ethtool.h: MDI setting support

This change modifies the core ethtool struct to allow a driver to
support setting of MDI/MDI-X state for twisted pair wiring.  This
change uses a previously reserved u8 and should not change any
binary compatibility of ethtool.

Also as per Ben Hutchings' suggestion, the capabilities are
stored in a separate byte so the driver can report if it supports
changing settings.

see thread: http://kerneltrap.org/mailarchive/linux-netdev/2010/11/17/6289820/thread

see ethtool patches titled:
ethtool: allow setting MDI-X state

Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
CC: Ben Hutchings <bhutchings@solarflare.com>
Tested-by: Aaron Brown aaron.f.brown@intel.com
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

team: add support for non-ethernet devices

This is resolved by two things:
1) allow dev_addr of different length than ETH_ALEN
2) during port add, check for dev->type and change it if necessary

Signed-off-by: Jiri Pirko <jiri@resnulli.us>
Signed-off-by: David S. Miller <davem@davemloft.net>

Merge branch 'master' of git://git./linux/kernel/git/ppwaskie/net-next

Peter P Waskiewicz Jr says:

====================
This series contains multiple updates to the ixgbe driver.

The following are changes since commit 02644a17457414f38e29f32d5c640b06d08fa092:
    sctp: fix bogus if statement in sctp_auth_recv_cid()

and are available in the git repository at:
  git://git.kernel.org/pub/scm/linux/kernel/git/ppwaskie/net-next master

Alexander Duyck (9):
  ixgbe: Remove code that was initializing Rx page offset
  ixgbe: combine ixgbe_add_rx_frag and ixgbe_can_reuse_page
  ixgbe: Only use double buffering if page size is less than 8K
  ixgbe: Have the CPU take ownership of the buffers sooner
  ixgbe: Make pull tail function separate from rest of cleanup_headers
  ixgbe: Copybreak sooner to avoid get_page/put_page and offset change
    overhead
  ixgbe: Make allocating skb and placing data in it a separate function
  ixgbe: Roll RSC code into non-EOP code
  ixgbe: Rewrite code related to configuring IFCS bit in Tx descriptor
====================

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

tipc: eliminate configuration for maximum number of name publications

Gets rid of the need for users to specify the maximum number of
name publications supported by TIPC. TIPC now automatically provides
support for the maximum number of name publications to 65535.

Signed-off-by: Ying Xue <ying.xue@windriver.com>
Signed-off-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

tipc: eliminate configuration for maximum number of name subscriptions

Gets rid of the need for users to specify the maximum number of
name subscriptions supported by TIPC. TIPC now automatically provides
support for the maximum number of name subscriptions to 65535.

Signed-off-by: Ying Xue <ying.xue@windriver.com>
Signed-off-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

tipc: add __read_mostly annotations to several global variables

Added to the following:

 - tipc_random
 - tipc_own_addr
 - tipc_max_ports
 - tipc_net_id
 - tipc_remote_management
 - handler_enabled

The above global variables are read often, but written rarely. Use
__read_mostly to prevent them being on the same cacheline as another
variable which is written to often, which would cause cacheline
bouncing.

Signed-off-by: Ying Xue <ying.xue@windriver.com>
Signed-off-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

tipc: convert tipc_nametbl_size type from variable to macro

There is nothing changing this variable dynamically, so change
it to a macro to make that more obvious when reading the code.

Signed-off-by: Ying Xue <ying.xue@windriver.com>
Signed-off-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

tipc: change tipc_net_start routine return value type

Since now tipc_net_start() always returns a success code - 0, its
return value type should be changed from integer to void, which can
avoid unnecessary check for its return value.

Signed-off-by: Ying Xue <ying.xue@windriver.com>
Signed-off-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

tipc: manually inline single use media_name_valid routine

After eliminating the mechanism which checks whether all letters
in media name string are within a given character set, the
media_name_valid routine becomes trivial.  It is also only
used once, so it is unnecessary to keep it as a separate function.

Signed-off-by: Ying Xue <ying.xue@windriver.com>
Signed-off-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

tipc: remove pointless name sanity check and tipc_alphabet array

There is no real reason to check whether all letters in the given
media name and network interface name are within the character set
defined in tipc_alphabet array. Even if we eliminate the checking,
the rest of checking conditions in tipc_enable_bearer() can ensure
we do not enable an invalid or illegal bearer.

