The error cleanup which is invoked when the hotplug state setup failed
tries to remove the failed state, which is broken.

Fixes: 8fba38c9 ("x86/msr: Convert to hotplug state machine")
Reported-by: kernel test robot <fengguang.wu@intel.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: Sebastian Siewior <bigeasy@linutronix.de>

In case the driver registration fails, the hotplug callback is leaked.

Not fatal, because it's never invoked as there are no instances registered,
but wrong nevertheless.

Fixes: fdc15a36 ("bus/arm-ccn: Convert to hotplug statemachine")
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Pawel Moll <pawel.moll@arm.com>
Cc: Suzuki K Poulose <suzuki.poulose@arm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Will Deacon <will.deacon@arm.com>

If the pmu registration fails the registered hotplug callbacks are not
removed. Wrong in any case, but fatal in case of a modular driver.

Replace the nonsensical state names with proper ones while at it.

Fixes: 77c34ef1 ("perf/x86/intel/cstate: Convert Intel CSTATE to hotplug state machine")
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: Sebastian Siewior <bigeasy@linutronix.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: stable@vger.kernel.org

The cpu hotplug support of this perf driver is broken in several ways:

1) It adds a instance before setting up the state.

2) The state for the instance is different from the state of the
   callback. It's just a randomly chosen state.

3) The instance registration is not error checked so nobody noticed that
   the call can never succeed.

4) The state for the multi install callbacks is chosen randomly and
   overwrites existing state. This is now prevented by the core code so the
   call is guaranteed to fail.

5) The error exit path in the init function leaves the instance registered
   and then frees the memory which contains the enqueued hlist node.

6) The remove function is removing the state and not the instance.

Fix it by:

- Setting up the state before adding instances. Use a dynamically allocated
  state for it.

- Installing instances after the state has been set up

- Removing the instance in the error path before freeing memory

- Removing the instance not the state in the driver remove callback

While at is use raw_cpu_processor_id(), because cpu_processor_id() cannot
be used in preemptible context, and set the driver data after successful
registration of the pmu.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: Shawn Guo <shawnguo@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Sebastian Siewior <bigeasy@linutronix.de>
Cc: Frank Li <frank.li@nxp.com>
Cc: Zhengyu Shen <zhengyu.shen@nxp.com>
Link: http://lkml.kernel.org/r/20161221192111.596204211@linutronix.deSigned-off-by: Thomas Gleixner <tglx@linutronix.de>

The CPU hotplug code is a trainwreck. It leaks a notifier in case of driver
registration error and the per cpu loop is racy against cpu hotplug. Aside
of that the driver should have been written and merged with the new state
machine interfaces in the first place.

Mop up the mess and Convert it to the hotplug state machine.
Signed-off-by: Thomas Grumpy Gleixner <tglx@linutronix.de>
Cc: Nilesh Javali <nilesh.javali@cavium.com>
Cc: Adheer Chandravanshi <adheer.chandravanshi@qlogic.com>
Cc: Chad Dupuis <chad.dupuis@cavium.com>
Cc: Saurav Kashyap <saurav.kashyap@cavium.com>
Cc: Arun Easi <arun.easi@cavium.com>
Cc: Manish Rangankar <manish.rangankar@cavium.com>
Cc: Johannes Thumshirn <jthumshirn@suse.de>
Cc: Hannes Reinecke <hare@suse.de>
Cc: Martin K. Petersen <martin.petersen@oracle.com>
Cc: James Bottomley <James.Bottomley@HansenPartnership.com>

Pull x86 fixes from Ingo Molnar:
 "There's a number of fixes:

   - a round of fixes for CPUID-less legacy CPUs
   - a number of microcode loader fixes
   - i8042 detection robustization fixes
   - stack dump/unwinder fixes
   - x86 SoC platform driver fixes
   - a GCC 7 warning fix
   - virtualization related fixes"

