  I did this because the
bonding driver has no napi instances to poll, it instead expects to check the
slave devices napi instances, which napi_poll was unaware of.  Looking at this
more closely however, I now see this isn't strictly needed.  As the bond driver
poll_controller calls the slaves poll_controller via netpoll_poll_dev, which
recursively calls poll_napi on each slave, allowing those napi instances to get
serviced.  The earlier patch isn't at all harmfull, its just not needed, so lets
revert it to make the code cleaner.  Sorry for the noise,

Signed-off-by: Neil Horman <nhorman@tuxdriver.com>
Reviewed-by: WANG Cong <amwang@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

netpoll: Remove netpoll blocking from uninit path

Some recent testing in netpoll with bonding showed this backtrace

 ------------[ cut here ]------------
 kernel BUG at drivers/net/bonding/bonding.h:134!
 invalid opcode: 0000 [#1] SMP
 last sysfs file: /sys/devices/pci0000:00/0000:00:1d.2/usb7/devnum
 CPU 0
 Pid: 1876, comm: rmmod Not tainted 2.6.36-rc3+ #10 D26928/
 RIP: 0010:[<ffffffffa0514ba4>]  [<ffffffffa0514ba4>] bond_uninit+0x6f4/0x7a0
 RSP: 0018:ffff88003b1b5d58  EFLAGS: 00010296
 RAX: ffff88003b9b6200 RBX: ffff8800373e8e00 RCX: 00000000000f4240
 RDX: 00000000ffffffff RSI: 0000000000000286 RDI: 0000000000000286
 RBP: ffff88003b1b5dc8 R08: 0000000000000000 R09: 00000001af7de920
 R10: 0000000000000000 R11: ffff880002495e98 R12: ffff880037922700
 R13: ffff880038c31000 R14: ffff880037922730 R15: 0000000000000286
 FS:  00007f90e6d72700(0000) GS:ffff880002400000(0000) knlGS:0000000000000000
 CS:  0010 DS: 0000 ES: 0000 CR0: 000000008005003b
 CR2: 000000346f0d9ad0 CR3: 000000003b263000 CR4: 00000000000006f0
 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
 DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400
 Process rmmod (pid: 1876, threadinfo ffff88003b1b4000, task ffff88003b36aa80)
 Stack:
 00000000ffffffff ffff88003b1b5d7a ffff8800379221e8 ffff880037922000
 <0> ffff88003b1b5dc8 ffffffff813eb5fb ffff88003b1b5da8 0000000031b177a3
 <0> ffff88003b1b5da8 ffff880037922000 ffff88003b1b5e48 ffff88003b1b5e48
 Call Trace:
 [<ffffffff813eb5fb>] ? rtmsg_ifinfo+0xcb/0xf0
 [<ffffffff813daad8>] rollback_registered_many+0x168/0x280
 [<ffffffff813dac09>] unregister_netdevice_many+0x19/0x80
 [<ffffffff813e97b3>] __rtnl_kill_links+0x63/0x90
 [<ffffffff813e980b>] __rtnl_link_unregister+0x2b/0x60
 [<ffffffff813e9bde>] rtnl_link_unregister+0x1e/0x30
 [<ffffffffa052124b>] bonding_exit+0x37/0x51 [bonding]
 [<ffffffff81098b2e>] sys_delete_module+0x19e/0x270
 [<ffffffff810bb2b2>] ? audit_syscall_entry+0x252/0x280
 [<ffffffff8100b0b2>] system_call_fastpath+0x16/0x1b
 RIP  [<ffffffffa0514ba4>] bond_uninit+0x6f4/0x7a0 [bonding]
 RSP <ffff88003b1b5d58>
 ---[ end trace 1395ad691cea24d1 ]---

It occurs because of my recent netpoll blocking patches, which I added to avoid
recursive deadlock in the bonding driver.  It relies on some per cpu bits, but
the shutdown path forces some rescheduling as we cancel workqueues for the
driver and wait for some device refcounts.  If after the forced reschedule, we
wind up on a different cpu we trigger the bughalt in unblock_netpoll_tx.

The fix is to remove the netpoll block/unblock calls from bond_release_all.
This is safe to do because bond_uninit, which is called via ndo_uninit in
rollback_registered_many, doesn't occur until we send a NETDEV_UNREGISTER event,
which triggers netconsole to remove us as a netpoll client, so we are guaranteed
not to recurse into our own tx path here.