Signed-off-by: Ying Xue <ying.xue@windriver.com>
Signed-off-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

tipc: fix lockdep warning during bearer initialization

When the lockdep validator is enabled, it will report the below
warning when we enable a TIPC bearer:

[ INFO: possible irq lock inversion dependency detected ]
---------------------------------------------------------
Possible interrupt unsafe locking scenario:

        CPU0                    CPU1
        ----                    ----
   lock(ptype_lock);
                                local_irq_disable();
                                lock(tipc_net_lock);
                                lock(ptype_lock);
   <Interrupt>
   lock(tipc_net_lock);

  *** DEADLOCK ***

the shortest dependencies between 2nd lock and 1st lock:
  -> (ptype_lock){+.+...} ops: 10 {
[...]
SOFTIRQ-ON-W at:
                      [<c1089418>] __lock_acquire+0x528/0x13e0
                      [<c108a360>] lock_acquire+0x90/0x100
                      [<c1553c38>] _raw_spin_lock+0x38/0x50
                      [<c14651ca>] dev_add_pack+0x3a/0x60
                      [<c182da75>] arp_init+0x1a/0x48
                      [<c182dce5>] inet_init+0x181/0x27e
                      [<c1001114>] do_one_initcall+0x34/0x170
                      [<c17f7329>] kernel_init+0x110/0x1b2
                      [<c155b6a2>] kernel_thread_helper+0x6/0x10
[...]
   ... key      at: [<c17e4b10>] ptype_lock+0x10/0x20
   ... acquired at:
    [<c108a360>] lock_acquire+0x90/0x100
    [<c1553c38>] _raw_spin_lock+0x38/0x50
    [<c14651ca>] dev_add_pack+0x3a/0x60
    [<c8bc18d2>] enable_bearer+0xf2/0x140 [tipc]
    [<c8bb283a>] tipc_enable_bearer+0x1ba/0x450 [tipc]
    [<c8bb3a04>] tipc_cfg_do_cmd+0x5c4/0x830 [tipc]
    [<c8bbc032>] handle_cmd+0x42/0xd0 [tipc]
    [<c148e802>] genl_rcv_msg+0x232/0x280
    [<c148d3f6>] netlink_rcv_skb+0x86/0xb0
    [<c148e5bc>] genl_rcv+0x1c/0x30
    [<c148d144>] netlink_unicast+0x174/0x1f0
    [<c148ddab>] netlink_sendmsg+0x1eb/0x2d0
    [<c1456bc1>] sock_aio_write+0x161/0x170
    [<c1135a7c>] do_sync_write+0xac/0xf0
    [<c11360f6>] vfs_write+0x156/0x170
    [<c11361e2>] sys_write+0x42/0x70
    [<c155b0df>] sysenter_do_call+0x12/0x38
[...]
}
  -> (tipc_net_lock){+..-..} ops: 4 {
[...]
    IN-SOFTIRQ-R at:
                     [<c108953a>] __lock_acquire+0x64a/0x13e0
                     [<c108a360>] lock_acquire+0x90/0x100
                     [<c15541cd>] _raw_read_lock_bh+0x3d/0x50
                     [<c8bb874d>] tipc_recv_msg+0x1d/0x830 [tipc]
                     [<c8bc195f>] recv_msg+0x3f/0x50 [tipc]
                     [<c146a5fa>] __netif_receive_skb+0x22a/0x590
                     [<c146ab0b>] netif_receive_skb+0x2b/0xf0
                     [<c13c43d2>] pcnet32_poll+0x292/0x780
                     [<c146b00a>] net_rx_action+0xfa/0x1e0
                     [<c103a4be>] __do_softirq+0xae/0x1e0
[...]
}

>From the log, we can see three different call chains between
CPU0 and CPU1:

Time 0 on CPU0:

  kernel_init()->inet_init()->dev_add_pack()

At time 0, the ptype_lock is held by CPU0 in dev_add_pack();

Time 1 on CPU1:

  tipc_enable_bearer()->enable_bearer()->dev_add_pack()

At time 1, tipc_enable_bearer() first holds tipc_net_lock, and then
wants to take ptype_lock to register TIPC protocol handler into the
networking stack.  But the ptype_lock has been taken by dev_add_pack()
on CPU0, so at this time the dev_add_pack() running on CPU1 has to be
busy looping.

Time 2 on CPU0:

  netif_receive_skb()->recv_msg()->tipc_recv_msg()

At time 2, an incoming TIPC packet arrives at CPU0, hence
tipc_recv_msg() will be invoked. In tipc_recv_msg(), it first wants
to hold tipc_net_lock.  At the moment, below scenario happens:

On CPU0, below is our sequence of taking locks:

  lock(ptype_lock)->lock(tipc_net_lock)

On CPU1, our sequence of taking locks looks like:

  lock(tipc_net_lock)->lock(ptype_lock)

Obviously deadlock may happen in this case.

But please note the deadlock possibly doesn't occur at all when the
first TIPC bearer is enabled.  Before enable_bearer() -- running on
CPU1 does not hold ptype_lock, so the TIPC receive handler (i.e.
recv_msg()) is not registered successfully via dev_add_pack(), so
the tipc_recv_msg() cannot be called by recv_msg() even if a TIPC
message comes to CPU0. But when the second TIPC bearer is
registered, the deadlock can perhaps really happen.

