inner_protocol was added to struct sk_buff in
0d89d203 ("MPLS: Add limited GSO support"),
which is scheduled to be included in v3.11.

That patch did not update __copy_skb_header to copy the inner_protocol.
Signed-off-by: NJoe Stringer <joe@wand.net.nz>
Signed-off-by: NSimon Horman <horms@verge.net.au>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The dev_forward_skb() assignment of pkt_type should be done
after the call to eth_type_trans().

ip-encapsulated packets can be handled by localhost. But skb->pkt_type
can be PACKET_OTHERHOST when packet comes via veth into ip tunnel device.
In that case, the packet is dropped by ip_rcv().
Although this example uses gretap. l2tp-eth also has same issue.
For l2tp-eth case, add dummy device for ip address and ip l2tp command.

netns A |                     root netns                      | netns B
   veth<->veth=bridge=gretap <-loop back-> gretap=bridge=veth<->veth

arp packet ->
pkt_type
         BROADCAST------------>ip_rcv()------------------------>

                                                             <- arp reply
                                                                pkt_type
                               ip_rcv()<-----------------OTHERHOST
                               drop

sample operations
  ip link add tapa type gretap remote 172.17.107.4 local 172.17.107.3
  ip link add tapb type gretap remote 172.17.107.3 local 172.17.107.4
  ip link set tapa up
  ip link set tapb up
  ip address add 172.17.107.3 dev tapa
  ip address add 172.17.107.4 dev tapb
  ip route get 172.17.107.3
  > local 172.17.107.3 dev lo  src 172.17.107.3
  >    cache <local>
  ip route get 172.17.107.4
  > local 172.17.107.4 dev lo  src 172.17.107.4
  >    cache <local>
  ip link add vetha type veth peer name vetha-peer
  ip link add vethb type veth peer name vethb-peer
  brctl addbr bra
  brctl addbr brb
  brctl addif bra tapa
  brctl addif bra vetha-peer
  brctl addif brb tapb
  brctl addif brb vethb-peer
  brctl show
  > bridge name     bridge id               STP enabled     interfaces
  > bra             8000.6ea21e758ff1       no              tapa
  >                                                         vetha-peer
  > brb             8000.420020eb92d5       no              tapb
  >                                                         vethb-peer
  ip link set vetha-peer up
  ip link set vethb-peer up
  ip link set bra up
  ip link set brb up
  ip netns add a
  ip netns add b
  ip link set vetha netns a
  ip link set vethb netns b
  ip netns exec a ip address add 10.0.0.3/24 dev vetha
  ip netns exec b ip address add 10.0.0.4/24 dev vethb
  ip netns exec a ip link set vetha up
  ip netns exec b ip link set vethb up
  ip netns exec a arping -I vetha 10.0.0.4
  ARPING 10.0.0.4 from 10.0.0.3 vetha
  ^CSent 2 probes (2 broadcast(s))
  Received 0 response(s)

Cc: Jason Wang <jasowang@redhat.com>
Cc: "Michael S. Tsirkin" <mst@redhat.com>
Cc: Eric Dumazet <edumazet@google.com>
Cc: Patrick McHardy <kaber@trash.net>
Cc: Hong Zhiguo <honkiko@gmail.com>
Cc: Rami Rosen <ramirose@gmail.com>
Cc: Tom Parkin <tparkin@katalix.com>
Cc: Cong Wang <xiyou.wangcong@gmail.com>
Cc: Pravin B Shelar <pshelar@nicira.com>
Cc: Jesse Gross <jesse@nicira.com>
Cc: dev@openvswitch.org
Signed-off-by: NIsaku Yamahata <yamahata@valinux.co.jp>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

As the patch "bnx2x: remove zeroing of dump data buffer" showed,
it is too easy implement .get_dump_data incorrectly in a driver.

Let's make sure drivers cannot get confused by userspace requesting
a too big dump.

