Embedded CPU bit doesn't change with PCI resume/suspend.
Hence read it only once while probing the PCI device.
Signed-off-by: Parav Pandit <parav@mellanox.com>
Reviewed-by: Bodong Wang <bodong@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>

netif_set_real_num_tx_queues and netif_set_real_num_rx_queues may fail.
Now that mlx5e supports handling errors in the preactivate hook, this
commit leverages that functionality to handle errors from those
functions and roll back all changes on failure.
Signed-off-by: Maxim Mikityanskiy <maximmi@mellanox.com>
Reviewed-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>

We use mapping to save and restore the tunnel options.
Save also the tunnel options mask.
Signed-off-by: Roi Dayan <roid@mellanox.com>
Reviewed-by: Paul Blakey <paulb@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>

In notify HW (ring doorbell) flow, we set the bit to request a completion
on the TX descriptor.
When doing so, we should not unset other bits in the same byte.
Currently, this does not fix a real issue, as we still don't have a flow
where both MLX5_WQE_CTRL_CQ_UPDATE and any adjacent bit are set together.

Fixes: 542578c6 ("net/mlx5e: Move helper functions to a new txrx datapath header")
Fixes: 864b2d71 ("net/mlx5e: Generalize tx helper functions for different SQ types")
Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Reviewed-by: Aya Levin <ayal@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>

Currently the SA handle is created and managed as part of the common
code for different IPsec supporting HW, this handle is passed to HW
to be used on Rx to identify the SA handle that was used to
return the xfrm state to stack.

The above implementation pose a limitation on managing this handle.

Refactor by moving management of this field to the specific HW code.

Downstream patches will introduce the Connect-X support for IPsec that
will use this handle differently than current implementation.
Signed-off-by: Raed Salem <raeds@mellanox.com>
Reviewed-by: Boris Pismenny <borisp@mellanox.com>
Reviewed-by: Huy Nguyen <huyn@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>

The current HW counters are supported only by Innova, split the ipsec
stats group into two groups, one for HW and one for SW. And expose
the HW counters to ethtool only if Innova HW is used for IPsec offload.
Signed-off-by: Raed Salem <raeds@mellanox.com>
Reviewed-by: Huy Nguyen <huyn@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>

Currently the FPGA IPsec is the only hw implementation of the IPsec
acceleration api, and so the mlx5_accel_esp_create_hw_context was
wrongly made to suit this HW api, among other in its parameter list
and some of its parameter endianness.

This implementation might not be suitable for different HW.

Refactor by group and pass all function arguments of
mlx5_accel_esp_create_hw_context in common mlx5_accel_esp_xfrm_attrs
struct field of mlx5_accel_esp_xfrm struct and correct the endianness
according to the HW being called.
Signed-off-by: Raed Salem <raeds@mellanox.com>
Reviewed-by: Boris Pismenny <borisp@mellanox.com>
Reviewed-by: Huy Nguyen <huyn@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>

The cited commit relies on include <net/geneve.h> being included
implicitly prior to include "en_accel/en_accel.h".
This mandates that all files that needs to include en_accel.h
to redantantly include net/geneve.h.

Include net/geneve.h explicitly at "en_accel/en_accel.h" to avoid
undesired constrain as above.

Fixes: e3cfc7e6 ("net/mlx5e: TX, Add geneve tunnel stateless offload support")
Signed-off-by: Raed Salem <raeds@mellanox.com>
Reviewed-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>

Currently the IPsec acceleration capability function is also used
at IPsec fpga capable device code.

This could cause a future bug as the acceleration layer is agnostic
to the device implementing its API.

Fix by using the IPsec FPGA capability function instead of acceleration
layer capability function in case of FPGA IPsec only related operations.