* 'x86-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (24 commits)
  Revert "x86/unwind: Detect bad stack return address"
  x86/paravirt: Mark unused patch_default label
  x86/microcode/AMD: Reload proper initrd start address
  x86/platform/intel/quark: Add printf attribute to imr_self_test_result()
  x86/platform/intel-mid: Switch MPU3050 driver to IIO
  x86/alternatives: Do not use sync_core() to serialize I$
  x86/topology: Document cpu_llc_id
  x86/hyperv: Handle unknown NMIs on one CPU when unknown_nmi_panic
  x86/asm: Rewrite sync_core() to use IRET-to-self
  x86/microcode/intel: Replace sync_core() with native_cpuid()
  Revert "x86/boot: Fail the boot if !M486 and CPUID is missing"
  x86/asm/32: Make sync_core() handle missing CPUID on all 32-bit kernels
  x86/cpu: Probe CPUID leaf 6 even when cpuid_level == 6
  x86/tools: Fix gcc-7 warning in relocs.c
  x86/unwind: Dump stack data on warnings
  x86/unwind: Adjust last frame check for aligned function stacks
  x86/init: Fix a couple of comment typos
  x86/init: Remove i8042_detect() from platform ops
  Input: i8042 - Trust firmware a bit more when probing on X86
  x86/init: Add i8042 state to the platform data
  ...

Pull timer fix from Ingo Molnar:
 "ARM/MOXA SoC clocksource driver fixes"

* 'timers-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  clocksource/drivers/moxart: Plug memory and mapping leaks

Pull perf fixes from Ingo Molnar:
 "On the kernel side there's two x86 PMU driver fixes and a uprobes fix,
  plus on the tooling side there's a number of fixes and some late
  updates"

* 'perf-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (36 commits)
  perf sched timehist: Fix invalid period calculation
  perf sched timehist: Remove hardcoded 'comm_width' check at print_summary
  perf sched timehist: Enlarge default 'comm_width'
  perf sched timehist: Honour 'comm_width' when aligning the headers
  perf/x86: Fix overlap counter scheduling bug
  perf/x86/pebs: Fix handling of PEBS buffer overflows
  samples/bpf: Move open_raw_sock to separate header
  samples/bpf: Remove perf_event_open() declaration
  samples/bpf: Be consistent with bpf_load_program bpf_insn parameter
  tools lib bpf: Add bpf_prog_{attach,detach}
  samples/bpf: Switch over to libbpf
  perf diff: Do not overwrite valid build id
  perf annotate: Don't throw error for zero length symbols
  perf bench futex: Fix lock-pi help string
  perf trace: Check if MAP_32BIT is defined (again)
  samples/bpf: Make perf_event_read() static
  uprobes: Fix uprobes on MIPS, allow for a cache flush after ixol breakpoint creation
  samples/bpf: Make samples more libbpf-centric
  tools lib bpf: Add flags to bpf_create_map()
  tools lib bpf: use __u32 from linux/types.h
  ...

Pull irq fix from Ingo Molnar:
 "A build warning fix with certain .config's"

* 'irq-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  irqchip/st: Mark st_irq_syscfg_resume() __maybe_unused

Revert the following commit:

  b6959a36 ("x86/unwind: Detect bad stack return address")

... because Andrey Konovalov reported an unwinder warning:

  WARNING: unrecognized kernel stack return address ffffffffa0000001 at ffff88006377fa18 in a.out:4467

The unwind was initiated from an interrupt which occurred while running in the
generated code for a kprobe.  The unwinder printed the warning because it
expected regs->ip to point to a valid text address, but instead it pointed to
the generated code.

Eventually we may want come up with a way to identify generated kprobe
code so the unwinder can know that it's a valid return address.  Until
then, just remove the warning.
Reported-by: Andrey Konovalov <andreyknvl@google.com>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
Cc: Andy Lutomirski <luto@kernel.org>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Brian Gerst <brgerst@gmail.com>
Cc: Denys Vlasenko <dvlasenk@redhat.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/02f296848fbf49fb72dfeea706413ecbd9d4caf6.1482418739.git.jpoimboe@redhat.comSigned-off-by: Ingo Molnar <mingo@kernel.org>

Merge tag 'perf-urgent-for-mingo-20161222' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux into perf/urgent

Pull perf/urgent fixes from Arnaldo Carvalho de Melo:

Fixes for 'perf sched timehist': (Namhyung Kim)

 - Define a larger initial alignment value for the COMM column and
   make it be more consistently honoured, for instance in the header.

 - Fix invalid period calculation when using the --time option to
   select a time slice, when events outside that slice were being
   considered for the per cpu idle stats summary.
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>

Pull networking fixes from David Miller:

 1) We have to be careful to not try and place a checksum after the end
    of a rawv6 packet, fix from Dave Jones with help from Hannes
    Frederic Sowa.

 2) Missing memory barriers in tcp_tasklet_func() lead to crashes, from
    Eric Dumazet.

