Currently the mlxsw driver supports an option for disabling syncing
the hardware learned FDBs with the software bridge. This behavior
breaks the bridge offload model and thus it is removed.
Signed-off-by: Arkadi Sharshevsky <arkadis@mellanox.com>
Reviewed-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 new static FDB is added to the bridge a notification is sent to
the driver for offload. In case of successful offload the driver should
notify the bridge back, which in turn should mark the FDB as offloaded.

Currently, externally learned is equivalent for being offloaded which is
not correct due to the fact that FDBs which are added from user-space are
also marked as externally learned. In order to specify if an FDB was
successfully offloaded a new flag is introduced.
Signed-off-by: Arkadi Sharshevsky <arkadis@mellanox.com>
Reviewed-by: Ido Schimmel <idosch@mellanox.com>
Reviewed-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Currently the bridge doesn't notify the underlying devices about new
FDBs learned. The FDB sync is placed on the switchdev notifier chain
because devices may potentially learn FDB that are not directly related
to their ports, for example:

1. Mixed SW/HW bridge - FDBs that point to the ASICs external devices
                        should be offloaded as CPU traps in order to
			perform forwarding in slow path.
2. EVPN - Externally learned FDBs for the vtep device.

Notification is sent only about static FDB add/del. This is done due
to fact that currently this is the only scenario supported by switch
drivers.
Signed-off-by: Arkadi Sharshevsky <arkadis@mellanox.com>
Reviewed-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Reviewed-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Reviewed-by: Ivan Vecera <ivecera@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

In order to use the switchdev notifier chain for FDB sync with the
device it has to be changed to atomic. The is done because the bridge
can learn new FDBs in atomic context.
Signed-off-by: Arkadi Sharshevsky <arkadis@mellanox.com>
Reviewed-by: Ido Schimmel <idosch@mellanox.com>
Reviewed-by: Ivan Vecera <ivecera@redhat.com>
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Reviewed-by: Ivan Vecera <ivecera@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

This is done as a preparation to moving the switchdev notifier chain
to be atomic. The FDB external learning should be called under rtnl
or rcu.
Signed-off-by: Arkadi Sharshevsky <arkadis@mellanox.com>
Reviewed-by: Ido Schimmel <idosch@mellanox.com>
Reviewed-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Currently the flood, learning and learning_sync port attributes are
offloaded by setting the SELF flag. Add support for offloading the
flood and learning attribute through the bridge code. In case of
setting an unsupported flag on a offloded port the operation will
fail.

The learning_sync attribute doesn't have any software representation
and cannot be offloaded through the bridge code.
Signed-off-by: Arkadi Sharshevsky <arkadis@mellanox.com>
Reviewed-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Reviewed-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Reviewed-by: Ivan Vecera <ivecera@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

This is done as a preparation stage before setting the bridge port flags
from the bridge code. Currently the device can be queried for the bridge
flags state, but the querier cannot distinguish if the flag is disabled
or if it is not supported at all. Thus, add new attr and a bit-mask which
include information regarding the support on a per-flag basis.

Drivers that support bridge offload but not support bridge flags should
return zeroed bitmask.
Signed-off-by: Arkadi Sharshevsky <arkadis@mellanox.com>
Reviewed-by: Ido Schimmel <idosch@mellanox.com>
Reviewed-by: Ivan Vecera <ivecera@redhat.com>
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Reviewed-by: Ivan Vecera <ivecera@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Vivien Didelot says:

====================
net: dsa: add cross-chip VLAN support

The current code in DSA does not support cross-chip VLAN. This means
that in a multi-chip environment such as this one (similar to ZII Rev B)

         [CPU].................... (mdio)
    (eth0) |   :       :          :
          _|_____    _______    _______
         [__sw0__]--[__sw1__]--[__sw2__]
          |  |  |    |  |  |    |  |  |
          v  v  v    v  v  v    v  v  v
          p1 p2 p3   p4 p5 p6   p7 p8 p9

adding a VLAN to p9 won't be enough to reach the CPU, until at least one
port of sw0 and sw1 join the VLAN as well and become aware of the VID.

This patchset makes the DSA core program the VLAN on the CPU and DSA
links itself, which brings seamlessly cross-chip VLAN support to DSA.

