TC actions are no longer freed in RCU callbacks and we should
always have RTNL lock, so this spinlock is no longer needed.

Cc: Eric Dumazet <eric.dumazet@gmail.com>
Cc: Jiri Pirko <jiri@mellanox.com>
Cc: Jamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: NCong Wang <xiyou.wangcong@gmail.com>
Acked-by: NJiri Pirko <jiri@mellanox.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

tcfm_dev always points to the correct netdev and we already
hold a refcnt, so no need to use tcfm_ifindex to lookup again.

If we would support moving target netdev across netns, using
pointer would be better than ifindex.

This also fixes dumping obsolete ifindex, now after the
target device is gone we just dump 0 as ifindex.

Cc: Jiri Pirko <jiri@mellanox.com>
Cc: Jamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: NCong Wang <xiyou.wangcong@gmail.com>
Acked-by: NJiri Pirko <jiri@mellanox.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

William Tu says:

====================
ipv6: add ip6erspan collect_md mode

Similar to erspan collect_md mode in ipv4, the first patch adds
support for ip6erspan collect metadata mode.  The second patch
adds the test case using bpf_skb_[gs]et_tunnel_key helpers.

The corresponding iproute2 patch:
https://marc.info/?l=linux-netdev&m=151251545410047&w=2
====================
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Extend the existing tests for ip6erspan.
Signed-off-by: NWilliam Tu <u9012063@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Similar to ip6 gretap and ip4 gretap, the patch allows
erspan tunnel to operate in collect metadata mode.
bpf_skb_[gs]et_tunnel_key() helpers can make use of
it right away.
Signed-off-by: NWilliam Tu <u9012063@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Smatch warns that:

    drivers/net/ethernet/broadcom/bnxt/bnxt_tc.c:160 bnxt_tc_parse_actions()
    error: uninitialized symbol 'rc'.

"rc" is either uninitialized or set to zero here so we can just remove
the check.

Fixes: 8c95f773 ("bnxt_en: add support for Flower based vxlan encap/decap offload")
Signed-off-by: NDan Carpenter <dan.carpenter@oracle.com>
Acked-by: NMichael Chan <michael.chan@broadcom.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Julia Cartwright says:

====================
macb rx filter cleanups

Here's a proper patchset based on net-next.

v1 -> v2:
  - Rebased on net-next
  - Add Nicolas's Acks
  - Reorder commits, putting the list_empty() cleanups prior to the
    others.
  - Added commit reverting the GFP_ATOMIC change.
====================
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Now that the rx_fs_lock is no longer held across allocation, it's safe
to use GFP_KERNEL for allocating new entries.

This reverts commit 81da3bf6 ("net: macb: change GFP_KERNEL to
GFP_ATOMIC").

Cc: Julia Lawall <julia.lawall@lip6.fr>
Signed-off-by: NJulia Cartwright <julia@ni.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Commit ae8223de ("net: macb: Added support for RX filtering")
introduces a lock, rx_fs_lock which is intended to protect the list of
rx_flow items and synchronize access to the hardware rx filtering
registers.

However, the region protected by this lock is overscoped, unnecessarily
including things like slab allocation.  Reduce this lock scope to only
include operations which must be performed atomically: list traversal,
addition, and removal, and hitting the macb filtering registers.

This fixes the use of kmalloc w/ GFP_KERNEL in atomic context.

Fixes: ae8223de ("net: macb: Added support for RX filtering")
Cc: Rafal Ozieblo <rafalo@cadence.com>
Cc: Julia Lawall <julia.lawall@lip6.fr>
Acked-by: NNicolas Ferre <nicolas.ferre@microchip.com>
Signed-off-by: NJulia Cartwright <julia@ni.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The list_for_each_entry() macro already handles the case where the list
is empty (by not executing the loop body).  It's not necessary to handle
this case specially, so stop doing so.

Cc: Rafal Ozieblo <rafalo@cadence.com>
Acked-by: NNicolas Ferre <nicolas.ferre@microchip.com>
Signed-off-by: NJulia Cartwright <julia@ni.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

No one actually uses it.

