timerfd_show() uses a 'struct itimerspec' internally, but that is
deprecated because of the time_t overflow and a conflict with the glibc
type of the same name that is now incompatible in user space.

Use a pair of timespec64 variables instead as a simple replacement.

As this removes the last use of itimerspec from the kernel, allowing the
removal of the definition from the uapi headers along with timespec and
timeval later.
Reviewed-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Arnd Bergmann <arnd@arndb.de>

We store elapsed time for a crashed process in struct elf_prstatus using
'timeval' structures. Once glibc starts using 64-bit time_t, this becomes
incompatible with the kernel's idea of timeval since the structure layout
no longer matches on 32-bit architectures.

This changes the definition of the elf_prstatus structure to use
__kernel_old_timeval instead, which is hardcoded to the currently used
binary layout. There is no risk of overflow in y2038 though, because
the time values are all relative times, and can store up to 68 years
of process elapsed time.

There is a risk of applications breaking at build time when they
use the new kernel headers and expect the type to be exactly 'timeval'
rather than a structure that has the same fields as before. Those
applications have to be modified to deal with 64-bit time_t anyway.
Signed-off-by: Arnd Bergmann <arnd@arndb.de>

This gets us one step closer to removing 'struct timeval' from the
kernel. We still keep __kernel_old_timeval for interfaces that we cannot
fix otherwise, and ns_to_compat_timeval() is provably safe for interfaces
that are legitimate users of __kernel_old_timeval on native kernels,
so this is an obvious change.
Signed-off-by: Arnd Bergmann <arnd@arndb.de>

The 'timespec' type definition and helpers like ktime_to_timespec()
or timespec64_to_timespec() should no longer be used in the kernel so
we can remove them and avoid introducing y2038 issues in new code.

Change the socket code that needs to pass a timespec to user space for
backward compatibility to use __kernel_old_timespec instead.  This type
has the same layout but with a clearer defined name.

Slightly reformat tcp_recv_timestamp() for consistency after the removal
of timespec64_to_timespec().
Acked-by: Deepa Dinamani <deepa.kernel@gmail.com>
Signed-off-by: Arnd Bergmann <arnd@arndb.de>

In order to remove the 'struct timespec' definition and the
timespec64_to_timespec() helper function, change over the in-kernel
definition of 'struct scm_timestamping' to use the __kernel_old_timespec
replacement and open-code the assignment.
Acked-by: Deepa Dinamani <deepa.kernel@gmail.com>
Signed-off-by: Arnd Bergmann <arnd@arndb.de>

All of the remaining syscalls that pass a timeval (gettimeofday, utime,
futimesat) can trivially be changed to pass a __kernel_old_timeval
instead, which has a compatible layout, but avoids ambiguity with
the timeval type in user space.
Acked-by: Christian Brauner <christian.brauner@ubuntu.com>
Acked-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Signed-off-by: Arnd Bergmann <arnd@arndb.de>

There are two 'struct timeval' fields in 'struct rusage'.

Unfortunately the definition of timeval is now ambiguous when used in
user space with a libc that has a 64-bit time_t, and this also changes
the 'rusage' definition in user space in a way that is incompatible with
the system call interface.

While there is no good solution to avoid all ambiguity here, change
the definition in the kernel headers to be compatible with the kernel
ABI, using __kernel_old_timeval as an unambiguous base type.

In previous discussions, there was also a plan to add a replacement
for rusage based on 64-bit timestamps and nanosecond resolution,
i.e. 'struct __kernel_timespec'. I have patches for that as well,
if anyone thinks we should do that.
Reviewed-by: Cyrill Gorcunov <gorcunov@gmail.com>
Signed-off-by: Arnd Bergmann <arnd@arndb.de>

This is mainly a patch for clarification, and to let us remove
the time_t definition from the kernel to prevent new users from
creeping in that might not be y2038-safe.

