WARN during transport destruction if ib_dealloc_pd() fails. This is
a sign that xprtrdma orphaned one or more RDMA API objects at some
point, which can pin lower layer kernel modules and cause shutdown
to hang.
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Reviewed-by: NSteve Wise <swise@opengridcomputing.com>
Reviewed-by: NSagi Grimberg <sagig@mellanox.com>
Reviewed-by: NDevesh Sharma <devesh.sharma@avagotech.com>
Tested-By: NDevesh Sharma <devesh.sharma@avagotech.com>
Reviewed-by: NDoug Ledford <dledford@redhat.com>
Signed-off-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>

We currently rely on the PMTU discovery of xfrm.
However if a packet is localy sent, the PMTU mechanism
of xfrm tries to to local socket notification what
might not work for applications like ping that don't
check for this. So add pmtu handling to vti6_xmit to
report MTU changes immediately.
Signed-off-by: NSteffen Klassert <steffen.klassert@secunet.com>
Signed-off-by: NAlexander Duyck <alexander.h.duyck@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This reverts commit f96dee13.

It isn't right, ethtool is meant to manage one PHY instance
per netdevice at a time, and this is selected by the SET
command.  Therefore by definition the GET command must only
return the settings for the configured and selected PHY.
Reported-by: NBen Hutchings <ben@decadent.org.uk>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This partially reverts commit 1086bbe9 ("netfilter: ensure number of
counters is >0 in do_replace()") in net/bridge/netfilter/ebtables.c.

Setting rules with ebtables does not work any more with 1086bbe9 place.

There is an error message and no rules set in the end.

e.g.

~# ebtables -t nat -A POSTROUTING --src 12:34:56:78:9a:bc -j DROP
Unable to update the kernel. Two possible causes:
1. Multiple ebtables programs were executing simultaneously. The ebtables
   userspace tool doesn't by default support multiple ebtables programs
running

Reverting the ebtables part of 1086bbe9 makes this work again.
Signed-off-by: NBernhard Thaler <bernhard.thaler@wvnet.at>
Signed-off-by: NPablo Neira Ayuso <pablo@netfilter.org>

While shuffling some code around, dsa_switch_setup_one() was introduced,
and it was modified to return either an error code using ERR_PTR() or a
NULL pointer when running out of memory or failing to setup a switch.

This is a problem for its caler: dsa_switch_setup() which uses IS_ERR()
and expects to find an error code, not a NULL pointer, so we still try
to proceed with dsa_switch_setup() and operate on invalid memory
addresses. This can be easily reproduced by having e.g: the bcm_sf2
driver built-in, but having no such switch, such that drv->setup will
fail.

Fix this by using PTR_ERR() consistently which is both more informative
and avoids for the caller to use IS_ERR_OR_NULL().

Fixes: df197195 ("net: dsa: split dsa_switch_setup into two functions")
Reported-by: NAndrew Lunn <andrew@lunn.ch>
Signed-off-by: NFlorian Fainelli <f.fainelli@gmail.com>
Tested-by: NAndrew Lunn <andrew@lunn.ch>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Linux 3.17 and earlier are explicitly engineered so that if the app
doesn't specifically request a CC module on a listener before the SYN
arrives, then the child gets the system default CC when the connection
is established. See tcp_init_congestion_control() in 3.17 or earlier,
which says "if no choice made yet assign the current value set as
default". The change ("net: tcp: assign tcp cong_ops when tcp sk is
created") altered these semantics, so that children got their parent
listener's congestion control even if the system default had changed
after the listener was created.

This commit returns to those original semantics from 3.17 and earlier,
since they are the original semantics from 2007 in 4d4d3d1e ("[TCP]:
Congestion control initialization."), and some Linux congestion
control workflows depend on that.

In summary, if a listener socket specifically sets TCP_CONGESTION to
"x", or the route locks the CC module to "x", then the child gets
"x". Otherwise the child gets current system default from
net.ipv4.tcp_congestion_control. That's the behavior in 3.17 and
earlier, and this commit restores that.

Fixes: 55d8694f ("net: tcp: assign tcp cong_ops when tcp sk is created")
Cc: Florian Westphal <fw@strlen.de>
Cc: Daniel Borkmann <dborkman@redhat.com>
Cc: Glenn Judd <glenn.judd@morganstanley.com>
Cc: Stephen Hemminger <stephen@networkplumber.org>
Signed-off-by: NNeal Cardwell <ncardwell@google.com>
Signed-off-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NYuchung Cheng <ycheng@google.com>
Acked-by: NDaniel Borkmann <daniel@iogearbox.net>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

We have two problems in UDP stack related to bogus checksums :

1) We return -EAGAIN to application even if receive queue is not empty.
   This breaks applications using edge trigger epoll()

2) Under UDP flood, we can loop forever without yielding to other
   processes, potentially hanging the host, especially on non SMP.

This patch is an attempt to make things better.

We might in the future add extra support for rt applications
wanting to better control time spent doing a recv() in a hostile
environment. For example we could validate checksums before queuing
packets in socket receive queue.
Signed-off-by: NEric Dumazet <edumazet@google.com>
Cc: Willem de Bruijn <willemb@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

br_multicast_query_expired() querier argument is a pointer to
a struct bridge_mcast_querier :

struct bridge_mcast_querier {
        struct br_ip addr;
        struct net_bridge_port __rcu    *port;
};

Intent of the code was to clear port field, not the pointer to querier.

