Merge back net-next to get wireless driver changes (from Kalle)
to be able to create the API change across all trees properly.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>

This patch fixes a bug in hwmp_preq_frame_process where the wrong metric
can be used when forwarding a PREQ. This happens because the code uses
the same metric variable to record the value of the metric to the source
of the PREQ and the value of the metric to the target of the PREQ.

This comes into play when both reply and forward are set which happens
when IEEE80211_PREQ_PROACTIVE_PREP_FLAG is set and when MP_F_DO | MP_F_RF
is set. The original code had a special case to handle the first case
but not the second.

The patch uses distinct variables for the two metrics which makes the
code flow much clearer and removes the need to restore the original
value of metric when forwarding.
Signed-off-by: Alexis Green <agreen@cococorp.com>
CC: Jesse Jones <jjones@cococorp.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>

Antonio Quartulli says:

====================
Included changes:
- use common Jenkins hash instead of private implementation
- extend internal routing API
- properly re-arrange header files inclusion
- clarify precedence between '&' and '?'
- remove unused ethhdr variable in batadv_gw_dhcp_recipient_get()
- ensure per-VLAN structs are updated upon MAC change
====================
Signed-off-by: David S. Miller <davem@davemloft.net>

In mesh mode there is a race between establishing links and processing
rates and capabilities in beacons. This is very noticeable with slow
beacons (e.g. beacon intervals of 1s) and manifested for us as stations
using minstrel when minstrel_ht should be used. Fixed by changing
mesh_sta_info_init so that it always checks rates and such if it has not
already done so.
Signed-off-by: Alexis Green <agreen@cococorp.com>
CC: Jesse Jones <jjones@cococorp.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>

The csa counter has moved from sdata to beacon/presp but
it is not updated accordingly for mesh and ibss. Fix this.

Fixes: af296bdb ("mac80211: move csa counters from sdata to beacon/presp")
Signed-off-by: Chun-Yeow Yeoh <yeohchunyeow@gmail.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>

The last hop metric should refer to link cost (this is how
hwmp_route_info_get uses it for example). But in mesh_rx_path_sel_frame
we are not dealing with link cost but with the total cost to the origin
of a PREQ or PREP.
Signed-off-by: Alexis Green <agreen@cococorp.com>
CC: Jesse Jones <jjones@cococorp.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>

Channels in 2.4GHz band overlap, this means that if we
send a probe request on channel 1 and then move to channel
2, we will hear the probe response on channel 2. In this
case, the RSSI will be lower than if we had heard it on
the channel on which it was sent (1 in this case).

The scan result ignores those invalid values and the
station last signal should not be updated as well.

In case the scan determines the signal to be invalid turn on
the flag so the station last signal will not be updated with
the value and thus user space probing for NL80211_STA_INFO_SIGNAL
and NL80211_STA_INFO_SIGNAL_AVG will not get this invalid RSSI
value.
Signed-off-by: Sara Sharon <sara.sharon@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>

There were a few rare cases when upon
authentication failure channel wasn't released.
This could cause stale pointers to remain in
chanctx assigned_vifs after interface removal and
trigger general protection fault later.

This could be triggered, e.g. on ath10k with the
following steps:

 1. start an AP
 2. create 2 extra vifs on ath10k host
 3. connect vif1 to the AP
 4. connect vif2 to the AP
    (auth fails because ath10k firmware isn't able
     to maintain 2 peers with colliding AP mac
     addresses across vifs and consequently
     refuses sta_info_insert() in
     ieee80211_prep_connection())
 5. remove the 2 extra vifs
 6. goto step 2; at step 3 kernel was crashing:

 general protection fault: 0000 [#1] SMP DEBUG_PAGEALLOC
 Modules linked in: ath10k_pci ath10k_core ath
 ...
 Call Trace:
  [<ffffffff81a2dabb>] ieee80211_check_combinations+0x22b/0x290
  [<ffffffff819fb825>] ? ieee80211_check_concurrent_iface+0x125/0x220
  [<ffffffff8180f664>] ? netpoll_poll_disable+0x84/0x100
  [<ffffffff819fb833>] ieee80211_check_concurrent_iface+0x133/0x220
  [<ffffffff81a0029e>] ieee80211_open+0x3e/0x80
  [<ffffffff817f2d26>] __dev_open+0xb6/0x130
  [<ffffffff817f3051>] __dev_change_flags+0xa1/0x170
 ...
 RIP  [<ffffffff81a23140>] ieee80211_chanctx_radar_detect+0xa0/0x170

 (gdb) l * ieee80211_chanctx_radar_detect+0xa0
 0xffffffff81a23140 is in ieee80211_chanctx_radar_detect (/devel/src/linux/net/mac80211/util.c:3182).
 3177             */
 3178            WARN_ON(ctx->replace_state == IEEE80211_CHANCTX_REPLACES_OTHER &&
 3179                    !list_empty(&ctx->assigned_vifs));
 3180
 3181            list_for_each_entry(sdata, &ctx->assigned_vifs, assigned_chanctx_list)
 3182                    if (sdata->radar_required)
 3183                            radar_detect |= BIT(sdata->vif.bss_conf.chandef.width);
 3184
 3185            return radar_detect;
Signed-off-by: Michal Kazior <michal.kazior@tieto.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>

FIF_PROMISC_IN_BSS was removed in commit df140465
("mac80211: remove support for IFF_PROMISC").
Signed-off-by: Jakub Kicinski <kubakici@wp.pl>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>

The conversion to the fast-xmit path lost proper aggregation session
timeout handling - the last_tx wasn't set on that path and the timer
would therefore incorrectly tear down the session periodically (with
those drivers/rate control algorithms that have a timeout.)

In case of iwlwifi, this was every 5 seconds and caused significant
throughput degradation.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>

bpf_tail_call() arguments:

 - ctx......: Context pointer
 - jmp_table: One of BPF_MAP_TYPE_PROG_ARRAY maps used as the jump table
 - index....: Index in the jump table

In this implementation s390x JIT does stack unwinding and jumps into the
callee program prologue. Caller and callee use the same stack.

With this patch a tail call generates the following code on s390x:

 if (index >= array->map.max_entries)
         goto out
 000003ff8001c7e4: e31030100016   llgf    %r1,16(%r3)
 000003ff8001c7ea: ec41001fa065   clgrj   %r4,%r1,10,3ff8001c828

 if (tail_call_cnt++ > MAX_TAIL_CALL_CNT)
         goto out;
 000003ff8001c7f0: a7080001       lhi     %r0,1
 000003ff8001c7f4: eb10f25000fa   laal    %r1,%r0,592(%r15)
 000003ff8001c7fa: ec120017207f   clij    %r1,32,2,3ff8001c828

 prog = array->prog[index];
 if (prog == NULL)
         goto out;
 000003ff8001c800: eb140003000d   sllg    %r1,%r4,3
 000003ff8001c806: e31310800004   lg      %r1,128(%r3,%r1)
 000003ff8001c80c: ec18000e007d   clgij   %r1,0,8,3ff8001c828

 Restore registers before calling function
 000003ff8001c812: eb68f2980004   lmg     %r6,%r8,664(%r15)
 000003ff8001c818: ebbff2c00004   lmg     %r11,%r15,704(%r15)

 goto *(prog->bpf_func + tail_call_start);
 000003ff8001c81e: e31100200004   lg      %r1,32(%r1,%r0)
 000003ff8001c824: 47f01006       bc      15,6(%r1)
Reviewed-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Signed-off-by: Michael Holzheu <holzheu@linux.vnet.ibm.com>
Acked-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Alexei Starovoitov <ast@plumgrid.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Pull networking fixes from David Miller:

 1) Fix stack allocation in s390 BPF JIT, from Michael Holzheu.