All remaining uses of 'time_t' or '__kernel_time_t' are part of
the user API that cannot be changed by that either have a
replacement or that do not suffer from the y2038 overflow.
Acked-by: Deepa Dinamani <deepa.kernel@gmail.com>
Acked-by: Christian Brauner <christian.brauner@ubuntu.com>
Signed-off-by: Arnd Bergmann <arnd@arndb.de>

The time_t definition may differ between user space and kernel space,
so replace time_t with an unambiguous 'long' for the mips and sparc.

The same structures also contain 'off_t', which has the same problem,
so replace that as well on those two architectures and powerpc.
Signed-off-by: Arnd Bergmann <arnd@arndb.de>

There are two structures based on time_t that conflict between libc and
kernel: timeval and timespec. Both are now renamed to __kernel_old_timeval
and __kernel_old_timespec.

For time_t, the old typedef is still __kernel_time_t. There is nothing
wrong with that name, but it would be nice to not use that going forward
as this type is used almost only in deprecated interfaces because of
the y2038 overflow.

In the IPC headers (msgbuf.h, sembuf.h, shmbuf.h), __kernel_time_t is only
used for the 64-bit variants, which are not deprecated.

Change these to a plain 'long', which is the same type as __kernel_time_t
on all 64-bit architectures anyway, to reduce the number of users of the
old type.
Signed-off-by: Arnd Bergmann <arnd@arndb.de>

As a preparation to stop using 'struct timespec' in the kernel,
change the powerpc vdso implementation:

- split up the vdso data definition to have equivalent members
   for seconds and nanoseconds instead of an xtime structure

- use timespec64 as an intermediate for the xtime update

- change the asm-offsets definition to be based the appropriate
  fixed-length types

This is only a temporary fix for changing the types, in order
to actually support a 64-bit safe vdso32 version of clock_gettime(),
the entire powerpc vdso should be replaced with the generic
lib/vdso/ implementation. If that happens first, this patch
becomes obsolete.
Signed-off-by: Arnd Bergmann <arnd@arndb.de>

The nds32 vdso is now the last user of the deprecated timespec_add_ns().

Change it to an open-coded version like the one it already uses in
do_realtime(). What we should really do though is to use the
generic vdso implementation that is now used in x86. arm and mips.
Signed-off-by: Arnd Bergmann <arnd@arndb.de>

Only x86 uses the 'time' syscall in vdso, so change that to
__kernel_old_time_t as a preparation for removing 'time_t' and
'__kernel_time_t' later.
Reviewed-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Arnd Bergmann <arnd@arndb.de>

In order to remove 'timespec' completely from the kernel, all
internal uses should be converted to a y2038-safe type, while
those that are only for compatibity with existing user space
should be marked appropriately.

Change vdso to use __kernel_old_timespec in order to avoid
the deprecated type and mark these interfaces as outdated.
Signed-off-by: Arnd Bergmann <arnd@arndb.de>

The gettimeofday() function in vdso uses the traditional 'timeval'
structure layout, which will be incompatible with future versions of
glibc on 32-bit architectures that use a 64-bit time_t.

This interface is problematic for y2038, when time_t overflows on 32-bit
architectures, but the plan so far is that a libc with 64-bit time_t
will not call into the gettimeofday() vdso helper at all, and only
have a method for entering clock_gettime().  This means we don't have
to fix it here, though we probably want to add a new clock_gettime()
entry point using a 64-bit version of 'struct timespec' at some point.

Changing the vdso code to use __kernel_old_timeval helps isolate
this usage from the other ones that still need to be fixed properly,
and it gets us closer to removing the 'timeval' definition from the
kernel sources.
Reviewed-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Arnd Bergmann <arnd@arndb.de>

The 'struct timespec' definition can no longer be part of the uapi headers
because it conflicts with a a now incompatible libc definition. Also,
we really want to remove it in order to prevent new uses from creeping in.

The same namespace conflict exists with time_t, which should also be
removed. __kernel_time_t could be used safely, but adding 'old' in the
name makes it clearer that this should not be used for new interfaces.

