cascardo/linux.git
8 years agomlxsw: Introduce Mellanox SwitchX-2 ASIC support
Jiri Pirko [Wed, 29 Jul 2015 21:33:49 +0000 (23:33 +0200)]
mlxsw: Introduce Mellanox SwitchX-2 ASIC support

Benefit from the previously introduced Mellanox Switch infrastructure and
add driver for SwitchX-2 ASIC. Note that this driver is very simple now.
It implements bare minimum for getting device to work on slow-path.
Fast-path offload functionality is going to be added soon.

Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: Elad Raz <eladr@mellanox.com>
Reviewed-by: Scott Feldman <sfeldma@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agomlxsw: Add interface to access registers and process events
Ido Schimmel [Wed, 29 Jul 2015 21:33:48 +0000 (23:33 +0200)]
mlxsw: Add interface to access registers and process events

Ethernet Management Datagrams (EMADs) are Ethernet packets sent between
the host and the device in order to configure the available device registers.
Another use case is notifications sent from the device to the host,
letting it know about certain events, such as port up / down.

Add the ability to construct EMADs with provisions to construct and
parse the registers' payloads. Implement EMAD transaction layer
which is responsible for the reliable transmission of EMADs. Also, add
an infrastructure used by the switch driver to register for particular
events generated by the device.

Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Elad Raz <eladr@mellanox.com>
Reviewed-by: Scott Feldman <sfeldma@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agomlxsw: Add PCI bus implementation
Jiri Pirko [Wed, 29 Jul 2015 21:33:47 +0000 (23:33 +0200)]
mlxsw: Add PCI bus implementation

Add PCI bus implementation for Mellanox Technologies Switch ASICs. This
includes firmware initialization, async queues manipulation and command
interface implementation.

Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: Elad Raz <eladr@mellanox.com>
Reviewed-by: Scott Feldman <sfeldma@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agomlxsw: Introduce Mellanox switch driver core
Jiri Pirko [Wed, 29 Jul 2015 21:33:46 +0000 (23:33 +0200)]
mlxsw: Introduce Mellanox switch driver core

Add core components of Mellanox switch driver infrastructure.
Core infrastructure is designed so that it can be used by multiple
bus drivers (PCI now, I2C and SGMII are planned to be implemented
in the future). Multiple switch kind drivers can be registered as well.
This core serves as a glue between buses and drivers.

Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: Elad Raz <eladr@mellanox.com>
Reviewed-by: Scott Feldman <sfeldma@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agobe2net: Support vxlan offload stats in the driver
Sriharsha Basavapatna [Wed, 29 Jul 2015 14:05:32 +0000 (19:35 +0530)]
be2net: Support vxlan offload stats in the driver

This patch adds vxlan offload specific counters to ethtool stats. We
provide tx/rx queue counters to show the number of vxlan offload pkts
sent/received.

Signed-off-by: Sriharsha Basavapatna <sriharsha.basavapatna@emulex.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agobnx2x: add vlan filtering offload
Yuval Mintz [Wed, 29 Jul 2015 12:52:46 +0000 (15:52 +0300)]
bnx2x: add vlan filtering offload

Current driver always uses vlan-promisc mode, i.e., it receives both
tagged and untagged traffic and lets the network stack drop packets
tagged with unrequested vlan tags.

This patch implements vlan-filtering offload in the driver -
Unless explicitly configured to promisc mode, only untagged packets or
packets tagged with requested vlans would reach the Rx flow.

Signed-off-by: Yuval Mintz <Yuval.Mintz@qlogic.com>
Signed-off-by: Ariel Elior <Ariel.Elior@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agoMerge branch 'mlx5e-next'
David S. Miller [Thu, 30 Jul 2015 06:04:47 +0000 (23:04 -0700)]
Merge branch 'mlx5e-next'

Amir Vadai says:

====================
net/mlx5e: Driver update 29-Jul-2015

This patchset contain bug fixes and code cleaning patches to the ConnectX-4
Ethernet driver.

Patchset was applied and tested over commit 8c1a91f ("Merge branch
'mlx4-802.1ad-accel'")
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agonet/mlx5e: Remove the mlx5e_update_priv_params() function
Achiad Shochat [Wed, 29 Jul 2015 12:05:46 +0000 (15:05 +0300)]
net/mlx5e: Remove the mlx5e_update_priv_params() function

It was used to update netdev priv parameters that require stopping
and re-opening the device in a generic way - it got the new
parameters and did: ndo_stop(), copy new parameters into current
parameters, ndo_open().

We chose to remove it for two reasons:
1) It requires additional instance of struct mlx5e_params on the
   stack and looking forward we expect this struct to grow.
2) Sometimes we want to do additional operations (besides
   just updating the priv parameters) while the netdev is stopped.
   For example, updating netdev->mtu @mlx5e_change_mtu() should
   be done while the netdev is stopped (done in this commit).

Signed-off-by: Achiad Shochat <achiad@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agonet/mlx5e: Introduce create/destroy RSS indir table access functions
Achiad Shochat [Wed, 29 Jul 2015 12:05:45 +0000 (15:05 +0300)]
net/mlx5e: Introduce create/destroy RSS indir table access functions

Introduce access functions to create/destroy RSS indrection table
and use it in the Ethernet driver.

Signed-off-by: Achiad Shochat <achiad@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agonet/mlx5e: Do not use netdev_err() before the netdev is registered
Achiad Shochat [Wed, 29 Jul 2015 12:05:44 +0000 (15:05 +0300)]
net/mlx5e: Do not use netdev_err() before the netdev is registered

Since it is un-named at this time.

Signed-off-by: Achiad Shochat <achiad@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agonet/mlx5e: Avoid redundant de-reference
Achiad Shochat [Wed, 29 Jul 2015 12:05:43 +0000 (15:05 +0300)]
net/mlx5e: Avoid redundant de-reference

Use the already defined rq pointer directly.

Signed-off-by: Achiad Shochat <achiad@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agonet/mlx5e: Remove redundant assignment of sq->user_index
Achiad Shochat [Wed, 29 Jul 2015 12:05:42 +0000 (15:05 +0300)]
net/mlx5e: Remove redundant assignment of sq->user_index

It is not needed by the mlx5 Eth driver since it has a CQ per RQ/SQ.

Signed-off-by: Achiad Shochat <achiad@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agonet/mlx5e: Remove redundant field mlx5e_priv->num_tc
Achiad Shochat [Wed, 29 Jul 2015 12:05:41 +0000 (15:05 +0300)]
net/mlx5e: Remove redundant field mlx5e_priv->num_tc

This field already exists under the mlx5e_params struct

Signed-off-by: Achiad Shochat <achiad@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agonet/mlx5e: Use hard-coded 4K page size for RQ/SQ/CQ
Achiad Shochat [Wed, 29 Jul 2015 12:05:40 +0000 (15:05 +0300)]
net/mlx5e: Use hard-coded 4K page size for RQ/SQ/CQ

The page size of the device's RQ/SQ/CQ objects is defined in 4K
units regardless of the system pages size.
Thus using the Linux's PAGE_SHIFT macro yields wrong device
configuration in systems where PAGE_SHIFT!=12.

Signed-off-by: Achiad Shochat <achiad@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agonet/mlx5_core: Check the return value of mlx5_command_exec()
Haggai Abramonvsky [Wed, 29 Jul 2015 12:05:39 +0000 (15:05 +0300)]
net/mlx5_core: Check the return value of mlx5_command_exec()

mlx5_cmd_exec() might fail - need to check return value.