Downstream patches will add support for Connect-X IPsec, this can avoid
a future bug.
Signed-off-by: Raed Salem <raeds@mellanox.com>
Reviewed-by: Boris Pismenny <borisp@mellanox.com>
Reviewed-by: Huy Nguyen <huyn@mellanox.com>
Reviewed-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>

Add PMTU discovery tests for these encapsulations:

- IPIP
- SIT, mode ip6ip
- ip6tnl, modes ip6ip6 and ipip6
Signed-off-by: Lourdes Pedrajas <lu@pplo.net>
Reviewed-by: Stefano Brivio <sbrivio@redhat.com>
Reviewed-by: David Ahern <dsahern@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Don't use internal knowledge of the mdio bus core, instead use
mdiobus_get_phy() which does the same thing.
Signed-off-by: Michael Walle <michael@walle.cc>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>

Ido Schimmel says:

====================
mlxsw: Two small changes

Patch #1 increases the scale of supported IPv6 nexthops groups when each
group has one nexthop and all are using the same nexthop device, but
with a different gateway IP.

Patch #2 adjusts a register definition in accordance with recent
firmware changes.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>

The Infrastructure Entry Delete Register (IEDR) is used to delete
entries stored in the KVD linear database. Currently, it is only
possible to delete entries of size up to 2048. Future firmware versions
will support deletion of entries of size up to 4096.

Increase the size of the field so that the driver will be able to
perform such deletions in the future, when required.
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Reviewed-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

As explained in commit fc25996e ("mlxsw: spectrum_router: Increase
scale of IPv6 nexthop groups"), each nexthop group is hashed by XOR-ing
the interface indexes of all the member nexthop devices.

To avoid many different nexthop groups ending up using the same key, the
above commit started hashing the interface indexes themselves before
they are XOR-ed.

However, in cases in which there are many nexthop groups that all use
the same nexthop device and only differ in the gateway IP, we can still
end up in a situation in which all the groups are using the same key.
This eventually leads to -EBUSY error from rhashtable during insertion.

Improve the situation by also making the gateway IP part of the key.
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Reported-by: Alex Veber <alexve@mellanox.com>
Reviewed-by: Jiri Pirko <jiri@mellanox.com>
Tested-by: Alex Veber <alexve@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

If a bus->reset() call for the mii_bus structure returns an error (e.g.:
-EPROE_DEFER) we should propagate it accordingly.
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>

Heiner Kallweit says:

====================
net: phy: realtek: move PHY resume delay from MAC to PHY driver

Internal PHY's from RTL8168h up may not be instantly ready after calling
genphy_resume(). So far r8169 network driver adds the needed delay, but
better handle this in the PHY driver. The network driver may miss other
places where the PHY is resumed.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>

The Realtek PHY driver takes care of adding the needed delay now,
therefore we can remove the delay here.
Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Internal PHY's from RTL8168h up may not be instantly ready after calling
genphy_resume(). So far r8169 network driver adds the needed delay, but
better handle this in the PHY driver. The network driver may miss other
places where the PHY is resumed.
Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Fix the following coccicheck warning:

drivers/net/ethernet/hisilicon/hns/hns_dsaf_mac.c:700:2-8: WARNING:
Assignment of 0/1 to bool variable
drivers/net/ethernet/hisilicon/hns/hns_dsaf_mac.c:702:2-8: WARNING:
Assignment of 0/1 to bool variable
Reported-by: Hulk Robot <hulkci@huawei.com>
Signed-off-by: Jason Yan <yanaijie@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Fix sparse warnings:

drivers/ptp/ptp_idt82p33.c:26:5: warning: symbol 'sync_tod_timeout' was not declared. Should it be static?
drivers/ptp/ptp_idt82p33.c:31:5: warning: symbol 'phase_snap_threshold' was not declared. Should it be static?
Reported-by: Hulk Robot <hulkci@huawei.com>
Signed-off-by: YueHaibing <yuehaibing@huawei.com>
Acked-by: Richard Cochran <richardcochran@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Fix sparse warning:

drivers/ptp/ptp_kvm.c:25:1: warning:
 symbol 'kvm_ptp_lock' was not declared. Should it be static?
Reported-by: Hulk Robot <hulkci@huawei.com>
Signed-off-by: YueHaibing <yuehaibing@huawei.com>
Acked-by: Richard Cochran <richardcochran@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Some boards do not have the RX_ER MII signal connected. Normally in such
situation, those pins would be grounded, but then again, some boards
left it electrically floating.