Add a replacement __kernel_old_timespec structure and __kernel_old_time_t
along the lines of __kernel_old_timeval.
Acked-by: Deepa Dinamani <deepa.kernel@gmail.com>
Signed-off-by: Arnd Bergmann <arnd@arndb.de>

The CONFIG_64BIT_TIME option is defined on all architectures, and can
be removed for simplicity now.
Signed-off-by: Arnd Bergmann <arnd@arndb.de>

Pull USB fixes from Greg KH:
 "The USB sub-maintainers woke up this past week and sent a bunch of
  tiny fixes. Here are a lot of small patches that that resolve a bunch
  of reported issues in the USB core, drivers, serial drivers, gadget
  drivers, and of course, xhci :)

  All of these have been in linux-next with no reported issues"

* tag 'usb-5.4-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb: (31 commits)
  usb: dwc3: gadget: fix race when disabling ep with cancelled xfers
  usb: cdns3: gadget: Fix g_audio use case when connected to Super-Speed host
  usb: cdns3: gadget: reset EP_CLAIMED flag while unloading
  USB: serial: whiteheat: fix line-speed endianness
  USB: serial: whiteheat: fix potential slab corruption
  USB: gadget: Reject endpoints with 0 maxpacket value
  UAS: Revert commit 3ae62a42 ("UAS: fix alignment of scatter/gather segments")
  usb-storage: Revert commit 747668db ("usb-storage: Set virt_boundary_mask to avoid SG overflows")
  usbip: Fix free of unallocated memory in vhci tx
  usbip: tools: Fix read_usb_vudc_device() error path handling
  usb: xhci: fix __le32/__le64 accessors in debugfs code
  usb: xhci: fix Immediate Data Transfer endianness
  xhci: Fix use-after-free regression in xhci clear hub TT implementation
  USB: ldusb: fix control-message timeout
  USB: ldusb: use unsigned size format specifiers
  USB: ldusb: fix ring-buffer locking
  USB: Skip endpoints with 0 maxpacket length
  usb: cdns3: gadget: Don't manage pullups
  usb: dwc3: remove the call trace of USBx_GFLADJ
  usb: gadget: configfs: fix concurrent issue between composite APIs
  ...

Pull cifs fix from Steve French:
 "A small smb3 memleak fix"

* tag '5.4-rc6-smb3-fix' of git://git.samba.org/sfrench/cifs-2.6:
  fix memory leak in large read decrypt offload

Merge tag 'hwmon-for-v5.4-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/groeck/linux-staging

Pull hwmon fixes from Guenter Roeck:

 - Fix read timeout problem in ina3221 driver

 - Fix wrong bitmask in nct7904 driver

* tag 'hwmon-for-v5.4-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/groeck/linux-staging:
  hwmon: (ina3221) Fix read timeout issue
  hwmon: (nct7904) Fix the incorrect value of vsen_mask & tcpu_mask & temp_mode in nct7904_data struct.

Merge tag 'pwm/for-5.4-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/thierry.reding/linux-pwm

Pull pwm fixes from Thierry Reding:
 "It turned out that relying solely on drivers storing all the PWM state
  in hardware was a little premature and causes a number of subtle (and
  some not so subtle) regressions. Revert the offending patch for now"

* tag 'pwm/for-5.4-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/thierry.reding/linux-pwm:
  Revert "pwm: Let pwm_get_state() return the last implemented state"

Pull SCSI fixes from James Bottomley:
 "Nine changes, eight in drivers [ufs, target, lpfc x 2, qla2xxx x 4]
  and one core change in sd that fixes an I/O failure on DIF type 3
  devices"

* tag 'scsi-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi:
  scsi: qla2xxx: stop timer in shutdown path
  scsi: sd: define variable dif as unsigned int instead of bool
  scsi: target: cxgbit: Fix cxgbit_fw4_ack()
  scsi: qla2xxx: Fix partial flash write of MBI
  scsi: qla2xxx: Initialized mailbox to prevent driver load failure
  scsi: lpfc: Honor module parameter lpfc_use_adisc
  scsi: ufs-bsg: Wake the device before sending raw upiu commands
  scsi: lpfc: Check queue pointer before use
  scsi: qla2xxx: fixup incorrect usage of host_byte

Pull powerpc fixes from Michael Ellerman:
 "Our recent cleanup of EEH led to an oops on bare metal machines when
  the cxl (CAPI) driver creates virtual devices for an attached FPGA
  accelerator.

