Rather than re-initialising the entire completion on every mdio access,
use reinit_completion() which only resets the completion count.  This
avoids possible reinitialisation of the contained spinlock and waitqueue
while they may be in use (eg, mid-completion.)

Such an event could occur if there's a long delay in interrupt handling
causing the mdio accessor to time out, then a second access comes in
while the interrupt handler on a different CPU has called complete().
Another scenario where this has been observed is while locking has
been missing at the phy layer, allowing concurrent attempts to access
the MDIO bus.
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>

The phy layer is missing locking for the above two functions - it
has been observed that two threads (userspace and the phy worker
thread) can race, entering the bus ->write or ->read functions
simultaneously.

This causes the FEC driver to initialise a completion while another
thread is waiting on it or while the interrupt is calling complete()
on it, which causes spinlock unlock-without-lock, spinlock lockups,
and completion timeouts.

Fixes: a59a4d19 ("phy: add the EEE support and the way to access to the MMD registers.")
Fixes: 0c1d77df ("net: libphy: Add phy specific function to access mmd phy registers")
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Acked-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Before 56ef9c90[1] it used to ignore all errors from igmp_join().
That commit enhanced that and made it error out whatever error happened
with igmp_join(), but that's not good because when using multicast
groups vxlan will try to join it multiple times if the socket is reused
and then the 2nd and further attempts will fail with EADDRINUSE.

As we don't track to which groups the socket is already subscribed, it's
okay to just ignore that error.

Fixes: 56ef9c90 ("vxlan: Move socket initialization to within rtnl scope")
Reported-by: John Nielsen <lists@jnielsen.net>
Signed-off-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Merge tag 'linux-can-fixes-for-4.2-20150825' of git://git.kernel.org/pub/scm/linux/kernel/git/mkl/linux-can

Marc Kleine-Budde says:

====================
this is the updated pull request of one patch by me for the peak_usb
driver. It fixes the driver, so that non FD adapters don't provide CAN
FD bittimings.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>

Currently the renesas ethernet driver directory is compiled if SH_ETH is
configured rather than NET_VENDOR_RENESAS. Although incorrect that was
quite harmless as until recently as SH_ETH configured the only driver in
the renesas directory. However, as of c156633f ("Renesas Ethernet AVB
driver proper") the renesas directory includes another driver, configured
by RAVB, and it makes little sense for it to have a hidden dependency on
SH_ETH.
Signed-off-by: Kazuya Mizuguchi <kazuya.mizuguchi.ks@renesas.com>
[horms: rewrote changelog]
Signed-off-by: Simon Horman <horms+renesas@verge.net.au>
Signed-off-by: David S. Miller <davem@davemloft.net>

When a tunnel is deleted, the cached dst entry should be released.

This problem may prevent the removal of a netns (seen with a x-netns IPv6
gre tunnel):
  unregister_netdevice: waiting for lo to become free. Usage count = 3

CC: Dmitry Kozlov <xeb@mail.ru>
Fixes: c12b395a ("gre: Support GRE over IPv6")
Signed-off-by: huaibin Wang <huaibin.wang@6wind.com>
Signed-off-by: Nicolas Dichtel <nicolas.dichtel@6wind.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

commit 18ee49dd ("phylib: rename mii_bus::dev to mii_bus::parent")
changed the parent of PHY devices from the bus to the bus parent.

Then, commit 4dea547f ("phylib: rework to prepare for OF
registration of PHYs") moved the code into phy_device.c

At this point, it is somewhat unclear why the change was seen as
necessary.  But, when we look at the device model tree in
/sys/devices, it is clearly incorrect.  The PHYs should be children of
their MDIO bus.

Change the PHY's parent device to be the MDIO bus device.

Cc: Lennert Buytenhek <buytenh@wantstofly.org>
Cc: Grant Likely <grant.likely@secretlab.ca>
Signed-off-by: David Daney <david.daney@cavium.com>
Acked-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

The CAN FD data bittiming constants are provided via netlink only when there
are valid CAN FD constants available in priv->data_bittiming_const.

Due to the indirection of pointer assignments in the peak_usb driver the
priv->data_bittiming_const never becomes NULL - not even for non-FD adapters.