Also WARN if the driver sets dump->len to something weird and make
sure the length reported to userspace is the actual length of data
copied to userspace.
Signed-off-by: NMichal Schmidt <mschmidt@redhat.com>
Reviewed-by: NBen Hutchings <ben@decadent.org.uk>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

There is a race in neighbour code, because neigh_destroy() uses
skb_queue_purge(&neigh->arp_queue) without holding neighbour lock,
while other parts of the code assume neighbour rwlock is what
protects arp_queue

Convert all skb_queue_purge() calls to the __skb_queue_purge() variant

Use __skb_queue_head_init() instead of skb_queue_head_init()
to make clear we do not use arp_queue.lock

And hold neigh->lock in neigh_destroy() to close the race.
Reported-by: NJoe Jin <joe.jin@oracle.com>
Signed-off-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The goal of this new function is to perform all needed cleanup before sending
an skb into another netns.
Signed-off-by: NNicolas Dichtel <nicolas.dichtel@6wind.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

When the kernel (compiled with CONFIG_PREEMPT=n) is performing the
rename of a network interface, it can end up waiting for a workqueue
to complete. If userland is able to invoke a SIOCGIFNAME ioctl or a
SO_BINDTODEVICE getsockopt in between, the kernel will deadlock due to
the fact that read_secklock_begin() will spin forever waiting for the
writer process (the one doing the interface rename) to update the
devnet_rename_seq sequence.

This patch fixes the problem by adding a helper (netdev_get_name())
and using it in the code handling the SIOCGIFNAME ioctl and
SO_BINDTODEVICE setsockopt.

The netdev_get_name() helper uses raw_seqcount_begin() to avoid
spinning forever, waiting for devnet_rename_seq->sequence to become
even. cond_resched() is used in the contended case, before retrying
the access to give the writer process a chance to finish.

The use of raw_seqcount_begin() will incur some unneeded work in the
reader process in the contended case, but this is better than
deadlocking the system.
Signed-off-by: NNicolas Schichan <nschichan@freebox.fr>
Acked-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

select/poll busy-poll support.

Split sysctl value into two separate ones, one for read and one for poll.
updated Documentation/sysctl/net.txt

Add a new poll flag POLL_LL. When this flag is set, sock_poll will call
sk_poll_ll if possible. sock_poll sets this flag in its return value
to indicate to select/poll when a socket that can busy poll is found.

When poll/select have nothing to report, call the low-level
sock_poll again until we are out of time or we find something.

Once the system call finds something, it stops setting POLL_LL, so it can
return the result to the user ASAP.
Signed-off-by: NEliezer Tamir <eliezer.tamir@linux.intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

commit 68c33163 ("v4 GRE: Add TCP segmentation offload for GRE")
added a possible skb leak, because it frees only the head of segment
list, in case a skb_linearize() call fails.

This patch adds a kfree_skb_list() helper to fix the bug.
Signed-off-by: NEric Dumazet <edumazet@google.com>
Cc: Pravin B Shelar <pshelar@nicira.com>
Cc: Daniel Borkmann <dborkman@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This is required for multiple default destinations management in VXLAN
Signed-off-by: NMike Rapoport <mike.rapoport@ravellosystems.com>
Signed-off-by: NStephen Hemminger <stephen@networkplumber.org>

Callers of skb_seq_read() are currently forced to call skb_abort_seq_read()
even when consuming all the data because the last call to skb_seq_read (the
one that returns 0 to indicate the end) fails to unmap the last fragment page.

With this patch callers will be allowed to traverse the SKB data by calling
skb_prepare_seq_read() once and repeatedly calling skb_seq_read() as originally
intended (and documented in the original commit 677e90ed), that is, only call
skb_abort_seq_read() if the sequential read is actually aborted.
Signed-off-by: NWedson Almeida Filho <wedsonaf@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

netif_alloc_netdev_queues() uses kcalloc() to allocate memory
for the "struct netdev_queue *_tx" array.

For large number of tx queues, kcalloc() might fail, so this
patch does a fallback to vzalloc().

As vmalloc() adds overhead on a critical network path, add __GFP_REPEAT
to kzalloc() flags to do this fallback only when really needed.
Signed-off-by: NEric Dumazet <edumazet@google.com>
Acked-by: NMichael S. Tsirkin <mst@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

thresh and interval are global resources,
only init net can change them.
Signed-off-by: NGao feng <gaofeng@cn.fujitsu.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Though we don't export the /proc/sys/net/ipv[4,6]/neigh/default/
directory to the un-init_net, but we can still use cmd such as
"ip ntable change name arp_cache locktime 129" to change the locktime
of default neigh_parms.