  The "secure virtual machine" support we added in v5.4 had a bug if the
  kernel was relocated (moved during boot), in those cases the signature
  of the kernel text wouldn't verify and the Ultravisor would refuse to
  run the VM.

  A recent change to disable interrupts before calling
  arch_cpu_idle_dead() caused a WARN_ON() in our bare metal CPU offline
  code to always trigger.

  The KUAP (SMAP) support we added for 32-bit Book3S had a bug if the
  address range crossed a segment (256MB) boundary which could lead to
  spurious faults.

  Thanks to: Christophe Leroy, Frederic Barrat, Michael Anderson,
  Nicholas Piggin, Sam Bobroff, Thiago Jung Bauermann"

* tag 'powerpc-5.4-4' of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux:
  powerpc/powernv: Fix CPU idle to be called with IRQs disabled
  powerpc/prom_init: Undo relocation before entering secure mode
  powerpc/powernv/eeh: Fix oops when probing cxl devices
  powerpc/32s: fix allow/prevent_user_access() when crossing segment boundaries.

Pull s390 fixes from Vasily Gorbik:

 - Fix cpu idle time accounting

 - Fix stack unwinder case when both pt_regs and sp are specified

 - Fix information leak via cmm timeout proc handler

* tag 's390-5.4-6' of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux:
  s390/idle: fix cpu idle time calculation
  s390/unwind: fix mixing regs and sp
  s390/cmm: fix information leak in cmm_timeout_handler()

Pull networking fixes from David Miller:

 1) Fix free/alloc races in batmanadv, from Sven Eckelmann.

 2) Several leaks and other fixes in kTLS support of mlx5 driver, from
    Tariq Toukan.

 3) BPF devmap_hash cost calculation can overflow on 32-bit, from Toke
    Høiland-Jørgensen.

 4) Add an r8152 device ID, from Kazutoshi Noguchi.

 5) Missing include in ipv6's addrconf.c, from Ben Dooks.

 6) Use siphash in flow dissector, from Eric Dumazet. Attackers can
    easily infer the 32-bit secret otherwise etc.

 7) Several netdevice nesting depth fixes from Taehee Yoo.

 8) Fix several KCSAN reported errors, from Eric Dumazet. For example,
    when doing lockless skb_queue_empty() checks, and accessing
    sk_napi_id/sk_incoming_cpu lockless as well.

 9) Fix jumbo packet handling in RXRPC, from David Howells.

10) Bump SOMAXCONN and tcp_max_syn_backlog values, from Eric Dumazet.

11) Fix DMA synchronization in gve driver, from Yangchun Fu.

12) Several bpf offload fixes, from Jakub Kicinski.

13) Fix sk_page_frag() recursion during memory reclaim, from Tejun Heo.

14) Fix ping latency during high traffic rates in hisilicon driver, from
    Jiangfent Xiao.

* git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net: (146 commits)
  net: fix installing orphaned programs
  net: cls_bpf: fix NULL deref on offload filter removal
  selftests: bpf: Skip write only files in debugfs
  selftests: net: reuseport_dualstack: fix uninitalized parameter
  r8169: fix wrong PHY ID issue with RTL8168dp
  net: dsa: bcm_sf2: Fix IMP setup for port different than 8
  net: phylink: Fix phylink_dbg() macro
  gve: Fixes DMA synchronization.
  inet: stop leaking jiffies on the wire
  ixgbe: Remove duplicate clear_bit() call
  Documentation: networking: device drivers: Remove stray asterisks
  e1000: fix memory leaks
  i40e: Fix receive buffer starvation for AF_XDP
  igb: Fix constant media auto sense switching when no cable is connected
  net: ethernet: arc: add the missed clk_disable_unprepare
  igb: Enable media autosense for the i350.
  igb/igc: Don't warn on fatal read failures when the device is removed
  tcp: increase tcp_max_syn_backlog max value
  net: increase SOMAXCONN to 4096
  netdevsim: Fix use-after-free during device dismantle
  ...