Signed-off-by: Haggai Abramovsky <hagaya@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agoopenvswitch: Re-add CONFIG_OPENVSWITCH_VXLAN
Thomas Graf [Wed, 29 Jul 2015 11:52:06 +0000 (13:52 +0200)]
openvswitch: Re-add CONFIG_OPENVSWITCH_VXLAN

This readds the config option CONFIG_OPENVSWITCH_VXLAN to avoid a
hard dependency of OVS on VXLAN. It moves the VXLAN config compat
code to vport-vxlan.c and allows compliation as a module.

Fixes: 614732eaa12d ("openvswitch: Use regular VXLAN net_device device")
Fixes: 2661371ace96 ("openvswitch: fix compilation when vxlan is a module")
Cc: Pravin B Shelar <pshelar@nicira.com>
Cc: Nicolas Dichtel <nicolas.dichtel@6wind.com>
Signed-off-by: Thomas Graf <tgraf@suug.ch>
Acked-by: Pravin B Shelar <pshelar@nicira.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agonet: pktgen: Remove unused 'allocated_skbs' field
Bogdan Hamciuc [Wed, 29 Jul 2015 09:47:16 +0000 (12:47 +0300)]
net: pktgen: Remove unused 'allocated_skbs' field

Field pktgen_dev.allocated_skbs had been written to, but never read
from. The number of allocated skbs can be deduced anyway, from the total
number of sent packets and the 'clone_skb' param.

Signed-off-by: Bogdan Hamciuc <bogdan.hamciuc@freescale.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agonet: pktgen: Observe needed_headroom of the device
Bogdan Hamciuc [Wed, 29 Jul 2015 09:45:58 +0000 (12:45 +0300)]
net: pktgen: Observe needed_headroom of the device

Allocate enough space so as not to force the outgoing net device to do
skb_realloc_headroom().

Signed-off-by: Bogdan Hamciuc <bogdan.hamciuc@freescale.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agolwtunnel: Make lwtun_encaps[] static
Thomas Graf [Wed, 29 Jul 2015 07:45:40 +0000 (09:45 +0200)]
lwtunnel: Make lwtun_encaps[] static

Any external user should use the registration API instead of
accessing this directly.

Cc: Roopa Prabhu <roopa@cumulusnetworks.com>
Signed-off-by: Thomas Graf <tgraf@suug.ch>
Acked-by: Roopa Prabhu <roopa@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agogianfar: Fix warnings when built on 64-bit
Scott Wood [Wed, 29 Jul 2015 13:13:06 +0000 (16:13 +0300)]
gianfar: Fix warnings when built on 64-bit

As part of defconfig consolidation using fragments, we'd like to be
able to have the same drivers enabled on 32-bit and 64-bit.  Gianfar
happens to only exist on 32-bit systems, and when building the
resulting 64-bit kernel warnings were produced.

A couple of the warnings are trivial, but the rfbptr code has deeper
issues.  It uses the virtual address as the DMA address, which again,
happens to work in the environments where this driver is currently
used, but is not the right thing to do.

Fixes: 45b679c9a3cc ("gianfar: Implement PAUSE frame generation
support")
Signed-off-by: Scott Wood <scottwood@freescale.com>
Signed-off-by: Claudiu Manoil <claudiu.manoil@freescale.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agoMerge branch 'sk_txhash'
David S. Miller [Thu, 30 Jul 2015 05:44:05 +0000 (22:44 -0700)]
Merge branch 'sk_txhash'

Tom Herbert says:

====================
net: Initialize sk_hash to random value and reset for failing cnxs

This patch set implements a common function to simply set sk_txhash to
a random number instead of going through the trouble to call flow
dissector. From dst_negative_advice we now reset the sk_txhash in hopes
of finding a better ECMP path through the network. Changing sk_txhash
affects:
  - IPv6 flow label and UDP source port which affect ECMP in the network
  - Local ECMP route selection (pending changes to use sk_txhash)
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agonet: Recompute sk_txhash on negative routing advice
Tom Herbert [Tue, 28 Jul 2015 23:02:06 +0000 (16:02 -0700)]
net: Recompute sk_txhash on negative routing advice

When a connection is failing a transport protocol calls
dst_negative_advice to try to get a better route. This patch includes
changing the sk_txhash in that function. This provides a rudimentary
method to try to find a different path in the network since sk_txhash
affects ECMP on the local host and through the network (via flow labels
or UDP source port in encapsulation).

Signed-off-by: Tom Herbert <tom@herbertland.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agonet: Set sk_txhash from a random number
Tom Herbert [Tue, 28 Jul 2015 23:02:05 +0000 (16:02 -0700)]
net: Set sk_txhash from a random number

This patch creates sk_set_txhash and eliminates protocol specific
inet_set_txhash and ip6_set_txhash. sk_set_txhash simply sets a
random number instead of performing flow dissection. sk_set_txash
is also allowed to be called multiple times for the same socket,
we'll need this when redoing the hash for negative routing advice.

Signed-off-by: Tom Herbert <tom@herbertland.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agoMerge branch 'thunderx_octeon_mdio'
David S. Miller [Thu, 30 Jul 2015 00:17:44 +0000 (17:17 -0700)]
Merge branch 'thunderx_octeon_mdio'

Radha Mohan Chintakuntla says:

====================
Add MDIO support to ThunderX NIC driver

This patch series adds MDIO support to ThunderX NIC driver by making use
of existing mdio-octeon driver. In the process modified the mdio-octeon
driver to work on both Octeon and ThunderX platforms.

* From v1:
  - Removed default selection in Kconfig for MDIO_OCTEON
  - Replace uint64 with u64 as suggested by David Daney
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agonet: thunderx: Select CONFIG_MDIO_OCTEON for ThunderX NIC
Radha Mohan Chintakuntla [Tue, 28 Jul 2015 22:12:13 +0000 (15:12 -0700)]
net: thunderx: Select CONFIG_MDIO_OCTEON for ThunderX NIC

The CONFIG_MDIO_OCTEON is required so that the ThunderX NIC driver can
talk to the PHY drivers.

Signed-off-by: Radha Mohan Chintakuntla <rchintakuntla@cavium.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agonet: mdio-octeon: Fix octeon_mdiobus_probe function for return values
Radha Mohan Chintakuntla [Tue, 28 Jul 2015 22:12:12 +0000 (15:12 -0700)]
net: mdio-octeon: Fix octeon_mdiobus_probe function for return values

This patch fixes a possible crash in the octeon_mdiobus_probe function
if the return values are not handled properly.

Signed-off-by: Radha Mohan Chintakuntla <rchintakuntla@cavium.com>
Signed-off-by: Tomasz Nowicki <tomasz.nowicki@linaro.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agonet: mdio-octeon: Modify driver to work on both ThunderX and Octeon
Radha Mohan Chintakuntla [Tue, 28 Jul 2015 22:12:11 +0000 (15:12 -0700)]
net: mdio-octeon: Modify driver to work on both ThunderX and Octeon

This patch modifies the mdio-octeon driver to work on both ThunderX and
Octeon SoCs from Cavium Inc.