The data_bittiming_const points to zero'ed data which leads to this result
when running 'ip -details link show can0':

35: can0: <NOARP,ECHO> mtu 16 qdisc noop state DOWN mode DEFAULT group default qlen 10
    link/can  promiscuity 0
    can state STOPPED restart-ms 0
	  pcan_usb: tseg1 1..16 tseg2 1..8 sjw 1..4 brp 1..64 brp-inc 1
	  : dtseg1 0..0 dtseg2 0..0 dsjw 1..0 dbrp 0..0 dbrp-inc 0  <== BROKEN!
	  clock 8000000

This patch changes the struct peak_usb_adapter::bittiming_const and struct
peak_usb_adapter::data_bittiming_const to pointers to fix the assignemnt
problems.

Cc: linux-stable <stable@vger.kernel.org> # >= 4.0
Reported-by: Oliver Hartkopp <socketcan@hartkopp.net>
Tested-by: Oliver Hartkopp <socketcan@hartkopp.net>
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>

Hit the following splat testing VRF change for ipsec:

[  113.475692] ===============================
[  113.476194] [ INFO: suspicious RCU usage. ]
[  113.476667] 4.2.0-rc6-1+deb7u2+clUNRELEASED #3.2.65-1+deb7u2+clUNRELEASED Not tainted
[  113.477545] -------------------------------
[  113.478013] /work/monster-14/dsa/kernel.git/include/linux/rcupdate.h:568 Illegal context switch in RCU read-side critical section!
[  113.479288]
[  113.479288] other info that might help us debug this:
[  113.479288]
[  113.480207]
[  113.480207] rcu_scheduler_active = 1, debug_locks = 1
[  113.480931] 2 locks held by setkey/6829:
[  113.481371]  #0:  (&net->xfrm.xfrm_cfg_mutex){+.+.+.}, at: [<ffffffff814e9887>] pfkey_sendmsg+0xfb/0x213
[  113.482509]  #1:  (rcu_read_lock){......}, at: [<ffffffff814e767f>] rcu_read_lock+0x0/0x6e
[  113.483509]
[  113.483509] stack backtrace:
[  113.484041] CPU: 0 PID: 6829 Comm: setkey Not tainted 4.2.0-rc6-1+deb7u2+clUNRELEASED #3.2.65-1+deb7u2+clUNRELEASED
[  113.485422] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.7.5.1-0-g8936dbb-20141113_115728-nilsson.home.kraxel.org 04/01/2014
[  113.486845]  0000000000000001 ffff88001d4c7a98 ffffffff81518af2 ffffffff81086962
[  113.487732]  ffff88001d538480 ffff88001d4c7ac8 ffffffff8107ae75 ffffffff8180a154
[  113.488628]  0000000000000b30 0000000000000000 00000000000000d0 ffff88001d4c7ad8
[  113.489525] Call Trace:
[  113.489813]  [<ffffffff81518af2>] dump_stack+0x4c/0x65
[  113.490389]  [<ffffffff81086962>] ? console_unlock+0x3d6/0x405
[  113.491039]  [<ffffffff8107ae75>] lockdep_rcu_suspicious+0xfa/0x103
[  113.491735]  [<ffffffff81064032>] rcu_preempt_sleep_check+0x45/0x47
[  113.492442]  [<ffffffff8106404d>] ___might_sleep+0x19/0x1c8
[  113.493077]  [<ffffffff81064268>] __might_sleep+0x6c/0x82
[  113.493681]  [<ffffffff81133190>] cache_alloc_debugcheck_before.isra.50+0x1d/0x24
[  113.494508]  [<ffffffff81134876>] kmem_cache_alloc+0x31/0x18f
[  113.495149]  [<ffffffff814012b5>] skb_clone+0x64/0x80
[  113.495712]  [<ffffffff814e6f71>] pfkey_broadcast_one+0x3d/0xff
[  113.496380]  [<ffffffff814e7b84>] pfkey_broadcast+0xb5/0x11e
[  113.497024]  [<ffffffff814e82d1>] pfkey_register+0x191/0x1b1
[  113.497653]  [<ffffffff814e9770>] pfkey_process+0x162/0x17e
[  113.498274]  [<ffffffff814e9895>] pfkey_sendmsg+0x109/0x213

In pfkey_sendmsg the net mutex is taken and then pfkey_broadcast takes
the RCU lock.

Since pfkey_broadcast takes the RCU lock the allocation argument is
pointless since GFP_ATOMIC must be used between the rcu_read_{,un}lock.
The one call outside of rcu can be done with GFP_KERNEL.