With this series applied*, the hardware VLAN tables of a 3-switch setup
look like this after adding a VLAN to only one port of the end switch:

    # cat /sys/class/net/br0/bridge/default_pvid
    42
    # cat /sys/kernel/debug/mv88e6xxx/sw{0,1,2}/vtu
    # ip link set up master br0 dev lan6
    # cat /sys/kernel/debug/mv88e6xxx/sw{0,1,2}/vtu
     VID  FID  SID  0  1  2  3  4  5  6
      42    1    0  x  x  x  x  x  =  =
     VID  FID  SID  0  1  2  3  4  5  6
      42    1    0  x  x  x  x  x  =  =
     VID  FID  SID  0  1  2  3  4  5  6  7  8  9
      42    1    0  u  x  x  x  x  x  x  x  x  =

('x' is excluded, 'u' is untagged, '=' is unmodified DSA and CPU ports.)

Completely removing a VLAN entry (which is currently the responsibility
of drivers anyway) is not supported yet since it requires some caching.

(*) the output is shown from this out-of-tree debugfs patch:
https://github.com/vivien/linux/commit/7b61a684b9d6b6a499135a587c7f62a1fddceb8b.patch

Changes in v2:
  - canonical incrementation (port++ instead of ++port)
  - check CPU and DSA ports before purging a VLAN
  - add Reviewed-by tags
====================
Signed-off-by: David S. Miller <davem@davemloft.net>

The mv88e6xxx driver currently tries to be smart and remove by itself a
VLAN entry from the VTU when the driven switch sees no user ports as
members of the VLAN.

This is bad in a multi-chip switch fabric, since a chip in between
others may have no bridge port members, but still needs to be aware of
the VID in order to correctly pass frames in the data path.

Now that the DSA core explicitly manages DSA and CPU ports, do not skip
them when checking remaining VLAN members.
Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Now that the DSA core adds the CPU and DSA ports itself to the new VLAN
entry, there is no need to include them as members of this VLAN when
initializing a new VTU entry.

As of now, initialize a new VTU entry with all ports excluded.
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

In a multi-chip switch fabric, it is currently the responsibility of the
driver to add the CPU or DSA (interconnecting chips together) ports as
members of a new VLAN entry. This makes the drivers more complicated.

We want the DSA drivers to be stupid and the DSA core being the one
responsible for caring about the abstracted switch logic and topology.

Make the DSA core program the CPU and DSA ports as part of the VLAN.

This makes all chips of the data path to be aware of VIDs spanning the
the whole fabric and thus, seamlessly add support for cross-chip VLAN.
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Now that the VLAN object is propagated to every switch chip of the
switch fabric, we can easily ensure that they all support the required
VLAN operations before modifying an entry on a single switch.

To achieve that, remove the condition skipping other target switches,
and add a bitmap of VLAN members, eventually containing the target port,
if we are programming the switch target.

This will allow us to easily add other VLAN members, such as the DSA or
CPU ports (to introduce cross-chip VLAN support) or the other port
members if we want to reduce hardware accesses later.
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Define the target port membership of the VLAN entry in
mv88e6xxx_port_vlan_add where ds is scoped.

Allow the DSA core to call later the port_vlan_add operation for CPU or
DSA ports, by using the Unmodified membership for these ports, as in the
current behavior.
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Merge tag 'rxrpc-rewrite-20170607-v2' of git://git.kernel.org/pub/scm/linux/kernel/git/dhowells/linux-fs

David Howells says:

====================
rxrpc: Tx length parameter

Here's a set of patches that allows someone initiating a client call with
AF_RXRPC to indicate upfront the total amount of data that will be
transmitted.  This will allow AF_RXRPC to encrypt directly from source
buffer to packet rather than having to copy into the buffer and only
encrypt when it's full (the encrypted portion of the packet starts with a
length and so we can't encrypt until we know what the length will be).

The three patches are:

 (1) Provide a means of finding out what control message types are actually
     supported.  EINVAL is reported if an unsupported cmsg type is seen, so
     we don't want to set the new cmsg unless we know it will be accepted.