Fixes: 2cd41431 ("bridge: memorize and export selected IGMP/MLD querier port")
Signed-off-by: NEric Dumazet <edumazet@google.com>
Acked-by: NThadeu Lima de Souza Cascardo <cascardo@redhat.com>
Acked-by: NLinus Lüssing <linus.luessing@c0d3.blue>
Cc: Linus Lüssing <linus.luessing@web.de>
Cc: Steinar H. Gunderson <sesse@samfundet.no>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The vti6_rcv_cb and vti_rcv_cb calls were leaving the skb->mark modified
after completing the function.  This resulted in the original skb->mark
value being lost.  Since we only need skb->mark to be set for
xfrm_policy_check we can pull the assignment into the rcv_cb calls and then
just restore the original mark after xfrm_policy_check has been completed.
Signed-off-by: NAlexander Duyck <alexander.h.duyck@redhat.com>
Signed-off-by: NSteffen Klassert <steffen.klassert@secunet.com>

This change makes it so that if a tunnel is defined we just use the mark
from the tunnel instead of the mark from the skb header.  By doing this we
can avoid the need to set skb->mark inside of the tunnel receive functions.
Signed-off-by: NAlexander Duyck <alexander.h.duyck@redhat.com>
Signed-off-by: NSteffen Klassert <steffen.klassert@secunet.com>

Instead of modifying skb->mark we can simply modify the flowi_mark that is
generated as a result of the xfrm_decode_session.  By doing this we don't
need to actually touch the skb->mark and it can be preserved as it passes
out through the tunnel.
Signed-off-by: NAlexander Duyck <alexander.h.duyck@redhat.com>
Signed-off-by: NSteffen Klassert <steffen.klassert@secunet.com>

For mq qdisc, we add per tx queue qdisc to root qdisc
for display purpose, however, that happens too early,
before the new dev->qdisc is finally set, this causes
q->list points to an old root qdisc which is going to be
freed right before assigning with a new one.

Fix this by moving ->attach() after setting dev->qdisc.

For the record, this fixes the following crash:

 ------------[ cut here ]------------
 WARNING: CPU: 1 PID: 975 at lib/list_debug.c:59 __list_del_entry+0x5a/0x98()
 list_del corruption. prev->next should be ffff8800d1998ae8, but was 6b6b6b6b6b6b6b6b
 CPU: 1 PID: 975 Comm: tc Not tainted 4.1.0-rc4+ #1019
 Hardware name: Bochs Bochs, BIOS Bochs 01/01/2011
  0000000000000009 ffff8800d73fb928 ffffffff81a44e7f 0000000047574756
  ffff8800d73fb978 ffff8800d73fb968 ffffffff810790da ffff8800cfc4cd20
  ffffffff814e725b ffff8800d1998ae8 ffffffff82381250 0000000000000000
 Call Trace:
  [<ffffffff81a44e7f>] dump_stack+0x4c/0x65
  [<ffffffff810790da>] warn_slowpath_common+0x9c/0xb6
  [<ffffffff814e725b>] ? __list_del_entry+0x5a/0x98
  [<ffffffff81079162>] warn_slowpath_fmt+0x46/0x48
  [<ffffffff81820eb0>] ? dev_graft_qdisc+0x5e/0x6a
  [<ffffffff814e725b>] __list_del_entry+0x5a/0x98
  [<ffffffff814e72a7>] list_del+0xe/0x2d
  [<ffffffff81822f05>] qdisc_list_del+0x1e/0x20
  [<ffffffff81820cd1>] qdisc_destroy+0x30/0xd6
  [<ffffffff81822676>] qdisc_graft+0x11d/0x243
  [<ffffffff818233c1>] tc_get_qdisc+0x1a6/0x1d4
  [<ffffffff810b5eaf>] ? mark_lock+0x2e/0x226
  [<ffffffff817ff8f5>] rtnetlink_rcv_msg+0x181/0x194
  [<ffffffff817ff72e>] ? rtnl_lock+0x17/0x19
  [<ffffffff817ff72e>] ? rtnl_lock+0x17/0x19
  [<ffffffff817ff774>] ? __rtnl_unlock+0x17/0x17
  [<ffffffff81855dc6>] netlink_rcv_skb+0x4d/0x93
  [<ffffffff817ff756>] rtnetlink_rcv+0x26/0x2d
  [<ffffffff818544b2>] netlink_unicast+0xcb/0x150
  [<ffffffff81161db9>] ? might_fault+0x59/0xa9
  [<ffffffff81854f78>] netlink_sendmsg+0x4fa/0x51c
  [<ffffffff817d6e09>] sock_sendmsg_nosec+0x12/0x1d
  [<ffffffff817d8967>] sock_sendmsg+0x29/0x2e
  [<ffffffff817d8cf3>] ___sys_sendmsg+0x1b4/0x23a
  [<ffffffff8100a1b8>] ? native_sched_clock+0x35/0x37
  [<ffffffff810a1d83>] ? sched_clock_local+0x12/0x72
  [<ffffffff810a1fd4>] ? sched_clock_cpu+0x9e/0xb7
  [<ffffffff810def2a>] ? current_kernel_time+0xe/0x32
  [<ffffffff810b4bc5>] ? lock_release_holdtime.part.29+0x71/0x7f
  [<ffffffff810ddebf>] ? read_seqcount_begin.constprop.27+0x5f/0x76
  [<ffffffff810b6292>] ? trace_hardirqs_on_caller+0x17d/0x199
  [<ffffffff811b14d5>] ? __fget_light+0x50/0x78
  [<ffffffff817d9808>] __sys_sendmsg+0x42/0x60
  [<ffffffff817d9838>] SyS_sendmsg+0x12/0x1c
  [<ffffffff81a50e97>] system_call_fastpath+0x12/0x6f
 ---[ end trace ef29d3fb28e97ae7 ]---

For long term, we probably need to clean up the qdisc_graft() code
in case it hides other bugs like this.