Signed-off-by: Neil Horman <nhorman@tuxdriver.com>
Reviewed-by: WANG Cong <amwang@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

bnx2x: update version to 1.60.00-3

Signed-off-by: Dmitry Kravkov <dmitry@broadcom.com>
Signed-off-by: Eilon Greenstein <eilong@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

bnx2x: prevent false parity error in MSI-X memory of HC block

Signed-off-by: Dmitry Kravkov <dmitry@broadcom.com>
Signed-off-by: Vladislav Zolotarov <vladz@broadcom.com>
Signed-off-by: Eilon Greenstein <eilong@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

bnx2x: fix possible deadlock in HC hw block

The possible deadlock (on 57710 devices only) will prevent from the
device to generate interrupts.

Signed-off-by: Dmitry Kravkov <dmitry@broadcom.com>
Signed-off-by: Eilon Greenstein <eilong@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

inet: RCU changes in inetdev_by_index()

Convert inetdev_by_index() to not increment in_dev refcount.

Callers hold RCU or RTNL, and should not decrement in_dev refcount.

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

net: avoid a dev refcount in ip_mc_find_dev()

We hold RTNL in ip_mc_find_dev(), no need to touch device refcount.

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

3c52x: remove IRQF_SAMPLE_RANDOM from legacy MCA drivers.

If you are genuinely using one of these legacy MCA drivers
then you are tragically on hardware where you really don't
have the extra CPU cycles to be wasting on this.

In addition, it makes two less cases for people to inadvertently
blindly copy flags from without explicitly thinking whether it
makes sense -- see the addition to feature-removal.txt as per
commit 9d9b8fb0e5ebf4b0398e579f6061d4451fea3242.

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

bonding: Re-enable netpoll over bonding

With the inclusion of previous fixup patches, netpoll over bonding apears to
work reliably with failover conditions.  This reverts Gospos previous commit
c22d7ac844f1cb9c6a5fd20f89ebadc2feef891b, and allows access again to the netpoll
functionality in the bonding driver.

Signed-off-by: Neil Horman <nhorman@tuxdriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

bonding: Fix netconsole to not deadlock on rmmod

Netconsole calls netpoll_cleanup on receipt of a NETDEVICE_UNREGISTER event.
The notifier subsystem calls these event handlers with rtnl_lock held, which
netpoll_cleanup also takes, resulting in deadlock.  Fix this by calling the
__netpoll_cleanup interior function instead, and fixing up the additional
pointers.

Signed-off-by: Neil Horman <nhorman@tuxdriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

bonding: Fix napi poll for bonding driver

Usually the netpoll path, when preforming a napi poll can get away with just
polling all the napi instances of the configured device.  Thats not the case for
the bonding driver however, as the napi instances which may wind up getting
flagged as needing polling after the poll_controller call don't belong to the
bonded device, but rather to the slave devices.  Fix this by checking the device
in question for the IFF_MASTER flag, if set, we know we need to check the full
poll list for this cpu, rather than just the devices napi instance list.

Signed-off-by: Neil Horman <nhorman@tuxdriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

bonding: Fix deadlock in bonding driver resulting from internal locking when using netpoll

The monitoring paths in the bonding driver take write locks that are shared by
the tx path.  If netconsole is in use, these paths can call printk which puts us
in the netpoll tx path, which, if netconsole is attached to the bonding driver,
result in deadlock (the xmit_lock guards are useless in netpoll_send_skb, as the
monitor paths in the bonding driver don't claim the xmit_lock, nor should they).
The solution is to use a per cpu flag internal to the driver to indicate when a
cpu is holding the lock in a path that might recusrse into the tx path for the
driver via netconsole.  By checking this flag on transmit, we can defer the
sending of the netconsole frames until a later time using the retransmit feature
of netpoll_send_skb that is triggered on the return code NETDEV_TX_BUSY.  I've
tested this and am able to transmit via netconsole while causing failover
conditions on the bond slave links.

Signed-off-by: Neil Horman <nhorman@tuxdriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

bonding: Fix bonding drivers improper modification of netpoll structure

The bonding driver currently modifies the netpoll structure in its xmit path
while sending frames from netpoll.  This is racy, as other cpus can access the
netpoll structure in parallel. Since the bonding driver points np->dev to a
slave device, other cpus can inadvertently attempt to send data directly to
slave devices, leading to improper locking with the bonding master, lost frames,
and deadlocks.  This patch fixes that up.

This patch also removes the real_dev pointer from the netpoll structure as that
data is really only used by bonding in the poll_controller, and we can emulate
its behavior by check each slave for IS_UP.