Fixes: 7f6b9dbd ("af_key: locking change")
Signed-off-by: David Ahern <dsa@cumulusnetworks.com>
Acked-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Clean the dma flags of multiq ring buffer int the interface stop
process. This patch fixes that the genet is not running while the
interface is re-enabled.

$ ifup eth0 - running after booting
$ ifdown eth0
$ ifup eth0 - not running and occur tx_timeout

The bcmgenet_dma_disable() in bcmgenet_open() do clean ring16 dma flag
only. If the genet has multiq, the dma register is not cleaned. and
bcmgenet_init_dma() is not done correctly. in case
GENET_V2(tx_queues=4), tdma_ctrl has 0x1e after running
bcmgenet_dma_disable().
Signed-off-by: Jaedon Shin <jaedon.shin@gmail.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

bcmgenet_timeout() executes in atomic context, yet we will invoke
napi_disable() which does sleep. Looking back at the changes, disabling
TX napi and re-enabling it is completely useless, since we reclaim all
TX buffers and re-enable interrupts, and wake up the TX queues.

Fixes: 13ea6578 ("net: bcmgenet: improve TX timeout")
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

I can't send netlink message via mmaped netlink socket since

    commit: a8866ff6
    netlink: make the check for "send from tx_ring" deterministic

msg->msg_iter.type is set to WRITE (1) at

    SYSCALL_DEFINE6(sendto, ...
        import_single_range(WRITE, ...
            iov_iter_init(1, WRITE, ...

call path, so that we need to check the type by iter_is_iovec()
to accept the WRITE.
Signed-off-by: Ken-ichirou MATSUZAWA <chamas@h4.dion.ne.jp>
Signed-off-by: David S. Miller <davem@davemloft.net>

Pull power management and ACPI fixes from Rafael Wysocki:
 "These fix a recent regression in the ACPI backlight code and a memory
  leak in the Exynos cpufreq driver.

  Specifics:

   - Fix a recently introduced issue in the ACPI backlight code which
     causes lockdep to complain about a circular lock dependency during
     initialization (Hans de Goede).

   - Fix a possible memory during initialization in the Exynos cpufreq
     driver (Shailendra Verma)"

* tag 'pm+acpi-4.2-rc8' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm:
  cpufreq: exynos: Fix for memory leak in case SoC name does not match
  ACPI / video: Fix circular lock dependency issue in the video-detect code

* acpi-video:
  ACPI / video: Fix circular lock dependency issue in the video-detect code

* cpufreq-fixes:
  cpufreq: exynos: Fix for memory leak in case SoC name does not match

Pull networking fixes from David Miller:

 1) Out of bounds array access in 802.11 minstrel code, from Adrien
    Schildknecht.

 2) Don't use skb_get() in IGMP/MLD code paths, as this makes
    pskb_may_pull() BUG.  From Linus Luessing.

 3) Fix off by one in ipv4 route dumping code, from Andy Whitcroft.

 4) Fix deadlock in reqsk_queue_unlink(), from Eric Dumazet.

 5) Fix ppp device deregistration wrt.  netns deletion, from Guillaume
    Nault.

 6) Fix deadlock when creating per-cpu ipv6 routes, from Martin KaFai
    Lau.

 7) Fix memory leak in batman-adv code, from Sven Eckelmann.

* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net:
  batman-adv: Fix memory leak on tt add with invalid vlan
  net: phy: fix semicolon.cocci warnings
  net: qmi_wwan: add HP lt4111 LTE/EV-DO/HSPA+ Gobi 4G Module
  be2net: avoid vxlan offloading on multichannel configs
  ipv6: Fix a potential deadlock when creating pcpu rt
  ipv6: Add rt6_make_pcpu_route()
  ipv6: Remove un-used argument from ip6_dst_alloc()
  net: phy: workaround for buggy cable detection by LAN8700 after cable plugging
  net: ethernet: micrel: fix an error code
  ppp: fix device unregistration upon netns deletion
  net: phy: fix PHY_RUNNING in phy_state_machine
  Revert "net: limit tcp/udp rmem/wmem to SOCK_{RCV,SND}BUF_MIN"
  inet: fix potential deadlock in reqsk_queue_unlink()
  gianfar: Restore link state settings after MAC reset
  ipv4: off-by-one in continuation handling in /proc/net/route
  net: fix wrong skb_get() usage / crash in IGMP/MLD parsing code
  mac80211: fix invalid read in minstrel_sort_best_tp_rates()

Pull xen build fix from David Vrabel:
 "Fix i386 build with an (uncommon) configuration"

* tag 'for-linus-4.2-rc7-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/xen/tip:
  x86/xen: make CONFIG_XEN depend on CONFIG_X86_LOCAL_APIC

Pull sound fixes from Takashi Iwai:
 "Here are a small collecton of sound fix patches.

