vlan: adds drops accounting

Its hard to tell if vlans are dropping frames, since
every frame given to vlan_???_start_xmit() functions
is accounted as fully transmitted by lower device.

We can test dev_queue_xmit() return values to
properly account for dropped frames.

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

macvlan: add multiqueue capability

macvlan devices are currently not multi-queue capable.

We can do that defining rtnl_link_ops method,
get_tx_queues(), called from rtnl_create_link()

This new method gets num_tx_queues/real_num_tx_queues
from lower device.

macvlan_get_tx_queues() is a copy of vlan_get_tx_queues().

Because macvlan_start_xmit() has to update netdev_queue
stats only (and not dev->stats), I chose to change
tx_errors/tx_aborted_errors accounting to tx_dropped,
since netdev_queue structure doesnt define tx_errors /
tx_aborted_errors.

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

netdev: Convert MDIO ioctl implementation to use struct mii_ioctl_data

A few drivers still access the arguments to MDIO ioctls as an array of
u16.  Convert them to use struct mii_ioctl_data.

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

netdev: Remove redundant checks for CAP_NET_ADMIN in MDIO implementations

dev_ioctl() already checks capable(CAP_NET_ADMIN) before calling the
driver's implementation of MDIO ioctls.

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

netdev: Remove SIOCDEVPRIVATE aliases for MDIO ioctls

The standard MDIO ioctl numbers are well-established and these should
no longer be needed.

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

sky2: only enable Vaux if capable of wakeup

While perusing vendor driver, I saw that it did not enable the Vaux
power unless device was able to wake from lan for D3cold.
This might help for Rene's power issue.

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

netxen: fix infinite loop on dma mapping failure

Fix a perpetual while() loop in unwinding partial
mapped tx skb on dma mapping failure.

Reported-by: "Juha Leppanen" <juha_motorsportcom@luukku.com>
Signed-off-by: Dhananjay Phadke <dhananjay@netxen.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

netxen: remove duplicate napi_add

Remove duplicate calls to netxen_napi_add().

Signed-off-by: Dhananjay Phadke <dhananjay@netxen.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

netxen: fix lro buffer allocation

Alloc 12k skbuffs so that firmware can aggregate more
packets into one buffer. This doesn't raise memory
consumption since 9k skbs use 16k slab cache anyway.

Signed-off-by: Dhananjay Phadke <dhananjay@netxen.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

ixgbe: Add support for using FCoE DDP in 82599 as FCoE targets

The FCoE DDP in 82599 can be used for both FCoE initiator as well as FCoE
target, depending on the indication of the exchange being the responder or
originator in the F_CTL (frame control) field in the encapsulated Fiber
Channel frame header (T10 Spec., FC-FS). For the initiator, OX_ID is used
for FCoE DDP, where for the target RX_ID is used for FCoE DDP.

Signed-off-by: Yi Zou <yi.zou@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

ixgbe: Distribute transmission of FCoE traffic in 82599

This adds a simple selection of a FCoE tx queue based on the current cpu id to
distribute transmission of FCoE traffic evenly among multiple FCoE transmit
queues.

Signed-off-by: Yi Zou <yi.zou@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

ixgbe: Add support for multiple Tx queues for FCoE in 82599

This patch adds support for multiple transmit queues to the Fiber Channel
over Ethernet (FCoE) feature found in 82599. Currently, FCoE has multiple
Rx queues available, along with a redirection table, that helps distribute
the I/O load across multiple CPUs based on the FC exchange ID. To make
this the most effective, we need to provide the same layout of transmit
queues to match receive.

Particularly, when Data Center Bridging (DCB) is enabled, the designated
traffic class for FCoE can have dedicated queues for just FCoE traffic,
while not affecting any other type of traffic flow.

Signed-off-by: Yi Zou <yi.zou@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

igb: set vf rlpml wasn't taking vlan tag into account

This patch updates things so that vlan tags are taken into account when
setting the receive large packet maximum length.  This allows the VF driver
to correctly receive full sized frames when vlans are enabled.

Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

igb: only disable/enable interrupt bits for igb physical function

The igb_irq_disable/enable calls were causing virtual functions associated
with the igb physical function to have their interrupts disabled.  In order
to prevent this from occuring we should only clear/set the bits related to
the physical function.

Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

igb: add support for set_rx_mode netdevice operation

This patch adds support for the set_rx_mode netdevice operation so that igb
can better support multiple unicast addresses.

Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

net: Remove debugging code

Remove a debugging aid I accidently left in previous 'cleanup' patch

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

vlan: enable multiqueue xmits

vlan_dev_hard_start_xmit() & vlan_dev_hwaccel_hard_start_xmit()
select txqueue number 0, instead of using index provided by
skb_get_queue_mapping().

This is not correct after commit 2e59af3dcbdf11635c03f
[vlan: multiqueue vlan device] because
txq->tx_packets  & txq->tx_bytes changes are performed on
a single location, and not the right locking.

Fix is to take the appropriate struct netdev_queue pointer

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

net: net/core/dev.c cleanups

Pure style cleanup patch before surgery :)

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

atm/br2684: netif_stop_queue() when atm device busy and netif_wake_queue() when we can send packets again.

This patch removes the call to dev_kfree_skb() when the atm device is busy.
Calling dev_kfree_skb() causes heavy packet loss then the device is under
heavy load, the more correct behavior should be to stop the upper layers,
then when the lower device can queue packets again wake the upper layers.