 (2) Consolidate some stuff into a struct to reduce the parameter count on
     the function that parses the cmsg buffer.

 (3) Introduce the RXRPC_TX_LENGTH cmsg.  This can be provided on the first
     sendmsg() that contributes data to a client call request or a service
     call reply.  If provided, the user must provide exactly that amount of
     data or an error will be incurred.

Changes in version 2:

 (*) struct rxrpc_send_params::tx_total_len should be s64 not u64.  Thanks to
     Julia Lawall for reporting this.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>

Bjorn Andersson says:

====================
Missing QRTR features

The QMUX specification covers packet routing as well as service life cycle and
discovery. The current implementation of qrtr supports the prior part, but in
order to fully implement service management on-top a few more parts are needed.

The first patch in the series serves the purpose of reducing duplication in
patch two and three.

The second and third patch adds two qrtr-level notifications required by the
specification, in order to notify local and remote service controllers about
dying clients.

The last patch serves the purpose of notifying local clients about the presence
of a local service register, allowing them to register services as well as
querying for remote registered services.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>

As the higher level communication only deals with "services" the
a service directory is required to keep track of local and remote
services. In order for qrtr clients to be informed about when the
service directory implementation is available some event needs to be
passed to them.

Rather than introducing support for broadcasting such a message in-band
to all open local sockets we flag each socket with ENETRESET, as there
are no other expected operations that would benefit from having support
from locally broadcasting messages.

Cc: Courtney Cavin <ccavin@gmail.com>
Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

Per the QMUXv2 protocol specificiation a DEL_CLIENT message should be
broadcasted when an endpoint is disconnected.

The protocol specification does suggest that the router can keep track
of which nodes the endpoint has been communicating with to not wake up
sleeping remotes unecessarily, but implementation of this suggestion is
left for the future.

Cc: Courtney Cavin <ccavin@gmail.com>
Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

Per the QMUX protocol specification a terminating node can send a BYE
control message to signal that the link is going down, upon receiving
this all information about remote services should be discarded and local
clients should be notified.

In the event that the link was brought down abruptly the router is
supposed to act like a BYE message has arrived. As there is no harm in
receiving an extra BYE from the remote this patch implements the latter
by injecting a BYE when the link to the remote is unregistered.

The name service will receive the BYE and can implement the notification
to the local clients.

Cc: Courtney Cavin <ccavin@gmail.com>
Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

Extract the allocation and filling in the control message header fields
to a separate function in order to reuse this in subsequent patches.

Cc: Courtney Cavin <ccavin@gmail.com>
Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

Remove unnecessary variable assignments.

Addresses-Coverity-ID: 1226917
Signed-off-by: Gustavo A. R. Silva <garsilva@embeddedor.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

DRAM supply shortage and poor memory pressure tracking in TCP
stack makes any change in SO_SNDBUF/SO_RCVBUF (or equivalent autotuning
limits) and tcp_mem[] quite hazardous.

TCPMemoryPressures SNMP counter is an indication of tcp_mem sysctl
limits being hit, but only tracking number of transitions.

If TCP stack behavior under stress was perfect :
1) It would maintain memory usage close to the limit.
2) Memory pressure state would be entered for short times.

We certainly prefer 100 events lasting 10ms compared to one event
lasting 200 seconds.

This patch adds a new SNMP counter tracking cumulative duration of
memory pressure events, given in ms units.

$ cat /proc/sys/net/ipv4/tcp_mem
3088    4117    6176
$ grep TCP /proc/net/sockstat
TCP: inuse 180 orphan 0 tw 2 alloc 234 mem 4140
$ nstat -n ; sleep 10 ; nstat |grep Pressure
TcpExtTCPMemoryPressures        1700
TcpExtTCPMemoryPressuresChrono  5209

v2: Used EXPORT_SYMBOL_GPL() instead of EXPORT_SYMBOL() as David
instructed.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Eric Dumazet says:

====================
tcp: Namespaceify 3 sysctls

Move tcp_sack, tcp_window_scaling and tcp_timestamps
sysctls to network namespaces.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

We want to move some TCP sysctls to net namespaces in the future.

tcp_window_scaling, tcp_sack and tcp_timestamps being fetched
from tcp_parse_options(), we need to pass an extra parameter.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

We need to push the chain index down to the drivers, so they have the
information to which chain the rule belongs. For now, no driver supports
multichain offload, so only chain 0 is supported. This is needed to
prevent chain squashes during offload for now. Later this will be used
to implement multichain offload.
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Tariq Toukan says:

====================
mlx4 drivers: version update

This patchset contains version updates for the MLX4 drivers:
Core, EN, and IB.