This patch disallows the un-init_net to find out the neigh_table.parms.
So the un-init_net will failed to influence the init_net.
Signed-off-by: NGao feng <gaofeng@cn.fujitsu.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

neigh_table.parms always exist and is initialized,kmemdup
can use it to create new neigh_parms, actually lookup_neigh_parms
here will return neigh_table.parms too.
Signed-off-by: NGao feng <gaofeng@cn.fujitsu.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

As part of the push to add 802.1ad server provider tagging support to the
kernel the VLAN features flags were renamed. Unfortunately the kernel name
for the VLAN hardware acceleration features that the kernel shows user space
was included in the rename, which broke ethtool (txvlan and rxvlan options
do not work). This patch restores the original names, i.e. the original ABI.
If we wanted to make clear to users that we are refering to CTAGs we can
always change ethtool's short_name and long_name for these features (for
example something along the lines of txvlan -> txvlan-ctag, tx-vlan-offload ->
tx-vlan-ctag-offload).

Cc: Patrick McHardy <kaber@trash.net>
Cc: David S. Miller <davem@davemloft.net>
Cc: netdev@vger.kernel.org
Signed-off-by: NFernando Luis Vazquez Cao <fernando@oss.ntt.co.jp>
Reviewed-by: NBen Hutchings <bhutchings@solarflare.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

adds a socket option for low latency polling.
This allows overriding the global sysctl value with a per-socket one.
Unexport sysctl_net_ll_poll since for now it's not needed in modules.
Signed-off-by: NEliezer Tamir <eliezer.tamir@linux.intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

There is no reason for sysctl_net_ll_poll to be an unsigned long.
Change it into an unsigned int.
Fix the proc handler.
Signed-off-by: NEliezer Tamir <eliezer.tamir@linux.intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Add netlink directives and ndo entry to allow for controling
VF link, which can be in one of three states:

Auto - VF link state reflects the PF link state (default)

Up - VF link state is up, traffic from VF to VF works even if
the actual PF link is down

Down - VF link state is down, no traffic from/to this VF, can be of
use while configuring the VF
Signed-off-by: NRony Efraim <ronye@mellanox.com>
Signed-off-by: NOr Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Caught by sparse:
- __rcu: missing annotation to sd->flow_limit
- __user: direct access in cpumask_scnprintf

Also
- add endline character when printing bitmap if room in buffer
- avoid bucket overflow by reducing FLOW_LIMIT_HISTORY

The last item warrants some explanation. The hashtable buckets are
subject to overflow if FLOW_LIMIT_HISTORY is larger than or equal
to bucket size, since all packets may end up in a single bucket. The
current (rather arbitrary) history value of 256 happens to match the
buffer size (u8).

As a result, with a single flow, the first 128 packets are accepted
(correct), the second 128 packets dropped (correct) and then the
history[] array has filled, so that each subsequent new packet
causes an increment in the bucket for new_flow plus a decrement
for old_flow: a steady state.

This is fine if packets are dropped, as the steady state goes away
as soon as a mix of traffic reappears. But, because the 256th packet
overflowed the bucket to 0: no packets are dropped.

Instead of explicitly adding an overflow check, this patch changes
FLOW_LIMIT_HISTORY to never be able to overflow a single bucket.
Reported-by: NFengguang Wu <fengguang.wu@intel.com>
(first item)
Signed-off-by: NWillem de Bruijn <willemb@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Reduce the uses of this unnecessary typedef.

Done via perl script:

$ git grep --name-only -w ctl_table net | \
  xargs perl -p -i -e '\
	sub trim { my ($local) = @_; $local =~ s/(^\s+|\s+$)//g; return $local; } \
        s/\b(?<!struct\s)ctl_table\b(\s*\*\s*|\s+\w+)/"struct ctl_table " . trim($1)/ge'

Reflow the modified lines that now exceed 80 columns.
Signed-off-by: NJoe Perches <joe@perches.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Since team functionality relies heavily on userspace daemon, we need to
deliver event to userspace via Netlink as quick as possible. So make all
team port device link events urgent.
Signed-off-by: NFlavio Leitner <fbl@redhat.com>
Signed-off-by: NJiri Pirko <jiri@resnulli.us>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

We currently allow for numa-node aware skb allocation only within the
fill_packet_ipv4() path, but not in fill_packet_ipv6(). Consolidate that
code to a common allocation helper to enable numa-node aware skb
allocation for ipv6, and use it in both paths. This also makes both
functions a bit more readable.
Signed-off-by: NDaniel Borkmann <dborkman@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

struct gnet_stats_rate_est contains u32 fields, so the bytes per second
field can wrap at 34360Mbit.

Add a new gnet_stats_rate_est64 structure to get 64bit bps/pps fields,
and switch the kernel to use this structure natively.