 3) Several bug fixes for the new XDP support in virtio_net, from Jason
    Wang.

 4) Increase headroom in RX skbs in be2net driver to accomodate
    encapsulations such as geneve. From Kalesh A P.

 5) Fix SKB frag unmapping on TX in mvpp2, from Thomas Petazzoni.

 6) Pre-pulling UDP headers created a regression in RECVORIGDSTADDR
    socket option support, from Willem de Bruijn.

 7) UID based routing added a potential OOPS in ip_do_redirect() when we
    see an SKB without a socket attached. We just need it for the
    network namespace which we can get from skb->dev instead. Fix from
    Lorenzo Colitti.

* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net: (30 commits)
  sctp: fix recovering from 0 win with small data chunks
  sctp: do not loose window information if in rwnd_over
  virtio-net: XDP support for small buffers
  virtio-net: remove big packet XDP codes
  virtio-net: forbid XDP when VIRTIO_NET_F_GUEST_UFO is support
  virtio-net: make rx buf size estimation works for XDP
  virtio-net: unbreak csumed packets for XDP_PASS
  virtio-net: correctly handle XDP_PASS for linearized packets
  virtio-net: fix page miscount during XDP linearizing
  virtio-net: correctly xmit linearized page on XDP_TX
  virtio-net: remove the warning before XDP linearizing
  mlxsw: spectrum_router: Correctly remove nexthop groups
  mlxsw: spectrum_router: Don't reflect dead neighs
  neigh: Send netevent after marking neigh as dead
  ipv6: handle -EFAULT from skb_copy_bits
  inet: fix IP(V6)_RECVORIGDSTADDR for udp sockets
  net/sched: cls_flower: Mandate mask when matching on flags
  net/sched: act_tunnel_key: Fix setting UDP dst port in metadata under IPv6
  stmmac: CSR clock configuration fix
  net: ipv4: Don't crash if passing a null sk to ip_do_redirect.
  ...

Currently if SCTP closes the receive window with window pressure, mostly
caused by excessive skb overhead on payload/overheads ratio, SCTP will
close the window abruptly while saving the delta on rwnd_press. It will
start recovering rwnd as the chunks are consumed by the application and
the rwnd_press will be only recovered after rwnd reach the same value as
of rwnd_press, mostly to prevent silly window syndrome.

Thing is, this is very inefficient with small data chunks, as with those
it will never reach back that value, and thus it will never recover from
such pressure. This means that we will not issue window updates when
recovering from 0 window and will rely on a sender retransmit to notice
it.

The fix here is to remove such threshold, as no value is good enough: it
depends on the (avg) chunk sizes being used.

Test with netperf -t SCTP_STREAM -- -m 1, and trigger 0 window by
sending SIGSTOP to netserver, sleep 1.2, and SIGCONT.
Rate limited to 845kbps, for visibility. Capture done at netserver side.

Previously:
01.500751 IP B.48277 > A.36925: sctp (1) [SACK] [cum ack 632372996] [a_rwnd 99153] [
01.500752 IP A.36925 > B.48277: sctp (1) [DATA] (B)(E) [TSN: 632372997] [SID: 0] [SS
01.517471 IP A.36925 > B.48277: sctp (1) [DATA] (B)(E) [TSN: 632373010] [SID: 0] [SS
01.517483 IP B.48277 > A.36925: sctp (1) [SACK] [cum ack 632373009] [a_rwnd 0] [#gap
01.517485 IP A.36925 > B.48277: sctp (1) [DATA] (B)(E) [TSN: 632373083] [SID: 0] [SS
01.517488 IP B.48277 > A.36925: sctp (1) [SACK] [cum ack 632373009] [a_rwnd 0] [#gap
01.534168 IP A.36925 > B.48277: sctp (1) [DATA] (B)(E) [TSN: 632373096] [SID: 0] [SS
01.534180 IP B.48277 > A.36925: sctp (1) [SACK] [cum ack 632373009] [a_rwnd 0] [#gap
01.534181 IP A.36925 > B.48277: sctp (1) [DATA] (B)(E) [TSN: 632373169] [SID: 0] [SS
01.534185 IP B.48277 > A.36925: sctp (1) [SACK] [cum ack 632373009] [a_rwnd 0] [#gap
02.525978 IP A.36925 > B.48277: sctp (1) [DATA] (B)(E) [TSN: 632373010] [SID: 0] [SS
02.526021 IP B.48277 > A.36925: sctp (1) [SACK] [cum ack 632373009] [a_rwnd 0] [#gap
  (window update missed)
04.573807 IP A.36925 > B.48277: sctp (1) [DATA] (B)(E) [TSN: 632373010] [SID: 0] [SS
04.779370 IP B.48277 > A.36925: sctp (1) [SACK] [cum ack 632373082] [a_rwnd 859] [#g
04.789162 IP A.36925 > B.48277: sctp (1) [DATA] (B)(E) [TSN: 632373083] [SID: 0] [SS
04.789323 IP A.36925 > B.48277: sctp (1) [DATA] (B)(E) [TSN: 632373156] [SID: 0] [SS
04.789372 IP B.48277 > A.36925: sctp (1) [SACK] [cum ack 632373228] [a_rwnd 786] [#g