Signed-off-by: Neil Horman <nhorman@tuxdriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

e1000e: Fix for offline diag test failure at first call

Move link test call to later in the offline sequence, move the
restore settings block to afterwards and add another reset to ensure
the hardware is in a known state afterwards.

Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com>
Acked-by: Bruce Allan <bruce.w.allan@intel.com>
Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

igbvf: Remove unneeded pm_qos* calls

Power Management Quality of Service is not supported or used by the VF
driver.

Signed-off-by: Greg Rose <gregory.v.rose@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

igb: fix stats handling

There are currently some problems with igb.

- On 32bit arches, maintaining 64bit counters without proper
synchronization between writers and readers.

- Stats updated every two seconds, as reported by Jesper.
   (Jesper provided a patch for this)

- Potential problem between worker thread and ethtool -S

This patch uses u64_stats_sync, and convert everything to be 64bit safe,
SMP safe, even on 32bit arches. It integrates Jesper idea of providing
accurate stats at the time user reads them.

Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Tested-by: Emil Tantilov <emil.s.tantilov@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

netxen: mask correctable error

HW workaround:
Disable logging of correctable error for some NX3031 based adapter.

Signed-off-by: Amit Kumar Salecha <amit.salecha@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

netxen: fix race in tx stop queue

There is race between netif_stop_queue and netif_stopped_queue
check.So check once again if buffers are available to avoid race.
With above logic we can also get rid of tx lock in process_cmd_ring.

Signed-off-by: Rajesh Borundia <rajesh.borundia@qlogic.com>
Signed-off-by: Amit Kumar Salecha <amit.salecha@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

qlcnic: update ethtool stats

Added statistics for Nic Partition supported adapter.
These statistics are maintined in device.

Signed-off-by: Amit Kumar Salecha <amit.salecha@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

IPv4: route.c: Change checks against 0xffffffff to ipv4_is_lbcast()

Change a few checks against the hardcoded broadcast address,
0xffffffff, to ipv4_is_lbcast().  Remove some existing checks
using ipv4_is_lbcast() that are now obviously superfluous.

Signed-off-by: Andy Walls <awalls@md.metrocast.net>
Signed-off-by: David S. Miller <davem@davemloft.net>

Merge branch 'can/mcp251x-for-net-next' of git://git.pengutronix.de/git/mkl/linux-2.6

bnx2x: update version to 1.60.00-2

Signed-off-by: Dmitry Kravkov <dmitry@broadcom.com>
Signed-off-by: Eilon Greenstein <eilong@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

bnx2x: remove unnecessary FUNC_FLG_RSS flag and related

As suggested by: Joe Perches <joe@perches.com>

Although RSS is meaningless when there is a single HW queue we
still need it enabled in order to have HW Rx hash generated.

Signed-off-by: Dmitry Kravkov <dmitry@broadcom.com>
Signed-off-by: Vladislav Zolotarov <vladz@broadcom.com>
Signed-off-by: Eilon Greenstein <eilong@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

bnx2x: Use correct FW constant for header padding

the value of the constant is the same, but it's clearer to use original
constant provided by HSI

Signed-off-by: Dmitry Kravkov <dmitry@broadcom.com>
Signed-off-by: Eilon Greenstein <eilong@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

bnx2x: do not deal with power if no capability

Signed-off-by: Dmitry Kravkov <dmitry@broadcom.com>
Signed-off-by: Vladislav Zolotarov <vladz@broadcom.com>
Signed-off-by: Eilon Greenstein <eilong@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

bnx2x: remove redundant commands during error handling

Signed-off-by: Dmitry Kravkov <dmitry@broadcom.com>
Signed-off-by: Eilon Greenstein <eilong@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

bnx2x: Optimized the branching in the bnx2x_rx_int()

Optimized the branching in the bnx2x_rx_int() based on the fact
that FP CQE will always have at least one of START or STOP flags set,
so if not both bits are set and START bit is not set,
then it's a STOP bit that is set.

Signed-off-by: Dmitry Kravkov <dmitry@broadcom.com>
Signed-off-by: Vladislav Zolotarov <vladz@broadcom.com>
Signed-off-by: Eilon Greenstein <eilong@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

can: mcp251x: optimize 2515, rx int gets cleared automatically

Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
Acked-by: Wolfgang Grandegger <wg@grandegger.com>

can: mcp251x: define helper functions mcp251x_is_2510, mcp251x_is_2515

Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
Acked-by: Wolfgang Grandegger <wg@grandegger.com>

can: mcp251x: Don't use pdata->model for chip selection anymore

Since commit e446630c960946b5c1762e4eadb618becef599e7, i.e. v2.6.35-rc1,
the mcp251x chip model can be selected via the modalias member in the
struct spi_board_info. The driver stores the actual model in the
struct mcp251x_platform_data.