This structure is dumped to user space as a new attribute :

TCA_STATS_RATE_EST64

Old tc command will now display the capped bps (to 34360Mbit), instead
of wrapped values, and updated tc command will display correct
information.

Old tc command output, after patch :

eric:~# tc -s -d qd sh dev lo
qdisc pfifo 8001: root refcnt 2 limit 1000p
 Sent 80868245400 bytes 1978837 pkt (dropped 0, overlimits 0 requeues 0)
 rate 34360Mbit 189696pps backlog 0b 0p requeues 0

This patch carefully reorganizes "struct Qdisc" layout to get optimal
performance on SMP.
Signed-off-by: NEric Dumazet <edumazet@google.com>
Cc: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Since commit 1a37e412(net: Use 16bits for *_headers fields of struct
skbuff), skb->*_header are relative to skb->head,
so copy_skb_header() should not call skb_headers_offset_update() now,
and we should pass correct parameter to skb_headers_offset_update() in
pskb_expand_head() and skb_copy_expand().
Signed-off-by: NWeiping Pan <panweiping3@gmail.com>
Reviewed-by: NSimon Horman <horms@verge.net.au>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Filters need to be translated to real BPF code for userland, like SO_GETFILTER.
Signed-off-by: NNicolas Dichtel <nicolas.dichtel@6wind.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Add upport for busy-polling on UDP sockets.
In __udp[46]_lib_rcv add a call to sk_mark_ll() to copy the napi_id
from the skb into the sk.
This is done at the earliest possible moment, right after we identify
which socket this skb is for.
In __skb_recv_datagram When there is no data and the user
tries to read we busy poll.
Signed-off-by: NAlexander Duyck <alexander.h.duyck@intel.com>
Signed-off-by: NJesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: NEliezer Tamir <eliezer.tamir@linux.intel.com>
Acked-by: NEric Dumazet <edumazet@google.com>
Tested-by: NWillem de Bruijn <willemb@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Adds an ndo_ll_poll method and the code that supports it.
This method can be used by low latency applications to busy-poll
Ethernet device queues directly from the socket code.
sysctl_net_ll_poll controls how many microseconds to poll.
Default is zero (disabled).
Individual protocol support will be added by subsequent patches.
Signed-off-by: NAlexander Duyck <alexander.h.duyck@intel.com>
Signed-off-by: NJesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: NEliezer Tamir <eliezer.tamir@linux.intel.com>
Acked-by: NEric Dumazet <edumazet@google.com>
Tested-by: NWillem de Bruijn <willemb@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Adds a napi_id and a hashing mechanism to lookup a napi by id.
This will be used by subsequent patches to implement low latency
Ethernet device polling.
Based on a code sample by Eric Dumazet.
Signed-off-by: NEliezer Tamir <eliezer.tamir@linux.intel.com>
Signed-off-by: NEric Dumazet <edumazet@google.com>
Tested-by: NWillem de Bruijn <willemb@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Similar to the problem in pktgen, netpoll uses skb_tail_offset()
too, as the code is copied from pktgen.

Also use return values of skb_put() directly, this will simiplify
the code.
Reported-by: NThomas Graf <tgraf@suug.ch>
Cc: Thomas Graf <tgraf@suug.ch>
Cc: Daniel Borkmann <dborkmann@redhat.com>
Cc: David S. Miller <davem@davemloft.net>
Signed-off-by: NCong Wang <amwang@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

skb_set_network_header() expects an offset based on the data pointer
whereas skb_tail_offset() also includes the headroom. This resulted
in the ip header being written in a wrong location.

Use return values of skb_put() directly and rely on skb->len to
set mac, network, and transport header.

Cc: Simon Horman <horms@verge.net.au>
Cc: Daniel Borkmann <dborkmann@redhat.com>
Assisted-by: NDaniel Borkmann <dborkmann@redhat.com>
Signed-off-by: NThomas Graf <tgraf@suug.ch>
Acked-by: NDaniel Borkmann <dborkmann@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Eric Dumazet spotted that we have to check skb->head instead
of skb->data as skb->head points to the beginning of the
data area of the skbuff. Similarly, we have to initialize the
skb->head pointer, not skb->data in __alloc_skb_head.

After this fix, netlink crashes in the release path of the
sk_buff, so let's fix that as well.