Cc: Jiri Pirko <jiri@mellanox.com>
Cc: Jamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: NCong Wang <xiyou.wangcong@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Vivien Didelot says:

====================
net: dsa: use per-port upstream port

An upstream port is a local switch port used to reach a CPU port.

DSA still considers a unique CPU port in the whole switch fabric and
thus return a unique upstream port for a given switch. This is wrong in
a multiple CPU ports environment.

We are now switching to using the dedicated CPU port assigned to each
port in order to get rid of the deprecated unique tree CPU port.

This patchset makes the dsa_upstream_port() helper take a port argument
and goes one step closer complete support for multiple CPU ports.

Changes in v2:
  - reverse-christmas-tree-fy variables
====================
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The current dsa_upstream_port() helper still assumes a unique CPU port
in the whole switch fabric. This is becoming wrong, as every port in the
fabric has its dedicated CPU port, thus every port has an upstream port.

Add a port argument to the dsa_upstream_port() helper and fetch its CPU
port instead of the deprecated unique fabric CPU port. A CPU or unused
port has no dedicated CPU port, so return itself in this case.

At the same time, change the return value from u8 to unsigned int since
there is no need to limit the size here.
Signed-off-by: NVivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

DSA ports also need to have a dedicated CPU port assigned to them,
because they need to know where to egress frames targeting the CPU,
e.g. To_Cpu frames received on a Marvell Tag port.
Signed-off-by: NVivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Move the setup of the global upstream port within the
mv88e6xxx_setup_upstream_port function.
Signed-off-by: NVivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Add a helper function to setup the upstream port of a given port.

This is the port used to reach the dedicated CPU port. This function
will be extended later to setup the global upstream port as well.
Signed-off-by: NVivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The mv88e6xxx driver currently assumes a single CPU port in the fabric
and thus floods frames with unknown DA on a single DSA port, the one
that is one hop closer to the CPU port.

With multiple CPU ports in mind, this isn't true anymore because CPU
ports could be found behind both DSA ports of a device in-between
others.

For example in a A <-> B <-> C fabric, both A and C having CPU ports,
device B will have to flood such frame to its two DSA ports.

This patch considers both CPU and DSA ports of a device as upstream
ports, where to flood frames with unknown DA addresses.
Signed-off-by: NVivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Alexander Aring says:

====================
net: sched: sch_api: fix coding style issues for extack

this patch prepares to handle extack for qdiscs and fixes checkpatch
issues.

There are a bunch of warnings issued by checkpatch which bothered me.
This first patchset is to get rid of those warnings to make way for
the next patchsets.

I plan to followup with qdiscs, classifiers and actions after this.
====================
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This patch fixes the following checkpatch error:

ERROR: do not use assignment in if condition

by rearranging the if condition to execute init callback only if init
callback exists. The whole setup afterwards is called in any case,
doesn't matter if init callback is set or not. This patch has the same
behaviour as before, just without assign err variable in if condition.
It also makes the code easier to read.
Reviewed-by: NJamal Hadi Salim <jhs@mojatatu.com>
Cc: David Ahern <dsahern@gmail.com>
Signed-off-by: NAlexander Aring <aring@mojatatu.com>
Acked-by: NJamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This patch fix checkpatch issues for upcomming patches according to the
sched api file. It changes checking on null pointer, remove unnecessary
brackets, add variable names for parameters and adjust 80 char width.

Cc: David Ahern <dsahern@gmail.com>
Signed-off-by: NAlexander Aring <aring@mojatatu.com>
Acked-by: NJamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Simon Horman says:

====================
nfp: enhanced debug dump via ethtool

Add debug dump implementation to the NFP driver. This makes use of
existing ethtool infrastructure.  ethtool -W is used to select the dump
level and ethtool -w is used to dump NFP state.

The existing behaviour of dump level 0, dumping the arm.diag resource, is
preserved. Dump levels greater than 0 are implemented by this patchset and
optionally supported by firmware providing a _abi_dump_spec rtsym. This
rtsym provides a specification, in TLV format, of the information to be
dumped from the NFP at each supported dump level.