Pull NFS client bugfixes from Anna Schumaker:
 "This contains two delegation fixes (with the RCU lock leak fix marked
  for stable), and three patches to fix destroying the the sunrpc back
  channel.

  Stable bugfixes:

   - Fix an RCU lock leak in nfs4_refresh_delegation_stateid()

  Other fixes:

   - The TCP back channel mustn't disappear while requests are
     outstanding

   - The RDMA back channel mustn't disappear while requests are
     outstanding

   - Destroy the back channel when we destroy the host transport

   - Don't allow a cached open with a revoked delegation"

* tag 'nfs-for-5.4-3' of git://git.linux-nfs.org/projects/anna/linux-nfs:
  NFS: Fix an RCU lock leak in nfs4_refresh_delegation_stateid()
  NFSv4: Don't allow a cached open with a revoked delegation
  SUNRPC: Destroy the back channel when we destroy the host transport
  SUNRPC: The RDMA back channel mustn't disappear while requests are outstanding
  SUNRPC: The TCP back channel mustn't disappear while requests are outstanding

Pull block fixes from Jens Axboe:

 - Two small nvme fixes, one is a fabrics connection fix, the other one
   a cleanup made possible by that fix (Anton, via Keith)

 - Fix requeue handling in umb ubd (Anton)

 - Fix spin_lock_irq() nesting in blk-iocost (Dan)

 - Three small io_uring fixes:
     - Install io_uring fd after done with ctx (me)
     - Clear ->result before every poll issue (me)
     - Fix leak of shadow request on error (Pavel)

* tag 'for-linus-20191101' of git://git.kernel.dk/linux-block:
  iocost: don't nest spin_lock_irq in ioc_weight_write()
  io_uring: ensure we clear io_kiocb->result before each issue
  um-ubd: Entrust re-queue to the upper layers
  nvme-multipath: remove unused groups_only mode in ana log
  nvme-multipath: fix possible io hang after ctrl reconnect
  io_uring: don't touch ctx in setup after ring fd install
  io_uring: Fix leaked shadow_req

Pull RISC-V fixes from Paul Walmsley:
 "One fix for PCIe users:

   - Fix legacy PCI I/O port access emulation

  One set of cleanups:

   - Resolve most of the warnings generated by sparse across arch/riscv.
     No functional changes

  And one MAINTAINERS update:

   - Update Palmer's E-mail address"

* tag 'riscv/for-v5.4-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/riscv/linux:
  MAINTAINERS: Change to my personal email address
  RISC-V: Add PCIe I/O BAR memory mapping
  riscv: for C functions called only from assembly, mark with __visible
  riscv: fp: add missing __user pointer annotations
  riscv: add missing header file includes
  riscv: mark some code and data as file-static
  riscv: init: merge split string literals in preprocessor directive
  riscv: add prototypes for assembly language functions from head.S

Pull parisc fix from Helge Deller:
 "Fix a parisc kernel crash with ftrace functions when compiled without
  frame pointers"

* 'parisc-5.4-3' of git://git.kernel.org/pub/scm/linux/kernel/git/deller/parisc-linux:
  parisc: fix frame pointer in ftrace_regs_caller()

Jakub Kicinski says:

====================
fix BPF offload related bugs

test_offload.py catches some recently added bugs.

First of a bug in test_offload.py itself after recent changes
to netdevsim is fixed.