Just like we've done in mlx5, we modify the outdated driver
version (reported in ethtool for example).
This better reflects the current driver state, and removes the
redundant date string.
We are not going to change this frequently or even use it.

I include the IB patch in this series as it has similar subject
and content.
It does not cause any kind of conflict with Doug's tree.
The rdma mailing list is CCed.
Please let me know if I need to submit this differently.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>

Remove date and bump version for mlx4_ib driver.
Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Remove date and bump version for mlx4_en driver.
Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Remove date and bump version for mlx4_core driver.
Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

The mv88e6161 and mv88e6123 are capable of using EDSA tags when
passing frames from the host to the switch and back.
Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Reviewed-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

When stopping the vxlan interface we detach it from the socket.
Use RCU_INIT_POINTER() and not rcu_assign_pointer() to do so.
Suggested-by: Stephen Hemminger <stephen@networkplumber.org>
Signed-off-by: Mark Bloch <markb@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

the adapter consumes two tids for every ipv6 offload
connection be it active or passive, calculate tid usage
count accordingly.

Also change the signatures of relevant functions to get
the address family.
Signed-off-by: Rizwan Ansari <rizwana@chelsio.com>
Signed-off-by: Varun Prakash <varun@chelsio.com>
Signed-off-by: Ganesh Goudar <ganeshgr@chelsio.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Jakub Kicinski says:

====================
nfp: ctrl vNIC

This series adds the ability to use one vNIC as a control channel
for passing messages to and from the application firmware.  The
implementation restructures the existing netdev vNIC code to be able
to deal with nfp_nets with netdev pointer set to NULL.  Control vNICs
are not visible to userspace (other than for dumping ring state), and
since they don't have netdevs we use a tasklet for RX and simple skb
list for TX queuing.

Due to special status of the control vNIC we have to reshuffle the
init code a bit to make sure control vNIC will be fully brought up
(and therefore communication with app FW can happen) before any netdev
or port is visible to user space.

FW will designate which vNIC is supposed to be used as control one
by setting _pf%u_net_ctrl_bar symbol.  Some FWs depend on metadata
being prepended to control message, some prefer to look at queue ID
to decide that something is a control message.  Our implementation
can cater to both.

First two users of this code will be eBPF maps and flower offloads.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>

NFD ABI 0.5 is equivalent to NFD ABI 3.0 but requires that the
driver checks the APP id symbol and makes sure it can support
given app.  Most advanced apps will likely require control vNIC
(ability to exchange control messages between the driver and
app FW).  Detailed app version checking and capability exchange
is left to app-specific code.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

When driver encounters an nfp_app which has a control message handler
defined, allocate a control vNIC.  This control channel will be used
to exchange data with the application FW such as flow table programming,
statistics and global datapath control.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Thus far the code assumed all vNICs will request similar number of IRQs.
This will be no longer true with control vNICs (where 1 IRQ will suffice).
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

We want to be able to create a special vNIC for control messages.
This vNIC should be created before any netdev is registered to allow
nfp_app logic to exchange messages with the FW app before any netdev
is visible to user space.  Unfortunately we can't enable IRQs until
we know how many vNICs we will need to spawn.

Divide the function which spawns netdevs for vNICs into three parts:
 - vNIC/memory allocation;
 - IRQ allocation;
 - netdev init and register.

This will help us insert the initialization of the control channel
after IRQ allocation but before netdev init and register.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Reading fw version from the BAR is trivial.  Don't pass it around
through layers of init functions, simply read it again where needed.

This commit has the side effect of each vNIC having the exact NFD
version from its own control memory, rather than all data vNICs
assuming the version of the first one.  This should not result in
user-visible changes, though.  Capabilities of data vNICs of trival
apps are identical.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>