This bug was introduced in (0ebd0ac5 net: add function to
allocate sk_buff head without data area).
Reported-by: NEric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: NPablo Neira Ayuso <pablo@netfilter.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Signed-off-by: NYan Burman <yanb@mellanox.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

netdev_create_hash() is only called from netdev_init() which is marked
__net_init.
Signed-off-by: NBaruch Siach <baruch@tkos.co.il>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

commit 1a37e412 (net: Use 16bits for *_headers
fields of struct skbuff) converts skb->*_header to u16,
some #if NET_SKBUFF_DATA_USES_OFFSET are now useless,
and to be safe, we could just use "X = (typeof(X)) ~0U;"
as suggested by David.

Cc: David S. Miller <davem@davemloft.net>
Cc: Simon Horman <horms@verge.net.au>
Signed-off-by: NCong Wang <amwang@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The dev_mc_sync_multiple function is currently calling
__hw_addr_sync, and not __hw_addr_sync_multiple.  This will result in
addresses only being synced to the first device from the set.

	Corrected by calling the _multiple variant.
Signed-off-by: NJay Vosburgh <fubar@us.ibm.com>
Reviewed-by: NVlad Yasevich <vyasevic@redhat.com>
Tested-by: NShawn Bohrer <sbohrer@rgmadvisors.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Currently, __hw_addr_sync_one is called in a loop by
__hw_addr_sync_multiple to sync each of a "from" device's hw addresses
to a "to" device.  __hw_addr_sync_one calls __hw_addr_add_ex to attempt
to add each address.  __hw_addr_add_ex is called with global=false, and
sync=true.

	__hw_addr_add_ex checks to see if the new address matches an
address already on the list.  If so, it tests global and sync.  In this
case, sync=true, and it then checks if the address is already synced,
and if so, returns 0.

	This 0 return causes __hw_addr_sync_one to increment the sync_cnt
and refcount for the "from" list's address entry, even though the address
is already synced and has a reference and sync_cnt.  This will cause
the sync_cnt and refcount to increment without bound every time an
addresses is added to the "from" device and synced to the "to" device.

	The fix here has two parts:

	First, when __hw_addr_add_ex finds the address already exists
and is synced, return -EEXIST instead of 0.

	Second, __hw_addr_sync_one checks the error return for -EEXIST,
and if so, it (a) does not add a refcount/sync_cnt, and (b) returns 0
itself so that __hw_addr_sync_multiple will not return an error.
Signed-off-by: NJay Vosburgh <fubar@us.ibm.com>
Reviewed-by: NVlad Yasevich <vyasevic@redhat.com>
Tested-by: NShawn Bohrer <sbohrer@rgmadvisors.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

When an address is added to a subordinate interface (the "to"
list), the address entry in the "from" list is not marked "synced" as
the entry added to the "to" list is.

	When performing the unsync operation (e.g., dev_mc_unsync),
__hw_addr_unsync_one calls __hw_addr_del_entry with the "synced"
parameter set to true for the case when the address reference is being
released from the "from" list.  This causes a test inside to fail,
with the result being that the reference count on the "from" address
is not properly decremeted and the address on the "from" list will
never be freed.

	Correct this by having __hw_addr_unsync_one call the
__hw_addr_del_entry function with the "sync" flag set to false for the
"remove from the from list" case.
Signed-off-by: NJay Vosburgh <fubar@us.ibm.com>
Reviewed-by: NVlad Yasevich <vyasevic@redhat.com>
Tested-by: NShawn Bohrer <sbohrer@rgmadvisors.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The sync_cnt field is not being initialized, which can result
in arbitrary values in the field.  Fixed by initializing it to zero.
Signed-off-by: NJay Vosburgh <fubar@us.ibm.com>
Reviewed-by: NVlad Yasevich <vyasevic@redhat.com>
Tested-by: NShawn Bohrer <sbohrer@rgmadvisors.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The three arrays of strings: af_family_key_strings,
af_family_slock_key_strings and af_family_clock_key_strings have not
VSOCK's string
Signed-off-by: NFederico Vaga <federico.vaga@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This corrects an regression introduced by "net: Use 16bits for *_headers
fields of struct skbuff" when NET_SKBUFF_DATA_USES_OFFSET is not set. In
that case skb->tail will be a pointer however skb->network_header is now
an offset.

This patch corrects the problem by adding a wrapper to return skb tail as
an offset regardless of the value of NET_SKBUFF_DATA_USES_OFFSET. It seems
that skb->tail that this offset may be more than 64k and some care has been
taken to treat such cases as an error.
Signed-off-by: NSimon Horman <horms@verge.net.au>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>