Fixes: 95dc1929 ("pkt_sched: give visibility to mq slave qdiscs")
Cc: Jamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: NCong Wang <xiyou.wangcong@gmail.com>
Acked-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

got a rare NULL pointer dereference in clear_bit
Signed-off-by: NMark Salyzyn <salyzyn@android.com>
Acked-by: NHannes Frederic Sowa <hannes@stressinduktion.org>
----
v2: switch to sock_flag(sk, SOCK_DEAD) and added net/caif/caif_socket.c
v3: return -ECONNRESET in upstream caller of wait function for SOCK_DEAD
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Following lockdep splat was reported :

[   29.382286] ===============================
[   29.382315] [ INFO: suspicious RCU usage. ]
[   29.382344] 4.1.0-0.rc0.git11.1.fc23.x86_64 #1 Not tainted
[   29.382380] -------------------------------
[   29.382409] net/bridge/br_private.h:626 suspicious
rcu_dereference_check() usage!
[   29.382455]
               other info that might help us debug this:

[   29.382507]
               rcu_scheduler_active = 1, debug_locks = 0
[   29.382549] 2 locks held by swapper/0/0:
[   29.382576]  #0:  (((&p->forward_delay_timer))){+.-...}, at:
[<ffffffff81139f75>] call_timer_fn+0x5/0x4f0
[   29.382660]  #1:  (&(&br->lock)->rlock){+.-...}, at:
[<ffffffffa0450dc1>] br_forward_delay_timer_expired+0x31/0x140
[bridge]
[   29.382754]
               stack backtrace:
[   29.382787] CPU: 0 PID: 0 Comm: swapper/0 Not tainted
4.1.0-0.rc0.git11.1.fc23.x86_64 #1
[   29.382838] Hardware name: LENOVO 422916G/LENOVO, BIOS A1KT53AUS 04/07/2015
[   29.382882]  0000000000000000 3ebfc20364115825 ffff880666603c48
ffffffff81892d4b
[   29.382943]  0000000000000000 ffffffff81e124e0 ffff880666603c78
ffffffff8110bcd7
[   29.383004]  ffff8800785c9d00 ffff88065485ac58 ffff880c62002800
ffff880c5fc88ac0
[   29.383065] Call Trace:
[   29.383084]  <IRQ>  [<ffffffff81892d4b>] dump_stack+0x4c/0x65
[   29.383130]  [<ffffffff8110bcd7>] lockdep_rcu_suspicious+0xe7/0x120
[   29.383178]  [<ffffffffa04520f9>] br_fill_ifinfo+0x4a9/0x6a0 [bridge]
[   29.383225]  [<ffffffffa045266b>] br_ifinfo_notify+0x11b/0x4b0 [bridge]
[   29.383271]  [<ffffffffa0450d90>] ? br_hold_timer_expired+0x70/0x70 [bridge]
[   29.383320]  [<ffffffffa0450de8>]
br_forward_delay_timer_expired+0x58/0x140 [bridge]
[   29.383371]  [<ffffffffa0450d90>] ? br_hold_timer_expired+0x70/0x70 [bridge]
[   29.383416]  [<ffffffff8113a033>] call_timer_fn+0xc3/0x4f0
[   29.383454]  [<ffffffff81139f75>] ? call_timer_fn+0x5/0x4f0
[   29.383493]  [<ffffffff8110a90f>] ? lock_release_holdtime.part.29+0xf/0x200
[   29.383541]  [<ffffffffa0450d90>] ? br_hold_timer_expired+0x70/0x70 [bridge]
[   29.383587]  [<ffffffff8113a6a4>] run_timer_softirq+0x244/0x490
[   29.383629]  [<ffffffff810b68cc>] __do_softirq+0xec/0x670
[   29.383666]  [<ffffffff810b70d5>] irq_exit+0x145/0x150
[   29.383703]  [<ffffffff8189f506>] smp_apic_timer_interrupt+0x46/0x60
[   29.383744]  [<ffffffff8189d523>] apic_timer_interrupt+0x73/0x80
[   29.383782]  <EOI>  [<ffffffff816f131f>] ? cpuidle_enter_state+0x5f/0x2f0
[   29.383832]  [<ffffffff816f131b>] ? cpuidle_enter_state+0x5b/0x2f0

Problem here is that br_forward_delay_timer_expired() is a timer
handler, calling br_ifinfo_notify() which assumes either rcu_read_lock()
or RTNL are held.

Simplest fix seems to add rcu read lock section.
Signed-off-by: NEric Dumazet <edumazet@google.com>
Reported-by: NJosh Boyer <jwboyer@fedoraproject.org>
Reported-by: NDominick Grift <dac.override@gmail.com>
Cc: Vlad Yasevich <vyasevich@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

When trying to configure the settings for PHY1, using commands
like 'ethtool -s eth0 phyad 1 speed 100', the 'ethtool' seems to
modify other settings apart from the speed of the PHY1, in the
above case.