Signed-off-by: Karl Hiramoto <karl@hiramoto.org>
Signed-off-by: Chas Williams <chas@cmf.nrl.navy.mil>
Signed-off-by: David S. Miller <davem@davemloft.net>

fec: don't enable irqs in hard irq context

fec_enet_mii, fec_enet_rx and fec_enet_tx are both only called by
fec_enet_interrupt in interrupt context.  So they must not use
spin_lock_irq/spin_unlock_irq.

This fixes:
WARNING: at kernel/lockdep.c:2140 trace_hardirqs_on_caller+0x130/0x194()
...

Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
Cc: Greg Ungerer <gerg@uclinux.org>
Cc: Ben Hutchings <ben@decadent.org.uk>
Cc: Patrick McHardy <kaber@trash.net>
Cc: Sascha Hauer <s.hauer@pengutronix.de>
Cc: Matt Waddel <Matt.Waddel@freescale.com>
Cc: netdev@vger.kernel.org
Cc: Tim Sander <tim01@vlsi.informatik.tu-darmstadt.de>
Acked-by: Greg Ungerer <gerg@uclinux.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

fec: fix recursive locking of mii_lock

mii_discover_phy is only called by fec_enet_mii (via mip->mii_func).  So
&fep->mii_lock is already held and mii_discover_phy must not call
mii_queue which locks &fep->mii_lock, too.

This was noticed by lockdep:

=============================================
[ INFO: possible recursive locking detected ]
2.6.31-rc8-00038-g37d0892 #109
---------------------------------------------
swapper/1 is trying to acquire lock:
 (&fep->mii_lock){-.....}, at: [<c01569f8>] mii_queue+0x2c/0xcc

but task is already holding lock:
 (&fep->mii_lock){-.....}, at: [<c0156328>] fec_enet_interrupt+0x78/0x460

other info that might help us debug this:
2 locks held by swapper/1:
 #0:  (rtnl_mutex){+.+.+.}, at: [<c0183534>] rtnl_lock+0x18/0x20
 #1:  (&fep->mii_lock){-.....}, at: [<c0156328>] fec_enet_interrupt+0x78/0x460

stack backtrace:
Backtrace:
[<c00226fc>] (dump_backtrace+0x0/0x108) from [<c01eac14>] (dump_stack+0x18/0x1c)
 r6:c781d118 r5:c03e41d8 r4:00000001
[<c01eabfc>] (dump_stack+0x0/0x1c) from [<c005bae4>] (__lock_acquire+0x1a20/0x1a88)
[<c005a0c4>] (__lock_acquire+0x0/0x1a88) from [<c005bbac>] (lock_acquire+0x60/0x74)
[<c005bb4c>] (lock_acquire+0x0/0x74) from [<c01edda8>] (_spin_lock_irqsave+0x54/0x68)
 r7:60000093 r6:c01569f8 r5:c785e468 r4:00000000
[<c01edd54>] (_spin_lock_irqsave+0x0/0x68) from [<c01569f8>] (mii_queue+0x2c/0xcc)
 r7:c785e468 r6:c0156b24 r5:600a0000 r4:c785e000
[<c01569cc>] (mii_queue+0x0/0xcc) from [<c0156b78>] (mii_discover_phy+0x54/0xa8)
 r8:00000002 r7:00000032 r6:c785e000 r5:c785e360 r4:c785e000
[<c0156b24>] (mii_discover_phy+0x0/0xa8) from [<c0156354>] (fec_enet_interrupt+0xa4/0x460)
 r5:c785e360 r4:c077a170
[<c01562b0>] (fec_enet_interrupt+0x0/0x460) from [<c0066674>] (handle_IRQ_event+0x48/0x120)
[<c006662c>] (handle_IRQ_event+0x0/0x120) from [<c0068438>] (handle_level_irq+0x94/0x11c)
...

Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
Cc: Greg Ungerer <gerg@uclinux.org>
Cc: Ben Hutchings <ben@decadent.org.uk>
Cc: Patrick McHardy <kaber@trash.net>
Cc: Sascha Hauer <s.hauer@pengutronix.de>
Cc: Matt Waddel <Matt.Waddel@freescale.com>
Cc: netdev@vger.kernel.org
Cc: Tim Sander <tim01@vlsi.informatik.tu-darmstadt.de>
Acked-by: Greg Ungerer <gerg@uclinux.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

NET: Fix possible corruption in bpqether driver

The bpq ether driver is modifying the data art of the skb by first
dropping the KISS byte (a command byte for the radio) then prepending the
length + 4 of the remaining AX.25 packet to be transmitted as a little
endian 16-bit number.  If the high byte of the length has a different
value than the dropped KISS byte users of clones of the skb may observe
this as corruption.  This was observed with by running listen(8) -a which
uses a packet socket which clones transmit packets.  The corruption will
then typically be displayed for as a KISS "TX Delay" command for AX.25
packets in the range of 252..508 bytes or any other KISS command for
yet larger packets.

Fixed by using skb_cow to create a private copy should the skb be cloned.
Using skb_cow also allows us to cleanup the old logic to ensure sufficient
headroom in the skb.

While at it, replace a return of 0 from bpq_xmit with the proper constant
NETDEV_TX_OK which is now being used everywhere else in this function.

Affected: all 2.2, 2.4 and 2.6 kernels.

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
Reported-by: Jann Traschewski <jann@gmx.de>
Signed-off-by: David S. Miller <davem@davemloft.net>

WARNING: some request_irq() failures ignored in el2_open()

Request_irq() may fail in different ways, handle accordingly.