 2) Disable LRO on openvswitch paths, from Jiri Benc.

 3) UDP early demux doesn't handle multicast group membership properly,
    fix from Shawn Bohrer.

 4) Fix TX queue hang due to incorrect handling of mixed sized fragments
    and linearlization in i40e driver, from Anjali Singhai Jain.

 5) Cannot use disable_irq() in timer handler of AMD xgbe driver, from
    Thomas Lendacky.

 6) b2net driver improperly assumes pci_alloc_consistent() gives zero'd
    out memory, use dma_zalloc_coherent().  From Sriharsha Basavapatna.

 7) Fix use-after-free in MPLS and ipv6, from Robert Shearman.

 8) Missing neif_napi_del() calls in cleanup paths of b44 driver, from
    Hauke Mehrtens.

* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net:
  net: replace last open coded skb_orphan_frags with function call
  net: bcmgenet: power on MII block for all MII modes
  ipv6: Fix protocol resubmission
  ipv6: fix possible use after free of dev stats
  b44: call netif_napi_del()
  bridge: disable softirqs around br_fdb_update to avoid lockup
  Revert "bridge: use _bh spinlock variant for br_fdb_update to avoid lockup"
  mpls: fix possible use after free of device
  be2net: Replace dma/pci_alloc_coherent() calls with dma_zalloc_coherent()
  bridge: use _bh spinlock variant for br_fdb_update to avoid lockup
  amd-xgbe: Use disable_irq_nosync from within timer function
  rhashtable: add missing import <linux/export.h>
  i40e: Make sure to be in VEB mode if SRIOV is enabled at probe
  i40e: start up in VEPA mode by default
  i40e/i40evf: Fix mixed size frags and linearization
  ipv4/udp: Verify multicast group is ours in upd_v4_early_demux()
  openvswitch: disable LRO
  s390/bpf: fix bpf frame pointer setup
  s390/bpf: fix stack allocation

Pull last-minute virtio fix from Michael Tsirkin:
 "This fixes a minor issue affecting multiqueue virtio net when user
  keeps changing the number of active queues and CPUs are added and
  removed by hotplug"

* tag 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mst/vhost:
  virtio_pci: Clear stale cpumask when setting irq affinity

Fix the device tree entries to modify the '_' to '-'.
Also changes the names of the internal delay properties
from -int- to -internal- as the -int- appeared as a keyword.
Signed-off-by: Dan Murphy <dmurphy@ti.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

API compliance scanning with coccinelle flagged:
./drivers/atm/iphase.c:2621:4-20:
        WARNING: timeout (50) seems HZ dependent

Numeric constants passed to schedule_timeout() make the effective
timeout HZ dependent which does not seem intended.
Fixed up by converting the constant to jiffies with msecs_to_jiffies()
As this driver was introduced in the early 2.3 series it is most
likely assuming HZ=100 so the constant 50 is converted to 500ms.
Signed-off-by: Nicholas Mc Guire <hofrat@osadl.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

Pull ARM SoC fixes from Kevin Hilman:
 "About 10 days worth of small bug fixes, and the (hopefully) final
  round fixes for from arm-soc land for the -rc cycle.  Nothing special
  to note, but here's a brief summary of fixes by SoC type:

   - OMAP:
        small set of misc DT fixes; boot fix for THUMB2 kernel

   - mediatek:
        PMIC fixes; DT fix for model name

   - exynos:
        wakeup interupt fixes for 3250

   - mvebu:
        revert mbus patch which broke DMA masters

* tag 'fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/arm/arm-soc:
  ARM: dts: am335x-boneblack: disable RTC-only sleep to avoid hardware damage
  ARM: dts: AM35xx: fix system control module clocks
  arm64: dts: mt8173-evb: fix model name
  ARM: exynos: Fix wake-up interrupts for Exynos3250
  ARM: dts: Fix n900 dts file to work around 4.1 touchscreen regression on n900
  ARM: dts: Fix dm816x to use right compatible flag for MUSB
  ARM: OMAP3: Fix booting with thumb2 kernel
  Revert "bus: mvebu-mbus: make sure SDRAM CS for DMA don't overlap the MBus bridge window"
  bus: mvebu-mbus: do not set WIN_CTRL_SYNCBARRIER on non io-coherent platforms.
  ARM: mvebu: armada-xp-linksys-mamba: Disable internal RTC
  soc: mediatek: Add compile dependency to pmic-wrapper
  soc: mediatek: PMIC wrap: Fix register state machine handling
  soc: mediatek: PMIC wrap: Fix clock rate handling

Update the AFE_TX_CONFIG value to solve marginal rise/fall issues
observed when the link is operating in 100BaseT. This workaround applies
to GPHY revisions D0, E0 and newer.
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Commit 70008aa5 ("skbuff: convert to skb_orphan_frags") replaced
open coded tests of SKBTX_DEV_ZEROCOPY and skb_copy_ubufs with calls
to helper function skb_orphan_frags. Apply that to the last remaining
open coded site.
Signed-off-by: Willem de Bruijn <willemb@google.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

The RGMII block is currently only powered on when using RGMII or
RGMII_NO_ID, which is not correct when using the GENET interface in MII
or Reverse MII modes. We always need to power on the RGMII interface for
this block to properly work, regardless of the MII mode in which we
operate.

Fixes: aa09677c ("net: bcmgenet: add MDIO routines")
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

UDP encapsulation is broken on IPv6. This is because the logic to resubmit
the nexthdr is inverted, checking for a ret value > 0 instead of < 0. Also,
the resubmit label is in the wrong position since we already get the
nexthdr value when performing decapsulation. In addition the skb pull is no
longer necessary either.

This changes the return value check to look for < 0, using it for the
nexthdr on the next iteration, and moves the resubmit label to the proper
location.

With these changes the v6 code now matches what we do in the v4 ip input
code wrt resubmitting when decapsulating.
Signed-off-by: Josh Hunt <johunt@akamai.com>
Acked-by: "Tom Herbert" <tom@herbertland.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

The memory pointed to by idev->stats.icmpv6msgdev,
idev->stats.icmpv6dev and idev->stats.ipv6 can each be used in an RCU
read context without taking a reference on idev. For example, through
IP6_*_STATS_* calls in ip6_rcv. These memory blocks are freed without
waiting for an RCU grace period to elapse. This could lead to the
memory being written to after it has been freed.

Fix this by using call_rcu to free the memory used for stats, as well
as idev after an RCU grace period has elapsed.
Signed-off-by: Robert Shearman <rshearma@brocade.com>
Acked-by: Hannes Frederic Sowa <hannes@stressinduktion.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

Merge tag 'omap-for-v4.1/fixes-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/tmlind/linux-omap into fixes

Merge omap fixes for v4.1, urgent fix to avoid potential hardware damage From Tony Lindgren:

Omap fixes for the -rc cycle, including a fix for potential hardware
breakage on BeagleBones:

- BeagleBones don't support RTC-only mode, it can cause hardware
  damage if system-power-controller is specified without
  ti,pmic-shutdown-controller

- Fix a recent regression to am3517 SoCs caused by the recent clock
  move that was not noticed until now despite automated boot
  testing

- Fix a regression for n900 touchscreen triggered by recent
  recent input changes

- Fix compatible property for dm816x USB to avoid errors with
  USB Ethernet

- Fix oops for omap3 when built with CONFIG_THUMB2_KERNEL

* tag 'omap-for-v4.1/fixes-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/tmlind/linux-omap:
  ARM: dts: am335x-boneblack: disable RTC-only sleep to avoid hardware damage
  ARM: dts: AM35xx: fix system control module clocks
  ARM: dts: Fix n900 dts file to work around 4.1 touchscreen regression on n900
  ARM: dts: Fix dm816x to use right compatible flag for MUSB
  ARM: OMAP3: Fix booting with thumb2 kernel

Pull Intel IOMMU fix from David Woodhouse:
 "This fixes an oops when attempting to enable 1:1 passthrough mode for
  devices on which VT-d translation was disabled anyway.