After:
02.568957 IP B.50536 > A.55173: sctp (1) [SACK] [cum ack 2490098728] [a_rwnd 99153]
02.568961 IP A.55173 > B.50536: sctp (1) [DATA] (B)(E) [TSN: 2490098729] [SID: 0] [S
02.585631 IP A.55173 > B.50536: sctp (1) [DATA] (B)(E) [TSN: 2490098742] [SID: 0] [S
02.585666 IP B.50536 > A.55173: sctp (1) [SACK] [cum ack 2490098741] [a_rwnd 0] [#ga
02.585671 IP A.55173 > B.50536: sctp (1) [DATA] (B)(E) [TSN: 2490098815] [SID: 0] [S
02.585683 IP B.50536 > A.55173: sctp (1) [SACK] [cum ack 2490098741] [a_rwnd 0] [#ga
02.602330 IP A.55173 > B.50536: sctp (1) [DATA] (B)(E) [TSN: 2490098828] [SID: 0] [S
02.602359 IP B.50536 > A.55173: sctp (1) [SACK] [cum ack 2490098741] [a_rwnd 0] [#ga
02.602363 IP A.55173 > B.50536: sctp (1) [DATA] (B)(E) [TSN: 2490098901] [SID: 0] [S
02.602372 IP B.50536 > A.55173: sctp (1) [SACK] [cum ack 2490098741] [a_rwnd 0] [#ga
03.600788 IP A.55173 > B.50536: sctp (1) [DATA] (B)(E) [TSN: 2490098742] [SID: 0] [S
03.600830 IP B.50536 > A.55173: sctp (1) [SACK] [cum ack 2490098741] [a_rwnd 0] [#ga
03.619455 IP B.50536 > A.55173: sctp (1) [SACK] [cum ack 2490098741] [a_rwnd 13508]
03.619479 IP B.50536 > A.55173: sctp (1) [SACK] [cum ack 2490098741] [a_rwnd 27017]
03.619497 IP B.50536 > A.55173: sctp (1) [SACK] [cum ack 2490098741] [a_rwnd 40526]
03.619516 IP B.50536 > A.55173: sctp (1) [SACK] [cum ack 2490098741] [a_rwnd 54035]
03.619533 IP B.50536 > A.55173: sctp (1) [SACK] [cum ack 2490098741] [a_rwnd 67544]
03.619552 IP B.50536 > A.55173: sctp (1) [SACK] [cum ack 2490098741] [a_rwnd 81053]
03.619570 IP B.50536 > A.55173: sctp (1) [SACK] [cum ack 2490098741] [a_rwnd 94562]
  (following data transmission triggered by window updates above)
03.633504 IP A.55173 > B.50536: sctp (1) [DATA] (B)(E) [TSN: 2490098742] [SID: 0] [S
03.836445 IP B.50536 > A.55173: sctp (1) [SACK] [cum ack 2490098814] [a_rwnd 100000]
03.843125 IP A.55173 > B.50536: sctp (1) [DATA] (B)(E) [TSN: 2490098815] [SID: 0] [S
03.843285 IP A.55173 > B.50536: sctp (1) [DATA] (B)(E) [TSN: 2490098888] [SID: 0] [S
03.843345 IP B.50536 > A.55173: sctp (1) [SACK] [cum ack 2490098960] [a_rwnd 99894]
03.856546 IP A.55173 > B.50536: sctp (1) [DATA] (B)(E) [TSN: 2490098961] [SID: 0] [S
03.866450 IP A.55173 > B.50536: sctp (1) [DATA] (B)(E) [TSN: 2490099011] [SID: 0] [S
Signed-off-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

It's possible that we receive a packet that is larger than current
window. If it's the first packet in this way, it will cause it to
increase rwnd_over. Then, if we receive another data chunk (specially as
SCTP allows you to have one data chunk in flight even during 0 window),
rwnd_over will be overwritten instead of added to.