When sending traffic to those switch ports, one can see that the
N_SOFERR statistics counter is incrementing once per each packet. The
user manual states for this counter that it may count the number of
frames "that have the MII error input being asserted prior to or
up to the SOF delimiter byte". So the switch MAC is sampling an
electrically floating signal, and preventing proper traffic reception
because of that.

As a workaround, enable the internal weak pull-downs on the input pads
for the MII control signals. This way, a floating signal would be
internally tied to ground.

The logic levels of signals which _are_ externally driven should not be
bothered by this 40-50 KOhm internal resistor. So it is not an issue to
enable the internal pull-down unconditionally, irrespective of PHY
interface type (MII, RMII, RGMII, SGMII) and of board layout.
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Heiner Kallweit says:

====================
r8169: series with improvements

Again a series with few improvements.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>

Add workaround for hw issues with TSO on RTL8168evl. This workaround is
based on information I got from Realtek, and *should* allow to safely
enable TSO on this chip version.
Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Simplify the code and avoid the overhead of calling vlan_get_protocol().
Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

At that place in rtl_init_one() we can safely use rtl8169_set_features()
to configure the chip according to the default features.
Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

So far we set RX_VLAN_8125 unconditionally, even if
NETIF_F_HW_VLAN_CTAG_RX may not be set. Don't touch these bits,
and let only rtl8169_set_features() control them.
Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

NETIF_F_HIGHDMA is added to vlan_features by register_netdev(),
therefore we can omit this here.
Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Move setting the ocp_base to rtl_init_one(). Where supported the value
is always the same, and if not supported it doesn't hurt.
Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Each ENETC station interface (SI) has a VLAN filter list and a port
flag (PSIPVMR) by which it can be put in "VLAN promiscuous" mode, which
enables the reception of VLAN-tagged traffic even if it is not in the
VLAN filtering list.

Currently the handling of this setting works like this: the port starts
off as VLAN promiscuous, then it switches to enabling VLAN filtering as
soon as the first VLAN is installed in its filter via
.ndo_vlan_rx_add_vid. In practice that does not work out very well,
because more often than not, the first VLAN to be installed is out of
the control of the user: the 8021q module, if loaded, adds its rule for
802.1p (VID 0) traffic upon bringing the interface up.

What the user is currently seeing in ethtool is this:
ethtool -k eno2
rx-vlan-filter: on [fixed]

which doesn't match the intention of the code, but the practical reality
of having the 8021q module install its VID which has the side-effect of
turning on VLAN filtering in this driver. All in all, a slightly
confusing experience.

So instead of letting this driver switch the VLAN filtering state by
itself, just wire it up with the rx-vlan-filter feature from ethtool,
and let it be user-configurable just through that knob, except for one
case, see below.

In promiscuous mode, it is more intuitive that all traffic is received,
including VLAN tagged traffic. It appears that it is necessary to set
the flag in PSIPVMR for that to be the case, so VLAN promiscuous mode is
also temporarily enabled. On exit from promiscuous mode, the setting
made by ethtool is restored.
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Claudiu Manoil <claudiu.manoil@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

By default, the VCAP IS2 will produce a single match for each frame, on
the most specific classification.

Example: a ping packet (ICMP over IPv4 over Ethernet) sent from an IP
address of 10.0.0.1 and a MAC address of 96:18:82:00:04:01 will match
this rule:

tc filter add dev swp0 ingress protocol ipv4 \
	flower skip_sw src_ip 10.0.0.1 action drop

but not this one:

tc filter add dev swp0 ingress \
	flower skip_sw src_mac 96:18:82:00:04:01 action drop

Currently the driver does not really warn the user in any way about
this, and the behavior is rather strange anyway.

The current patch is a workaround to force matches on MAC_ETYPE keys
(DMAC and SMAC) for all packets irrespective of higher layer protocol.
The setting is made at the port level.

Of course this breaks all other non-src_mac and non-dst_mac matches, so
rule exclusivity checks have been added to the driver, in order to never
have rules of both types on any ingress port.