Signed-off-by: Sunil Goutham <sgoutham@cavium.com>
Signed-off-by: Radha Mohan Chintakuntla <rchintakuntla@cavium.com>
Signed-off-by: David Daney <david.daney@cavium.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agonet: netcp: Fixes efuse mac addr swap on k2e and k2l
WingMan Kwok [Tue, 28 Jul 2015 20:01:11 +0000 (16:01 -0400)]
net: netcp: Fixes efuse mac addr swap on k2e and k2l

On some of the K2E and K2L platforms, the two DWORDs in
efuse occupied by the pre-programmed mac address for
slave port 1 are swapped.  To workaround this issue,
this patch adds a new define NETCP_EFUSE_ADDR_SWAP (2)
which signifies the occurrence of such swapping so that
the driver can take proper action.  The flag can be
enabled in the corresponding netcp interface dts binding
as efuse-mac = <2>  under the corresponding netcp
interface node.

Signed-off-by: WingMan Kwok <w-kwok2@ti.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agotipc: fix bug in broadcast synch message create function
Jon Maloy [Wed, 29 Jul 2015 22:28:01 +0000 (18:28 -0400)]
tipc: fix bug in broadcast synch message create function

In commit d999297c3dbbe7fdd832f7fa4ec84301e170b3e6
("tipc: reduce locking scope during packet reception") we introduced
a new function tipc_build_bcast_sync_msg(), which carries initial
synchronization data between two nodes at first contact and at
re-contact. In this function, we missed to add synchronization data,
with the effect that the broadcast link endpoints will fail to
synchronize correctly at re-contact between a running and a restarted
node. All other cases work as intended.

With this commit, we fix this bug.

Signed-off-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agopacket: remove handling of tx_ring from prb_shutdown_retire_blk_timer()
Tobias Klauser [Tue, 28 Jul 2015 12:21:26 +0000 (14:21 +0200)]
packet: remove handling of tx_ring from prb_shutdown_retire_blk_timer()

Follow e8e85cc5eb57 ("packet: remove handling of tx_ring") and remove
the tx_ring parameter from prb_shutdown_retire_blk_timer() as it is only
called with tx_ring = 0.

Signed-off-by: Tobias Klauser <tklauser@distanz.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agoMerge branch 's390-bpf-push-pop'
David S. Miller [Wed, 29 Jul 2015 21:59:59 +0000 (14:59 -0700)]
Merge branch 's390-bpf-push-pop'

Michael Holzheu says:

====================
s390/bpf: recache skb->data/hlen for skb_vlan_push/pop

Here the s390 backend for Alexei's patch 4e10df9a60d9 ("bpf: introduce
bpf_skb_vlan_push/pop() helpers") plus two bugfixes and two minor
improvements.

The first patch "s390/bpf: clear correct BPF accumulator register" will
also go upstream via Martin's "fixes" branch.

* v2: Integrated suggestions from Joe Perches
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agos390/bpf: recache skb->data/hlen for skb_vlan_push/pop
Michael Holzheu [Wed, 29 Jul 2015 19:15:18 +0000 (21:15 +0200)]
s390/bpf: recache skb->data/hlen for skb_vlan_push/pop

Allow eBPF programs attached to TC qdiscs call skb_vlan_push/pop
via helper functions. These functions may change skb->data/hlen.
This data is cached by s390 JIT to improve performance of ld_abs/ld_ind
instructions. Therefore after a change we have to reload the data.

In case of usage of skb_vlan_push/pop, in the prologue we store
the SKB pointer on the stack and restore it after BPF_JMP_CALL
to skb_vlan_push/pop.

Signed-off-by: Michael Holzheu <holzheu@linux.vnet.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agos390/bpf: Only clear A and X for converted BPF programs
Michael Holzheu [Wed, 29 Jul 2015 19:15:17 +0000 (21:15 +0200)]
s390/bpf: Only clear A and X for converted BPF programs

Only classic BPF programs that have been converted to eBPF need to clear
the A and X registers. We can check for converted programs with:

  bpf_prog->type == BPF_PROG_TYPE_UNSPEC

So add the check and skip initialization for real eBPF programs.

Signed-off-by: Michael Holzheu <holzheu@linux.vnet.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agos390/bpf: increase BPF_SIZE_MAX
Michael Holzheu [Wed, 29 Jul 2015 19:15:16 +0000 (21:15 +0200)]
s390/bpf: increase BPF_SIZE_MAX

Currently we have the restriction that jitted BPF programs can
have a maximum size of one page. The reason is that we use short
displacements for the literal pool.

The 20 bit displacements are available since z990 and BPF requires
z196 as minimum. Therefore we can remove this restriction and use
everywhere 20 bit signed long displacements.

Acked-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Signed-off-by: Michael Holzheu <holzheu@linux.vnet.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agos390/bpf: Fix multiple macro expansions
Michael Holzheu [Wed, 29 Jul 2015 19:15:15 +0000 (21:15 +0200)]
s390/bpf: Fix multiple macro expansions

The EMIT6_DISP_LH macro passes the "disp" parameter to the _EMIT6_DISP_LH
macro. The _EMIT6_DISP_LH macro uses the "disp" parameter twice:

 unsigned int __disp_h = ((u32)disp) & 0xff000;
 unsigned int __disp_l = ((u32)disp) & 0x00fff;

The EMIT6_DISP_LH is used several times with EMIT_CONST_U64() as "disp"
parameter. Therefore always two constants are created per usage of
EMIT6_DISP_LH.

Fix this and add variable "_disp" to avoid multiple expansions.

* v2: Move "_disp" to _EMIT6_DISP_LH as suggested by Joe Perches

Fixes: 054623105728 ("s390/bpf: Add s390x eBPF JIT compiler backend")
Signed-off-by: Michael Holzheu <holzheu@linux.vnet.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agos390/bpf: clear correct BPF accumulator register
Michael Holzheu [Wed, 29 Jul 2015 19:15:14 +0000 (21:15 +0200)]
s390/bpf: clear correct BPF accumulator register

Currently we assumed the following BPF to eBPF register mapping:

 - BPF_REG_A -> BPF_REG_7
 - BPF_REG_X -> BPF_REG_8

Unfortunately this mapping is wrong. The correct mapping is:

 - BPF_REG_A -> BPF_REG_0
 - BPF_REG_X -> BPF_REG_7

So clear the correct registers and use the BPF_REG_A and BPF_REG_X
macros instead of BPF_REG_0/7.

Fixes: 054623105728 ("s390/bpf: Add s390x eBPF JIT compiler backend")
Cc: stable@vger.kernel.org # 4.0+
Signed-off-by: Michael Holzheu <holzheu@linux.vnet.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agonet/mlx4_en: Hardware accelerated 802.1ad works only on the first port
Amir Vadai [Tue, 28 Jul 2015 10:19:19 +0000 (13:19 +0300)]
net/mlx4_en: Hardware accelerated 802.1ad works only on the first port

Fix mistakenly used, hard coded, port number in get_phv_bit()

Fixes: 77fc29c ("net/mlx4_core: Preparations for 802.1ad VLAN support")
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agoMerge branch 'stmmac-cleanup'
David S. Miller [Wed, 29 Jul 2015 07:13:25 +0000 (00:13 -0700)]
Merge branch 'stmmac-cleanup'

Joachim Eastwood says:

====================
stmmac clean up for 4.3 part2

This patch set creates proper probe functions in the remaining
dwmac-* drivers, lets the drivers manage their own of match data
structure, and cleans up some of the drivers a bit. Note that
should be no functional change in any of the drivers.