To fix it, we will push the work of registering TIPC protocol
handler into workqueue context. After the change, both paths taking
ptype_lock are always in process contexts, thus, the deadlock should
never occur.

Signed-off-by: Ying Xue <ying.xue@windriver.com>
Signed-off-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

tipc: optimize the initialization of network device notifier

Ethernet media initialization is only done when TIPC is started or
switched to network mode. So the initialization of the network device
notifier structure can be moved out of this function and done
statically instead.

Signed-off-by: Ying Xue <ying.xue@windriver.com>
Signed-off-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

packet: Report fanout status via diag engine

Reported value is the same reported by the FANOUT getsockoption, but
unlike it, the absent fanout setup results in absent nlattr, rather
than in nlattr with zero value. This is done so, since zero fanout
report may mean both -- no fanout, and fanout with both id and type zero.

Signed-off-by: Pavel Emelyanov <xemul@parallels.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

packet: Report rings cfg via diag engine

One extension bit may result in two nlattrs -- one per ring type.
If some ring type is not configured, then the respective nlatts
will be empty.

The structure reported contains the data, that is given to the
corresponding ring setup socket option.

Signed-off-by: Pavel Emelyanov <xemul@parallels.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

gre: information leak in ip6_tnl_ioctl()

There is a one byte hole between p->hop_limit and p->flowinfo where
stack memory is leaked to the user.  This was introduced in c12b395a46
"gre: Support GRE over IPv6".

Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>

xfrm: Use rcu_dereference_bh to deference pointer protected by rcu_read_lock_bh

Signed-off-by: Fan Du <fan.du@windriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

ixgbe: Rewrite code related to configuring IFCS bit in Tx descriptor

This change updates the code related to configuring the transmit frame
checksum.  Specifically I have updated the code so that we can only skip
inserting the checksum in the case that we are not performing some other
offload that will modify the frame data.

Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
Tested-by: Phil Schmitt <phillip.j.schmitt@intel.com>
Signed-off-by: Peter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com>

ixgbe: Roll RSC code into non-EOP code

This change moves the RSC code into the non-EOP descriptor handling
function. The main motivation behind this change is to help reduce the
overhead in the non-RSC case. Previously the non-RSC path code would
always be checking for append count even if RSC had been disabled. Now
this code is completely skipped in a single conditional check instead of
having to make two separate checks.

Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
Tested-by: Phil Schmitt <phillip.j.schmitt@intel.com>
Signed-off-by: Peter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com>

ixgbe: Make allocating skb and placing data in it a separate function

This patch creates a function named ixgbe_fetch_rx_buffer. The sole
purpose of this function is to retrieve a single buffer off of the ring and
to place it in an skb.

The advantage to doing this is that it helps improve the readability since
I can decrease the indentation and for the code in this section.

Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
Tested-by: Phil Schmitt <phillip.j.schmitt@intel.com>
Signed-off-by: Peter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com>

ixgbe: Copybreak sooner to avoid get_page/put_page and offset change overhead

This change makes it so that if only the first 256 bytes of a buffer are
used we just copy the data out and leave the offset and page count
unchanged. There are multiple advantages to this. First it allows us to
reuse the page much more in the case of pages larger than 4K. It also
allows us to avoid some expensive atomic operations in the form of
get_page/put_page. In perf I have seen CPU utilization for put_page drop
from 3.5% to 1.8% as a result of this patch when doing small packet routing,
and packet rates increased by about 3%.

Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
Tested-by: Phil Schmitt <phillip.j.schmitt@intel.com>
Signed-off-by: Peter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com>

ixgbe: Make pull tail function separate from rest of cleanup_headers

This change creates a separate function for functionality similar to
pskb_pull_tail.  The main motivation for moving it to a separate function
is so that later I can just skip this function in the case where we have
already copied the buffer into skb->head.

Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
Tested-by: Phil Schmitt <phillip.j.schmitt@intel.com>
Signed-off-by: Peter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com>

ixgbe: Have the CPU take ownership of the buffers sooner

This patch makes it so that we will always have ownership of the buffers by
the time we get to ixgbe_add_rx_frag. This is necessary as I am planning to
add a copy-break to ixgbe_add_rx_frag and in order for that to function
correctly we need the CPU to have ownership of the buffer.

Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
Tested-by: Phil Schmitt <phillip.j.schmitt@intel.com>
Signed-off-by: Peter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com>

ixgbe: Only use double buffering if page size is less than 8K

This change makes it so that we do not use double buffering if the page
size is larger than 4K.  Instead we will simply walk through the page using
up to 3K per receive, and if we receive less than we only move the offset
by that amount.  We will free the page when there is no longer any space
left that we can use instead of checking the page count to see if we can
cycle back to the start.