From the driver point of view the platform_data should be read only.
Since all in-tree users of the mcp251x have already been converted to
the modalias method, this patch moves the "model" member from the
struct mcp251x_platform_data to the driver's private data structure.

Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
Cc: Christian Pellegrin <chripell@fsfe.org>
Cc: Marc Zyngier <maz@misterjones.org>

can: mcp251x: write intf only when needed

This patch introduces a variable "clear_intf" that hold the bits that
should be cleared. Only read-modify-write register if "clear_intf"
is set.

Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
Acked-by: Wolfgang Grandegger <wg@grandegger.com>

can: mcp251x: read-modify-write eflag only when needed

Use read-modify-write instead of a simple write to change the register
contents, to close existing the race window between the original manual
read and write.

Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
Acked-by: Wolfgang Grandegger <wg@grandegger.com>

can: mcp251x: allow to read two registers in one spi transfer

This patch bases on work done earlier by David Jander.

Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
Acked-by: David Jander <david@protonic.nl>
Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
Acked-by: Wolfgang Grandegger <wg@grandegger.com>

can: mcp251x: increase rx_errors on overflow, not only rx_over_errors

Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
Acked-by: Wolfgang Grandegger <wg@grandegger.com>

can: mcp251x: fix NOHZ local_softirq_pending 08 warning

This patch replaces netif_rx() with netif_rx_ni() which has to be used
from the threaded interrupt i.e. process context context.

Thanks to Christian Pellegrin for pointing at the right fix:
481a8199142c050b72bff8a1956a49fd0a75bbe0 by Oliver Hartkopp.

Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
Acked-by: Wolfgang Grandegger <wg@grandegger.com>

ns83820: spin_lock_irq() => spin_lock()

This is essentially cosmetic.  At this point the IRQs are already
disabled because we called spin_lock_irq(&dev->rx_info.lock).

The real bug here was fixed back in 2006 in 3a10ccebe: "[PATCH] lock
validator: fix ns83820.c irq-flags bug".  Prior to that patch, it was
a "spin_lock_irq is not nestable" type bug.  The 2006 patch changes the
unlock to not re-enable IRQs, which eliminates the potential deadlock.

But this bit was missed.  We should change the lock function as well so
it balances nicely.

Signed-off-by: Dan Carpenter <error27@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

tipc: Simplify bearer shutdown logic

Optimize processing in TIPC's bearer shutdown code, including:

1. Remove an unnecessary check to see if TIPC bearer's can exist.
2. Don't release spinlocks before calling a media-specific disabling
routine, since the routine can't sleep.
3. Make bearer_disable() operate directly on a struct bearer, instead
of needlessly taking a name and then mapping that to the struct.

Signed-off-by: Allan Stephens <allan.stephens@windriver.com>
Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
Reviewed-by: Neil Horman <nhorman@tuxdriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

tipc: Kill tipc_get_mode() completely.

It's completely unused and exporting a static symbol
makes no sense and breaks the build.

Reported-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>

tg3: Update version to 3.115

This patch updates the tg3 version to 3.115.

Signed-off-by: Matt Carlson <mcarlson@broadcom.com>
Reviewed-by: Michael Chan <mchan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

tg3: Report invalid link from tg3_get_settings()

Currently the tg3 driver leaves the speed and duplex fields
uninitialized in tg3_get_settings() if the device is not up.  This can
lead to some strange deductions in certain versions of ethtool.  When
the device is up and the link is down, the driver reports SPEED_INVALID
and DUPLEX_INVALID for these fields.  This patch makes the presentation
consistent by returning SPEED_INVALID and DUPLEX_INVALID when the device
has not been brought up as well.

Signed-off-by: Matt Carlson <mcarlson@broadcom.com>
Reviewed-by: Michael Chan <mchan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

tg3: Don't allocate jumbo ring for 5780 class devs

The 5714, 5715, and 5780 devices do not have a separate rx jumbo
producer ring.  This patch changes the code so that resources are not
allocated for it.

Signed-off-by: Matt Carlson <mcarlson@broadcom.com>
Reviewed-by: Michael Chan <mchan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

tg3: Cleanup tg3_alloc_rx_skb()

src_map is no longer used in tg3_alloc_rx_skb().  Remove it.