Signed-off-by: Roel Kluin <roel.kluin@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

tcp: replace hard coded GFP_KERNEL with sk_allocation

This fixed a lockdep warning which appeared when doing stress
memory tests over NFS:

inconsistent {RECLAIM_FS-ON-W} -> {IN-RECLAIM_FS-W} usage.

page reclaim => nfs_writepage => tcp_sendmsg => lock sk_lock

mount_root => nfs_root_data => tcp_close => lock sk_lock =>
tcp_send_fin => alloc_skb_fclone => page reclaim

David raised a concern that if the allocation fails in tcp_send_fin(), and it's
GFP_ATOMIC, we are going to yield() (which sleeps) and loop endlessly waiting
for the allocation to succeed.

But fact is, the original GFP_KERNEL also sleeps. GFP_ATOMIC+yield() looks
weird, but it is no worse the implicit sleep inside GFP_KERNEL. Both could
loop endlessly under memory pressure.

CC: Arnaldo Carvalho de Melo <acme@ghostprotocols.net>
CC: David S. Miller <davem@davemloft.net>
CC: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: Wu Fengguang <fengguang.wu@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

net/ethtool: Add support for the ethtool feature to flash firmware image from a specified file.

This patch adds support to flash a firmware image to a device using ethtool.
The driver gets the filename of the firmware image and flashes the image
using the request firmware path.

The region "on the chip" to be flashed can be specified by an option.
It is upto the device driver to enumerate the region number passed by ethtool,
to the region to be flashed.

The default behavior is to flash all the regions on the chip.

Signed-off-by: Ajit Khaparde <ajitk@serverengines.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

drivers: Kill now superfluous ->last_rx stores

The generic packet receive code takes care of setting
netdev->last_rx when necessary, for the sake of the
bonding ARP monitor.

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

ip: Report qdisc packet drops

Christoph Lameter pointed out that packet drops at qdisc level where not
accounted in SNMP counters. Only if application sets IP_RECVERR, drops
are reported to user (-ENOBUFS errors) and SNMP counters updated.

IP_RECVERR is used to enable extended reliable error message passing,
but these are not needed to update system wide SNMP stats.

This patch changes things a bit to allow SNMP counters to be updated,
regardless of IP_RECVERR being set or not on the socket.

Example after an UDP tx flood
# netstat -s
...
IP:
    1487048 outgoing packets dropped
...
Udp:
...
    SndbufErrors: 1487048

send() syscalls, do however still return an OK status, to not
break applications.

Note : send() manual page explicitly says for -ENOBUFS error :

 "The output queue for a network interface was full.
  This generally indicates that the interface has stopped sending,
  but may be caused by transient congestion.
  (Normally, this does not occur in Linux. Packets are just silently
  dropped when a device queue overflows.) "

This is not true for IP_RECVERR enabled sockets : a send() syscall
that hit a qdisc drop returns an ENOBUFS error.

Many thanks to Christoph, David, and last but not least, Alexey !

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

vlan: multiqueue vlan device

vlan devices are currently not multi-queue capable.

We can do that with a new rtnl_link_ops method,
get_tx_queues(), called from rtnl_create_link()

This new method gets num_tx_queues/real_num_tx_queues
from real device.

register_vlan_device() is also handled.

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

net: drop_monitor: make last_rx timestamp private

It was recently pointed out to me that the last_rx field of the
net_device structure wasn't updated regularly.  In fact only the
bonding driver really uses it currently.  Since the drop_monitor code
relies on the last_rx field to detect drops on recevie in hardware, We
need to find a more reliable way to rate limit our drop checks (so
that we don't check for drops on every frame recevied, which would be
inefficient.  This patch makes a last_rx timestamp that is private to
the drop monitor code and is updated for every device that we track.

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

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

ath9k: Reconfigure beacon timers after the scan is completed.

Signed-off-by: Vivek Natarajan <vnatarajan@atheros.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

cfg80211: fix looping soft lockup in find_ie()

The find_ie() function uses a size_t for the len parameter, and
directly uses len as a loop variable.  If any received packets
are malformed, it is possible for the decrease of len to overflow,
and since the result is unsigned, the loop will not terminate.
Change it to a signed int so the loop conditional works for
negative values.

This fixes the following soft lockup:

[38573.102007] BUG: soft lockup - CPU#0 stuck for 61s! [phy0:2230]
[38573.102007] Modules linked in: aes_i586 aes_generic fuse af_packet ipt_REJECT xt_tcpudp nf_conntrack_ipv4 nf_defrag_ipv4 xt_state iptable_filter ip_tables x_tables acpi_cpufreq binfmt_misc dm_mirror dm_region_hash dm_log dm_multipath dm_mod kvm_intel kvm uinput i915 arc4 ecb drm snd_hda_codec_idt ath5k snd_hda_intel hid_apple mac80211 usbhid appletouch snd_hda_codec snd_pcm ath cfg80211 snd_timer i2c_algo_bit ohci1394 video snd processor ieee1394 rfkill ehci_hcd sg sky2 backlight snd_page_alloc uhci_hcd joydev output ac thermal button battery sr_mod applesmc cdrom input_polldev evdev unix [last unloaded: scsi_wait_scan]
[38573.102007] irq event stamp: 2547724535
[38573.102007] hardirqs last  enabled at (2547724534): [<c1002ffc>] restore_all_notrace+0x0/0x18
[38573.102007] hardirqs last disabled at (2547724535): [<c10038f4>] apic_timer_interrupt+0x28/0x34
[38573.102007] softirqs last  enabled at (92950144): [<c103ab48>] __do_softirq+0x108/0x210
[38573.102007] softirqs last disabled at (92950274): [<c1348e74>] _spin_lock_bh+0x14/0x80
[38573.102007]
[38573.102007] Pid: 2230, comm: phy0 Tainted: G        W  (2.6.31-rc7-wl #8) MacBook1,1
[38573.102007] EIP: 0060:[<f8ea2d50>] EFLAGS: 00010292 CPU: 0
[38573.102007] EIP is at cmp_ies+0x30/0x180 [cfg80211]
[38573.102007] EAX: 00000082 EBX: 00000000 ECX: ffffffc1 EDX: d8efd014
[38573.102007] ESI: ffffff7c EDI: 0000004d EBP: eee2dc50 ESP: eee2dc3c
[38573.102007]  DS: 007b ES: 007b FS: 00d8 GS: 0000 SS: 0068
[38573.102007] CR0: 8005003b CR2: d8efd014 CR3: 01694000 CR4: 000026d0
[38573.102007] DR0: 00000000 DR1: 00000000 DR2: 00000000 DR3: 00000000
[38573.102007] DR6: ffff0ff0 DR7: 00000400
[38573.102007] Call Trace:
[38573.102007]  [<f8ea2f8d>] cmp_bss+0xed/0x100 [cfg80211]
[38573.102007]  [<f8ea33e4>] cfg80211_bss_update+0x84/0x410 [cfg80211]
[38573.102007]  [<f8ea3884>] cfg80211_inform_bss_frame+0x114/0x180 [cfg80211]
[38573.102007]  [<f97255ff>] ieee80211_bss_info_update+0x4f/0x180 [mac80211]
[38573.102007]  [<f972b118>] ieee80211_rx_bss_info+0x88/0xf0 [mac80211]
[38573.102007]  [<f9739297>] ? ieee802_11_parse_elems+0x27/0x30 [mac80211]
[38573.102007]  [<f972b224>] ieee80211_rx_mgmt_probe_resp+0xa4/0x1c0 [mac80211]
[38573.102007]  [<f972bc59>] ieee80211_sta_rx_queued_mgmt+0x919/0xc50 [mac80211]
[38573.102007]  [<c1009707>] ? sched_clock+0x27/0xa0
[38573.102007]  [<c1009707>] ? sched_clock+0x27/0xa0
[38573.102007]  [<c105ffd0>] ? mark_held_locks+0x60/0x80
[38573.102007]  [<c1348be5>] ? _spin_unlock_irqrestore+0x55/0x70
[38573.102007]  [<c134baa5>] ? sub_preempt_count+0x85/0xc0
[38573.102007]  [<c1348bce>] ? _spin_unlock_irqrestore+0x3e/0x70
[38573.102007]  [<c12c1c0f>] ? skb_dequeue+0x4f/0x70
[38573.102007]  [<f972c021>] ieee80211_sta_work+0x91/0xb80 [mac80211]
[38573.102007]  [<c1009707>] ? sched_clock+0x27/0xa0
[38573.102007]  [<c134baa5>] ? sub_preempt_count+0x85/0xc0
[38573.102007]  [<c10479af>] worker_thread+0x18f/0x320
[38573.102007]  [<c104794e>] ? worker_thread+0x12e/0x320
[38573.102007]  [<c1348be5>] ? _spin_unlock_irqrestore+0x55/0x70
[38573.102007]  [<f972bf90>] ? ieee80211_sta_work+0x0/0xb80 [mac80211]
[38573.102007]  [<c104cbb0>] ? autoremove_wake_function+0x0/0x50
[38573.102007]  [<c1047820>] ? worker_thread+0x0/0x320
[38573.102007]  [<c104c854>] kthread+0x84/0x90
[38573.102007]  [<c104c7d0>] ? kthread+0x0/0x90
[38573.102007]  [<c1003ab7>] kernel_thread_helper+0x7/0x10

Cc: stable@kernel.org
Signed-off-by: Bob Copeland <me@bobcopeland.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

wireless: remove mac80211 rate selection extra menu

We can just display this upon enabling mac80211 with an
'if MAC80211 != n' check.

Cc: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

wireless: update reg debug kconfig entry

Refer to the wireless wiki for more information.

Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

wireless: update top level wireless driver entry

Change it to a menuconfig to give it some documentation, to
refer users to our wireless wiki for extra resources and
documentation. It seems our wiki is still obscure to some.

Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

net: file_operations should be const

All instances of file_operations should be const.

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

inet: inet_connection_sock_af_ops const

The function block inet_connect_sock_af_ops contains no data
make it constant.

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

tcp: MD5 operations should be const

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

net: seq_operations should be const

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

netdev: drivers should make ethtool_ops const

No need to put ethtool_ops in data, they should be const.

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

tg3: Update version to 3.102

This patch updates the tg3 version to 3.102.

Signed-off-by: Matt Carlson <mcarlson@broadcom.com>
Reviewed-by: Benjamin Li <benli@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

tg3: Add MDIO bus address assignments

The 5717 is a dual port chip that has a shared MDIO bus design.  While
it is impossible for one function to interface with the wrong phy, that
function still needs to know which MDIO bus address to use when
interfacing with its own phy.  This patch adds code to determine which
MDIO bus address to use.

Signed-off-by: Matt Carlson <mcarlson@broadcom.com>
Reviewed-by: Benjamin Li <benli@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

tg3: Add 5717 NVRAM detection routines

This patch adds NVRAM detection routines for the 5717.