First patch from Fengguang fixes a Coccinelle that was found after
the previous patch set when into next.

Probe functions are best placed just above the platform_driver
struct since it usually needs to call other functions in the
driver but this requires some code movement in the drivers. To
keep the conversion itself easier to review the code movement is
done in separate patches. This creates some extra churn but I
think it's worth it in this case.

As only a couple of drivers actually needs to use of match data
this pushed down from stmmac_platform and into the drivers. This
also makes it possible for drivers to have their own match data
structure. dwmac-rk benefits most from this.

After each driver has it's own probe function and manages it's
own match data the setup/free hooks and common match data struct
are removed.

Sorry for the slightly largish patch set this time around.

Since I can only test the core changes on my platform and not the
specific dwmac-drivers I greatly appreciate if people could test
this on their respective platforms. I did spend quite a bit of
time visually checking all these patches, but nothing beats real
hardware testing. Thanks!

Patch set based on net-next can also be picked up from here:
https://github.com/manabian/linux-lpc.git stmmac_4.3_part2

For those who want to have a look at all the patches for
v4.3 a branch based on Linus master can be found here:
https://github.com/manabian/linux-lpc.git stmmac_4.2

Previous parts can be found here:
http://www.spinics.net/lists/netdev/msg328997.html
http://www.spinics.net/lists/netdev/msg329932.html
http://www.spinics.net/lists/netdev/msg336097.html
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agostmmac: dwmac-sti: refactor the init glue callbacks
Joachim Eastwood [Tue, 28 Jul 2015 22:09:04 +0000 (00:09 +0200)]
stmmac: dwmac-sti: refactor the init glue callbacks

Remove the two platform specific init callbacks and make
them use a common one by creating a function member in
the internal data structure. This allow us to remove the
layer of indirection and simplify the code a bit.

Signed-off-by: Joachim Eastwood <manabian@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agostmmac: remove setup/free glue callbacks
Joachim Eastwood [Tue, 28 Jul 2015 22:09:03 +0000 (00:09 +0200)]
stmmac: remove setup/free glue callbacks

As all dwmac-* drivers have been converted to have a proper probe
function the setup callback can now be removed. Also remove the
free callback that wasn't used by any driver.

New dwmac-* drivers should implement standard probe and remove
functions to preform any needed setup and teardown.

Signed-off-by: Joachim Eastwood <manabian@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agostmmac: remove unused stmmac_of_data struct
Joachim Eastwood [Tue, 28 Jul 2015 22:09:02 +0000 (00:09 +0200)]
stmmac: remove unused stmmac_of_data struct

As dwmac-* drivers that need OF match have been converted
to use their own internal OF match data structure this can
now be removed.

Signed-off-by: Joachim Eastwood <manabian@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agostmmac: dwmac-rk: use rk_gmac_ops as of match data
Joachim Eastwood [Tue, 28 Jul 2015 22:09:01 +0000 (00:09 +0200)]
stmmac: dwmac-rk: use rk_gmac_ops as of match data

Remove the setup glue callback and use rk_gmac_ops as OF match
data so it can used directly when calling rk_gmac_setup.

Signed-off-by: Joachim Eastwood <manabian@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agostmmac: dwmac-rk: make rk_gmac_ops structs static const
Joachim Eastwood [Tue, 28 Jul 2015 22:09:00 +0000 (00:09 +0200)]
stmmac: dwmac-rk: make rk_gmac_ops structs static const

Mark the rk_gmac_ops structures as static const as they should be.

Signed-off-by: Joachim Eastwood <manabian@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agostmmac: dwmac-sti: use custom of match structure
Joachim Eastwood [Tue, 28 Jul 2015 22:08:59 +0000 (00:08 +0200)]
stmmac: dwmac-sti: use custom of match structure

Create a new private structure for OF match data in the
dwmac-sti driver. This enables us to eventually drop the
common OF match data structure which contains a lot of
unused fields.

Signed-off-by: Joachim Eastwood <manabian@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agostmmac: let dwmac-* drivers handle their own match data
Joachim Eastwood [Tue, 28 Jul 2015 22:08:58 +0000 (00:08 +0200)]
stmmac: let dwmac-* drivers handle their own match data

Since only a few of the dwmac-* drivers actually need to use
the OF match move handling into the dwmac-* drivers that need
it. This will also allow dwmac-* drivers to use their own
custom match data structure.

Signed-off-by: Joachim Eastwood <manabian@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agostmmac: move stmmac_pltfr_probe into dwmac-generic
Joachim Eastwood [Tue, 28 Jul 2015 22:08:57 +0000 (00:08 +0200)]
stmmac: move stmmac_pltfr_probe into dwmac-generic

As all dwmac-* drivers now have their own probe function move
the common one into dwmac-generic driver and drop the EXPORT.

Signed-off-by: Joachim Eastwood <manabian@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agostmmac: dwmac-rk: create a new probe function
Joachim Eastwood [Tue, 28 Jul 2015 22:08:56 +0000 (00:08 +0200)]
stmmac: dwmac-rk: create a new probe function

Create a new probe functions that call the necessary setup
functions. This is done in preparation for a code refactor
in this driver and dropping the common probe function in
stmmac_platform.c.

Signed-off-by: Joachim Eastwood <manabian@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agostmmac: dwmac-sti: turn setup callback into a probe function
Joachim Eastwood [Tue, 28 Jul 2015 22:08:55 +0000 (00:08 +0200)]
stmmac: dwmac-sti: turn setup callback into a probe function

By using a few functions from stmmac_platform a proper probe
function can be created from the setup glue callback. This
makes it look more like a standard driver and prepares the
driver for further clean ups.

Signed-off-by: Joachim Eastwood <manabian@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agostmmac: dwmac-sunxi: turn setup callback into a probe function
Joachim Eastwood [Tue, 28 Jul 2015 22:08:54 +0000 (00:08 +0200)]
stmmac: dwmac-sunxi: turn setup callback into a probe function

By using a few functions from stmmac_platform a proper probe
function can be created from the setup glue callback. This
makes it look more like a standard driver and the OF match
data can also be dropped.

Signed-off-by: Joachim Eastwood <manabian@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agostmmac: dwmac-sunxi: move sun7i_gmac_setup function
Joachim Eastwood [Tue, 28 Jul 2015 22:08:53 +0000 (00:08 +0200)]
stmmac: dwmac-sunxi: move sun7i_gmac_setup function

Move sun7i_gmac_setup in preparation for turning it into
a proper probe function.

Signed-off-by: Joachim Eastwood <manabian@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agostmmac: dwmac-socfpga: turn setup callback into a probe function
Joachim Eastwood [Tue, 28 Jul 2015 22:08:52 +0000 (00:08 +0200)]
stmmac: dwmac-socfpga: turn setup callback into a probe function

By using a few functions from stmmac_platform a proper probe
function can be created from the setup glue callback. This
makes it look more like a standard driver and the OF match
data can also be dropped.

Signed-off-by: Joachim Eastwood <manabian@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agostmmac: dwmac-socfpga: move socfpga_dwmac_probe function
Joachim Eastwood [Tue, 28 Jul 2015 22:08:51 +0000 (00:08 +0200)]
stmmac: dwmac-socfpga: move socfpga_dwmac_probe function

Move socfpga_dwmac_probe in preparation for turning it into
a proper probe function.

