This makes use of the local fc variable in bit more places and uses a
common helper macro. The part of RX process that delivers skb's to
mac80211 is moved to a separate function in preparation for future
changes that will need to do this from two places. The modifications
here should not result in any functional changes.
Signed-off-by: NJouni Malinen <jouni.malinen@atheros.com>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

The broadcast bit is in the first, not the last octet..
Signed-off-by: NJouni Malinen <jouni.malinen@atheros.com>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

While the probe request poll is expected to work, it looks like it
does not always result in getting a response. The exact reason for
this is unclear, but anyway, if we do receive a Beacon frame from our
AP, there is no need to disconnect based on the probereq poll. This
seems to help keep the connection bit more stable in cases where
beacon loss is occurring semi-frequently.
Signed-off-by: NJouni Malinen <jouni.malinen@atheros.com>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

According to my tests, all that ZD_CS_MULTICAST does is to
disable retrying/waiting for an ACK. Reflect this by renaming
the bit to ZD_CS_NO_ACK and setting it based on
IEEE80211_TX_CTL_NO_ACK, instead of is_multicast_ether_addr.
Signed-off-by: NGábor Stefanik <netrolller.3d@gmail.com>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

The STA may drop the very first frame if it happens to be a retried
frame. This is because we maintian the last received sequence number
per TID for QoS frames and it is initialized to zero through kzalloc
during sta_info_alloc and the sequence number of the very first date
frame received would be ZERO (as per IEEE 802.11-2007, 7.1.3.4.1).

If the frame dropped happens to be an EAP Request Identity(very first
frame from the AP), then wpa_supplicnat disconnects the STA and the
whole procedure starts again.
Signed-off-by: NSenthil Balasubramanian <senthilkumar@atheros.com>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

Here's a screenshot of what this looks like with ath9k:

mcgrof@pogo /debug/ieee80211/phy0 $ cat ht40allow_map
2412 HT40  +
2417 HT40  +
2422 HT40  +
2427 HT40  +
2432 HT40 -+
2437 HT40 -+
2442 HT40 -+
2447 HT40 -
2452 HT40 -
2457 HT40 -
2462 HT40 -
2467 Disabled
2472 Disabled
2484 Disabled
5180 HT40  +
5200 HT40 -+
5220 HT40 -+
5240 HT40 -+
5260 HT40 -+
5280 HT40 -+
5300 HT40 -+
5320 HT40 -
5500 HT40  +
5520 HT40 -+
5540 HT40 -+
5560 HT40 -+
5580 HT40 -+
5600 HT40 -+
5620 HT40 -+
5640 HT40 -+
5660 HT40 -+
5680 HT40 -+
5700 HT40 -
5745 HT40  +
5765 HT40 -+
5785 HT40 -+
5805 HT40 -+
5825 HT40 -
Signed-off-by: NLuis R. Rodriguez <lrodriguez@atheros.com>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

This moves the cfg80211 specific stuff to new cfg80211 debugfs
entries. Non-mac80211 will also get these entries now. There were
only 4 which we take:

rts_threshold
fragmentation_threshold
short_retry_limit
long_retry_limit
Signed-off-by: NLuis R. Rodriguez <lrodriguez@atheros.com>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

Thanks to nl80211 userspace can be very specific upon device
configuration. Before processing the request for the new HT40
channel types (HT40- or HT40+) we need to ensure we can use them
regulatory-wise. This wasn't required with wireless extensions as
specifying the channel type wasn't not available and configuration
was done towards the end implicitly upon association or reception
of beacons from the AP. For the new nl80211 we have to check this
when configuring the interfaces explicitly.
Signed-off-by: NLuis R. Rodriguez <lrodriguez@atheros.com>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

We weren't checking this at all.
Signed-off-by: NLuis R. Rodriguez <lrodriguez@atheros.com>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

This is more consistent with our nl80211 naming convention
for HT40-/+.
Signed-off-by: NLuis R. Rodriguez <lrodriguez@atheros.com>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

We are not correctly listening to the regulatory max bandwidth
settings. To actually make use of it we need to redesign things
a bit. This patch does the work for that. We do this to so we
can obey to regulatory rules accordingly for use of HT40.

We end up dealing with HT40 by having two passes for each channel.

The first check will see if a 20 MHz channel fits into the channel's
center freq on a given frequency range. We check for a 20 MHz
banwidth channel as that is the maximum an individual channel
will use, at least for now. The first pass will go ahead and
check if the regulatory rule for that given center of frequency
allows 40 MHz bandwidths and we use this to determine whether
or not the channel supports HT40 or not. So to support HT40 you'll
need at a regulatory rule that allows you to use 40 MHz channels
but you're channel must also be enabled and support 20 MHz by itself.