Signed-off-by: Matt Carlson <mcarlson@broadcom.com>
Reviewed-by: Michael Chan <mchan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

tg3: Add EEE support

This patch adds Energy Efficient Ethernet (EEE) support for the 5718
device ID and the 57765 B0 asic revision.

Signed-off-by: Matt Carlson <mcarlson@broadcom.com>
Reviewed-by: Michael Chan <mchan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

tg3: Add clause 45 register accessor methods

This patch adds clause 45 register access methods.  They will be used in
the following patch.

Signed-off-by: Matt Carlson <mcarlson@broadcom.com>
Reviewed-by: Michael Chan <mchan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

tg3: Disable unused transmit rings

This patch allows the driver to disable the additional transmit rings
available on the 5717 and 5719 devices.  This is not strictly necessary,
but is done anyways for correctness.

Signed-off-by: Matt Carlson <mcarlson@broadcom.com>
Reviewed-by: Michael Chan <mchan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

tg3: Add support for selfboot format 1 v6

5718 B0 and 5719 devices will use a new selfboot firmware format.  This
patch adds code to detect the new format so that bootcode versions get
reported correctly.

Signed-off-by: Matt Carlson <mcarlson@broadcom.com>
Reviewed-by: Michael Chan <mchan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

fib_hash: RCU conversion phase 2

Get rid of fib_hash_lock rwlock.

The fn_zone hash table resize is the noticeable part of this patch.

I added a seqlock per fn_zone, so that readers can restart their lookup
in the (very rare) case a writer expanded the hash table.

Add rcu heads in fib_alias and fib_node, use call_rcu() to defer their
freeing, and use appropriate _rcu list manipulations.

Stress test (160.000.000 udp frames sent, IP route cache disabled to
mimic DDOS attack, FIB_HASH)

Before:
real 0m41.191s
user 0m13.137s
sys 8m55.241s

After:
real 0m38.091s
user 0m13.189s
sys 7m53.018s

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

fib_hash: RCU conversion phase 1

First step for RCU conversion of fib_hash :

struct fn_zone are created and never deleted.

Very classic conversion, using rcu_assign_pointer(), rcu_dereference()
and rtnl_dereference() verbs.

__rcu markers on fz_next and fn_zone_list

They are created under RTNL, we dont need fib_hash_lock anymore in
fn_new_zone().

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

fib_hash: embed initial hash table in fn_zone

While looking for false sharing problems, I noticed
sizeof(struct fn_zone) was small (28 bytes) and possibly sharing a cache
line with an often written kernel structure.

Most of the time, fn_zone uses its initial hash table of 16 slots.

We can avoid the false sharing problem by embedding this initial hash
table in fn_zone itself, so that sizeof(fn_zone) > L1_CACHE_BYTES

We did a similar optimization in commit a6501e080c (Reduce memory needs
and speedup lookups)

Add a fz_revorder field to speedup fn_hash() a bit.

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

drivers/net/pch_gbe: Use DEFINE_PCI_DEVICE_TABLE

Use the standard macro to put this table in __devinitconst.

Compiled, untested.

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

netns: reorder fields in struct net

In a network bench, I noticed an unfortunate false sharing between
'loopback_dev' and 'count' fields in "struct net".

'count' is written each time a socket is created or destroyed, while
loopback_dev might be often read in routing code.

Move loopback_dev in a read mostly section of "struct net"

Note: struct netns_xfrm is cache line aligned on SMP.
(It contains a "struct dst_ops")
Move it at the end to avoid holes, and reduce sizeof(struct net) by 128
bytes on ia32.

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

tcp: use correct counters in CA_CWR state too

As CWR is stronger than CA_Disorder state, we can miscount
SACK/Reno failure into other timeouts. Not a bad problem as
it can happen only due to ECN, FRTO detecting spurious RTO
or xmit error which are the only callers of tcp_enter_cwr.
And even then losses and RTO must still follow thereafter
to actually end up into the relevant code paths.

Compile tested.

Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>

tcp: sack lost marking fixes

When only fast rexmit should be done, tcp_mark_head_lost marks
L too far. Also, sacked_upto below 1 is perfectly valid number,
the packets == 0 then needs to be trapped elsewhere.

Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>

stmmac: remove ifdef NETIF_F_TSO from stmmac_ethtool.c

Signed-off-by: Giuseppe Cavallaro <peppe.cavallaro@st.com>
Reported-by: Armando Visconti <armando.visconti@st.com>
Signed-off-by: David S. Miller <davem@davemloft.net>