Signed-off-by: Joachim Eastwood <manabian@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agostmmac: dwmac-ipq806x: turn setup callback into a probe function
Joachim Eastwood [Tue, 28 Jul 2015 22:08:50 +0000 (00:08 +0200)]
stmmac: dwmac-ipq806x: turn setup callback into a probe function

By using a few functions from stmmac_platform a proper probe
function can be created from the setup glue callback. This
makes it look more like a standard driver and the OF match
data can also be dropped.

Signed-off-by: Joachim Eastwood <manabian@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agostmmac: dwmac-ipq806x: move ipq806x_gmac_fix_mac_speed function
Joachim Eastwood [Tue, 28 Jul 2015 22:08:49 +0000 (00:08 +0200)]
stmmac: dwmac-ipq806x: move ipq806x_gmac_fix_mac_speed function

Move ipq806x_gmac_fix_mac_speed in preparation for turning
the setup glue callback in a proper probe function.

Signed-off-by: Joachim Eastwood <manabian@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agostmmac: fix ptr_ret.cocci warning
Fengguang Wu [Tue, 28 Jul 2015 22:08:48 +0000 (00:08 +0200)]
stmmac: fix ptr_ret.cocci warning

drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c:304:1-3: WARNING: PTR_ERR_OR_ZERO can be used

Use PTR_ERR_OR_ZERO rather than if(IS_ERR(...)) + PTR_ERR

Generated by: scripts/coccinelle/api/ptr_ret.cocci

Signed-off-by: Fengguang Wu <fengguang.wu@intel.com>
[je: rebase and insert newline before return]
Signed-off-by: Joachim Eastwood <manabian@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agoMerge branch 'dwc_ethernet'
David S. Miller [Wed, 29 Jul 2015 06:55:01 +0000 (23:55 -0700)]
Merge branch 'dwc_ethernet'

Lars Persson says:

====================
dwc_eth_qos: Add support for Synopsys DWC Ethernet QoS

This is a driver supporting version 4.10a of the Synopsys DWC Ethernet QoS
gigabit ethernet controller. The IP has changed significantly compared to the
dwmac1000 so a separate driver is justified.

The IP is highly configurable at synthesis time. This driver has been
developed for a subset of the total available feature set. Currently
it supports:
* TSO
* Checksum offload for RX and TX.
* Energy efficient ethernet.
* GMII phy interface.
* The statistics module.
* Single RX and TX queue.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agodwc_eth_qos: Add maintainer info
Lars Persson [Tue, 28 Jul 2015 10:01:50 +0000 (12:01 +0200)]
dwc_eth_qos: Add maintainer info

Add maintainer information for the Synopsys DWC Ethernet QOS driver.

Signed-off-by: Lars Persson <larper@axis.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agodwc_eth_qos: Add the synopsys folder to the build system.
Lars Persson [Tue, 28 Jul 2015 10:01:49 +0000 (12:01 +0200)]
dwc_eth_qos: Add the synopsys folder to the build system.

Signed-off-by: Lars Persson <larper@axis.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agodwc_eth_qos: Add support for Synopsys DWC Ethernet QoS
Lars Persson [Tue, 28 Jul 2015 10:01:48 +0000 (12:01 +0200)]
dwc_eth_qos: Add support for Synopsys DWC Ethernet QoS

This patch adds a platform driver for the new generation of the
gigabit ethernet IP from Synopsys. It is developed for version 4.10a
of the IP core.

Signed-off-by: Lars Persson <larper@axis.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agodwc_eth_qos: Add Synopsys DWC Ethernet QoS bindings
Lars Persson [Tue, 28 Jul 2015 10:01:47 +0000 (12:01 +0200)]
dwc_eth_qos: Add Synopsys DWC Ethernet QoS bindings

Add device tree binding documentation for the Synopsys DWC Ethernet
QoS driver supporting revision 4.10a of the hardware IP.

Signed-off-by: Lars Persson <larper@axis.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agoMerge branch 'mlx4-802.1ad-accel'
David S. Miller [Mon, 27 Jul 2015 22:00:37 +0000 (15:00 -0700)]
Merge branch 'mlx4-802.1ad-accel'

Amir Vadai says:

====================
net/mlx4_en: Hardware accelerated 802.1ad

This patchset by Hadar introduces support in Hardware accelerated 802.1ad, for
ConnectX-3pro NIC's.  In order to support existing deployment, and due to some
hardware limitations, the feature is disabled by default, and needed to be
enabled using a private flag in ethtool. Ofcourse user can enable the private
flag only if hardware has support.
After being enabled, the standard ethtool -k/-K can be used.

Patchset was applied and tested over commit 71790a2 ("hv_netvsc: Add structs
and handlers for VF messages")
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agonet/mlx4_en: Add support for hardware accelerated 802.1ad vlan
Hadar Hen Zion [Mon, 27 Jul 2015 11:46:34 +0000 (14:46 +0300)]
net/mlx4_en: Add support for hardware accelerated 802.1ad vlan

To enable device support in accelerated 802.1ad vlan, the port
capability "packet has vlan enable" (phv_en) should be set.
Firmware won't work properly, in case phv_en is not set.

The user can enable "phv_en" port capability with the new ethtool
private flag phv-bit. The phv-bit private flag default value is OFF,
users who are interested in 802.1ad hardware acceleration should turn ON
the phv-bit private flag:
$ ethtool --set-priv-flags eth1 phv-bit on

Once the private flag is set, the device is ready for 802.1ad vlan
acceleration.

The user should also change the interface device features and turn on
"tx-vlan-stag-hw-insert" which is off by default:
$ ethtool -K eth1  tx-vlan-stag-hw-insert on

"phv-bit" private flag setting is available only for Physical
Functions(PF), the Virtual Function (VF) will be able to use the feature
by setting "tx-vlan-stag-hw-insert" ethtool device feature only if the
feature was enabled by the Hypervisor.

Signed-off-by: Hadar Hen Zion <hadarh@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agonet/mlx4: Prepare VLAN macros for 802.1ad Hardware accelerated support
Hadar Hen Zion [Mon, 27 Jul 2015 11:46:33 +0000 (14:46 +0300)]
net/mlx4: Prepare VLAN macros for 802.1ad Hardware accelerated support

To add Hardware accelerated support in 802.1ad vlan, replace
Current VLAN macros to CVLAN.
Replace:
MLX4_WQE_CTRL_INS_VLAN
MLX4_CQE_VLAN_PRESENT_MASK
With:
MLX4_WQE_CTRL_INS_CVLAN
MLX4_CQE_CVLAN_PRESENT_MASK

Signed-off-by: Hadar Hen Zion <hadarh@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agonet/mlx4_en: Prepare ethtool private flags to support more flags
Hadar Hen Zion [Mon, 27 Jul 2015 11:46:32 +0000 (14:46 +0300)]
net/mlx4_en: Prepare ethtool private flags to support more flags

Currently we support only one ethtool private flag. Prepare
mlx4_en_set_priv_flags function to support more than one private flag.
Will be used in the next patch to support hardware accelerated 802.1ad
vlan.