The ethtool seems to query the settings for PHY0, and use this
as the base to apply the new settings to the PHY1. This is
causing the other settings of the PHY 1 to be wrongly
configured.

The issue is caused by the '_ethtool_get_settings()' API, which
gets called because of the 'ETHTOOL_GSET' command, is clearing
the 'cmd' pointer (of type 'struct ethtool_cmd') by calling
memset. This clears all the parameters (if any) passed for the
'ETHTOOL_GSET' cmd. So the driver's callback is always invoked
with 'cmd->phy_address' as '0'.

The '_ethtool_get_settings()' is called from other files in the
'net/core'. So the fix is applied to the 'ethtool_get_settings()'
which is only called in the context of the 'ethtool'.
Signed-off-by: NArun Parameswaran <aparames@broadcom.com>
Reviewed-by: NRay Jui <rjui@broadcom.com>
Reviewed-by: NScott Branden <sbranden@broadcom.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

When more than a multicast address is present in a MLDv2 report, all but
the first address is ignored, because the code breaks out of the loop if
there has not been an error adding that address.

This has caused failures when two guests connected through the bridge
tried to communicate using IPv6. Neighbor discoveries would not be
transmitted to the other guest when both used a link-local address and a
static address.

This only happens when there is a MLDv2 querier in the network.

The fix will only break out of the loop when there is a failure adding a
multicast address.

The mdb before the patch:

dev ovirtmgmt port vnet0 grp ff02::1:ff7d:6603 temp
dev ovirtmgmt port vnet1 grp ff02::1:ff7d:6604 temp
dev ovirtmgmt port bond0.86 grp ff02::2 temp

After the patch:

dev ovirtmgmt port vnet0 grp ff02::1:ff7d:6603 temp
dev ovirtmgmt port vnet1 grp ff02::1:ff7d:6604 temp
dev ovirtmgmt port bond0.86 grp ff02::fb temp
dev ovirtmgmt port bond0.86 grp ff02::2 temp
dev ovirtmgmt port bond0.86 grp ff02::d temp
dev ovirtmgmt port vnet0 grp ff02::1:ff00:76 temp
dev ovirtmgmt port bond0.86 grp ff02::16 temp
dev ovirtmgmt port vnet1 grp ff02::1:ff00:77 temp
dev ovirtmgmt port bond0.86 grp ff02::1:ff00:def temp
dev ovirtmgmt port bond0.86 grp ff02::1:ffa1:40bf temp

Fixes: 08b202b6 ("bridge br_multicast: IPv6 MLD support.")
Reported-by: NRik Theys <Rik.Theys@esat.kuleuven.be>
Signed-off-by: NThadeu Lima de Souza Cascardo <cascardo@redhat.com>
Tested-by: NRik Theys <Rik.Theys@esat.kuleuven.be>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

When replacing an IPv4 route, tb_id member of the new fib_alias
structure is not set in the replace code path so that the new route is
ignored.

Fixes: 0ddcf43d ("ipv4: FIB Local/MAIN table collapse")
Signed-off-by: NMichal Kubecek <mkubecek@suse.cz>
Acked-by: NAlexander Duyck <alexander.h.duyck@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

ip_error does not check if in_dev is NULL before dereferencing it.

IThe following sequence of calls is possible:
CPU A                          CPU B
ip_rcv_finish
    ip_route_input_noref()
        ip_route_input_slow()
                               inetdev_destroy()
    dst_input()

With the result that a network device can be destroyed while processing
an input packet.

A crash was triggered with only unicast packets in flight, and
forwarding enabled on the only network device.   The error condition
was created by the removal of the network device.

As such it is likely the that error code was -EHOSTUNREACH, and the
action taken by ip_error (if in_dev had been accessible) would have
been to not increment any counters and to have tried and likely failed
to send an icmp error as the network device is going away.

Therefore handle this weird case by just dropping the packet if
!in_dev.  It will result in dropping the packet sooner, and will not
result in an actual change of behavior.

Fixes: 251da413 ("ipv4: Cache ip_error() routes even when not forwarding.")
Reported-by: NVittorio Gambaletta <linuxbugs@vittgam.net>
Tested-by: NVittorio Gambaletta <linuxbugs@vittgam.net>
Signed-off-by: NVittorio Gambaletta <linuxbugs@vittgam.net>
Signed-off-by: N"Eric W. Biederman" <ebiederm@xmission.com>
Acked-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Taking socket spinlock in tcp_get_info() can deadlock, as
inet_diag_dump_icsk() holds the &hashinfo->ehash_locks[i],
while packet processing can use the reverse locking order.

We could avoid this locking for TCP_LISTEN states, but lockdep would
certainly get confused as all TCP sockets share same lockdep classes.