  The most significant one is the disablement of newly introduced
  topology API.  Its ABI couldn't be stabilized enough, so we decided to
  delay for 4.3 in the end.  Other than that, all oneliner fixes: a
  USB-audio runtime PM fix and a couple of HD-audio quirks"

* tag 'sound-4.2' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound:
  ALSA: hda - Add dock support for Thinkpad W541 (17aa:2211)
  ALSA: usb-audio: Fix runtime PM unbalance
  ASoC: topology: Disable use from userspace
  ASoC: topology: Add Kconfig option for topology
  ALSA: hda - Fix the white noise on Dell laptop

Pull SCSI target fixes from Nicholas Bellinger:
 "This contains a v4.2-rc specific RCU module unload regression bug-fix,
  a long-standing iscsi-target bug-fix for duplicate target_xfer_tags
  during NOP processing from Alexei, and two more small REPORT_LUNs
  emulation related patches to make Solaris FC host LUN scanning happy
  from Roland.

  There is also one patch not included that allows target-core to limit
  the number of fabric driver SGLs per I/O request using residuals, that
  is currently required as a work-around for FC hosts which don't honor
  EVPD block-limits settings.  At this point, it will most likely become
  for-next material"

* git://git.kernel.org/pub/scm/linux/kernel/git/nab/target-pending:
  target: Fix handling of small allocation lengths in REPORT LUNS
  target: REPORT LUNS should return LUN 0 even for dynamic ACLs
  target/iscsi: Fix double free of a TUR followed by a solicited NOPOUT
  target: Perform RCU callback barrier before backend/fabric unload

Pull thermal fixes from Eduardo Valentin:
 "Last minute fixes on the thermal-soc tree.  There is a fix of a long
  lasting bug in cpu cooling device, thanks for RMK for being pushing
  this"

* 'fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/evalenti/linux-soc-thermal:
  thermal/cpu_cooling: update policy limits if clipped_freq < policy->max
  thermal/cpu_cooling: rename max_freq as clipped_freq in notifier
  thermal/cpu_cooling: rename cpufreq_val as clipped_freq
  thermal/cpu_cooling: convert 'switch' block to 'if' block in notifier
  thermal/cpu_cooling: quit early after updating policy
  thermal/cpu_cooling: No need to initialize max_freq to 0
  thermal: cpu_cooling: fix lockdep problems in cpu_cooling
  thermal: power_allocator: do not use devm* interfaces

Since commit feb44f1f (x86/xen:
Provide a "Xen PV" APIC driver to support >255 VCPUs) Xen guests need
a full APIC driver and thus should depend on X86_LOCAL_APIC.

This fixes an i386 build failure with !SMP && !CONFIG_X86_UP_APIC by
disabling Xen support in this configuration.

Users needing Xen support in a non-SMP i386 kernel will need to enable
CONFIG_X86_UP_APIC.
Signed-off-by: David Vrabel <david.vrabel@citrix.com>
Cc: <stable@vger.kernel.org>

Merge tag 'asoc-v4.2-disable-topology' of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/sound into for-linus

ASoC: Disable topology support for v4.2

The topology code merged in the v4.2 merge window introduced a new ABI
which was believed to be suitable for use but subsequently additional
work by the developers of this feature have revealed some problems that
need to be addressed.  In order to allow this to be done without having
to support the initial ABI add Kconfig to disable the build and also add
some #error statements to the UAPI header so users can't use them.

Tested with W541 and Ultra Dock 170w
Signed-off-by: Rick Sherman <rick@shermdog.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>

The fix for deadlock in PM in commit [1ee23fe0: ALSA: usb-audio:
Fix deadlocks at resuming] introduced a new check of in_pm flag.
However, the brainless patch author evaluated it in a wrong way
(logical AND instead of logical OR), thus usb_autopm_get_interface()
is wrongly called at probing, leading to unbalance of runtime PM
refcount.