Dumps are also structured using a TLVs. They consist a prolog and the data
described int he corresponding dump.
====================
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

- The spec defines CSR address ranges for indirect ME CSRs. For Each TLV
  chunk in the spec, dump a chunk that includes the spec and the data
  over the defined address range.
- Each indirect CSR has 8 contexts. To read one context, first write the
  context to a specific derived address, read it back, and then read the
  register value.
- For each address, read and dump all 8 contexts in this manner.
Signed-off-by: NCarl Heymann <carl.heymann@netronome.com>
Reviewed-by: NJakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: NSimon Horman <simon.horman@netronome.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

- The spec defines CSR address ranges for these types.
- Dump each TLV chunk in the spec as a chunk that includes the spec and
  the data over the defined address range.
Signed-off-by: NCarl Heymann <carl.heymann@netronome.com>
Reviewed-by: NJakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: NSimon Horman <simon.horman@netronome.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Dump FW name as TLV, based on dump specification.
Signed-off-by: NCarl Heymann <carl.heymann@netronome.com>
Reviewed-by: NJakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: NSimon Horman <simon.horman@netronome.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

- Add spec TLV for hwinfo field, containing key string as data.
- Add dump TLV for hwinfo field, with data being key and value as packed
  zero-terminated strings.
- If specified hwinfo field is not found, dump the spec TLV as -ENOENT
  error.
Signed-off-by: NCarl Heymann <carl.heymann@netronome.com>
Reviewed-by: NJakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: NSimon Horman <simon.horman@netronome.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

- Dump hwinfo as separate TLV chunk, in a packed format containing
  zero-separated key and value strings.
- This provides additional debug context, if requested by the dumpspec.
Signed-off-by: NCarl Heymann <carl.heymann@netronome.com>
Reviewed-by: NJakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: NSimon Horman <simon.horman@netronome.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

- Support rtsym TLVs.
- If specified rtsym is not found, dump the spec TLV as -ENOENT error.
Signed-off-by: NCarl Heymann <carl.heymann@netronome.com>
Reviewed-by: NJakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: NSimon Horman <simon.horman@netronome.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

- Perform dumpspec traversals for calculating size and populating the
  dump.
- Initially, wrap all spec TLVs in dump error TLVs (changed by later
  patches in the series).
Signed-off-by: NCarl Heymann <carl.heymann@netronome.com>
Reviewed-by: NJakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: NSimon Horman <simon.horman@netronome.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

- Use a TLV structure, with the typed chunks aligned to 8-byte sizes.
- Dump numeric fields as big-endian.
- Prolog contains the dump level.
Signed-off-by: NCarl Heymann <carl.heymann@netronome.com>
Reviewed-by: NJakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: NSimon Horman <simon.horman@netronome.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Load the TLV-based binary specification of what needs to be included in
a dump, from the "_abi_dump_spec" rtsymbol. If the symbol is not defined,
then dumps for levels >= 1 are not supported.
Signed-off-by: NCarl Heymann <carl.heymann@netronome.com>
Reviewed-by: NJakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: NSimon Horman <simon.horman@netronome.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

- Skeleton code to perform a binary debug dump via ethtoolops
  "set_dump", "get_dump_flags" and "get_dump_data", i.e. the ethtool
  -W/w mechanism.
- Skeleton functions for debugdump operations provided.
- An integer "dump level" can be specified, this is stored between
  ethtool invocations. Dump level 0 is still the "arm.diag" resource for
  backward compatibility. Other dump levels each define a set of state
  information to include in the dump, driven by a spec from FW.
Signed-off-by: NCarl Heymann <carl.heymann@netronome.com>
Reviewed-by: NJakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: NSimon Horman <simon.horman@netronome.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Both Eric and Paolo noticed the rcu_barrier() we use in
tcf_block_put_ext() could be a performance bottleneck when
we have a lot of tc classes.

Paolo provided the following to demonstrate the issue:

tc qdisc add dev lo root htb
for I in `seq 1 1000`; do
        tc class add dev lo parent 1: classid 1:$I htb rate 100kbit
        tc qdisc add dev lo parent 1:$I handle $((I + 1)): htb
        for J in `seq 1 10`; do
                tc filter add dev lo parent $((I + 1)): u32 match ip src 1.1.1.$J
        done
done
time tc qdisc del dev root

real    0m54.764s
user    0m0.023s
sys     0m0.000s

The rcu_barrier() there is to ensure we free the block after all chains
are gone, that is, to queue tcf_block_put_final() at the tail of workqueue.
We can achieve this ordering requirement by refcnt'ing tcf block instead,
that is, the tcf block is freed only when the last chain in this block is
gone. This also simplifies the code.