[  523.722504] ======================================================
[  523.728706] [ INFO: possible circular locking dependency detected ]
[  523.734990] 4.1.0-dbg-DEV #1676 Not tainted
[  523.739202] -------------------------------------------------------
[  523.745474] ss/18032 is trying to acquire lock:
[  523.750002]  (slock-AF_INET){+.-...}, at: [<ffffffff81669d44>] tcp_get_info+0x2c4/0x360
[  523.758129]
[  523.758129] but task is already holding lock:
[  523.763968]  (&(&hashinfo->ehash_locks[i])->rlock){+.-...}, at: [<ffffffff816bcb75>] inet_diag_dump_icsk+0x1d5/0x6c0
[  523.774661]
[  523.774661] which lock already depends on the new lock.
[  523.774661]
[  523.782850]
[  523.782850] the existing dependency chain (in reverse order) is:
[  523.790326]
-> #1 (&(&hashinfo->ehash_locks[i])->rlock){+.-...}:
[  523.796599]        [<ffffffff811126bb>] lock_acquire+0xbb/0x270
[  523.802565]        [<ffffffff816f5868>] _raw_spin_lock+0x38/0x50
[  523.808628]        [<ffffffff81665af8>] __inet_hash_nolisten+0x78/0x110
[  523.815273]        [<ffffffff816819db>] tcp_v4_syn_recv_sock+0x24b/0x350
[  523.822067]        [<ffffffff81684d41>] tcp_check_req+0x3c1/0x500
[  523.828199]        [<ffffffff81682d09>] tcp_v4_do_rcv+0x239/0x3d0
[  523.834331]        [<ffffffff816842fe>] tcp_v4_rcv+0xa8e/0xc10
[  523.840202]        [<ffffffff81658fa3>] ip_local_deliver_finish+0x133/0x3e0
[  523.847214]        [<ffffffff81659a9a>] ip_local_deliver+0xaa/0xc0
[  523.853440]        [<ffffffff816593b8>] ip_rcv_finish+0x168/0x5c0
[  523.859624]        [<ffffffff81659db7>] ip_rcv+0x307/0x420

Lets use u64_sync infrastructure instead. As a bonus, 64bit
arches get optimized, as these are nop for them.

Fixes: 0df48c26 ("tcp: add tcpi_bytes_acked to tcp_info")
Signed-off-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Vijay reported that a loop as simple as ...

  while true; do
    tc qdisc add dev foo root handle 1: prio
    tc filter add dev foo parent 1: u32 match u32 0 0  flowid 1
    tc qdisc del dev foo root
    rmmod cls_u32
  done

... will panic the kernel. Moreover, he bisected the change
apparently introducing it to 78fd1d0a ("netlink: Re-add
locking to netlink_lookup() and seq walker").

The removal of synchronize_net() from the netlink socket
triggering the qdisc to be removed, seems to have uncovered
an RCU resp. module reference count race from the tc API.
Given that RCU conversion was done after e341694e ("netlink:
Convert netlink_lookup() to use RCU protected hash table")
which added the synchronize_net() originally, occasion of
hitting the bug was less likely (not impossible though):

When qdiscs that i) support attaching classifiers and,
ii) have at least one of them attached, get deleted, they
invoke tcf_destroy_chain(), and thus call into ->destroy()
handler from a classifier module.

After RCU conversion, all classifier that have an internal
prio list, unlink them and initiate freeing via call_rcu()
deferral.

Meanhile, tcf_destroy() releases already reference to the
tp->ops->owner module before the queued RCU callback handler
has been invoked.

Subsequent rmmod on the classifier module is then not prevented
since all module references are already dropped.

By the time, the kernel invokes the RCU callback handler from
the module, that function address is then invalid.

One way to fix it would be to add an rcu_barrier() to
unregister_tcf_proto_ops() to wait for all pending call_rcu()s
to complete.

synchronize_rcu() is not appropriate as under heavy RCU
callback load, registered call_rcu()s could be deferred
longer than a grace period. In case we don't have any pending
call_rcu()s, the barrier is allowed to return immediately.

Since we came here via unregister_tcf_proto_ops(), there
are no users of a given classifier anymore. Further nested
call_rcu()s pointing into the module space are not being
done anywhere.

Only cls_bpf_delete_prog() may schedule a work item, to
unlock pages eventually, but that is not in the range/context
of cls_bpf anymore.

Fixes: 25d8c0d5 ("net: rcu-ify tcf_proto")
Fixes: 9888faef ("net: sched: cls_basic use RCU")
Reported-by: NVijay Subramanian <subramanian.vijay@gmail.com>
Signed-off-by: NDaniel Borkmann <daniel@iogearbox.net>
Cc: John Fastabend <john.r.fastabend@intel.com>
Cc: Eric Dumazet <edumazet@google.com>
Cc: Thomas Graf <tgraf@suug.ch>
Cc: Jamal Hadi Salim <jhs@mojatatu.com>
Cc: Alexei Starovoitov <ast@plumgrid.com>
Tested-by: NVijay Subramanian <subramanian.vijay@gmail.com>
Acked-by: NAlexei Starovoitov <ast@plumgrid.com>
Acked-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

As we're now always including the high bits of the sequence number
in the IV generation process we need to ensure that they don't
contain crap.

This patch ensures that the high sequence bits are always zeroed
so that we don't leak random data into the IV.
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: NSteffen Klassert <steffen.klassert@secunet.com>

This reverts commit ba9d114e.

.. which introduced a regression that prevented all lingering requests
requeued in kick_requests() from ever being sent to the OSDs, resulting
in a lot of missed notifies.  In retrospect it's pretty obvious that
r_req_lru_item item in the case of lingering requests can be used not
only for notarget, but also for unsent linkage due to how tightly
actual map and enqueue operations are coupled in __map_request().