This patch fixes it by correcting the logic.
Reported-by: Hans Yang <hansy@nvidia.com>
Fixes: 1ee23fe0 ('ALSA: usb-audio: Fix deadlocks at resuming')
Cc: <stable@vger.kernel.org> [v3.15+]
Signed-off-by: Takashi Iwai <tiwai@suse.de>

REPORT LUNS should not fail just because the allocation length is less
than 16.  The relevant section of SPC-4 is:

  4.2.5.6 Allocation length

  The ALLOCATION LENGTH field specifies the maximum number of bytes or
  blocks that an application client has allocated in the Data-In
  Buffer. The ALLOCATION LENGTH field specifies bytes unless a
  different requirement is stated in the command definition.

  An allocation length of zero specifies that no data shall be
  transferred. This condition shall not be considered an error.

So we should just truncate our response rather than return an error.
Signed-off-by: Roland Dreier <roland@purestorage.com>
Signed-off-by: Spencer Baugh <sbaugh@catern.com>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>

The object tt_local is allocated with kmalloc and not initialized when the
function batadv_tt_local_add checks for the vlan. But this function can
only cleanup the object when the (not yet initialized) reference counter of
the object is 1. This is unlikely and thus the object would leak when the
vlan could not be found.

Instead the uninitialized object tt_local has to be freed manually and the
pointer has to set to NULL to avoid calling the function which would try to
decrement the reference counter of the not existing object.

CID: 1316518
Fixes: 354136bc ("batman-adv: fix kernel crash due to missing NULL checks")
Signed-off-by: Sven Eckelmann <sven@narfation.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

Pull dmaengine fix from Vinod Koul:
 "We recently found issue with dma_request_slave_channel() API causing
  privatecnt value to go bad.  This is fixed by balancing the privatecnt"

* tag 'dmaengine-fix-4.2-rc8' of git://git.infradead.org/users/vkoul/slave-dma:
  dmaengine: fix balance of privatecnt inc/dec operations

Since the topology API is still in sufficient flux for changes to be
identified disable the use of the userspace ABI by adding #error
statements to the code, ensuring that nobody relies on the headers as
currently defined.  It is expected that this change will be reverted for
v4.3.
Signed-off-by: Mark Brown <broonie@kernel.org>

Pull drm fixes from Dave Airlie:
 "These came in late last week, I wanted to look over the mst one before
  forwarding, but it seems good.

  Just three i915 and one MST fix"

* 'drm-fixes' of git://people.freedesktop.org/~airlied/linux:
  drm/i915: Commit planes on each crtc separately.
  drm/i915: calculate primary visibility changes instead of calling from set_config
  drm/i915: Only dither on 6bpc panels
  drm/dp/mst: Remove port after removing connector.

Allow the topology code to be compiled out so that users who don't need
topology don't need to havve the code compiled in, saving them some
memory.

Some more configuration could be added to remove some of the hooks into
the core data structures but that is probably best done with some
refactoring to use functions to do the updates of the data structures
rather than ifdefing in the code as we'd need to do at the minute.
Suggested-by: Takashi Iwai <tiwai@suse.de>
Signed-off-by: Mark Brown <broonie@kernel.org>

Pull rdma bugfix from Doug Ledford:
 "Bugfix in iw_cxgb4"

* tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/dledford/rdma:
  iw_cxgb4: gracefully handle unknown CQE status errors

Pull libata fixes from Tejun Heo:
 "Three minor device-specific fixes and revert of NCQ autosense added
  during this -rc1.

  It turned out that NCQ autosense as currently implemented interferes
  with the usual error handling behavior.  It will be revisited in the
  near future"

* 'for-4.2-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/libata:
  ata: ahci_brcmstb: Fix misuse of IS_ENABLED
  sata_sx4: Check return code from pdc20621_i2c_read()
  Revert "libata: Implement NCQ autosense"
  Revert "libata: Implement support for sense data reporting"
  Revert "libata-eh: Set 'information' field for autosense"
  ata: ahci_brcmstb: Fix warnings with CONFIG_PM_SLEEP=n

Pull cgroup fix from Tejun Heo:
 "A fix for a subtle bug introduced back during 3.17 cycle which
  interferes with setting configurations under specific conditions"

* 'for-4.2-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/cgroup:
  cpuset: use trialcs->mems_allowed as a temp variable

drivers/net/phy/smsc.c:127:3-4: Unneeded semicolon

 Remove unneeded semicolon.