Second patch fixes a bug in cls_bpf, and last one addresses
a problem with the recently added XDP installation optimization.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>

When netdevice with offloaded BPF programs is destroyed
the programs are orphaned and removed from the program
IDA - their IDs get released (the programs may remain
accessible via existing open file descriptors and pinned
files). After IDs are released they are set to 0.

This confuses dev_change_xdp_fd() because it compares
the __dev_xdp_query() result where 0 means no program
with prog->aux->id where 0 means orphaned.

dev_change_xdp_fd() would have incorrectly returned success
even though it had not installed the program.

Since drivers already catch this case via bpf_offload_dev_match()
let them handle this case. The error message drivers produce in
this case ("program loaded for a different device") is in fact
correct as the orphaned program must had to be loaded for a
different device.

Fixes: c14a9f63 ("net: Don't call XDP_SETUP_PROG when nothing is changed")
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Commit 40119211 ("net: sched: refactor block offloads counter
usage") missed the fact that either new prog or old prog may be
NULL.

Fixes: 40119211 ("net: sched: refactor block offloads counter usage")
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

DebugFS for netdevsim now contains some "action trigger" files
which are write only. Don't try to capture the contents of those.

Note that we can't use os.access() because the script requires
root.

Fixes: 4418f862 ("netdevsim: implement support for devlink region and snapshots")
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

This test reports EINVAL for getsockopt(SOL_SOCKET, SO_DOMAIN)
occasionally due to the uninitialized length parameter.
Initialize it to fix this, and also use int for "test_family" to comply
with the API standard.

Fixes: d6a61f80 ("soreuseport: test mixed v4/v6 sockets")
Reported-by: Maciej Żenczykowski <maze@google.com>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Wei Wang <weiwan@google.com>
Cc: Craig Gallek <cgallek@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

As reported in [0] at least one RTL8168dp version has problems
establishing a link. This chip version has an integrated RTL8211b PHY,
however the chip seems to report a wrong PHY ID, resulting in a wrong
PHY driver (for Generic Realtek PHY) being loaded.
Work around this issue by adding a hook to r8168dp_2_mdio_read()
for returning the correct PHY ID.

[0] https://bbs.archlinux.org/viewtopic.php?id=246508

Fixes: 242cd9b5 ("r8169: use phy_resume/phy_suspend")
Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Since it became possible for the DSA core to use a CPU port different
than 8, our bcm_sf2_imp_setup() function was broken because it assumes
that registers are applicable to port 8. In particular, the port's MAC
is going to stay disabled, so make sure we clear the RX_DIS and TX_DIS
bits if we are not configured for port 8.

Fixes: 9f91484f ("net: dsa: make "label" property optional for dsa2")
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

The phylink_dbg() macro does not follow dynamic debug or defined(DEBUG)
and as a result, it spams the kernel log since a PR_DEBUG level is
currently used. Fix it to be defined appropriately whether
CONFIG_DYNAMIC_DEBUG or defined(DEBUG) are set.

Fixes: 17091180 ("net: phylink: Add phylink_{printk, err, warn, info, dbg} macros")
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Synces the DMA buffer properly in order for CPU and device to see
the most up-to-data data.
Signed-off-by: Yangchun Fu <yangchun@google.com>
Reviewed-by: Catherine Sullivan <csully@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Historically linux tried to stick to RFC 791, 1122, 2003
for IPv4 ID field generation.

RFC 6864 made clear that no matter how hard we try,
we can not ensure unicity of IP ID within maximum
lifetime for all datagrams with a given source
address/destination address/protocol tuple.

Linux uses a per socket inet generator (inet_id), initialized
at connection startup with a XOR of 'jiffies' and other
fields that appear clear on the wire.

Thiemo Nagel pointed that this strategy is a privacy
concern as this provides 16 bits of entropy to fingerprint
devices.

Let's switch to a random starting point, this is just as
good as far as RFC 6864 is concerned and does not leak
anything critical.

Fixes: 1da177e4 ("Linux-2.6.12-rc2")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reported-by: Thiemo Nagel <tnagel@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>