The second pass is done after we do the regulatory checks over
an device's supported channel list. On each channel we'll check
if the control channel and the extension both:

 o exist
 o are enabled
 o regulatory allows 40 MHz bandwidth on its frequency range

This work allows allows us to idependently check for HT40- and
HT40+.
Signed-off-by: NLuis R. Rodriguez <lrodriguez@atheros.com>
Signed-off-by: NJohn W. Linville <linville@tuxdriver.com>

GRO/LRO can be controlled through ethtool so this is unnecessary.
Signed-off-by: NBen Hutchings <bhutchings@solarflare.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

be sent periodically. The rs_delay can be speficied when adding the
PRL entry and defaults to 15 minutes.

The RS is sent from every link local adress that's assigned to the
tunnel interface. It's directed to the (guessed) linklocal address
of the router and is sent through the tunnel.

Better: send to ff02::2 encapsuled in unicast directed to router-v4.
Signed-off-by: NSascha Hlusiak <contact@saschahlusiak.de>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

A tunnel with no local ipv4 endpoint would otherwise use the
ISATAP linklocal address fe80::5efe:0:0, which is invalid. Rather not
add a linklocal address at all.
Signed-off-by: NSascha Hlusiak <contact@saschahlusiak.de>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Typo. When deleting a PRL entry, return status to userspace
instead of success.
Signed-off-by: NSascha Hlusiak <contact@saschahlusiak.de>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Check link device when looking up a tunnel. When a tunnel is
linked to a interface, traffic from a different interface must not
reach the tunnel.

This also allows creating of multiple tunnels with the same
endpoints, if the link device differs.
Signed-off-by: NSascha Hlusiak <contact@saschahlusiak.de>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

When locating the tunnel, do not continue if it is found. Otherwise
a different tunnel with similar configuration would be returned and
parts could be overwritten.
Signed-off-by: NSascha Hlusiak <contact@saschahlusiak.de>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The DHCP spec allows the server to specify the MTU.  This can be useful
for netbooting with UDP-based NFS-root on a network using jumbo frames.
This patch allows the kernel IP autoconfiguration to handle this option
correctly.

It would be possible to use initramfs and add a script to set the MTU,
but that seems like a complicated solution if no initramfs is otherwise
necessary, and would bloat the kernel image more than this code would.

This patch was originally submitted to LKML in 2003 by Hans-Peter Jansen.
Signed-off-by: NChris Friesen <cfriesen@nortel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

nlmsg_new() adds the size of the netlink header to the value
that has been passed as parameter. If NLMSG_GOODSIZE is selected,
we request an allocation of one memory page plus the size of the
header. Instead, NLMSG_DEFAULT_SIZE should be used since it
already substracts the size of the Netlink header.

I have the impression that the similar naming in both constant
is error prone when using it with nlmsg_new(). This is already
documented in include/net/netlink.h
Signed-off-by: NPablo Neira Ayuso <pablo@netfilter.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Update myri10ge driver version to 1.5.0-1.415.
Signed-off-by: NBrice Goglin <brice@myri.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Allow myri10ge LRO to be enabled/disabled via ethtool
(and by the stack for packet forwarding).
Signed-off-by: NBrice Goglin <brice@myri.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

We can slightly reduce size of teqlN structure, not duplicating stats
structure in teql_master but using stats field from net_device.stats
for tx_errors and from netdev_queue for tx_bytes/tx_packets/tx_dropped
values.
Signed-off-by: NEric Dumazet <dada1@cosmosbay.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This is purely a cleanup patch.  This collapses some of the code required
when we configure our Tx and Rx feature sets, and makes the code more
readable and maintainable.
Signed-off-by: NPeter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com>
Signed-off-by: NJesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The SFF specification for Direct Attach cable detection has now been
ratified.  Previously, DA cable detect was looking at the Twinaxial bit in
byte 9 of the SFP+ EEPROM.  The spec now defines active and passive DA
cables in byte 8 of the SFP+ EEPROM.  This patch changes the cable
detection for both 82598 and 82599 SFP+ adapters to conform to the new
spec.
Signed-off-by: NPeter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The SFP+ NIC (device id 0x10fb) needs a semaphore to serialize
PHY access, so our PHY init code must honor that same semaphore.
Signed-off-by: NPeter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Due to mostly historic reasons, including a lack of reliability
of the link handling (especially with the older 8169), the
current r8169 driver emulates forced mode setting by limiting
the advertised modes.