In the long run, this could cause the window to grow bigger than its
initial size, as rwnd_over would be charged only for the last received
data chunk while the code will try open the window for all packets that
were received and had its value in rwnd_over overwritten. This, then,
can lead to the worsening of payload/buffer ratio and cause rwnd_press
to kick in more often.

The fix is to sum it too, same as is done for rwnd_press, so that if we
receive 3 chunks after closing the window, we still have to release that
same amount before re-opening it.

Log snippet from sctp_test exhibiting the issue:
[  146.209232] sctp: sctp_assoc_rwnd_decrease: asoc:ffff88013928e000
rwnd decreased by 1 to (0, 1, 114221)
[  146.209232] sctp: sctp_assoc_rwnd_decrease:
association:ffff88013928e000 has asoc->rwnd:0, asoc->rwnd_over:1!
[  146.209232] sctp: sctp_assoc_rwnd_decrease: asoc:ffff88013928e000
rwnd decreased by 1 to (0, 1, 114221)
[  146.209232] sctp: sctp_assoc_rwnd_decrease:
association:ffff88013928e000 has asoc->rwnd:0, asoc->rwnd_over:1!
[  146.209232] sctp: sctp_assoc_rwnd_decrease: asoc:ffff88013928e000
rwnd decreased by 1 to (0, 1, 114221)
[  146.209232] sctp: sctp_assoc_rwnd_decrease:
association:ffff88013928e000 has asoc->rwnd:0, asoc->rwnd_over:1!
[  146.209232] sctp: sctp_assoc_rwnd_decrease: asoc:ffff88013928e000
rwnd decreased by 1 to (0, 1, 114221)
Signed-off-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Pull final vfs updates from Al Viro:
 "Assorted cleanups and fixes all over the place"

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs:
  sg_write()/bsg_write() is not fit to be called under KERNEL_DS
  ufs: fix function declaration for ufs_truncate_blocks
  fs: exec: apply CLOEXEC before changing dumpable task flags
  seq_file: reset iterator to first record for zero offset
  vfs: fix isize/pos/len checks for reflink & dedupe
  [iov_iter] fix iterate_all_kinds() on empty iterators
  move aio compat to fs/aio.c
  reorganize do_make_slave()
  clone_private_mount() doesn't need to touch namespace_sem
  remove a bogus claim about namespace_sem being held by callers of mnt_alloc_id()

Jason Wang says:

====================
several fixups for virtio-net XDP

Merry Xmas and a Happy New year to all:

This series tries to fixes several issues for virtio-net XDP which
could be categorized into several parts:

- fix several issues during XDP linearizing
- allow csumed packet to work for XDP_PASS
- make EWMA rxbuf size estimation works for XDP
- forbid XDP when GUEST_UFO is support
- remove big packet XDP support
- add XDP support or small buffer

Please see individual patches for details.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>

Commit f600b690 ("virtio_net: Add XDP support") leaves the case of
small receive buffer untouched. This will confuse the user who want to
set XDP but use small buffers. Other than forbid XDP in small buffer
mode, let's make it work. XDP then can only work at skb->data since
virtio-net create skbs during refill, this is sub optimal which could
be optimized in the future.

Cc: John Fastabend <john.r.fastabend@intel.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
Acked-by: John Fastabend <john.r.fastabend@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Now we in fact don't allow XDP for big packets, remove its codes.

Cc: John Fastabend <john.r.fastabend@intel.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

When VIRTIO_NET_F_GUEST_UFO is negotiated, host could still send UFO
packet that exceeds a single page which could not be handled
correctly by XDP. So this patch forbids setting XDP when GUEST_UFO is
supported. While at it, forbid XDP for ECN (which comes only from GRO)
too to prevent user from misconfiguration.

Cc: John Fastabend <john.r.fastabend@intel.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
Acked-by: John Fastabend <john.r.fastabend@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

We don't update ewma rx buf size in the case of XDP. This will lead
underestimation of rx buf size which causes host to produce more than
one buffers. This will greatly increase the possibility of XDP page
linearization.