The assertion that was being silenced is taken care of in the previous
("libceph: request a new osdmap if lingering request maps to no osd")
commit: by always kicking homeless lingering requests we ensure that
none of them ends up on the notarget list outside of the critical
section guarded by request_mutex.

Cc: stable@vger.kernel.org # 3.18+, needs b0494532 "libceph: request a new osdmap if lingering request maps to no osd"
Signed-off-by: NIlya Dryomov <idryomov@gmail.com>
Reviewed-by: NSage Weil <sage@redhat.com>

This commit does two things.  First, if there are any homeless
lingering requests, we now request a new osdmap even if the osdmap that
is being processed brought no changes, i.e. if a given lingering
request turned homeless in one of the previous epochs and remained
homeless in the current epoch.  Not doing so leaves us with a stale
osdmap and as a result we may miss our window for reestablishing the
watch and lose notifies.

MON=1 OSD=1:

    # cat linger-needmap.sh
    #!/bin/bash
    rbd create --size 1 test
    DEV=$(rbd map test)
    ceph osd out 0
    rbd map dne/dne # obtain a new osdmap as a side effect (!)
    sleep 1
    ceph osd in 0
    rbd resize --size 2 test
    # rbd info test | grep size -> 2M
    # blockdev --getsize $DEV -> 1M

N.B.: Not obtaining a new osdmap in between "osd out" and "osd in"
above is enough to make it miss that resize notify, but that is a
bug^Wlimitation of ceph watch/notify v1.

Second, homeless lingering requests are now kicked just like those
lingering requests whose mapping has changed.  This is mainly to
recognize that a homeless lingering request makes no sense and to
preserve the invariant that a registered lingering request is not
sitting on any of r_req_lru_item lists.  This spares us a WARN_ON,
which commit ba9d114e ("libceph: clear r_req_lru_item in
__unregister_linger_request()") tried to fix the _wrong_ way.

Cc: stable@vger.kernel.org # 3.10+
Signed-off-by: NIlya Dryomov <idryomov@gmail.com>
Reviewed-by: NSage Weil <sage@redhat.com>

When replacing an IPv6 multipath route with "ip route replace", i.e.
NLM_F_CREATE | NLM_F_REPLACE, fib6_add_rt2node() replaces only first
matching route without fixing its siblings, resulting in corrupted
siblings linked list; removing one of the siblings can then end in an
infinite loop.

IPv6 ECMP implementation is a bit different from IPv4 so that route
replacement cannot work in exactly the same way. This should be a
reasonable approximation:

1. If the new route is ECMP-able and there is a matching ECMP-able one
already, replace it and all its siblings (if any).

2. If the new route is ECMP-able and no matching ECMP-able route exists,
replace first matching non-ECMP-able (if any) or just add the new one.

3. If the new route is not ECMP-able, replace first matching
non-ECMP-able route (if any) or add the new route.

We also need to remove the NLM_F_REPLACE flag after replacing old
route(s) by first nexthop of an ECMP route so that each subsequent
nexthop does not replace previous one.

Fixes: 51ebd318 ("ipv6: add support of equal cost multipath (ECMP)")
Signed-off-by: NMichal Kubecek <mkubecek@suse.cz>
Acked-by: NNicolas Dichtel <nicolas.dichtel@6wind.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

If adding a nexthop of an IPv6 multipath route fails, comment in
ip6_route_multipath() says we are going to delete all nexthops already
added. However, current implementation deletes even the routes it
hasn't even tried to add yet. For example, running

  ip route add 1234:5678::/64 \
      nexthop via fe80::aa dev dummy1 \
      nexthop via fe80::bb dev dummy1 \
      nexthop via fe80::cc dev dummy1

twice results in removing all routes first command added.

Limit the second (delete) run to nexthops that succeeded in the first
(add) run.

Fixes: 51ebd318 ("ipv6: add support of equal cost multipath (ECMP)")
Signed-off-by: NMichal Kubecek <mkubecek@suse.cz>
Acked-by: NNicolas Dichtel <nicolas.dichtel@6wind.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Some splats I was seeing:

 (a) WARNING: CPU: 1 PID: 0 at /devel/src/linux/net/mac80211/wep.c:102 ieee80211_wep_add_iv
 (b) WARNING: CPU: 1 PID: 0 at /devel/src/linux/net/mac80211/wpa.c:73 ieee80211_tx_h_michael_mic_add
 (c) WARNING: CPU: 3 PID: 0 at /devel/src/linux/net/mac80211/wpa.c:433 ieee80211_crypto_ccmp_encrypt

I've seen (a) and (b) with ath9k hw crypto and (c)
with ath9k sw crypto. All of them were related to
insufficient skb tailroom and I was able to
trigger these with ping6 program.

AP_VLANs may inherit crypto keys from parent AP.
This wasn't considered and yielded problems in
some setups resulting in inability to transmit
data because mac80211 wouldn't resize skbs when
necessary and subsequently drop some packets due
to insufficient tailroom.

For efficiency purposes don't inspect both AP_VLAN
and AP sdata looking for tailroom counter. Instead
update AP_VLAN tailroom counters whenever their
master AP tailroom counter changes.
Signed-off-by: NMichal Kazior <michal.kazior@tieto.com>
Signed-off-by: NJohannes Berg <johannes.berg@intel.com>

Due to remain-on-channel scheduling delays, when we split an ROC
while coalescing, we'll usually get a picture like this:

existing ROC:  |------------------|
current time:              ^
new ROC:                   |------|              |-------|

If the expected response frames are then transmitted by the peer
in the hole between the two fragments of the new ROC, we miss
them and the process (e.g. ANQP query) fails.

mac80211 expects that the window to miss something is small:

existing ROC:  |------------------|
new ROC:                   |------||-------|

but that's normally not the case.