The main motivation behind this is to avoid the unnecessary truesize cost
for using a half page when most packets are 2K or smaller. With this new
approach the largest possible truesize for a page fragment will be 3K when
PAGE_SIZE is larger than 4K.

Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
Tested-by: Phil Schmitt <phillip.j.schmitt@intel.com>
Signed-off-by: Peter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com>

ixgbe: combine ixgbe_add_rx_frag and ixgbe_can_reuse_page

This patch combines ixgbe_add_rx_frag and ixgbe_can_reuse_page into a
single function. The main motivation behind this is to make better use of
the values so that we don't have to load them from memory and into
registers twice.

Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
Tested-by: Phil Schmitt <phillip.j.schmitt@intel.com>
Signed-off-by: Peter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com>

ixgbe: Remove code that was initializing Rx page offset

This change reverts an earlier patch that introduced
ixgbe_init_rx_page_offset. The idea behind the function was to provide
some variation in the starting offset for the page in order to reduce
hot-spots in the cache. However it doesn't appear to provide any
significant benefit in the testing I have done. It has however been a
source of several bugs, and it blocks us from being able to use 2K
fragments on larger page sizes. So the decision I made was to remove it.

Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
Tested-by: Phil Schmitt <phillip.j.schmitt@intel.com>
Signed-off-by: Peter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com>

sctp: fix bogus if statement in sctp_auth_recv_cid()

There is an extra semi-colon here, so we always return 0 instead of
calling __sctp_auth_cid().

Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Reviewed-by: "Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

sctp: fix compile issue with disabled CONFIG_NET_NS

struct seq_net_private has no struct net
if CONFIG_NET_NS is not enabled

Signed-off-by: Ulrich Weber <ulrich.weber@sophos.com>
Reviewed-by: "Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

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

John W. Linville says:

====================
This is a batch of updates intended for 3.7.  The ath9k, mwifiex,
and b43 drivers get the bulk of the commits this time, with a handful
of other driver bits thrown-in.  It is mostly just minor fixes and
cleanups, etc.

Also included is a Bluetooth pull, with a lot of refactoring.
Gustavo says:

"These are the changes I queued for 3.7. There are a many
small fixes/improvements by Andre Guedes. A l2cap channel
refcounting refactor by Jaganath. Bluetooth sockets now
appears in /proc/net, by Masatake Yamato and Sachin Kamat
changes ours drivers to use devm_kzalloc()."
====================

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

net: remove wrong initialization for snd_wl1

The field tp->snd_wl1 is twice initialized, the second time
seems to be wrong as it may overwrite any update in tcp_ack.

Signed-off-by: Razvan Ghitulete <rghitulete@ixiacom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

xfrm: remove redundant parameter "int dir" in struct xfrm_mgr.acquire

Sematically speaking, xfrm_mgr.acquire is called when kernel intends to ask
user space IKE daemon to negotiate SAs with peers. IOW the direction will
*always* be XFRM_POLICY_OUT, so remove int dir for clarity.

Signed-off-by: Fan Du <fan.du@windriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

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

sctp: fix a compile error in sctp.h

I got the following compile error:

In file included from include/net/sctp/checksum.h:46:0,
                 from net/ipv4/netfilter/nf_nat_proto_sctp.c:14:
include/net/sctp/sctp.h: In function ‘sctp_dbg_objcnt_init’:
include/net/sctp/sctp.h:370:88: error: parameter name omitted
include/net/sctp/sctp.h: In function ‘sctp_dbg_objcnt_exit’:
include/net/sctp/sctp.h:371:88: error: parameter name omitted

which is caused by

commit 13d782f6b4fbbaf9d0380a9947deb45a9de46ae7
Author: Eric W. Biederman <ebiederm@xmission.com>
Date:   Mon Aug 6 08:45:15 2012 +0000

    sctp: Make the proc files per network namespace.

This patch could fix it.

Cc: David S. Miller <davem@davemloft.net>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: Cong Wang <amwang@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

sctp: Make sysctl tunables per net

Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

sctp: Push struct net down into sctp_verify_ext_param

Add struct net as a parameter to sctp_verify_param so it can be passed
to sctp_verify_ext_param where struct net will be needed when the sctp
tunables become per net tunables.

Add struct net as a parameter to sctp_verify_init so struct net can be
passed to sctp_verify_param.

Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

sctp: Push struct net down into all of the state machine functions

There are a handle of state machine functions primarily those dealing
with processing INIT packets where there is neither a valid endpoint nor
a valid assoication from which to derive a struct net.  Therefore add
struct net * to the parameter list of sctp_state_fn_t and update all of
the state machine functions.

Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

sctp: Push struct net down into sctp_in_scope

struct net will be needed shortly when the tunables are made per network
namespace.

Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

sctp: Push struct net down into sctp_transport_init

Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

sctp: Push struct net down to sctp_chunk_event_lookup

This trickles up through sctp_sm_lookup_event up to sctp_do_sm
and up further into sctp_primitiv_NAME before the code reaches
places where struct net can be reliably found.

Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

sctp: Add infrastructure for per net sysctls

Start with an empty sctp_net_table that will be populated as the various
tunable sysctls are made per net.

Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
Acked-by: Vlad Yasevich <vyasevich@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

sctp: Make the mib per network namespace

Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
Acked-by: Vlad Yasevich <vyasevich@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

sctp: Enable sctp in all network namespaces

- Fix the sctp_af operations to work in all namespaces
- Enable sctp socket creation in all network namespaces.

Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
Acked-by: Vlad Yasevich <vyasevich@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

sctp: Make the proc files per network namespace.

- Convert all of the files under /proc/net/sctp to be per
  network namespace.

- Don't print anything for /proc/net/sctp/snmp except in
  the initial network namespaces as the snmp counters still
  have to be converted to be per network namespace.

Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
Acked-by: Vlad Yasevich <vyasevich@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

sctp: Move the percpu sockets counter out of sctp_proc_init

The percpu sctp socket counter has nothing at all to do with the sctp
proc files, and having it in the wrong initialization is confusing,
and makes network namespace support a pain.

Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
Acked-by: Vlad Yasevich <vyasevich@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

sctp: Make the ctl_sock per network namespace

- Kill sctp_get_ctl_sock, it is useless now.
- Pass struct net where needed so net->sctp.ctl_sock is accessible.

Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
Acked-by: Vlad Yasevich <vyasevich@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

sctp: Make the address lists per network namespace

- Move the address lists into struct net
- Add per network namespace initialization and cleanup
- Pass around struct net so it is everywhere I need it.
- Rename all of the global variable references into references
  to the variables moved into struct net

Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
Acked-by: Vlad Yasevich <vyasevich@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

sctp: Make the association hashtable handle multiple network namespaces

- Use struct net in the hash calculation
- Use sock_net(association.base.sk) in the association lookups.
- On receive calculate the network namespace from skb->dev.
- Pass struct net from receive down to the functions that actually
  do the association lookup.

Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
Acked-by: Vlad Yasevich <vyasevich@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

sctp: Make the endpoint hashtable handle multiple network namespaces

- Use struct net in the hash calculation
- Use sock_net(endpoint.base.sk) in the endpoint lookups.
- On receive calculate the network namespace from skb->dev.

Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
Acked-by: Vlad Yasevich <vyasevich@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

sctp: Make the port hash table use struct net in it's key.

- Add struct net into the port hash table hash calculation
- Add struct net inot the struct sctp_bind_bucket so there
  is a memory of which network namespace a port is allocated in.
  No need for a ref count because sctp_bind_bucket only exists
  when there are sockets in the hash table and sockets can not
  change their network namspace, and sockets already ref count
  their network namespace.
- Add struct net into the key comparison when we are testing
  to see if we have found the port hash table entry we are
  looking for.

With these changes lookups in the port hash table becomes
safe to use in multiple network namespaces.

Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
Acked-by: Vlad Yasevich <vyasevich@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

packet: Report socket mclist info via diag module

The info is reported as an array of packet_diag_mclist structures. Each
includes not only the directly configured values (index, type, etc), but
also the "count".

Signed-off-by: Pavel Emelyanov <xemul@parallels.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

packet: Report more packet sk info via diag module

This reports in one rtattr message all the other scalar values, that can be
set on a packet socket with setsockopt.

Signed-off-by: Pavel Emelyanov <xemul@parallels.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

packet: Diag core and basic socket info dumping

The diag module can be built independently from the af_packet.ko one,
just like it's done in unix sockets.

The core dumping message carries the info available at socket creation
time, i.e. family, type and protocol (in the same byte order as shown in
the proc file).

The socket inode number and cookie is reserved for future per-socket info
retrieving. The per-protocol filtering is also reserved for future by
requiring the sdiag_protocol to be zero.

Signed-off-by: Pavel Emelyanov <xemul@parallels.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

packet: Introduce net/packet/internal.h header

The diag module will need to access some private packet_sock data, so
move it to a header in advance. This file will be shared between the
af_packet.c and the diag.c

Signed-off-by: Pavel Emelyanov <xemul@parallels.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

net: ipv4: fib_trie: Don't unnecessarily search for already found fib leaf

We've already found leaf, don't search for it again. Same is for fib leaf info.

Signed-off-by: Igor Maravic <igorm@etf.rs>
Signed-off-by: David S. Miller <davem@davemloft.net>

Replace rwlock on xfrm_policy_afinfo with rcu

xfrm_policy_afinfo is read mosly data structure.
Write on xfrm_policy_afinfo is done only at the
time of configuration.
So rwlocks can be safely replaced with RCU.