Generated by: scripts/coccinelle/misc/semicolon.cocci

CC: Igor Plyatov <plyatov@gmail.com>
Signed-off-by: Fengguang Wu <fengguang.wu@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

This is an HP-branded Sierra Wireless EM7355:
https://bugzilla.redhat.com/show_bug.cgi?id=1223646#c2Signed-off-by: David Ward <david.ward@ll.mit.edu>
Signed-off-by: David S. Miller <davem@davemloft.net>

VxLAN offloading is not functional if the NIC is running in multichannel
mode (UMC, FLEX-10, VNIC...). Enabling this additionally kills whole
connectivity through the NIC and the device needs to be down and up to
restore it. The firmware should take care about it and does not allow
the conversion of interface to tunnel type (be_cmd_manage_iface) or should
support VxLAN offloading if multichannel config is enabled.
I have tested this on the latest available firmware (10.6.144.21).

Result:
[root@sm-04 ~]# ip link set enp5s0f0 up[root@sm-04 ~]# ip addr add 172.30.10.50/24 dev enp5s0f0
[root@sm-04 ~]# ping -c 3 172.30.10.254PING 172.30.10.254 (172.30.10.254) 56(84) bytes of data.
64 bytes from 172.30.10.254: icmp_seq=1 ttl=64 time=0.317 ms
64 bytes from 172.30.10.254: icmp_seq=2 ttl=64 time=0.187 ms
64 bytes from 172.30.10.254: icmp_seq=3 ttl=64 time=0.188 ms

 --- 172.30.10.254 ping statistics ---
3 packets transmitted, 3 received, 0% packet loss, time 2000ms
rtt min/avg/max/mdev = 0.187/0.230/0.317/0.063 ms
[root@sm-04 ~]# ip link add link enp5s0f0 vxlan10 type vxlan id 10 remote 172.30.10.60 dstport 4789
[root@sm-04 ~]# ip link set vxlan10 up
[ 7900.442811] be2net 0000:05:00.0: Enabled VxLAN offloads for UDP port 4789
[ 7900.455722] be2net 0000:05:00.1: Enabled VxLAN offloads for UDP port 4789
[ 7900.468635] be2net 0000:05:00.2: Enabled VxLAN offloads for UDP port 4789
[ 7900.481553] be2net 0000:05:00.3: Enabled VxLAN offloads for UDP port 4789
[root@sm-04 ~]# ping -c 3 172.30.10.254
PING 172.30.10.254 (172.30.10.254) 56(84) bytes of data.

 --- 172.30.10.254 ping statistics ---
3 packets transmitted, 0 received, 100% packet loss, time 1999ms

[root@sm-04 ~]# ip link set vxlan10 down
[ 7959.434093] be2net 0000:05:00.0: Disabled VxLAN offloads for UDP port 4789
[ 7959.444792] be2net 0000:05:00.1: Disabled VxLAN offloads for UDP port 4789
[ 7959.455592] be2net 0000:05:00.2: Disabled VxLAN offloads for UDP port 4789
[ 7959.466416] be2net 0000:05:00.3: Disabled VxLAN offloads for UDP port 4789
[root@sm-04 ~]# ip link del vxlan10
[root@sm-04 ~]# ping -c 3 172.30.10.254
PING 172.30.10.254 (172.30.10.254) 56(84) bytes of data.

 --- 172.30.10.254 ping statistics ---
3 packets transmitted, 0 received, 100% packet loss, time 1999ms

[root@sm-04 ~]# ip link set enp5s0f0 down
[root@sm-04 ~]# ip link set enp5s0f0 up
[ 8071.019003] be2net 0000:05:00.0 enp5s0f0: Link is Up
[root@sm-04 ~]# ping -c 3 172.30.10.254
PING 172.30.10.254 (172.30.10.254) 56(84) bytes of data.
64 bytes from 172.30.10.254: icmp_seq=1 ttl=64 time=0.318 ms
64 bytes from 172.30.10.254: icmp_seq=2 ttl=64 time=0.196 ms
64 bytes from 172.30.10.254: icmp_seq=3 ttl=64 time=0.194 ms

 --- 172.30.10.254 ping statistics ---
3 packets transmitted, 3 received, 0% packet loss, time 2000ms
rtt min/avg/max/mdev = 0.194/0.236/0.318/0.057 ms