The bits that discard higher-level protocol information are set only
once a MAC_ETYPE rule is added to a filter block, and only for the ports
that are bound to that filter block. Then all further non-MAC_ETYPE
rules added to that filter block should be denied by the ports bound to
it.
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Jeff Kirsher says:

====================
1GbE Intel Wired LAN Driver Updates 2020-04-17

This series contains updates to e1000e and igc only.

Sasha adds partial generic segmentation offload (GSO partial) support to
the igc driver.  Also added support for translating taprio schedules
into i225 cycles in igc.  Did clean up of dead code or unused defines in
the igc driver.  Refactored the code to avoid forward declarations where
possible.  Enables the NETIF_F_HW_TC flag for igc by default.

Vinicius adds support for ETF offloading using the similar approach that
taprio offload used.

Kees Cook fixes a clang warning in the e1000e driver by moving the
declared variable either into the switch case that uses the variable or
lift them up into the main function body, to help the compiler.

Andre fixed some register overwriting when dumping registers via ethtool
for igc driver.  Also fixed support for ethtool Network Flow
Classification (NFC) queue redirection by adding the missing code needed
to enable the queue selection feature from Receive Address High (RAH)
register.  Cleans up code to remove the code bits designed to support
tc-flower filters, since this client part does not support it.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>

BCM53125 has internal Gigabit PHYs which support interrupts as well as
statistics, make it possible to configure both of those features with a
PHY driver entry.
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Displaying the virtual address at which the MDIO base register address
has been mapped is not useful and is not visible with pointer hashing in
place, replace the message with something indicating successful
registration instead.
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>

Make use of platform_get_irq_byname_optional() to avoid printing
messages on the kernel console that interrupts cannot be found.
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

This adds support for the Qualcomm Atheros AR8032 Fast Ethernet PHY.

It shares many similarities with the already supported AR8030 PHY but
additionally supports MII connection to the MAC.
Signed-off-by: David Bauer <mail@david-bauer.net>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>

Merge branch 'for-upstream' of git://git.kernel.org/pub/scm/linux/kernel/git/bluetooth/bluetooth-next

Johan Hedberg says:

====================
pull request: bluetooth-next 2020-04-17

Here's the first bluetooth-next pull request for the 5.8 kernel:

 - Added debugfs option to control MITM flag usage during pairing
 - Added new BT_MODE socket option
 - Added support for Qualcom QCA6390 device
 - Added support for Realtek RTL8761B device
 - Added support for mSBC audio codec over USB endpoints
 - Added framework for Microsoft HCI vendor extensions
 - Added new Read Security Information management command
 - Fixes/cleanup to link layer privacy related code
 - Various other smaller cleanups & fixes
====================
Signed-off-by: David S. Miller <davem@davemloft.net>

This patch fixes a bug when the user adds the first MAC address filter
via ethtool NFC mechanism.

When the first MAC address filter is added, it overwrites the default
MAC address filter configured at RAL[0] and RAH[0]. As consequence,
frames addressed to the interface MAC address are not sent to host
anymore.

This patch fixes the bug by calling igc_set_default_mac_filter() during
adapter init so the position 0 of adapter->mac_table[] is assigned to
the default MAC address.
Signed-off-by: Andre Guedes <andre.guedes@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

IGC driver has no support for tc-flower filters so this patch removes
some leftover code, probably copied from IGB driver by mistake.
Signed-off-by: Andre Guedes <andre.guedes@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

The support for ethtool Network Flow Classification (NFC) queue
redirection based on destination MAC address is currently broken in IGC.
For instance, if we add the following rule, matching frames aren't
enqueued on the expected rx queue.

$ ethtool -N IFNAME flow-type ether dst 3c:fd:fe:9e:7f:71 queue 2

The issue here is due to the fact that igc_rar_set_index() is missing
code to enable the queue selection feature from Receive Address High
(RAH) register. This patch adds the missing code and fixes the issue.
Signed-off-by: Andre Guedes <andre.guedes@intel.com>
Acked-by: Sasha Neftin <sasha.neftin@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>