Cc: John Fastabend <john.r.fastabend@intel.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
Acked-by: John Fastabend <john.r.fastabend@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

We drop csumed packet when do XDP for packets. This breaks
XDP_PASS when GUEST_CSUM is supported. Fix this by allowing csum flag
to be set. With this patch, simple TCP works for XDP_PASS.

Cc: John Fastabend <john.r.fastabend@intel.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
Acked-by: John Fastabend <john.r.fastabend@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

When XDP_PASS were determined for linearized packets, we try to get
new buffers in the virtqueue and build skbs from them. This is wrong,
we should create skbs based on existed buffers instead. Fixing them by
creating skb based on xdp_page.

With this patch "ping 192.168.100.4 -s 3900 -M do" works for XDP_PASS.

Cc: John Fastabend <john.r.fastabend@intel.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
Acked-by: John Fastabend <john.r.fastabend@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

We don't put page during linearizing, the would cause leaking when
xmit through XDP_TX or the packet exceeds PAGE_SIZE. Fix them by
put page accordingly. Also decrease the number of buffers during
linearizing to make sure caller can free buffers correctly when packet
exceeds PAGE_SIZE. With this patch, we won't get OOM after linearize
huge number of packets.

Cc: John Fastabend <john.r.fastabend@intel.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
Acked-by: John Fastabend <john.r.fastabend@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

After we linearize page, we should xmit this page instead of the page
of first buffer which may lead unexpected result. With this patch, we
can see correct packet during XDP_TX.

Cc: John Fastabend <john.r.fastabend@intel.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
Acked-by: John Fastabend <john.r.fastabend@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Since we use EWMA to estimate the size of rx buffer. When rx buffer
size is underestimated, it's usual to have a packet with more than one
buffers. Consider this is not a bug, remove the warning and correct
the comment before XDP linearizing.

Cc: John Fastabend <john.r.fastabend@intel.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Pull befs updates from Luis de Bethencourt:
 "A series of small fixes and adding NFS export support"

* tag 'befs-v4.10-rc1' of git://github.com/luisbg/linux-befs:
  befs: add NFS export support
  befs: remove trailing whitespaces
  befs: remove signatures from comments
  befs: fix style issues in header files
  befs: fix style issues in linuxvfs.c
  befs: fix typos in linuxvfs.c
  befs: fix style issues in io.c
  befs: fix style issues in inode.c
  befs: fix style issues in debug.c

Pull drm fixes from Dave Airlie:
 "Some fixes came in while I was out, mostly intel and amdgpu ones, with
  one ast fix"

Daniel Vetter says:
 "This should also shut up the WARN_ON(!intel_dp->lane_count) noise"

* tag 'drm-fixes-for-4.10-rc1' of git://people.freedesktop.org/~airlied/linux: (35 commits)
  drm/amdgpu: update tile table for oland/hainan
  drm/amdgpu: update tile table for verde
  drm/amdgpu: update rev id for verde
  drm/amdgpu: update golden setting for verde
  drm/amdgpu: update rev id for oland
  drm/amdgpu: update golden setting for oland
  drm/amdgpu: update rev id for hainan
  drm/amdgpu: update golden setting for hainan
  drm/amdgpu: update rev id for pitcairn
  drm/amdgpu: update golden setting for pitcairn
  drm/amdgpu: update golden setting/tiling table of tahiti
  drm/i915: skip the first 4k of stolen memory on everything >= gen8
  drm/i915: Fallback to single PAGE_SIZE segments for DMA remapping
  drm/i915: Fix use after free in logical_render_ring_init
  drm/i915: disable PSR by default on HSW/BDW
  drm/i915: Fix setting of boost freq tunable
  drm/i915: tune down the fast link training vs boot fail
  drm/i915: Reorder phys backing storage release
  drm/i915/gen9: Fix PCODE polling during SAGV disabling
  drm/i915/gen9: Fix PCODE polling during CDCLK change notification
  ...

