ovpn: implement basic RX path (UDP)

Packets received over the socket are forwarded to the user device.

Implementation is UDP only. TCP will be added by a later patch.

Note: no decryption/decapsulation exists yet, packets are forwarded as
they arrive without much processing.

Signed-off-by: Antonio Quartulli <antonio@openvpn.net>
Link: https://patch.msgid.link/20250415-b4-ovpn-v26-8-577f6097b964@openvpn.net
Reviewed-by: Sabrina Dubroca <sd@queasysnail.net>
Tested-by: Oleksandr Natalenko <oleksandr@natalenko.name>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>

ovpn: implement basic TX path (UDP)

Packets sent over the ovpn interface are processed and transmitted to the
connected peer, if any.

Implementation is UDP only. TCP will be added by a later patch.

Note: no crypto/encapsulation exists yet. Packets are just captured and
sent.

Signed-off-by: Antonio Quartulli <antonio@openvpn.net>
Link: https://patch.msgid.link/20250415-b4-ovpn-v26-7-577f6097b964@openvpn.net
Reviewed-by: Sabrina Dubroca <sd@queasysnail.net>
Tested-by: Oleksandr Natalenko <oleksandr@natalenko.name>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>

ovpn: introduce the ovpn_socket object

This specific structure is used in the ovpn kernel module
to wrap and carry around a standard kernel socket.

ovpn takes ownership of passed sockets and therefore an ovpn
specific objects is attached to them for status tracking
purposes.

Initially only UDP support is introduced. TCP will come in a later
patch.

Cc: willemdebruijn.kernel@gmail.com
Signed-off-by: Antonio Quartulli <antonio@openvpn.net>
Link: https://patch.msgid.link/20250415-b4-ovpn-v26-6-577f6097b964@openvpn.net
Reviewed-by: Sabrina Dubroca <sd@queasysnail.net>
Tested-by: Oleksandr Natalenko <oleksandr@natalenko.name>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>

ovpn: introduce the ovpn_peer object

An ovpn_peer object holds the whole status of a remote peer
(regardless whether it is a server or a client).

This includes status for crypto, tx/rx buffers, napi, etc.

Only support for one peer is introduced (P2P mode).
Multi peer support is introduced with a later patch.

Along with the ovpn_peer, also the ovpn_bind object is introcued
as the two are strictly related.
An ovpn_bind object wraps a sockaddr representing the local
coordinates being used to talk to a specific peer.

Signed-off-by: Antonio Quartulli <antonio@openvpn.net>
Link: https://patch.msgid.link/20250415-b4-ovpn-v26-5-577f6097b964@openvpn.net
Reviewed-by: Sabrina Dubroca <sd@queasysnail.net>
Tested-by: Oleksandr Natalenko <oleksandr@natalenko.name>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>

ovpn: keep carrier always on for MP interfaces

An ovpn interface configured in MP mode will keep carrier always
on and let the user decide when to bring it administratively up and
down.

This way a MP node (i.e. a server) will keep its interface always
up and running, even when no peer is connected.

Signed-off-by: Antonio Quartulli <antonio@openvpn.net>
Link: https://patch.msgid.link/20250415-b4-ovpn-v26-4-577f6097b964@openvpn.net
Reviewed-by: Sabrina Dubroca <sd@queasysnail.net>
Tested-by: Oleksandr Natalenko <oleksandr@natalenko.name>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>

ovpn: add basic interface creation/destruction/management routines

Add basic infrastructure for handling ovpn interfaces.

Tested-by: Donald Hunter <donald.hunter@gmail.com>
Signed-off-by: Antonio Quartulli <antonio@openvpn.net>
Link: https://patch.msgid.link/20250415-b4-ovpn-v26-3-577f6097b964@openvpn.net
Reviewed-by: Sabrina Dubroca <sd@queasysnail.net>
Tested-by: Oleksandr Natalenko <oleksandr@natalenko.name>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>

ovpn: add basic netlink support

This commit introduces basic netlink support with family
registration/unregistration functionalities and stub pre/post-doit.

More importantly it introduces the YAML uAPI description along
with its auto-generated files:
- include/uapi/linux/ovpn.h
- drivers/net/ovpn/netlink-gen.c
- drivers/net/ovpn/netlink-gen.h

Reviewed-by: Donald Hunter <donald.hunter@gmail.com>
Signed-off-by: Antonio Quartulli <antonio@openvpn.net>
Link: https://patch.msgid.link/20250415-b4-ovpn-v26-2-577f6097b964@openvpn.net
Reviewed-by: Sabrina Dubroca <sd@queasysnail.net>
Tested-by: Oleksandr Natalenko <oleksandr@natalenko.name>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>

net: introduce OpenVPN Data Channel Offload (ovpn)

OpenVPN is a userspace software existing since around 2005 that allows
users to create secure tunnels.

So far OpenVPN has implemented all operations in userspace, which
implies several back and forth between kernel and user land in order to
process packets (encapsulate/decapsulate, encrypt/decrypt, rerouting..).

With `ovpn` we intend to move the fast path (data channel) entirely
in kernel space and thus improve user measured throughput over the
tunnel.

`ovpn` is implemented as a simple virtual network device driver, that
can be manipulated by means of the standard RTNL APIs. A device of kind
`ovpn` allows only IPv4/6 traffic and can be of type:
* P2P (peer-to-peer): any packet sent over the interface will be
  encapsulated and transmitted to the other side (typical OpenVPN
  client or peer-to-peer behaviour);
* P2MP (point-to-multipoint): packets sent over the interface are
  transmitted to peers based on existing routes (typical OpenVPN
  server behaviour).

After the interface has been created, OpenVPN in userspace can
configure it using a new Netlink API. Specifically it is possible
to manage peers and their keys.