Signed-off-by: Matt Carlson <mcarlson@broadcom.com>
Reviewed-by: Benjamin Li <benli@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

tg3: Add 5717 asic rev

This patch adds the 5717 asic rev.

Signed-off-by: Matt Carlson <mcarlson@broadcom.com>
Reviewed-by: Benjamin Li <benli@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

tg3: Assign rx ret producer indexes by vector

When RSS is enabled, the status block format changes slightly.  The
"rx_jumbo_consumer", "reserved", and "rx_mini_consumer" members get
mapped to the other three rx return ring producer indexes.  This patch
introduces a new per-interrupt member which identifies which location
in the status block a particular vector should look for return ring
updates.

Signed-off-by: Matt Carlson <mcarlson@broadcom.com>
Reviewed-by: Benjamin Li <benli@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

tg3: Adjust RSS ring allocation strategies

When multivector RSS is enabled, the first interrupt vector is only used
to report link interrupts and error conditions.  This patch changes the
code so that rx and tx ring resources are not allocated for this vector.

Signed-off-by: Matt Carlson <mcarlson@broadcom.com>
Reviewed-by: Benjamin Li <benli@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

tg3: Add RSS support

This patch adds code needed to enable RSS.

Signed-off-by: Matt Carlson <mcarlson@broadcom.com>
Reviewed-by: Benjamin Li <benli@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

tg3: Add coalesce parameters for msix vectors

This patch adds code to tune the coalescing parameters for the other
msix vectors.

Signed-off-by: Matt Carlson <mcarlson@broadcom.com>
Reviewed-by: Benjamin Li <benli@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

tg3: Enable NAPI instances for other int vectors

This patch adds code to enable and disable the rest of the NAPI
instances.

Signed-off-by: Matt Carlson <mcarlson@broadcom.com>
Reviewed-by: Benjamin Li <benli@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

tg3: Add TSS support

This patch exposes the additional transmit rings to the kernel and makes
the necessary modifications to transmit, open, and close paths.

Signed-off-by: Matt Carlson <mcarlson@broadcom.com>
Reviewed-by: Benjamin Li <benli@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

tg3: Update intmbox and coal_now for msix

This patch fixes up two spots that need attention now that msix support
has been added.

Signed-off-by: Matt Carlson <mcarlson@broadcom.com>
Reviewed-by: Benjamin Li <benli@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

tg3: Add tx and rx ring resource tracking

This patch adds code to assign status block, tx producer ring and rx
return ring resources needed for the other interrupt vectors.

Signed-off-by: Matt Carlson <mcarlson@broadcom.com>
Reviewed-by: Benjamin Li <benli@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

tg3: Add mailbox assignments

The 5717 assigns mailbox locations to interrupt vectors in a rather
non-intuitive way.  (Much of the complexity stems from legacy
compatibility issues.)  This patch implements the assignment scheme.

Signed-off-by: Matt Carlson <mcarlson@broadcom.com>
Reviewed-by: Benjamin Li <benli@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

tg3: Add MSI-X support

This patch adds MSI-X support.

Signed-off-by: Matt Carlson <mcarlson@broadcom.com>
Reviewed-by: Benjamin Li <benli@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

tg3: Add support code around kernel interrupt API

This patch adds code to support multiple interrupt vectors around the
kernel's interrupt API.

Signed-off-by: Matt Carlson <mcarlson@broadcom.com>
Reviewed-by: Benjamin Li <benli@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

tg3: Create tg3_rings_reset()

This patch moves most of the chip ring setup logic into a separate
function.  This will make it easier to verify the multi ring setup
changes.

Signed-off-by: Matt Carlson <mcarlson@broadcom.com>
Reviewed-by: Benjamin Li <benli@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

tg3: Add per-int coalesce now member

Each interrupt vector has its own bit in the host coalescing register to
force that vector's status block to be updated and generate an
interrupt.  This patch adds a member to the per-interrupt structure
that records which bit belongs to that vector.

Signed-off-by: Matt Carlson <mcarlson@broadcom.com>
Reviewed-by: Benjamin Li <benli@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

tg3: inline tg3_cond_int()

This patch inlines the code of tg3_cond_int() into the function's only
callsite.  This prep work makes the following patch cleaner.

Signed-off-by: Matt Carlson <mcarlson@broadcom.com>
Reviewed-by: Benjamin Li <benli@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Merge branch 'master' of /linux/kernel/git/davem/net-2.6

Conflicts:
drivers/net/yellowfin.c

ipv6: ip6_push_pending_frames() should increment IPSTATS_MIB_OUTDISCARDS

qdisc drops should be notified to IP_RECVERR enabled sockets, as done in IPV4.

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

drop_monitor: fix trace_napi_poll_hit()

The net_dev of backlog napi is NULL, like below:

__get_cpu_var(softnet_data).backlog.dev == NULL

So, we should check it in napi tracepoint's probe function

Acked-by: Neil Horman <nhorman@tuxdriver.com>
Signed-off-by: Xiao Guangrong <xiaoguangrong@cn.fujitsu.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

pkt_sched: Revert tasklet_hrtimer changes.

These are full of unresolved problems, mainly that conversions don't
work 1-1 from hrtimers to tasklet_hrtimers because unlike hrtimers
tasklets can't be killed from softirq context.

And when a qdisc gets reset, that's exactly what we need to do here.

We'll work this out in the net-next-2.6 tree and if warranted we'll
backport that work to -stable.