  It's actually a long-standing bug but recent changes (commit
  18436afd: "iommu/vt-d: Allow RMRR on graphics devices too") have
  made it much easier to trigger with 'iommu=pt intel_iommu=igfx_off' on
  the command line"

* git://git.infradead.org/intel-iommu:
  iommu/vt-d: Fix passthrough mode with translation-disabled devices

Pull libata fixes from Tejun Heo:
 "Two driver fixes.  One is for an ahci_mvebu controller config bug and
  the other fixes pata_octeon_cf build issue"

* 'for-4.1-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/libata:
  pata_octeon_cf: fix broken build
  ata: ahci_mvebu: Fix wrongly set base address for the MBus window setting

MODULE_DEVICE_TABLE is referring to wrong driver's table and breaks the
build. Fix that.

Cc: stable@vger.kernel.org
Signed-off-by: Aaro Koskinen <aaro.koskinen@nokia.com>
Signed-off-by: Tejun Heo <tj@kernel.org>

Jaeden Amero says:

====================
net/phy: micrel: Center FLP timing at 16ms

In v2, we add an additional cleanup commit to make an array of strings
static const and to improve const correctness generally. We also no longer
unnecessarily initialize the result variable in
ksz9031_center_flp_timing().

In v3, we remove the unnecessary result variable from ksz9031_config_init()
introduced by a previous version of "net/phy: micrel: Center FLP timing at
16ms".

In v4, we modify the commit message of "net/phy: micrel: Center FLP timing
at 16ms" to replace the awkward quotation of the data sheet's programming
procedure with an explanation of why we program the FLP burst registers and
restart auto-negotiation where we do (config_init).
====================
Signed-off-by: David S. Miller <davem@davemloft.net>

Link failures have been observed when using the KSZ9031 with HP 1810-8G
and HP 1910-8G network switches. Center the FLP timing at 16ms to help
avoid intermittent link failures.

>From the KSZ9031RNX and KSZ9031MNX data sheets revision 2.2, section
"Auto-Negotiation Timing":
	The KSZ9031[RNX or MNX] Fast Link Pulse (FLP) burst-to-burst
	transmit timing for Auto-Negotiation defaults to 8ms. IEEE 802.3
	Standard specifies this timing to be 16ms +/-8ms. Some PHY link
	partners need to receive the FLP with 16ms centered timing;
	otherwise, there can be intermittent link failures and long
	link-up times.

The PHY data sheet recommends configuring the FLP burst registers after
power-up/reset and immediately thereafter restarting auto-negotiation, so
we center the FLP timing at 16ms and then restart auto-negotiation in the
config_init for KSZ9031.
Signed-off-by: Jaeden Amero <jaeden.amero@ni.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

There are some defines for a few pad skew related extended registers.
Specify for which MMD Address (dev_addr) they are for.
Signed-off-by: Jaeden Amero <jaeden.amero@ni.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

In a few places in this driver, we weren't using const where we could
have. Use const more.

In addition, change the arrays of strings in ksz9031_config_init() to be
not only const, but also static.
Signed-off-by: Jaeden Amero <jaeden.amero@ni.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

As Alexander Duyck pointed out that:
struct tnode {
        ...
        struct key_vector kv[1];
}
The kv[1] member of struct tnode is an arry that refernced by
a null pointer will not crash the system, like this:
struct tnode *p = NULL;
struct key_vector *kv = p->kv;
As such p->kv doesn't actually dereference anything, it is simply a
means for getting the offset to the array from the pointer p.