Paolo reported after this patch we get:

real    0m0.017s
user    0m0.000s
sys     0m0.017s
Tested-by: NPaolo Abeni <pabeni@redhat.com>
Cc: Eric Dumazet <edumazet@google.com>
Cc: Jiri Pirko <jiri@mellanox.com>
Cc: Jamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: NCong Wang <xiyou.wangcong@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Merge branch 'ieee802154-for-davem-2017-12-04' of git://git.kernel.org/pub/scm/linux/kernel/git/sschmidt/wpan-next

Stefan Schmidt says:

====================
pull-request: ieee802154-next 2017-12-04

Some update from ieee802154 to *net-next*

Jian-Hong Pan updated our docs to match the APIs in code.
Michael Hennerichs enhanced the adf7242 driver to work with adf7241
devices and reworked the IRQ and packet handling in the driver.
====================
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Functions nsim_bpf_create_prog and nsim_bpf_destroy_prog are local to the
source and do not need to be in global scope, so make them static.

Cleans up sparse warnings:
symbol 'nsim_bpf_create_prog' was not declared. Should it be static?
symbol 'nsim_bpf_destroy_prog' was not declared. Should it be static?
Signed-off-by: NColin Ian King <colin.king@canonical.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Geert Uytterhoeven says:

====================
Teach phylib hard-resetting devices

This patch series adds optional PHY reset support to phylib.

The first two patches are destined for David's net-next tree. They add
core PHY reset code, and update a driver that currently uses its own
reset code.

The last two patches are destined for Simon's renesas tree.  They add
properties to describe the EthernetAVB PHY reset topology to the common
Salvator-X/XS and ULCB DTS files, which solves two issues:
  1. On Salvator-XS, the enable pin of the regulator providing PHY power
     is connected to PRESETn, and PSCI powers down the SoC during system
     suspend.  Hence a PHY reset is needed to restore network
     functionality after system resume.
  2. Linux should not rely on the boot loader having reset the PHY, but
     should reset the PHY during driver probe.

Changes compared to v3:
  - Remove Florian's Acked-by,
  - Add missing #include <linux/gpio/consumer.h>,
  - Re-add the gpiod check, as the dummy gpiod_set_value() for !GPIOLIB
    does not ignore NULL, and calls WARN_ON(1),
  - Do not reassert the reset signal if {mdio,phy}_probe() or
    phy_device_register() succeeded, as that may destroy initial setup,
  - Do not deassert the reset signal in {mdio,phy}_remove(), as it
    should already be deasserted,
  - Bring the PHY back into reset state in phy_device_remove(),
  - Move/consolidate GPIO descriptor acquiring code from
    of_mdiobus_register_phy() and of_mdiobus_register_device() to
    mdiobus_register_device().
    Note that this changes behavior slightly, in that the reset signal
    is now also asserted when called from of_mdiobus_register_device().
  - Add Reviewed-by,

Changes compared to v2, as sent by Sergei Shtylyov:
  - Fix fwnode_get_named_gpiod() call due to added parameters (which
    allowed to eliminate the gpiod_direction_output() call),
  - Rebased, refreshed, reworded,
  - Take over from Sergei,
  - Add Acked-by,
  - Remove unneeded gpiod check, as gpiod_set_value() handles NULL fine,
  - Handle fwnode_get_named_gpiod() errors correctly:
      - -ENOENT is ignored (the GPIO is optional), and turned into NULL,
	which allowed to remove all later !IS_ERR() checks,
      - Other errors (incl. -EPROBE_DEFER) are propagated,
  - Extract DTS patches from series "[PATCH 0/4] ravb: Add PHY reset
    support" (https://www.spinics.net/lists/netdev/msg457308.html), and
    incorporate in this series, after moving reset-gpios from the
    ethernet to the ethernet-phy node.