This reverts the following 3 changesets:

a2cb6a4dd470d7a64255a10b843b0d188416b78f
("pkt_sched: Fix bogon in tasklet_hrtimer changes.")

38acce2d7983632100a9ff3fd20295f6e34074a8
("pkt_sched: Convert CBQ to tasklet_hrtimer.")

ee5f9757ea17759e1ce5503bdae2b07e48e32af9
("pkt_sched: Convert qdisc_watchdog to tasklet_hrtimer")

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

xilinx_emaclite: Fix permissions on driver sources.

Noticed by Michal Simek.

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

net: sk_free() should be allowed right after sk_alloc()

After commit 2b85a34e911bf483c27cfdd124aeb1605145dc80
(net: No more expensive sock_hold()/sock_put() on each tx)
sk_free() frees socks conditionally and depends
on sk_wmem_alloc being set e.g. in sock_init_data(). But in some
cases sk_free() is called earlier, usually after other alloc errors.

Fix is to move sk_wmem_alloc initialization from sock_init_data()
to sk_alloc() itself.

Signed-off-by: Jarek Poplawski <jarkao2@gmail.com>
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

net: make neigh_ops constant

These tables are never modified at runtime. Move to read-only
section.

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

au1000_eth: possible NULL dereference of aup->mii_bus->irq in au1000_probe()

aup->mii_bus->irq allocation may fail, prevent a dereference of NULL.

Signed-off-by: Roel Kluin <roel.kluin@gmail.com>
Acked-by: Florian Fainelli <florian@openwrt.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

RTO connection timeout: sysctl documentation update

This patch updates the sysctl documentation concerning the interpretation
of tcp_retries{1,2} and tcp_orphan_retries.

Signed-off-by: Damian Lukowski <damian@tvk.rwth-aachen.de>
Signed-off-by: David S. Miller <davem@davemloft.net>

RTO connection timeout: coding style fixes and comments

This patch affects the retransmits_timed_out() function.

Changes:
1) Variables have more meaningful names
2) retransmits_timed_out() has an introductionary comment.
3) Small coding style changes.

Signed-off-by: Damian Lukowski <damian@tvk.rwth-aachen.de>
Signed-off-by: David S. Miller <davem@davemloft.net>

sky2: Use 32bit read to read Y2_VAUX_AVAIL

B0_CTST is a 24bit register according to the vendor driver (sk98lin).
A 16bit read on B0_CTST will always return 0 for Y2_VAUX_AVAIL (1<<16),
 so use a 32bit read when testing Y2_VAUX_AVAIL

Signed-off-by: Mike McCormack <mikem@ring3k.org>
Acked-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

sky2: Create buffer alloc and free helpers

Refactor similar two sections of code that free buffers into one.
Only call tx_init if all buffer allocations succeed.

Signed-off-by: Mike McCormack <mikem@ring3k.org>
Acked-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

sky2: fix management of driver LED

Observed by Mike McCormack.

The LED bit here is just a software controlled value used to
turn on one of the LED's on some boards. The register value was wrong,
which could have been causing some power control issues.
Get rid of problematic define use the correct mask.

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

tun: reuse struct sock fields

As tun always has an embeedded struct sock,
use sk and sk_receive_queue fields instead of
duplicating them in tun_struct.

Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

netns: embed ip6_dst_ops directly

struct net::ipv6.ip6_dst_ops is separatedly dynamically allocated,
but there is no fundamental reason for it. Embed it directly into
struct netns_ipv6.

For that:
* move struct dst_ops into separate header to fix circular dependencies
I honestly tried not to, it's pretty impossible to do other way
* drop dynamical allocation, allocate together with netns

For a change, remove struct dst_ops::dst_net, it's deducible
by using container_of() given dst_ops pointer.

Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

bonding: use compare_ether_addr_64bits() in ALB

We can speedup ether addresses compares using compare_ether_addr_64bits()
instead of memcmp(). We make sure all operands are at least 8 bytes long and
16bits aligned (or better, long word aligned if possible)

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

macvlan: Use compare_ether_addr_64bits()

To speedup ether addresses compares, we can use compare_ether_addr_64bits()
(all operands are guaranteed to be at least 8 bytes long)

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

ixgbe: Patch to fix 82599 multispeed fiber link issues when driver is loaded without any cable and reconnecting it to 1G partner

In 82599 multi speed fiber case when driver is loaded without any
cable and reconnecting the cable with a 1G partner does not bring
up the link in 1Gb mode. When there is no link we first setup the link
at 10G & 1G and then try to re-establish the link at highest speed 10G
and thereby changing autoneg_advertised value to highest speed 10G.
After connecting back the cable to a 1G link partner we never try 1G
as autoneg advertised value is changed to link at 10G only. The
following patch fixes the issue by properly initializing the
autoneg_advertised value just before exiting from link setup routine.

Signed-off-by: Mallikarjuna R Chilakala <mallikarjuna.chilakala@intel.com>
Acked-by: Peter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

ixgbe: Properly disable DCB arbiters prior to applying changes

When disabling the Rx and Tx data arbiters prior to configuration changes,
the arbiters were not being shut down properly.  This can create a race
in the DCB hardware blocks, and potentially hang the arbiters.  Also, the
Tx descriptor arbiter shouldn't be disabled when applying configuration
changes; disabling this arbiter can cause a Tx hang.