To avoid this problem, coalesce only if the new ROC's duration
is <= the remaining time on the existing one:

existing ROC:  |------------------|
new ROC:                   |-----|

and never split a new one but schedule it afterwards instead:

existing ROC:  |------------------|
new ROC:                                       |-------------|

type=bugfix
bug=not-tracked
fixes=unknown
Reported-by: NMatti Gottlieb <matti.gottlieb@intel.com>
Reviewed-by: NEliadX Peller <eliad@wizery.com>
Reviewed-by: NMatti Gottlieb <matti.gottlieb@intel.com>
Tested-by: NMatti Gottlieb <matti.gottlieb@intel.com>
Signed-off-by: NJohannes Berg <johannes.berg@intel.com>

This reverts commit c055d5b0.

There are two issues:
'dnat_took_place' made me think that this is related to
-j DNAT/MASQUERADE.

But thats only one part of the story.  This is also relevant for SNAT
when we undo snat translation in reverse/reply direction.

Furthermore, I originally wanted to do this mainly to avoid
storing ipv6 addresses once we make DNAT/REDIRECT work
for ipv6 on bridges.

However, I forgot about SNPT/DNPT which is stateless.

So we can't escape storing address for ipv6 anyway. Might as
well do it for ipv4 too.
Reported-and-tested-by: NBernhard Thaler <bernhard.thaler@wvnet.at>
Signed-off-by: NFlorian Westphal <fw@strlen.de>
Signed-off-by: NPablo Neira Ayuso <pablo@netfilter.org>

After improving setsockopt() coverage in trinity, I started triggering
vmalloc failures pretty reliably from this code path:

warn_alloc_failed+0xe9/0x140
__vmalloc_node_range+0x1be/0x270
vzalloc+0x4b/0x50
__do_replace+0x52/0x260 [ip_tables]
do_ipt_set_ctl+0x15d/0x1d0 [ip_tables]
nf_setsockopt+0x65/0x90
ip_setsockopt+0x61/0xa0
raw_setsockopt+0x16/0x60
sock_common_setsockopt+0x14/0x20
SyS_setsockopt+0x71/0xd0

It turns out we don't validate that the num_counters field in the
struct we pass in from userspace is initialized.

The same problem also exists in ebtables, arptables, ipv6, and the
compat variants.
Signed-off-by: NDave Jones <davej@codemonkey.org.uk>
Signed-off-by: NPablo Neira Ayuso <pablo@netfilter.org>

nfnetlink_{log,queue}_init() register the netlink callback nf*_rcv_nl_event
before registering the pernet_subsys, but the callback relies on data
structures allocated by pernet init functions.

When nfnetlink_{log,queue} is loaded, if a netlink message is received after
the netlink callback is registered but before the pernet_subsys is registered,
the kernel will panic in the sequence

nfulnl_rcv_nl_event
  nfnl_log_pernet
    net_generic
      BUG_ON(id == 0)  where id is nfnl_log_net_id.

The panic can be easily reproduced in 4.0.3 by:

while true ;do modprobe nfnetlink_log ; rmmod nfnetlink_log ; done &
while true ;do ip netns add dummy ; ip netns del dummy ; done &

This patch moves register_pernet_subsys to earlier in nfnetlink_log_init.

Notice that the BUG_ON hit in 4.0.3 was recently removed in 2591ffd3
["netns: remove BUG_ONs from net_generic()"].
Signed-off-by: NFrancesco Ruggeri <fruggeri@arista.com>
Signed-off-by: NPablo Neira Ayuso <pablo@netfilter.org>

After sending the new data packets to probe (step 2), F-RTO may
incorrectly send more probes if the next ACK advances SND_UNA and
does not sack new packet. However F-RTO RFC 5682 probes at most
once. This bug may cause sender to always send new data instead of
repairing holes, inducing longer HoL blocking on the receiver for
the application.
Signed-off-by: NYuchung Cheng <ycheng@google.com>
Signed-off-by: NNeal Cardwell <ncardwell@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Undo based on TCP timestamps should only happen on ACKs that advance
SND_UNA, according to the Eifel algorithm in RFC 3522:

Section 3.2:

  (4) If the value of the Timestamp Echo Reply field of the
      acceptable ACK's Timestamps option is smaller than the
      value of RetransmitTS, then proceed to step (5),

Section Terminology:
   We use the term 'acceptable ACK' as defined in [RFC793].  That is an
   ACK that acknowledges previously unacknowledged data.

This is because upon receiving an out-of-order packet, the receiver
returns the last timestamp that advances RCV_NXT, not the current
timestamp of the packet in the DUPACK. Without checking the flag,
the DUPACK will cause tcp_packet_delayed() to return true and
tcp_try_undo_loss() will revert cwnd reduction.

Note that we check the condition in CA_Recovery already by only
calling tcp_try_undo_partial() if FLAG_SND_UNA_ADVANCED is set or
tcp_try_undo_recovery() if snd_una crosses high_seq.
Signed-off-by: NYuchung Cheng <ycheng@google.com>
Signed-off-by: NNeal Cardwell <ncardwell@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Commit <5cf3d461> ("udp: Simplify__udp*_lib_mcast_deliver")
simplified the filter for incoming IPv6 multicast but removed
the check of the local socket address and the UDP destination
address.