Signed-off-by: Hadar Hen Zion <hadarh@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agonet/mlx4_core: Preparations for 802.1ad VLAN support
Hadar Hen Zion [Mon, 27 Jul 2015 11:46:31 +0000 (14:46 +0300)]
net/mlx4_core: Preparations for 802.1ad VLAN support

mlx4_core preparation to support hardware accelerated 802.1ad VLAN
device.

To allow 802.1ad accelerated device, "packet has vlan" (phv)
Firmware capability should be available. Firmware without the
phv capability won't behave properly and can't support 802.1ad device
acceleration.

The driver checks the Firmware capability and sets the phv bit
accordingly in SET_PORT command.

Signed-off-by: Hadar Hen Zion <hadarh@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agoMerge branch 'arm-bpf-next'
David S. Miller [Mon, 27 Jul 2015 21:57:41 +0000 (14:57 -0700)]
Merge branch 'arm-bpf-next'

Nicolas Schichan says:

====================
ARM BPF JIT features

This series adds support for more instructions to the ARM BPF JIT
namely skb netdevice type retrieval, skb payload offset retrieval, and
skb packet type retrieval.

This allows 35 tests to use the JIT instead of 29 before.

This series depends on the "BPF JIT fixes for ARM" serie sent earlier.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agoARM: net: add support for BPF_ANC | SKF_AD_HATYPE in ARM JIT.
Nicolas Schichan [Mon, 27 Jul 2015 13:06:51 +0000 (15:06 +0200)]
ARM: net: add support for BPF_ANC | SKF_AD_HATYPE in ARM JIT.

Signed-off-by: Nicolas Schichan <nschichan@freebox.fr>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agoARM: net: add support for BPF_ANC | SKF_AD_PAY_OFFSET in ARM JIT.
Nicolas Schichan [Mon, 27 Jul 2015 13:06:50 +0000 (15:06 +0200)]
ARM: net: add support for BPF_ANC | SKF_AD_PAY_OFFSET in ARM JIT.

Signed-off-by: Nicolas Schichan <nschichan@freebox.fr>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agoARM: net: add support for BPF_ANC | SKF_AD_PKTTYPE in ARM JIT.
Nicolas Schichan [Mon, 27 Jul 2015 13:06:49 +0000 (15:06 +0200)]
ARM: net: add support for BPF_ANC | SKF_AD_PKTTYPE in ARM JIT.

Signed-off-by: Nicolas Schichan <nschichan@freebox.fr>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agolwtunnel: use kfree_skb() instead of vanilla kfree()
Dan Carpenter [Mon, 27 Jul 2015 08:07:47 +0000 (11:07 +0300)]
lwtunnel: use kfree_skb() instead of vanilla kfree()

kfree_skb() is correct here.

Fixes: ffce41962ef6 ('lwtunnel: support dst output redirect function')
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agotcp: tso: allow deferring under reordering state
Eric Dumazet [Sun, 26 Jul 2015 07:45:24 +0000 (09:45 +0200)]
tcp: tso: allow deferring under reordering state

While doing experiments with reordering resilience, we found
linux senders were not able to send at full speed under reordering,
because every incoming SACK was releasing one MSS.

This patch removes the limitation, as we did for CWR state
in commit a0ea700e409 ("tcp: tso: allow CA_CWR state in
tcp_tso_should_defer()")

Neal Cardwell had a concern about limited transmit so
Yuchung conducted experiments on GFE and found nothing
worth adding an extra check on fast path :

  if (icsk->icsk_ca_state == TCP_CA_Disorder &&
      tcp_sk(sk)->reordering == sysctl_tcp_reordering)
          goto send_now;

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Yuchung Cheng <ycheng@google.com>
Cc: Neal Cardwell <ncardwell@google.com>
Acked-by: Neal Cardwell <ncardwell@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agoravb: minimize TX data copying
Sergei Shtylyov [Sat, 25 Jul 2015 20:42:01 +0000 (23:42 +0300)]
ravb: minimize TX data copying

Renesas Ethernet AVB controller requires that all data are aligned on 4-byte
boundary.  While it's  easily achievable for  the RX  data with  the help of
skb_reserve() (we even align on 128-byte boundary as recommended by the manual),
we  can't  do the same with the TX data, and it always comes  unaligned from
the networking core. Originally we solved it an easy way, copying all packet
to  a  preallocated  aligned buffer; however, it's enough to copy only up to
3 first bytes from each packet, doing the transfer using 2 TX descriptors
instead of just 1. Here's an implementation of the new  TX algorithm that
significantly reduces the driver's memory requirements.

Signed-off-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agodsa: mv88e6352/mv88e6xxx: Move temperature sensor code to mv88e6xxx.c
Guenter Roeck [Sat, 25 Jul 2015 16:42:28 +0000 (09:42 -0700)]
dsa: mv88e6352/mv88e6xxx: Move temperature sensor code to mv88e6xxx.c

Move the temperature sensing code for mv88e6352 and mv88e6320 families
into mv88e6xxx.c to simplify adding support for additional chips.

With this change, mv88e6xxx_6320_family() no longer needs to be
a global function and is made static.

Cc: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
Acked-by: Andrew Lunn <andrew@lunn.ch>
Tested-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agohv_netvsc: Add structs and handlers for VF messages
Haiyang Zhang [Fri, 24 Jul 2015 17:08:40 +0000 (10:08 -0700)]
hv_netvsc: Add structs and handlers for VF messages

This patch adds data structures and handlers for messages related
to SRIOV Virtual Function.

Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Reviewed-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agoMerge branch 'rt6_probe_write_lock'
David S. Miller [Mon, 27 Jul 2015 08:08:26 +0000 (01:08 -0700)]
Merge branch 'rt6_probe_write_lock'

Martin KaFai Lau says:

====================
ipv6: Avoid rt6_probe() taking writer lock in the fast path

v1 -> v2:
1. Separate the code re-arrangement into another patch
2. Fix style
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agoipv6: Avoid rt6_probe() taking writer lock in the fast path
Martin KaFai Lau [Fri, 24 Jul 2015 16:57:43 +0000 (09:57 -0700)]
ipv6: Avoid rt6_probe() taking writer lock in the fast path

The patch checks neigh->nud_state before acquiring the writer lock.
Note that rt6_probe() is only used in CONFIG_IPV6_ROUTER_PREF.

40 udpflood processes and a /64 gateway route are used.
The gateway has NUD_PERMANENT.  Each of them is run for 30s.
At the end, the total number of finished sendto():

Before: 55M
After: 95M

Signed-off-by: Martin KaFai Lau <kafai@fb.com>
Cc: Hannes Frederic Sowa <hannes@stressinduktion.org>
CC: Julian Anastasov <ja@ssi.bg>
CC: YOSHIFUJI Hideaki <hideaki.yoshifuji@miraclelinux.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agoipv6: Re-arrange code in rt6_probe()
Martin KaFai Lau [Fri, 24 Jul 2015 16:57:42 +0000 (09:57 -0700)]
ipv6: Re-arrange code in rt6_probe()

It is a prep work for the next patch to remove write_lock
from rt6_probe().

1. Reduce the number of if(neigh) check.  From 4 to 1.
2. Bring the write_(un)lock() closer to the operations that the
   lock is protecting.

Hopefully, the above make rt6_probe() more readable.