Signed-off-by: Peter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

ixgbe: refactor link setup code

Link code cleanup: a number of redundant functions and MAC variables are cleaned up,
with some functions being consolidated into a single-purpose code path.
Removed following deprecated link functions and mac variables
 * ixgbe_setup_copper_link_speed_82598
 * ixgbe_setup_mac_link_speed_multispeed_fiber
 * ixgbe_setup_mac_link_speed_82599
 * mac.autoneg, mac.autoneg_succeeded, phy.autoneg_wait_to_complete

Signed-off-by: Mallikarjuna R Chilakala <mallikarjuna.chilakala@intel.com>
Acked-by: Peter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

e1000: Fix for e1000 kills IPMI on a tagged vlan.

Enabling VLAN filters (VFE) when the primary interface is brought up
(per commit 78ed11a) has caused problems for some users who manage
their systems using IPMI over a VLAN. This is because when the driver
enables the VLAN filter, this same filter table is enabled for the
management channel, and the table is initially empty, which means that
the IPMI/VLAN packets are filtered out and not received by the BMC.
This is a problem only on e1000 class adapters, as it is only
on e1000 that the filter table is common to the management and host
streams.

With this change, filtering is only enabled when one or more host VLANs
exist, and is disabled when the last host VLAN is removed. VLAN filtering
is always disabled when the primary interface is in promiscuous mode,
and will be (re)enabled if VLANs exist when the interface exits
promiscuous mode.

Note that this does not completely resolve the issue for those using VLAN
management, because if the host adds a VLAN, then the above problem
occurs when that VLAN is enabled. However, it does mean the there is no
problem for configurations where management is on a VLAN and the host is
not.

A complete solution to this issue would require further driver changes.
The driver would need to discover if (and which) management VLANs are
active before enabling VLAN filtering, so that it could ensure that the
managed VLANs are included in the VLAN filter table. This discovery
requires that the BMC identifies its VLAN in registers accessible
to the driver, and at least on Dell PE2850 systems the BMC does not
identify its VLAN to allow such discovery. Intel is pursuing this issue
with the BMC vendor.

Signed-off-by: Dave Graham <david.graham@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Tested-by: Krzysztof Piotr Oledzki <ole@ans.pl>
Signed-off-by: David S. Miller <davem@davemloft.net>

iwmc3200wifi: Add a last_fw_err debugfs entry

In order to check what was the last fw error we got accross resets, we add
this debugfs entry. It displays the complete ASSERT information.

Signed-off-by: Samuel Ortiz <sameo@linux.intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

iwmc3200wifi: Handle UMAC stalls and UMAC assert properly

When UMAC stalls or asserts, we want to reset the device. But when we're
associated, the current reset worker will end up calling
cfg80211_connect_result() with the cfg80211 sme layer knowing that we're
reassociating. That ends up with some ugly warnings.
With this patch we're telling the upper layer that we've roamed if
reassociation succeeds, and that we're disconnected if it fails.

Signed-off-by: Samuel Ortiz <sameo@linux.intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

iwmc3200wifi: New initial LMAC calibration

The LMAC calibration API got broken mostly by having a configuration bitmap
being different than the result one.
This patch tries to address that issue by correctly running calibrations with
the newest firmwares, and keeping a backward compatibility fallback path for
older firmwares, where the configuration and result bitmaps were identical.

Signed-off-by: Samuel Ortiz <sameo@linux.intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

iwmc3200wifi: fix misuse of le16_to_cpu

Also mark some functions static.

Signed-off-by: Zhu Yi <yi.zhu@intel.com>
Signed-off-by: Samuel Ortiz <sameo@linux.intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

iwmc3200wifi: add disconnect work

When the driver receives "connection terminated" event from device,
it could be caused by 2 reasons: the firmware is roaming or the
connection is lost (AP disappears). For the former, an association
complete event is supposed to come within 3 seconds. For the latter,
the driver won't receive any event except the connection terminated.
So we kick a delayed work (5*HZ) when we receive the connection
terminated event. It will be canceled if it turns out to be a roaming
event later. Otherwise we notify SME and userspace the disconnection.

Signed-off-by: Zhu Yi <yi.zhu@intel.com>
Signed-off-by: Samuel Ortiz <sameo@linux.intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

iwmc3200wifi: use cfg80211_roamed to send roam event

The device sends connection terminated and [re]association success
(or failure) events when roaming occours. The patch uses
cfg80211_roamed instead of cfg80211_connect_result to notify SME
for roaming.

Signed-off-by: Zhu Yi <yi.zhu@intel.com>
Signed-off-by: Samuel Ortiz <sameo@linux.intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

iwmc3200wifi: Fix sparse warning

iwm_cfg80211_get_station() should be static.

Signed-off-by: Samuel Ortiz <sameo@linux.intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

iwmc3200wifi: Set WEP key from connect

When connect is called with the LEGACY_PSK authentication type set, and a
proper sme->key, we need to set the WEP key straight after setting the
profile otherwise the authentication will never start.

Signed-off-by: Samuel Ortiz <sameo@linux.intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

iwmc3200wifi: invalidate profile when necessary before connect

If cfg80211 requests to connect when we have already had an active
profile, invalidate the current profile first before sending a new
profile to UMAC.

Signed-off-by: Zhu Yi <yi.zhu@intel.com>
Signed-off-by: Samuel Ortiz <sameo@linux.intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

rndis_wlan: remove 'select WIRELESS_EXT' in Kconfig

Since rndis_wlan is now converted to cfg80211, WIRELESS_EXT isn't
required anymore.