RCUs usage optimizes the performance.

Signed-off-by: Priyanka Jain <Priyanka.Jain@freescale.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

gre: Support GRE over IPv6

GRE over IPv6 implementation.

Signed-off-by: Dmitry Kozlov <xeb@mail.ru>
Signed-off-by: David S. Miller <davem@davemloft.net>

net: remove netdev_bonding_change()

I don't see any benifits to use netdev_bonding_change() than
using call_netdevice_notifiers() directly.

Cc: David S. Miller <davem@davemloft.net>
Signed-off-by: Cong Wang <amwang@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

net: move and rename netif_notify_peers()

I believe net/core/dev.c is a better place for netif_notify_peers(),
because other net event notify functions also stay in this file.

And rename it to netdev_notify_peers().

Cc: David S. Miller <davem@davemloft.net>
Cc: Ian Campbell <Ian.Campbell@citrix.com>
Signed-off-by: Cong Wang <amwang@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

p54: fix powerpc gcc warnings

My commit "p54: parse output power table" introduced
the following compiler warnings for powerpc-allmodconfig

eeprom.c: In function 'p54_get_maxpower':
eeprom.c:291 warning: comparison of distinct pointer types lacks a cast
eeporm.c:292 warning: comparison of distinct pointer types lacks a cast
eeprom.c:293 warning: comparison of distinct pointer types lacks a cast
eeprom.c:294 warning: comparison of distinct pointer types lacks a cast

This patch fixes those by using max_t(u16
which forces a type cast.

Reported-by: Fengguang Wu <fengguang.wu@intel.com>
Tested-by: Fengguang Wu <fengguang.wu@intel.com>
Signed-off-by: Christian Lamparter <chunkeey@googlemail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

mac80211: fix unnecessary beacon update after peering status change

ieee80211_bss_info_change_notify is called everytime a peer link is established
or closed, because the accepting_plinks flag in the meshconf IE *might* have changed.

With this patch the corresponding functions return the BSS_CHANGED_BEACON flag when a beacon update is necessary.

Also it makes mesh_accept_plinks_update the common place to update the accepting_plinks flag.
mesh_accept_plinks_update is called upon plink change and also periodically from ieee80211_mesh_housekeeping.
Thus, it also picks up changes of local->num_sta.

Signed-off-by: Marco Porsch <marco.porsch@etit.tu-chemnitz.de>
Acked-by: Thomas Pedersen <thomas@cozybit.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

brcmsmac: document firmware dependencies

The brcmsmac driver requests firmware but doesn't document the
dependency. This means that software that analyzes the modules to
determine if firmware is needed won't detect it.

Specifically, (at least) openSUSE won't install the kernel-firmware
package if no hardware requires it.

This patch adds the MODULE_FIRMWARE directives.

Signed-off-by: Jeff Mahoney <jeffm@suse.com>
Acked-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

ath9k: tune rc_stats to display only valid rates

This could make rc_stats more simpler and ease the debugging.

Signed-off-by: Rajkumar Manoharan <rmanohar@qca.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

ath9k: Trim rate table

Remove ctrl_rate, cw40index, sgi_index, ht_index and calculate
the rate index for TX status from the valid_rate_index that
is populated at initialization time.

Signed-off-by: Sujith Manoharan <c_manoha@qca.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

ath9k: Remove MIMO-PS specific code

Signed-off-by: Sujith Manoharan <c_manoha@qca.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

ath9k: Cleanup TX status API

Calculate the final rate index inside ath_rc_tx_status().

Signed-off-by: Sujith Manoharan <c_manoha@qca.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

ath9k: Bail out properly before calculating rate index

Signed-off-by: Sujith Manoharan <c_manoha@qca.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

ath9k: Fix RTS/CTS rate selection

The current method of assigning the RTS/CTS rate is completely
broken for HT mode and breaks P2P operation. Fix this by using
the basic_rates provided to the driver by mac80211. For now,
choose the lowest supported basic rate for HT frames.

Signed-off-by: Sujith Manoharan <c_manoha@qca.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

ath9k: Do not set IEEE80211_TX_RC_USE_SHORT_PREAMBLE

mac80211 does it for us.

Signed-off-by: Sujith Manoharan <c_manoha@qca.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

ath9k: Unify valid rate calculation routines

Signed-off-by: Sujith Manoharan <c_manoha@qca.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

ath9k: Remove ath_rc_set_valid_rate_idx

Signed-off-by: Sujith Manoharan <c_manoha@qca.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

ath9k: Change rateset calculation

Commit "ath9k: Change rate control to use legacy rate as last MRR"
resulted in the mixing of HT/legacy rates in a single rateset,
which is undesirable. Revert this behavior.