This patch restores the filter to prevent sockets bound to a IPv6
multicast IP to receive other UDP traffic link unicast.
Signed-off-by: NHenning Rogge <hrogge@gmail.com>
Fixes: 5cf3d461 ("udp: Simplify__udp*_lib_mcast_deliver")
Cc: "David S. Miller" <davem@davemloft.net>
Acked-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

No matter how the driver manages its NAPI context, there's no way
sending frames to it from a timer can be correct, since it would
corrupt the internal GRO lists.

To avoid that, always use the non-NAPI path when releasing frames
from the timer.

Cc: stable@vger.kernel.org
Reported-by: NJean Trivelly <jean.trivelly@intel.com>
Signed-off-by: NJohannes Berg <johannes.berg@intel.com>

commit 1d13a96c ("ipv6: tcp: fix flowlabel value in ACK messages
send from TIME_WAIT") added the flow label in the last TCP packets.
Unfortunately, it was not casted properly.

This patch replace the buggy shift with be32_to_cpu/cpu_to_be32.

Fixes: 1d13a96c ("ipv6: tcp: fix flowlabel value in ACK messages")
Reported-by: NEric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: NFlorent Fourcot <florent.fourcot@enst-bretagne.fr>
Acked-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Before the patch, the command 'ip link add bond2 type bond mode 802.3ad'
causes the kernel to send a rtnl message for the bond2 interface, with an
ifindex 0.

'ip monitor' shows:
0: bond2: <BROADCAST,MULTICAST,MASTER> mtu 1500 state DOWN group default
    link/ether 00:00:00:00:00:00 brd ff:ff:ff:ff:ff:ff
9: bond2@NONE: <BROADCAST,MULTICAST,MASTER> mtu 1500 qdisc noop state DOWN group default
    link/ether ea:3e:1f:53:92:7b brd ff:ff:ff:ff:ff:ff
[snip]

The patch fixes the spotted bug by checking in bond driver if the interface
is registered before calling the notifier chain.
It also adds a check in rtmsg_ifinfo() to prevent this kind of bug in the
future.

Fixes: d4261e56 ("bonding: create netlink event when bonding option is changed")
CC: Jiri Pirko <jiri@resnulli.us>
Reported-by: NJulien Meunier <julien.meunier@6wind.com>
Signed-off-by: NNicolas Dichtel <nicolas.dichtel@6wind.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The commit c5adde94 ("netlink:
eliminate nl_sk_hash_lock") breaks the autobind retry mechanism
because it doesn't reset portid after a failed netlink_insert.

This means that should autobind fail the first time around, then
the socket will be stuck in limbo as it can never be bound again
since it already has a non-zero portid.

Fixes: c5adde94 ("netlink: eliminate nl_sk_hash_lock")
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The values 0x00000000-0xfffffeff are reserved for userspace datatype. When,
deleting set elements with maps, a bogus warning is triggered.

WARNING: CPU: 0 PID: 11133 at net/netfilter/nf_tables_api.c:4481 nft_data_uninit+0x35/0x40 [nf_tables]()

This fixes the check accordingly to enum definition in
include/linux/netfilter/nf_tables.h

Fixes: https://bugzilla.netfilter.org/show_bug.cgi?id=1013Signed-off-by: NMirek Kratochvil <exa.exa@gmail.com>
Signed-off-by: NPablo Neira Ayuso <pablo@netfilter.org>

In compliance with RFC5961, the network stack send challenge ACK in
response to spurious SYN packets, since commit 0c228e83 ("tcp:
Restore RFC5961-compliant behavior for SYN packets").

This pose a problem for netfilter conntrack in state LAST_ACK, because
this challenge ACK is (falsely) seen as ACKing last FIN, causing a
false state transition (into TIME_WAIT).

The challenge ACK is hard to distinguish from real last ACK.  Thus,
solution introduce a flag that tracks the potential for seeing a
challenge ACK, in case a SYN packet is let through and current state
is LAST_ACK.

When conntrack transition LAST_ACK to TIME_WAIT happens, this flag is
used for determining if we are expecting a challenge ACK.

Scapy based reproducer script avail here:
 https://github.com/netoptimizer/network-testing/blob/master/scapy/tcp_hacks_3WHS_LAST_ACK.py

Fixes: 0c228e83 ("tcp: Restore RFC5961-compliant behavior for SYN packets")
Signed-off-by: NJesper Dangaard Brouer <brouer@redhat.com>
Acked-by: NJozsef Kadlecsik <kadlec@blackhole.kfki.hu>
Signed-off-by: NPablo Neira Ayuso <pablo@netfilter.org>

With TPROXY=y but DEFRAG_IPV6=m we get build failure:

net/built-in.o: In function `tproxy_tg_init':
net/netfilter/xt_TPROXY.c:588: undefined reference to `nf_defrag_ipv6_enable'

If DEFRAG_IPV6 is modular, TPROXY must be too.
(or both must be builtin).

This enforces =m for both.
Reported-and-tested-by: NLiu Hua <liusdu@126.com>
Signed-off-by: NFlorian Westphal <fw@strlen.de>
Signed-off-by: NPablo Neira Ayuso <pablo@netfilter.org>