Pull rdma fixes from Doug Ledford:
 "First round of -rc fixes for 4.10 kernel:

   - a series of qedr fixes
   - a series of rxe fixes
   - one i40iw fix
   - one cma fix
   - one cxgb4 fix"

* tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/dledford/rdma:
  IB/rxe: Don't check for null ptr in send()
  IB/rxe: Drop future atomic/read packets rather than retrying
  IB/rxe: Use BTH_PSN_MASK when ACKing duplicate sends
  qedr: Always notify the verb consumer of flushed CQEs
  qedr: clear the vendor error field in the work completion
  qedr: post_send/recv according to QP state
  qedr: ignore inline flag in read verbs
  qedr: modify QP state to error when destroying it
  qedr: return correct value on modify qp
  qedr: return error if destroy CQ failed
  qedr: configure the number of CQEs on CQ creation
  i40iw: Set 128B as the only supported RQ WQE size
  IB/cma: Fix a race condition in iboe_addr_get_sgid()
  IB/rxe: Fix a memory leak in rxe_qp_cleanup()
  iw_cxgb4: set correct FetchBurstMax for QPs

Pull late SCSI updates from James Bottomley:
 "This is mostly stuff which missed the initial pull.

  There's a new driver: qedi, and some ufs, ibmvscsis and ncr5380
  updates plus some assorted driver fixes and also a fix for the bug
  where if a device goes into a blocked state between configuration and
  sysfs device add (which can be a long time under async probing) it
  would become permanently blocked"

* tag 'scsi-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi: (30 commits)
  scsi: avoid a permanent stop of the scsi device's request queue
  scsi: mpt3sas: Recognize and act on iopriority info
  scsi: qla2xxx: Fix Target mode handling with Multiqueue changes.
  scsi: qla2xxx: Add Block Multi Queue functionality.
  scsi: qla2xxx: Add multiple queue pair functionality.
  scsi: qla2xxx: Utilize pci_alloc_irq_vectors/pci_free_irq_vectors calls.
  scsi: qla2xxx: Only allow operational MBX to proceed during RESET.
  scsi: hpsa: remove memory allocate failure message
  scsi: Update 3ware driver email addresses
  scsi: zfcp: fix rport unblock race with LUN recovery
  scsi: zfcp: do not trace pure benign residual HBA responses at default level
  scsi: zfcp: fix use-after-"free" in FC ingress path after TMF
  scsi: libcxgbi: return error if interface is not up
  scsi: cxgb4i: libcxgbi: add missing module_put()
  scsi: cxgb4i: libcxgbi: cxgb4: add T6 iSCSI completion feature
  scsi: cxgb4i: libcxgbi: add active open cmd for T6 adapters
  scsi: cxgb4i: use cxgb4_tp_smt_idx() to get smt_idx
  scsi: qedi: Add QLogic FastLinQ offload iSCSI driver framework.
  scsi: aacraid: remove wildcard for series 9 controllers
  scsi: ibmvscsi: add write memory barrier to CRQ processing
  ...

Pull more ARC updates from Vineet Gupta:

 - Fix for aliasing VIPT dcache in old ARC700 cores

 - micro-optimization in ARC700 ProtV handler

 - Enable SG_CHAIN  [Vladimir]

 - ARC HS38 core intc default to prio 1

* tag 'arc-4.10-rc1-part2' of git://git.kernel.org/pub/scm/linux/kernel/git/vgupta/arc:
  ARC: mm: arc700: Don't assume 2 colours for aliasing VIPT dcache
  ARC: mm: No need to save cache version in @cpuinfo
  ARC: enable SG chaining
  ARCv2: intc: default all interrupts to priority 1
  ARCv2: entry: document intr disable in hard isr
  ARC: ARCompact entry: elide re-reading ECR in ProtV handler

Jiri Pirko says:

====================
mlxsw: Router fixes

Ido says:

First two patches ensure we remove from the device's table neighbours
that are considered to be dead by the neighbour core.

The last patch removes nexthop groups from the device when they are no
longer valid.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>

At the end of the nexthop initialization process we determine whether
the nexthop should be offloaded or not based on the NUD state of the
neighbour representing it. After all the nexthops were initialized we
refresh the nexthop group and potentially offload it to the device, in
case some of the nexthops were resolved.

Make the destruction of a nexthop group symmetric with its creation by
marking all nexthops as invalid and then refresh the nexthop group to
make sure it was removed from the device's tables.

Fixes: b2157149 ("mlxsw: spectrum_router: Add the nexthop neigh activity update")
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

When a neighbour is considered to be dead, we should remove it from the
device's table regardless of its NUD state.

Without this patch, after setting a port to be administratively down we
get the following errors when we periodically try to update the kernel
about neighbours activity:

[  461.947268] mlxsw_spectrum 0000:03:00.0 sw1p3: Failed to find
matching neighbour for IP=192.168.100.2

Fixes: a6bf9e93 ("mlxsw: spectrum_router: Offload neighbours based on NUD state change")
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

neigh_cleanup_and_release() is always called after marking a neighbour
as dead, but it only notifies user space and not in-kernel listeners of
the netevent notification chain.