Signed-off-by: Sujith Manoharan <c_manoha@qca.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

ath9k: Cleanup index retrieval routines

Trim API and remove unused variables.

Signed-off-by: Sujith Manoharan <c_manoha@qca.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

ath9k: Cleanup ath_rc_setvalid_htrates

Signed-off-by: Sujith Manoharan <c_manoha@qca.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

ath9k: Cleanup ath_rc_setvalid_rates

Signed-off-by: Sujith Manoharan <c_manoha@qca.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

ath9k: Cleanup RC init API

A reference to the rate table is stored inside the
private structure, so there is no need to pass "rate_table"
around.

Signed-off-by: Sujith Manoharan <c_manoha@qca.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

ath9k: Simplify rate table initialization

Remove various local variables that duplicate information
already stored in mac80211.

Signed-off-by: Sujith Manoharan <c_manoha@qca.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

mwifiex: use GFP_ATOMIC under spin lock

We're holding the sta_list_spinlock here so we can't sleep.

Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Acked-by: Bing Zhao <bzhao@marvell.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

mwifiex: notify cfg80211 about MIC failures

Call cfg80211_michael_mic_failure() handler when there is a MIC error
event from firmware.

Signed-off-by: Amitkumar Karwar <akarwar@marvell.com>
Signed-off-by: Kiran Divekar <dkiran@marvell.com>
Signed-off-by: Bing Zhao <bzhao@marvell.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

mwifiex: fix 'smatch' warning in preparing key_material cmd

The key length can be 32 bytes for TKIP and 16 bytes for AES_CMAC.
'smatch' warns on memcpy using key_len variable to copy data to
a 16 bytes buffer. Use fixed length to avoid the warning.

Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Bing Zhao <bzhao@marvell.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

ssb: check for flash presentence

We can not assume parallel flash is always present, there are boards
with *serial* flash and probably some without flash at all.
Define some bits by the way.

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Reviewed-by: Hauke Mehrtens <hauke@hauke-m.de>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

b43legacy: fix logic in GPIO init

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

b43: N-PHY: fix 0x2057 radio calib copy/paste mistake

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Reported-by: Fengguang Wu <fengguang.wu@intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

bcma: add (mostly) NAND defines

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

mwifiex: fix powerpc64-linux- compilation warnings

These warnings can be detected by using powerpc64-linux toolchain
(gcc-4.6.3-nolibc).

  CC [M]  drivers/net/wireless/mwifiex/sta_event.o
drivers/net/wireless/mwifiex/sta_event.c: In function 'mwifiex_process_sta_event':
drivers/net/wireless/mwifiex/sta_event.c:388:4: warning: comparison of distinct pointer types lacks a cast [enabled by default]
  CC [M]  drivers/net/wireless/mwifiex/uap_event.o
drivers/net/wireless/mwifiex/uap_event.c: In function 'mwifiex_process_uap_event':
drivers/net/wireless/mwifiex/uap_event.c:258:11: warning: comparison of distinct pointer types lacks a cast [enabled by default]

Use min_t() instead of min() to fix the warnings.

Reported-by: Fengguang Wu <fengguang.wu@intel.com>
Signed-off-by: Bing Zhao <bzhao@marvell.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

iwlegacy: clean up suspend/resume

There's no need to export the il_pci_suspend
and il_pci_resume functions since they're only
referenced from il_pm_ops. The latter can also
be defined using SIMPLE_DEV_PM_OPS instead of
open-coding it.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

mwifiex: fix code mis-alignment after the if statement

Signed-off-by: Fengguang Wu <fengguang.wu@intel.com>
Acked-by: Bing Zhao <bzhao@marvell.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

ath5k: Put power_level where it belongs and rename it

Put power_level to ah_txpower struct with the rest tx power infos and
also rename it to txp_requested to make more sense.

v2 make sure we don't memset it to zero on reset

Signed-off-by: Nick Kossifidis <mickflemm@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

ath5k: Preserve tx power level requested from above on phy_init

By using cur_pwr on phy_init we re-use the power level previously set by the
driver, not the one we got from above.

Signed-off-by: Nick Kossifidis <mickflemm@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

ath5k: Fix range scaling when setting rate power table

rates[i] is unsigned but txp_offset can be negative for newer parts
with PDADC table. We cover the case when rates[i] + txp_offset > 63
but we must also cover the case when its < 0 or else rates[i] will overflow.

Signed-off-by: Nick Kossifidis <mickflemm@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

ath5k: Use correct value for min_pwr and cur_pwr

Make sure we don't store the table offsets for min and cur power levels,
store the 0.25dB values instead. This way we don't clamp the tx power level
to max (because now cur_pwr holds the 0.25dB value, not the table offset) after
re-using cur_pwr on reset.