Signed-off-by: Martin KaFai Lau <kafai@fb.com>
Cc: Hannes Frederic Sowa <hannes@stressinduktion.org>
Cc: Julian Anastasov <ja@ssi.bg>
Cc: YOSHIFUJI Hideaki <hideaki.yoshifuji@miraclelinux.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agobonding: convert num_grat_arp to the new bonding option API
Nikolay Aleksandrov [Fri, 24 Jul 2015 13:50:31 +0000 (15:50 +0200)]
bonding: convert num_grat_arp to the new bonding option API

num_grat_arp wasn't converted to the new bonding option API, so do this
now and remove the specific sysfs store option in order to use the
standard one. num_grat_arp is the same as num_unsol_na so add it as an
alias with the same option settings. An important difference is the option
name which is matched in bond_sysfs_store_option().

Signed-off-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Acked-by: Veaceslav Falico <vfalico@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agonet: phy: fix auto negotiation checking for teranetics
Shaohui Xie [Fri, 24 Jul 2015 11:26:02 +0000 (19:26 +0800)]
net: phy: fix auto negotiation checking for teranetics

When using fiber port, the phy cannot report it's auto negotiation state,
driver should always report auto negotiation is done when using fiber port.

Signed-off-by: Shaohui Xie <Shaohui.Xie@freescale.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agolwtunnel: change prototype of lwtunnel_state_get()
Nicolas Dichtel [Fri, 24 Jul 2015 10:28:36 +0000 (12:28 +0200)]
lwtunnel: change prototype of lwtunnel_state_get()

It saves some lines and simplify a bit the code when the state is returning
by this function. It's also useful to handle a NULL entry.

To avoid too long lines, I've also renamed lwtunnel_state_get() and
lwtunnel_state_put() to lwtstate_get() and lwtstate_put().

CC: Thomas Graf <tgraf@suug.ch>
CC: Roopa Prabhu <roopa@cumulusnetworks.com>
Signed-off-by: Nicolas Dichtel <nicolas.dichtel@6wind.com>
Acked-by: Thomas Graf <tgraf@suug.ch>
Acked-by: Roopa Prabhu <roopa@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agoipv6: copy lwtstate in ip6_rt_copy_init()
Nicolas Dichtel [Fri, 24 Jul 2015 10:28:35 +0000 (12:28 +0200)]
ipv6: copy lwtstate in ip6_rt_copy_init()

We need to copy this field (ip6_rt_cache_alloc() and ip6_rt_pcpu_alloc()
use ip6_rt_copy_init() to build a dst).

CC: Thomas Graf <tgraf@suug.ch>
CC: Roopa Prabhu <roopa@cumulusnetworks.com>
Fixes: 19e42e451506 ("ipv6: support for fib route lwtunnel encap attributes")
Signed-off-by: Nicolas Dichtel <nicolas.dichtel@6wind.com>
Acked-by: Thomas Graf <tgraf@suug.ch>
Acked-by: Roopa Prabhu <roopa@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agoipv6: use lwtunnel_output6() only if flag redirect is set
Nicolas Dichtel [Fri, 24 Jul 2015 08:59:41 +0000 (10:59 +0200)]
ipv6: use lwtunnel_output6() only if flag redirect is set

This function make sense only when LWTUNNEL_STATE_OUTPUT_REDIRECT is set.
The check is already done in IPv4.

CC: Thomas Graf <tgraf@suug.ch>
CC: Roopa Prabhu <roopa@cumulusnetworks.com>
Fixes: 74a0f2fe8ed5 ("ipv6: rt6_info output redirect to tunnel output")
Signed-off-by: Nicolas Dichtel <nicolas.dichtel@6wind.com>
Acked-by: Thomas Graf <tgraf@suug.ch>
Acked-by: Roopa Prabhu <roopa@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agonet: phy: dp83867: fix simple_return.cocci warnings
Wu Fengguang [Fri, 24 Jul 2015 06:16:10 +0000 (14:16 +0800)]
net: phy: dp83867: fix simple_return.cocci warnings

drivers/net/phy/dp83867.c:126:1-4: WARNING: end returns can be simpified
drivers/net/phy/dp83867.c:74:5-8: WARNING: end returns can be simpified if tested value is negative or 0

 Simplify a trivial if-return sequence.  Possibly combine with a
 preceding function call.

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

CC: Dan Murphy <dmurphy@ti.com>
Signed-off-by: Fengguang Wu <fengguang.wu@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agodev: Spelling fix in comments
subashab@codeaurora.org [Fri, 24 Jul 2015 03:03:29 +0000 (03:03 +0000)]
dev: Spelling fix in comments

Fix the following typo
- unchainged -> unchanged

Signed-off-by: Subash Abhinov Kasiviswanathan <subashab@codeaurora.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agoebpf: Allow dereferences of PTR_TO_STACK registers
Alex Gartrell [Thu, 23 Jul 2015 21:24:40 +0000 (14:24 -0700)]
ebpf: Allow dereferences of PTR_TO_STACK registers

mov %rsp, %r1           ; r1 = rsp
        add $-8, %r1            ; r1 = rsp - 8
        store_q $123, -8(%rsp)  ; *(u64*)r1 = 123  <- valid
        store_q $123, (%r1)     ; *(u64*)r1 = 123  <- previously invalid
        mov $0, %r0
        exit                    ; Always need to exit

And we'd get the following error:

0: (bf) r1 = r10
1: (07) r1 += -8
2: (7a) *(u64 *)(r10 -8) = 999
3: (7a) *(u64 *)(r1 +0) = 999
R1 invalid mem access 'fp'

Unable to load program

We already know that a register is a stack address and the appropriate
offset, so we should be able to validate those references as well.

Signed-off-by: Alex Gartrell <agartrell@fb.com>
Acked-by: Alexei Starovoitov <ast@plumgrid.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agoMerge branch 'mlx5e-next'
David S. Miller [Mon, 27 Jul 2015 07:29:18 +0000 (00:29 -0700)]
Merge branch 'mlx5e-next'

Amir Vadai says:

====================
ConnectX-4 driver update 2015-07-23

This patchset introduce some performance enhancements to the ConnectX-4 driver.
1. Improving RSS distribution, and make RSS function controlable using ethtool.
2. Make memory that is written by NIC and read by host CPU allocate in the
   local NUMA to the processing CPU
3. Support tx copybreak
4. Using hardware feature called blueflame to save DMA reads when possible

Another patch by Achiad fix some cosmetic issues in the driver.

Patchset was applied and tested on top of commit 045a0fa ("ip_tunnel: Call
ip_tunnel_core_init() from inet_init()")
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agonet/mlx5e: Input IPSEC.SPI into the RX RSS hash function
Achiad Shochat [Thu, 23 Jul 2015 20:36:01 +0000 (23:36 +0300)]
net/mlx5e: Input IPSEC.SPI into the RX RSS hash function

In addition to the source/destination IP which are already hashed.
Only for unicast traffic for now.

Signed-off-by: Achiad Shochat <achiad@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agonet/mlx5e: Cosmetics: use BIT() instead of "1 <<", and others
Achiad Shochat [Thu, 23 Jul 2015 20:36:00 +0000 (23:36 +0300)]
net/mlx5e: Cosmetics: use BIT() instead of "1 <<", and others

No logical change in this commit.

Signed-off-by: Achiad Shochat <achiad@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agonet/mlx5e: TX latency optimization to save DMA reads
Achiad Shochat [Thu, 23 Jul 2015 20:35:59 +0000 (23:35 +0300)]
net/mlx5e: TX latency optimization to save DMA reads

A regular TX WQE execution involves two or more DMA reads -
one to fetch the WQE, and another one per WQE gather entry.