With this change the driver allows real 10/100 forced mode
settings on the 8169 and 8101/8102.

Original idea by Vincent Steenhoute. The RTL_GIGA_MAC_VER_03
tweak was extracted from Realtek's r8169 v6.010.00 driver.
Signed-off-by: NFrancois Romieu <romieu@fr.zoreil.com>
Tested-by: NJean Delvare <jdelvare@suse.de>
Cc: Edward Hsu <edward_hsu@realtek.com.tw>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Signed-off-by: NFrancois Romieu <romieu@fr.zoreil.com>
Cc: Edward Hsu <edward_hsu@realtek.com.tw>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Jarek pointed pppoe can call back dev_queue_xmit(), and might need
skb->dst, so its safer to unset IFF_XMIT_DST_RELEASE on ppp devices.
Signed-off-by: NEric Dumazet <dada1@cosmosbay.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

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

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

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

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

List of devices that must clear this flag is :

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

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

Sysfs files for a network device can not unconditionally take the
rtnl_lock as the bonding sysfs files do.  If someone accesses those
sysfs files while the network device is being unregistered with the
rtnl_lock held we will deadlock.

So use trylock and restart_syscall to avoid this problem.
Signed-off-by: NEric W. Biederman <ebiederm@aristanetworks.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Network device sysfs files that grab the rtnl_lock unconditionally
will deadlock if accessed when the network device is being
unregistered.  So use trylock and syscall_restart to avoid this
deadlock.
Signed-off-by: NEric W. Biederman <ebiederm@aristanetworks.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Holding rtnl_lock when we are unregistering the sysfs files can
deadlock if we unconditionally take rtnl_lock in a sysfs file.  So fix
it with the now familiar patter of: rtnl_trylock and syscall_restart()
Signed-off-by: NEric W. Biederman <ebiederm@aristanetworks.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

sysctls are unregistered with the rntl_lock held making
it unsafe to unconditionally grab the the rtnl_lock.  Instead
we need to call rtnl_trylock and restart the system call
if we can not grab it.  Otherwise we could deadlock at unregistration
time.
Signed-off-by: NEric W. Biederman <ebiederm@aristanetworks.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Just returning -ERESTARTSYS without a signal pending is not
good that will just leak it to userspace.  We need return
-ERESTARTNOINTR so we always restart and set signal pending
so that we fall of the fast path of syscall return and setup
the system call restart.

So use restart_syscall() which does all of this for us.
Signed-off-by: NEric W. Biederman <ebiederm@aristanetworks.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The earlier patch to fix the deadlock between a network device going
away and writing to sysfs attributes was incomplete.
- It did not set signal_pending so we would leak ERSTARTSYS to user space.
- It used ERESTARTSYS which only restarts if sigaction configures it to.
- It did not cover store and show for ifalias.

So fix all of these up and use the new helper restart_syscall so we get
the details correct on what it takes.
Signed-off-by: NEric W. Biederman <ebiederm@aristanetworks.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Currently when we have a signal pending we have the functionality
to restart that the current system call.  There are other cases
such as nasty lock ordering issues where it makes sense to have
a simple fix that uses try lock and restarts the system call.
Buying time to figure out how to rework the locking strategy.
Signed-off-by: NEric W. Biederman <ebiederm@aristanetworks.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

New packet socket feature that makes packet socket more efficient for
transmission.

- It reduces number of system call through a PACKET_TX_RING mechanism,
  based on PACKET_RX_RING (Circular buffer allocated in kernel space
  which is mmapped from user space).

- It minimizes CPU copy using fragmented SKB (almost zero copy).
Signed-off-by: NJohann Baudy <johann.baudy@gnu-log.net>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The pdev->irq was not saved in netxen_adapter, causing request_irq()
with invalid irq number.

This was broken in commit be339aee
("netxen: fix irq tear down and msix leak.").
Signed-off-by: NDhananjay Phadke <dhananjay@netxen.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Conflicts:
	drivers/scsi/fcoe/fcoe.c

gen_estimator can overflow bps (bytes per second) with Gb links, while
it was designed with a u32 API, with a theorical limit of 34360Mbit
(2^32 bytes)

Using 64 bit intermediate avbps/brate counters can allow us to reach
this theorical limit.
Signed-off-by: NEric Dumazet <dada1@cosmosbay.com>
Signed-off-by: NJarek Poplawski <jarkao2@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>