Signed-off-by: Nick Kossifidis <mickflemm@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

bcma: make some functions static

The functions and structs are not used in an other file and the
prototypes are in no header file, just make them static so the compiler
is able to optimize them better.

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

p54: parse output power table

For the upcoming tpc changes, the driver needs
to provide sensible max output values for each
supported channel.

And while the eeprom always had a output_limit
table, which defines the upper limit for each
frequency and modulation, it was never really
useful for anything... until now.

Note: For anyone wondering about what your card
is calibrated for: check "iw list".
* 2412 MHz [1] (18.0 dBm)
* 2437 MHz [6] (19.0 dBm)
[...]
* 5180 MHz [36] (18.0 dBm)
* 5260 MHz [52] (17.0 dBm) (radar detection)
* 5680 MHz [136] (19.0 dBm) (radar detection)
(for a Dell Wireless 1450 USB Adapter)

Signed-off-by: Christian Lamparter <chunkeey@googlemail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

Merge branch 'master' of git://git./linux/kernel/git/bluetooth/bluetooth-next

hyperv: Add comments for the extended buffer after RNDIS message

Reported-by: Olaf Hering <olaf@aepfle.de>
Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

net: Loopback ifindex is constant now

As pointed out, there are places, that access net->loopback_dev->ifindex
and after ifindex generation is made per-net this value becomes constant
equals 1. So go ahead and introduce the LOOPBACK_IFINDEX constant and use
it where appropriate.

Signed-off-by: Pavel Emelyanov <xemul@parallels.com>
Acked-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

net: Make ifindex generation per-net namespace

Strictly speaking this is only _really_ required for checkpoint-restore to
make loopback device always have the same index.

This change appears to be safe wrt "ifindex should be unique per-system"
concept, as all the ifindex usage is either already made per net namespace
of is explicitly limited with init_net only.

There are two cool side effects of this. The first one -- ifindices of
devices in container are always small, regardless of how many containers
we've started (and re-started) so far. The second one is -- we can speed
up the loopback ifidex access as shown in the next patch.

v2: Place ifindex right after dev_base_seq : avoid two holes and use the
    same cache line, dirtied in list_netdevice()/unlist_netdevice()

Signed-off-by: Pavel Emelyanov <xemul@parallels.com>
Acked-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

veth: Allow to create peer link with given ifindex

The ifinfomsg is in there (thanks kaber@ for foreseeing this long time ago),
so take the given ifidex and register netdev with it.

Ben noticed, that this code path previously ignored ifmp->ifi_index and
userland could be passing in garbage. Thus it may now fail occasionally
because the value clashes with an existing interface.

To address this it's assumed that if the caller specifies the ifindex for
the veth master device, then it's aware of this possibility and should
explicitly specify (or set to 0 for auto-assignment) the peer's ifindex as
well. With this the compatibility with old tools not setting ifindex is
preserved.

Signed-off-by: Pavel Emelyanov <xemul@parallels.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

net: Allow to create links with given ifindex

Currently the RTM_NEWLINK results in -EOPNOTSUPP if the ifinfomsg->ifi_index
is not zero. I propose to allow requesting ifindices on link creation. This
is required by the checkpoint-restore to correctly restore a net namespace
(i.e. -- a container).

Signed-off-by: Pavel Emelyanov <xemul@parallels.com>
Acked-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

net: Dont use ifindices in hash fns

Eric noticed, that when there will be devices with equal indices, some
hash functions that use them will become less effective as they could.
Fix this in advance by mixing the net_device address into the hash value
instead of the device index.

This is true for arp and ndisc hash fns. The netlabel, can and llc ones
are also ifindex-based, but that three are init_net-only, thus will not
be affected.

Many thanks to David and Eric for the hash32_ptr implementation!

Signed-off-by: Pavel Emelyanov <xemul@parallels.com>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

time: jiffies_delta_to_clock_t() helper to the rescue

Various /proc/net files sometimes report crazy timer values, expressed
in clock_t units.

This happens when an expired timer delta (expires - jiffies) is passed
to jiffies_to_clock_t().

This function has an overflow in :

return div_u64((u64)x * TICK_NSEC, NSEC_PER_SEC / USER_HZ);

commit cbbc719fccdb8cb (time: Change jiffies_to_clock_t() argument type
to unsigned long) only got around the problem.

As we cant output negative values in /proc/net/tcp without breaking
various tools, I suggest adding a jiffies_delta_to_clock_t() wrapper
that caps the negative delta to a 0 value.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Reported-by: Maciej Żenczykowski <maze@google.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Paul Gortmaker <paul.gortmaker@windriver.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: hank <pyu@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

fib: use __fls() on non null argument

__fls(x) is a bit faster than fls(x), granted we know x is non null.

As Ben Hutchings pointed out, fls(x) = __fls(x) + 1

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