Cc: Sathya Perla <sathya.perla@avagotech.com>
Cc: Ajit Khaparde <ajit.khaparde@avagotech.com>
Cc: Padmanabh Ratnakar <padmanabh.ratnakar@avagotech.com>
Cc: Sriharsha Basavapatna <sriharsha.basavapatna@avagotech.com>
Signed-off-by: Ivan Vecera <ivecera@redhat.com>
Acked-by: Ajit Khaparde <ajit.khaparde@avagotech.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Martin KaFai Lau says:

====================
ipv6: Fix a potential deadlock when creating pcpu rt

v1 -> v2:
A minor change in the commit message of patch 2.

This patch series fixes a potential deadlock when creating a pcpu rt.
It happens when dst_alloc() decided to run gc. Something like this:

read_lock(&table->tb6_lock);
ip6_rt_pcpu_alloc()
=> dst_alloc()
=> ip6_dst_gc()
=> write_lock(&table->tb6_lock); /* oops */

Patch 1 and 2 are some prep works.
Patch 3 is the fix.

Original report: https://bugzilla.kernel.org/show_bug.cgi?id=102291
====================
Signed-off-by: David S. Miller <davem@davemloft.net>

rt6_make_pcpu_route() is called under read_lock(&table->tb6_lock).
rt6_make_pcpu_route() calls ip6_rt_pcpu_alloc(rt) which then
calls dst_alloc().  dst_alloc() _may_ call ip6_dst_gc() which takes
the write_lock(&tabl->tb6_lock).  A visualized version:

read_lock(&table->tb6_lock);
rt6_make_pcpu_route();
=> ip6_rt_pcpu_alloc();
=> dst_alloc();
=> ip6_dst_gc();
=> write_lock(&table->tb6_lock); /* oops */

The fix is to do a read_unlock first before calling ip6_rt_pcpu_alloc().

A reported stack:

[141625.537638] INFO: rcu_sched self-detected stall on CPU { 27}  (t=60000 jiffies g=4159086 c=4159085 q=2139)
[141625.547469] Task dump for CPU 27:
[141625.550881] mtr             R  running task        0 22121  22081 0x00000008
[141625.558069]  0000000000000000 ffff88103f363d98 ffffffff8106e488 000000000000001b
[141625.565641]  ffffffff81684900 ffff88103f363db8 ffffffff810702b0 0000000008000000
[141625.573220]  ffffffff81684900 ffff88103f363de8 ffffffff8108df9f ffff88103f375a00
[141625.580803] Call Trace:
[141625.583345]  <IRQ>  [<ffffffff8106e488>] sched_show_task+0xc1/0xc6
[141625.589650]  [<ffffffff810702b0>] dump_cpu_task+0x35/0x39
[141625.595144]  [<ffffffff8108df9f>] rcu_dump_cpu_stacks+0x6a/0x8c
[141625.601320]  [<ffffffff81090606>] rcu_check_callbacks+0x1f6/0x5d4
[141625.607669]  [<ffffffff810940c8>] update_process_times+0x2a/0x4f
[141625.613925]  [<ffffffff8109fbee>] tick_sched_handle+0x32/0x3e
[141625.619923]  [<ffffffff8109fc2f>] tick_sched_timer+0x35/0x5c
[141625.625830]  [<ffffffff81094a1f>] __hrtimer_run_queues+0x8f/0x18d
[141625.632171]  [<ffffffff81094c9e>] hrtimer_interrupt+0xa0/0x166
[141625.638258]  [<ffffffff8102bf2a>] local_apic_timer_interrupt+0x4e/0x52
[141625.645036]  [<ffffffff8102c36f>] smp_apic_timer_interrupt+0x39/0x4a
[141625.651643]  [<ffffffff8140b9e8>] apic_timer_interrupt+0x68/0x70
[141625.657895]  <EOI>  [<ffffffff81346ee8>] ? dst_destroy+0x7c/0xb5
[141625.664188]  [<ffffffff813d45b5>] ? fib6_flush_trees+0x20/0x20
[141625.670272]  [<ffffffff81082b45>] ? queue_write_lock_slowpath+0x60/0x6f
[141625.677140]  [<ffffffff8140aa33>] _raw_write_lock_bh+0x23/0x25
[141625.683218]  [<ffffffff813d4553>] __fib6_clean_all+0x40/0x82
[141625.689124]  [<ffffffff813d45b5>] ? fib6_flush_trees+0x20/0x20
[141625.695207]  [<ffffffff813d6058>] fib6_clean_all+0xe/0x10
[141625.700854]  [<ffffffff813d60d3>] fib6_run_gc+0x79/0xc8
[141625.706329]  [<ffffffff813d0510>] ip6_dst_gc+0x85/0xf9
[141625.711718]  [<ffffffff81346d68>] dst_alloc+0x55/0x159
[141625.717105]  [<ffffffff813d09b5>] __ip6_dst_alloc.isra.32+0x19/0x63
[141625.723620]  [<ffffffff813d1830>] ip6_pol_route+0x36a/0x3e8
[141625.729441]  [<ffffffff813d18d6>] ip6_pol_route_output+0x11/0x13
[141625.735700]  [<ffffffff813f02c8>] fib6_rule_action+0xa7/0x1bf
[141625.741698]  [<ffffffff813d18c5>] ? ip6_pol_route_input+0x17/0x17
[141625.748043]  [<ffffffff81357c48>] fib_rules_lookup+0xb5/0x12a
[141625.754050]  [<ffffffff81141628>] ? poll_select_copy_remaining+0xf9/0xf9
[141625.761002]  [<ffffffff813f0535>] fib6_rule_lookup+0x37/0x5c
[141625.766914]  [<ffffffff813d18c5>] ? ip6_pol_route_input+0x17/0x17
[141625.773260]  [<ffffffff813d008c>] ip6_route_output+0x7a/0x82
[141625.779177]  [<ffffffff813c44c8>] ip6_dst_lookup_tail+0x53/0x112
[141625.785437]  [<ffffffff813c45c3>] ip6_dst_lookup_flow+0x2a/0x6b
[141625.791604]  [<ffffffff813ddaab>] rawv6_sendmsg+0x407/0x9b6
[141625.797423]  [<ffffffff813d7914>] ? do_ipv6_setsockopt.isra.8+0xd87/0xde2
[141625.804464]  [<ffffffff8139d4b4>] inet_sendmsg+0x57/0x8e
[141625.810028]  [<ffffffff81329ba3>] sock_sendmsg+0x2e/0x3c
[141625.815588]  [<ffffffff8132be57>] SyS_sendto+0xfe/0x143
[141625.821063]  [<ffffffff813dd551>] ? rawv6_setsockopt+0x5e/0x67
[141625.827146]  [<ffffffff8132c9f8>] ? sock_common_setsockopt+0xf/0x11
[141625.833660]  [<ffffffff8132c08c>] ? SyS_setsockopt+0x81/0xa2
[141625.839565]  [<ffffffff8140ac17>] entry_SYSCALL_64_fastpath+0x12/0x6a

Fixes: d52d3997 ("pv6: Create percpu rt6_info")
Signed-off-by: Martin KaFai Lau <kafai@fb.com>
CC: Hannes Frederic Sowa <hannes@stressinduktion.org>
Reported-by: Steinar H. Gunderson <sgunderson@bigfoot.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

It is a prep work for fixing a potential deadlock when creating
a pcpu rt.

The current rt6_get_pcpu_route() will also create a pcpu rt if one does not
exist.  This patch moves the pcpu rt creation logic into another function,
rt6_make_pcpu_route().
Signed-off-by: Martin KaFai Lau <kafai@fb.com>
CC: Hannes Frederic Sowa <hannes@stressinduktion.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

After 4b32b5ad ("ipv6: Stop rt6_info from using inet_peer's metrics"),
ip6_dst_alloc() does not need the 'table' argument.  This patch
cleans it up.
Signed-off-by: Martin KaFai Lau <kafai@fb.com>
CC: Hannes Frederic Sowa <hannes@stressinduktion.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

* Due to HW bug, LAN8700 sometimes does not detect presence of energy in the
  Ethernet cable in Energy Detect Power-Down mode (e.g while EDPWRDOWN bit is
  set, the ENERGYON bit does not asserted sometimes). This is a common bug of
  LAN87xx family of PHY chips.
* The lan87xx_read_status() was improved to acquire ENERGYON bit. Its previous
  algorythm still not reliable on 100 % and sometimes skip cable plugging.
Signed-off-by: Igor Plyatov <plyatov@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>