This patch make the code more regular to avoid making people feel
odd when they look at the code.
Signed-off-by: Firo Yang <firogm@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

API compliance scanning with coccinelle flagged:
./drivers/net/wan/dscc4.c:1036:1-33:
        WARNING: timeout (10) seems HZ dependent
./drivers/net/wan/dscc4.c:554:2-34:
        WARNING: timeout (10) seems HZ dependent
./drivers/net/wan/dscc4.c:599:2-34:
        WARNING: timeout (10) seems HZ dependent

Numeric constants passed to schedule_timeout_*() make the effective
timeout HZ dependent which does not seem to be the intent here.
Fixed up by converting the constant to jiffies with msecs_to_jiffies(),
passing 100ms (assuming HZ==100 in the original code).
Signed-off-by: Nicholas Mc Guire <hofrat@osadl.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

API compliance scanning with coccinelle flagged:
./drivers/net/wan/cosa.c:520:2-18: WARNING:
	timeout (30) seems HZ dependent

Numeric constants passed to schedule_timeout() make the effective
timeout HZ dependent which makes little sense in a device probe.
Fixed up by converting the constant to jiffies with msecs_to_jiffies()
Signed-off-by: Nicholas Mc Guire <hofrat@osadl.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

Fix static checker warnings in the flow of system guid query.

Fixes: 707c4602 ('net/mlx5_core: Add new query HCA vport commands')
Signed-off-by: Majd Dibbiny <majd@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

When the driver gets unregistered a call to netif_napi_del() was
missing, this all was also missing in the error paths of
b44_init_one().
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Signed-off-by: David S. Miller <davem@davemloft.net>

br_fdb_update() can be called in process context in the following way:
br_fdb_add() -> __br_fdb_add() -> br_fdb_update() (if NTF_USE flag is set)
so we need to disable softirqs because there are softirq users of the
hash_lock. One easy way to reproduce this is to modify the bridge utility
to set NTF_USE, enable stp and then set maxageing to a low value so
br_fdb_cleanup() is called frequently and then just add new entries in
a loop. This happens because br_fdb_cleanup() is called from timer/softirq
context. The spin locks in br_fdb_update were _bh before commit f8ae737d
("[BRIDGE]: forwarding remove unneeded preempt and bh diasables")
and at the time that commit was correct because br_fdb_update() couldn't be
called from process context, but that changed after commit:
292d1398 ("bridge: add NTF_USE support")
Using local_bh_disable/enable around br_fdb_update() allows us to keep
using the spin_lock/unlock in br_fdb_update for the fast-path.
Signed-off-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Fixes: 292d1398 ("bridge: add NTF_USE support")
Signed-off-by: David S. Miller <davem@davemloft.net>

This reverts commit 1d7c4903.

Nikolay Aleksandrov has a better version of this fix.
Signed-off-by: David S. Miller <davem@davemloft.net>

In big endian cases, macro cpu_to_le16 unfolds to __swab16 which
provides special case for constants. In little endian cases,
__constant_cpu_to_le16 and cpu_to_le16 expand directly to the
same expression. So, replace __constant_cpu_to_le16 with
cpu_to_le16 with the goal of getting rid of the definition of
__constant_cpu_to_le16 completely.

The semantic patch that performs this transformation is as follows:

@@expression x;@@

- __constant_cpu_to_le16(x)
+ cpu_to_le16(x)
Signed-off-by: Vaishali Thakkar <vthakkar1994@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

The mpls device is used in an RCU read context without a lock being
held. As the memory is freed without waiting for the RCU grace period
to elapse, the freed memory could still be in use.

Address this by using kfree_rcu to free the memory for the mpls device
after the RCU grace period has elapsed.

Fixes: 03c57747 ("mpls: Per-device MPLS state")
Signed-off-by: Robert Shearman <rshearma@brocade.com>
Acked-by: "Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: David S. Miller <davem@davemloft.net>