Signed-off-by: Jussi Kivilinna <jussi.kivilinna@mbnet.fi>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

rndis_wlan: fix sparse endianess warnings

Signed-off-by: Jussi Kivilinna <jussi.kivilinna@mbnet.fi>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

rndis_wlan: cleanup

- remove double newlines between functions
- remove commented out function (rndis_set_config_parameter_u32())
- coding style fix in rndis_set_config_parameter_str()
- add comment banners between function sections

Signed-off-by: Jussi Kivilinna <jussi.kivilinna@mbnet.fi>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

rndis_wlan: use bool for on/off switches

Signed-off-by: Jussi Kivilinna <jussi.kivilinna@mbnet.fi>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

ath9k: Call spin_lock_bh() on btcoex_lock

As generic hw timer interrupt handler is moved to tasklet,
we no more need to call spin_lock_irqsave().

Signed-off-by: Vasanthakumar Thiagarajan <vasanth@atheros.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

ath9k: Move generic hw timer intr handler to bottom-half

There is no point handling this in hard irq, move it to
tasklet.

Signed-off-by: Vasanthakumar Thiagarajan <vasanth@atheros.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

rfkill: relicense header file

This header file is copied into userspace tools that
need not be GPLv2 licensed, make that easier.

Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
Acked-by: Alan Jenkins <alan-jenkins@tuffmail.co.uk>
Acked-by: Henrique de Moraes Holschuh <hmh@hmh.eng.br>
Acked-by: Iñaky Pérez-González <inaky@linux.intel.com>
Acked-by: Ivo van Doorn <IvDoorn@gmail.com>
Acked-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Acked-by: Michael Buesch <mb@bu3sch.de>
Acked-by: Tomas Winkler <tomas.winkler@intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

Revert Backoff [v3]: Calculate TCP's connection close threshold as a time value.

RFC 1122 specifies two threshold values R1 and R2 for connection timeouts,
which may represent a number of allowed retransmissions or a timeout value.
Currently linux uses sysctl_tcp_retries{1,2} to specify the thresholds
in number of allowed retransmissions.

For any desired threshold R2 (by means of time) one can specify tcp_retries2
(by means of number of retransmissions) such that TCP will not time out
earlier than R2. This is the case, because the RTO schedule follows a fixed
pattern, namely exponential backoff.

However, the RTO behaviour is not predictable any more if RTO backoffs can be
reverted, as it is the case in the draft
"Make TCP more Robust to Long Connectivity Disruptions"
(http://tools.ietf.org/html/draft-zimmermann-tcp-lcd).

In the worst case TCP would time out a connection after 3.2 seconds, if the
initial RTO equaled MIN_RTO and each backoff has been reverted.

This patch introduces a function retransmits_timed_out(N),
which calculates the timeout of a TCP connection, assuming an initial
RTO of MIN_RTO and N unsuccessful, exponentially backed-off retransmissions.

Whenever timeout decisions are made by comparing the retransmission counter
to some value N, this function can be used, instead.

The meaning of tcp_retries2 will be changed, as many more RTO retransmissions
can occur than the value indicates. However, it yields a timeout which is
similar to the one of an unpatched, exponentially backing off TCP in the same
scenario. As no application could rely on an RTO greater than MIN_RTO, there
should be no risk of a regression.

Signed-off-by: Damian Lukowski <damian@tvk.rwth-aachen.de>
Acked-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>

Revert Backoff [v3]: Revert RTO on ICMP destination unreachable

Here, an ICMP host/network unreachable message, whose payload fits to
TCP's SND.UNA, is taken as an indication that the RTO retransmission has
not been lost due to congestion, but because of a route failure
somewhere along the path.
With true congestion, a router won't trigger such a message and the
patched TCP will operate as standard TCP.

This patch reverts one RTO backoff, if an ICMP host/network unreachable
message, whose payload fits to TCP's SND.UNA, arrives.
Based on the new RTO, the retransmission timer is reset to reflect the
remaining time, or - if the revert clocked out the timer - a retransmission
is sent out immediately.
Backoffs are only reverted, if TCP is in RTO loss recovery, i.e. if
there have been retransmissions and reversible backoffs, already.

Changes from v2:
1) Renaming of skb in tcp_v4_err() moved to another patch.
2) Reintroduced tcp_bound_rto() and __tcp_set_rto().
3) Fixed code comments.

Signed-off-by: Damian Lukowski <damian@tvk.rwth-aachen.de>
Acked-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>

Revert Backoff [v3]: Rename skb to icmp_skb in tcp_v4_err()

This supplementary patch renames skb to icmp_skb in tcp_v4_err() in order to
disambiguate from another sk_buff variable, which will be introduced
in a separate patch.

Signed-off-by: Damian Lukowski <damian@tvk.rwth-aachen.de>
Acked-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>

ixgbe: Add support for dcbnl_rtnl_ops.setapp/getapp

Add support for dcbnl_rtnl_ops.setapp/getapp to set or get the current user
priority bitmap for the given application protocol. Currently, 82599 only
supports setapp/getapp for Fiber Channel over Ethernet (FCoE) protocol.

Signed-off-by: Yi Zou <yi.zou@intel.com>
Acked-by: Peter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

dcbnl: Add implementations of dcbnl setapp/getapp commands

Implements the dcbnl netlink setapp/getapp pair. When a setapp/getapp
is received, dcbnl would just pass on to dcbnl_rtnl_op.setapp/getapp
that are supposed to be implemented by the low level drivers.

Signed-off-by: Yi Zou <yi.zou@intel.com>
Acked-by: Peter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>