This can cause multiple problems. In my specific use case, it causes the
listener (a switch driver capable of L3 offloads) to believe a neighbour
entry is still valid, and is thus erroneously kept in the device's
table.

Fix that by sending a netevent after marking the neighbour as dead.

Fixes: a6bf9e93 ("mlxsw: spectrum_router: Offload neighbours based on NUD state change")
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

By setting certain socket options on ipv6 raw sockets, we can confuse the
length calculation in rawv6_push_pending_frames triggering a BUG_ON.

RIP: 0010:[<ffffffff817c6390>] [<ffffffff817c6390>] rawv6_sendmsg+0xc30/0xc40
RSP: 0018:ffff881f6c4a7c18  EFLAGS: 00010282
RAX: 00000000fffffff2 RBX: ffff881f6c681680 RCX: 0000000000000002
RDX: ffff881f6c4a7cf8 RSI: 0000000000000030 RDI: ffff881fed0f6a00
RBP: ffff881f6c4a7da8 R08: 0000000000000000 R09: 0000000000000009
R10: ffff881fed0f6a00 R11: 0000000000000009 R12: 0000000000000030
R13: ffff881fed0f6a00 R14: ffff881fee39ba00 R15: ffff881fefa93a80

Call Trace:
 [<ffffffff8118ba23>] ? unmap_page_range+0x693/0x830
 [<ffffffff81772697>] inet_sendmsg+0x67/0xa0
 [<ffffffff816d93f8>] sock_sendmsg+0x38/0x50
 [<ffffffff816d982f>] SYSC_sendto+0xef/0x170
 [<ffffffff816da27e>] SyS_sendto+0xe/0x10
 [<ffffffff81002910>] do_syscall_64+0x50/0xa0
 [<ffffffff817f7cbc>] entry_SYSCALL64_slow_path+0x25/0x25

Handle by jumping to the failure path if skb_copy_bits gets an EFAULT.

Reproducer:

#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <unistd.h>
#include <sys/types.h>
#include <sys/socket.h>
#include <netinet/in.h>

#define LEN 504

int main(int argc, char* argv[])
{
	int fd;
	int zero = 0;
	char buf[LEN];

	memset(buf, 0, LEN);

	fd = socket(AF_INET6, SOCK_RAW, 7);

	setsockopt(fd, SOL_IPV6, IPV6_CHECKSUM, &zero, 4);
	setsockopt(fd, SOL_IPV6, IPV6_DSTOPTS, &buf, LEN);

	sendto(fd, buf, 1, 0, (struct sockaddr *) buf, 110);
}
Signed-off-by: Dave Jones <davej@codemonkey.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>

Socket cmsg IP(V6)_RECVORIGDSTADDR checks that port range lies within
the packet. For sockets that have transport headers pulled, transport
offset can be negative. Use signed comparison to avoid overflow.

Fixes: e6afc8ac ("udp: remove headers from UDP packets before queueing")
Reported-by: Nisar Jagabar <njagabar@cloudmark.com>
Signed-off-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Or Gerlitz says:

====================
net/sched fixes for cls_flower and act_tunnel_key

This small series contain a fix to the matching flags support
in flower and to the tunnel key action MD prep for IPv6.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>

When matching on flags, we should require the user to provide the
mask and avoid using an all-ones mask. Not doing so causes matching
on flags provided w.o mask to hit on the value being unset for all
flags, which may not what the user wanted to happen.

Fixes: faa3ffce ('net/sched: cls_flower: Add support for matching on flags')
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Reported-by: Paul Blakey <paulb@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

The UDP dst port was provided to the helper function which sets the
IPv6 IP tunnel meta-data under a wrong param order, fix that.

Fixes: 75bfbca0 ('net/sched: act_tunnel_key: Add UDP dst port option')
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Reviewed-by: Hadar Hen Zion <hadarh@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

When testing stmmac with my QoS reference design I checked a problem in the
CSR clock configuration that was impossibilitating the phy discovery, since
every read operation returned 0x0000ffff. This patch fixes the issue.
Signed-off-by: Joao Pinto <jpinto@synopsys.com>
Signed-off-by: David S. Miller <davem@davemloft.net>