Given (1) the new reset-gpios DT property in the PHY node follows
established practises, (2) the DT binding change in the first patch has
been acked by Rob, and (3) the DTS patch does not cause any regressions
if it is applied before the PHY driver patches, the DTS patches can be
applied independently.
====================
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

With the phylib now being aware of the "reset-gpios" PHY node property,
there should be no need to frob the PHY reset in this driver anymore...
Signed-off-by: NSergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Acked-by: NNicolas Ferre <nicolas.ferre@atmel.com>
Reviewed-by: NFlorian Fainelli <f.fainelli@gmail.com>
Signed-off-by: NGeert Uytterhoeven <geert+renesas@glider.be>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The PHY devices sometimes do have their reset signal (maybe even power
supply?) tied to some GPIO and sometimes it also does happen that a boot
loader does not leave it deasserted. So far this issue has been attacked
from (as I believe) a wrong angle: by teaching the MAC driver to manipulate
the GPIO in question; that solution, when applied to the device trees, led
to adding the PHY reset GPIO properties to the MAC device node, with one
exception: Cadence MACB driver which could handle the "reset-gpios" prop
in a PHY device subnode. I believe that the correct approach is to teach
the 'phylib' to get the MDIO device reset GPIO from the device tree node
corresponding to this device -- which this patch is doing...

Note that I had to modify the AT803x PHY driver as it would stop working
otherwise -- it made use of the reset GPIO for its own purposes...
Signed-off-by: NSergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Acked-by: NRob Herring <robh@kernel.org>
[geert: Propagate actual errors from fwnode_get_named_gpiod()]
[geert: Avoid destroying initial setup]
[geert: Consolidate GPIO descriptor acquiring code]
Signed-off-by: NGeert Uytterhoeven <geert+renesas@glider.be>
Tested-by: NRichard Leitner <richard.leitner@skidata.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Move dissection of tunnel info to outside of the main flow dissection
function, __skb_flow_dissect(). The sole user of this feature, the flower
classifier, is updated to call tunnel info dissection directly, using
skb_flow_dissect_tunnel_info().

This results in a slightly less complex implementation of
__skb_flow_dissect(), in particular removing logic from that call path
which is not used by the majority of users. The expense of this is borne by
the flower classifier which now has to make an extra call for tunnel info
dissection.

This patch should not result in any behavioural change.
Signed-off-by: NSimon Horman <simon.horman@netronome.com>
Reviewed-by: NJakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This patch introduces an eBPF based queue selection method. With this,
the policy could be offloaded to userspace completely through a new
ioctl TUNSETSTEERINGEBPF.
Signed-off-by: NJason Wang <jasowang@redhat.com>
Acked-by: NWillem de Bruijn <willemb@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Salil Mehta says:

====================
net: hns3: Refactors "reset" handling code in HCLGE layer of HNS3 driver

This patch refactors the code of the reset feature in HCLGE layer
of HNS3 PF driver. Prime motivation to do this change is:
1. To reduce the time for which common miscellaneous Vector 0
   interrupt is disabled because of the reset. Simplification
   of the common miscellaneous interrupt handler routine(for
   Vector 0) used to handle reset and other sources of Vector
   0 interrupt.
2. Separate the task for handling the reset
3. Simplification of reset request submission and pending reset
   logic.

To achieve above below few things have been done:
1. Interrupt is disabled while common miscellaneous interrupt
   handler is entered and re-enabled before it is exit. This
   reduces the interrupt handling latency as compared to older
   interrupt handling scheme where interrupt was being disabled
   in interrupt handler context and re-enabled in task context
   some time later. Made Miscellaneous interrupt handler more
   generic to handle all sources including reset interrupt source.
2. New reset service task has been introduced to service the
   reset handling.
3. Introduces new reset service task for honoring software reset
   requests like from network stack related to timeout and serving
   the pending reset request(to reset the driver and associated
   clients).

Change Log:
Patch V2: Addressed comment by Andrew Lunn
   Link: https://lkml.org/lkml/2017/12/1/366
Patch V1: Initial Submit
====================
Signed-off-by: NDavid S. Miller <davem@davemloft.net>