These DMA reads obviously increase the TX latency.
There are two mlx5 mechanisms to bypass these DMA reads:
1) Inline WQE
2) Blue Flame (BF)

An inline WQE contains a whole packet, thus saves the DMA read/s
of the regular WQE gather entry/s. Inline WQE support was already
added in the previous commit.

A BF WQE is written directly to the device I/O mapped memory, thus
enables saving the DMA read that fetches the WQE.

The BF WQE I/O write must be in cache line granularity, thus uses
the CPU write combining mechanism.
A BF WQE I/O write acts also as a TX doorbell for notifying the
device of new TX WQEs.
A BF WQE is written to the same I/O mapped address as the regular TX
doorbell, thus this address is being mapped twice - once by ioremap()
and once by io_mapping_map_wc().

While both mechanisms reduce the TX latency, they both consume more CPU
cycles than a regular WQE:
- A BF WQE must still be written to host memory, in addition to being
  written directly to the device I/O mapped memory.
- An inline WQE involves copying the SKB data into it.

To handle this tradeoff, we introduce here a heuristic algorithm that
strives to avoid using these two mechanisms in case the TX queue is
being back-pressured by the device, and limit their usage rate otherwise.

An inline WQE will always be "Blue Flamed" (written directly to the
device I/O mapped memory) while a BF WQE may not be inlined (may contain
gather entries).

Preliminary testing using netperf UDP_RR shows that the latency goes down
from 17.5us to 16.9us, while the message rate (tested with pktgen) stays
the same.

Signed-off-by: Achiad Shochat <achiad@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agonet/mlx5e: Support TX packet copy into WQE
Achiad Shochat [Thu, 23 Jul 2015 20:35:58 +0000 (23:35 +0300)]
net/mlx5e: Support TX packet copy into WQE

AKA inline WQE.
A TX latency optimization to save data gather DMA reads.
Controlled by ETHTOOL_TX_COPYBREAK.

Signed-off-by: Achiad Shochat <achiad@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agonet/mlx5e: Allocate DMA coherent memory on reader NUMA node
Saeed Mahameed [Thu, 23 Jul 2015 20:35:57 +0000 (23:35 +0300)]
net/mlx5e: Allocate DMA coherent memory on reader NUMA node

By affinity hints and XPS, each mlx5e channel is assigned a CPU
core.

Channel DMA coherent memory that is written by the NIC and read
by SW (e.g CQ buffer) is allocated on the NUMA node of the CPU
core assigned for the channel.

Channel DMA coherent memory that is written by SW and read by the
NIC (e.g SQ/RQ buffer) is allocated on the NUMA node of the NIC.

Doorbell record (written by SW and read by the NIC) is an
exception since it is accessed by SW more frequently.

Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agonet/mlx5e: Support ETH_RSS_HASH_XOR
Saeed Mahameed [Thu, 23 Jul 2015 20:35:56 +0000 (23:35 +0300)]
net/mlx5e: Support ETH_RSS_HASH_XOR

The ConnectX-4 HW implements inverted XOR8.
To make it act as XOR we re-order the HW RSS indirection table.

Set XOR to be the default RSS hash function and add ethtool API to
control it.

Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agoMerge branch 'netcp-next'
David S. Miller [Mon, 27 Jul 2015 07:18:40 +0000 (00:18 -0700)]
Merge branch 'netcp-next'

WingMan Kwok says:

====================
net: netcp: Bug fixes of CPSW statistics collection

This patch set contains bug fixes and enhencements of hw ethernet
statistics processing on TI's Keystone2 CPSW ethernet switches.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agonet: netcp: Adds missing statistics for K2L and K2E
WingMan Kwok [Thu, 23 Jul 2015 19:57:24 +0000 (15:57 -0400)]
net: netcp: Adds missing statistics for K2L and K2E

This patch adds the missing statistics for the host
and slave ports of the CPSW on K2L and K2E platforms.

Signed-off-by: WingMan Kwok <w-kwok2@ti.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agonet: netcp: Fixes to CPSW statistics collection
WingMan Kwok [Thu, 23 Jul 2015 19:57:23 +0000 (15:57 -0400)]
net: netcp: Fixes to CPSW statistics collection

In certain applications it's beneficial to allow the CPSW h/w
stats counters to continue to increment even while the kernel
polls them. This patch implements this behavior for both 1G
and 10G ethernet subsystem modules.

Signed-off-by: WingMan Kwok <w-kwok2@ti.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agonet: netcp: Consolidates statistics collection code
WingMan Kwok [Thu, 23 Jul 2015 19:57:22 +0000 (15:57 -0400)]
net: netcp: Consolidates statistics collection code

Different Keystone2 platforms have different number and
layouts of hw statistics modules.  This patch consolidates
the statistics processing of different Keystone2 platforms
for easy maintenance.

Signed-off-by: WingMan Kwok <w-kwok2@ti.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agonet: netcp: Fixes error in oversized memory allocation for statistics storage
WingMan Kwok [Thu, 23 Jul 2015 19:57:21 +0000 (15:57 -0400)]
net: netcp: Fixes error in oversized memory allocation for statistics storage

The CPSW driver keeps internally some, but not all, of
the statistics available in the hw statistics modules.  Furthermore,
some of the locations in the hw statistics modules are reserved and
contain no useful information.  Prior to this patch, the driver
allocates memory of the size of the the whole hw statistics modules,
instead of the size of statistics-entries-interested-in (i.e. et_stats),
for internal storage.  This patch fixes that.

Signed-off-by: WingMan Kwok <w-kwok2@ti.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agonet: netcp: Fixes hw statistics module base setting error
WingMan Kwok [Thu, 23 Jul 2015 19:57:20 +0000 (15:57 -0400)]
net: netcp: Fixes hw statistics module base setting error

This patch fixes error in the setting of the hw statistics
module base for K2HK platform.  In K2HK although there are
4 hw statistics modules, but only 2 are visible at a time.
Thus when setting up the pointers to the base of the
corresponding hw statistics modules, modules 0 and 2 should
point to one base, while modules 1 and 3 should point to the
other.

Signed-off-by: WingMan Kwok <w-kwok2@ti.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agonet: netcp: Fixes the use of spin_lock_bh in timer function
WingMan Kwok [Thu, 23 Jul 2015 19:57:19 +0000 (15:57 -0400)]
net: netcp: Fixes the use of spin_lock_bh in timer function

This patch fixes a bug in which the timer routine synchronized
against the ethtool-triggered statistics updates with spin_lock_bh().
A timer function is itself a bottom-half, so this should be
spin_lock().

Signed-off-by: WingMan Kwok <w-kwok2@ti.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
8 years agocxgb4vf: Read correct FL congestion threshold for T5 and T6
Hariprasad Shenai [Thu, 23 Jul 2015 17:11:14 +0000 (22:41 +0530)]
cxgb4vf: Read correct FL congestion threshold for T5 and T6

VF driver was reading incorrect freelist congestion notification threshold
for FLM queues when packing is enabled for T5 and T6 adapter. Fixing it
now.

Signed-off-by: Hariprasad Shenai <hariprasad@chelsio.com>
Signed-off-by: David S. Miller <davem@davemloft.net>