The OpenVPN control channel is multiplexed over the same transport
socket by means of OP codes. Anything that is not DATA_V2 (OpenVPN
OP code for data traffic) is sent to userspace and handled there.
This way the `ovpn` codebase is kept as compact as possible while
focusing on handling data traffic only (fast path).

Any OpenVPN control feature (like cipher negotiation, TLS handshake,
rekeying, etc.) is still fully handled by the userspace process.

When userspace establishes a new connection with a peer, it first
performs the handshake and then passes the socket to the `ovpn` kernel
module, which takes ownership. From this moment on `ovpn` will handle
data traffic for the new peer.
When control packets are received on the link, they are forwarded to
userspace through the same transport socket they were received on, as
userspace is still listening to them.

Some events (like peer deletion) are sent to a Netlink multicast group.

Although it wasn't easy to convince the community, `ovpn` implements
only a limited number of the data-channel features supported by the
userspace program.

Each feature that made it to `ovpn` was attentively vetted to
avoid carrying too much legacy along with us (and to give a clear cut to
old and probalby-not-so-useful features).

Notably, only encryption using AEAD ciphers (specifically
ChaCha20Poly1305 and AES-GCM) was implemented. Supporting any other
cipher out there was not deemed useful.

Both UDP and TCP sockets are supported.

As explained above, in case of P2MP mode, OpenVPN will use the main system
routing table to decide which packet goes to which peer. This implies
that no routing table was re-implemented in the `ovpn` kernel module.

This kernel module can be enabled by selecting the CONFIG_OVPN entry
in the networking drivers section.

NOTE: this first patch introduces the very basic framework only.
Features are then added patch by patch, however, although each patch
will compile and possibly not break at runtime, only after having
applied the full set it is expected to see the ovpn module fully working.

Cc: steffen.klassert@secunet.com
Cc: antony.antony@secunet.com
Signed-off-by: Antonio Quartulli <antonio@openvpn.net>
Link: https://patch.msgid.link/20250415-b4-ovpn-v26-1-577f6097b964@openvpn.net
Reviewed-by: Sabrina Dubroca <sd@queasysnail.net>
Tested-by: Oleksandr Natalenko <oleksandr@natalenko.name>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>

net: airoha: Add matchall filter offload support

Introduce tc matchall filter offload support in airoha_eth driver.
Matchall hw filter is used to implement hw rate policing via tc action
police:

$tc qdisc add dev eth0 handle ffff: ingress
$tc filter add dev eth0 parent ffff: matchall action police \
 rate 100mbit burst 1000k drop

The current implementation supports just drop/accept as exceed/notexceed
actions. Moreover, rate and burst are the only supported configuration
parameters.

Reviewed-by: Davide Caratti <dcaratti@redhat.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Signed-off-by: Lorenzo Bianconi <lorenzo@kernel.org>
Link: https://patch.msgid.link/20250415-airoha-hw-rx-ratelimit-v4-1-03458784fbc3@kernel.org
Signed-off-by: Paolo Abeni <pabeni@redhat.com>

Merge branch '10GbE' of git://git./linux/kernel/git/tnguy/next-queue

Tony Nguyen says:

====================
ixgbe: Add basic devlink support

Jedrzej Jagielski says:

Create devlink specific directory for more convenient future feature
development.

Flashing and reloading are supported only by E610 devices.

Introduce basic FW/NVM validation since devlink reload introduces
possibility of runtime NVM update. Check FW API version, FW recovery
mode and FW rollback mode. Introduce minimal recovery probe to let
user to reload the faulty FW when recovery mode is detected.

* '10GbE' of git://git.kernel.org/pub/scm/linux/kernel/git/tnguy/next-queue:
  ixgbe: add support for FW rollback mode
  ixgbe: add E610 implementation of FW recovery mode
  ixgbe: add FW API version check
  ixgbe: add support for devlink reload
  ixgbe: add device flash update via devlink
  ixgbe: extend .info_get() with stored versions
  ixgbe: add E610 functions getting PBA and FW ver info
  ixgbe: add .info_get extension specific for E610 devices
  ixgbe: read the netlist version information
  ixgbe: read the OROM version information
  ixgbe: add E610 functions for acquiring flash data
  ixgbe: add handler for devlink .info_get()
  ixgbe: add initial devlink support
  ixgbe: wrap netdev_priv() usage
  devlink: add value check to devlink_info_version_put()
====================

Link: https://patch.msgid.link/20250415221301.1633933-1-anthony.l.nguyen@intel.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

Merge branch 'adopting-nlmsg_payload-in-ipv4-ipv6'

Breno Leitao says:

====================
Adopting nlmsg_payload() in IPv4/IPv6

The commit 95d06e92a4019 ("netlink: Introduce nlmsg_payload helper")
introduced the nlmsg_payload() helper function.

This patchset aims to replace manual implementations with the
nlmsg_payload() helper in IPv4 and IPv6 files, one file per patch.
====================

Link: https://patch.msgid.link/20250415-nlmsg_v2-v1-0-a1c75d493fd7@debian.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

vxlan: Use nlmsg_payload in vxlan_vnifilter_dump

Leverage the new nlmsg_payload() helper to avoid checking for message
size and then reading the nlmsg data.

Signed-off-by: Breno Leitao <leitao@debian.org>
Reviewed-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Link: https://patch.msgid.link/20250415-nlmsg_v2-v1-8-a1c75d493fd7@debian.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

ipv4: Use nlmsg_payload in ipmr file

Leverage the new nlmsg_payload() helper to avoid checking for message
size and then reading the nlmsg data.

Signed-off-by: Breno Leitao <leitao@debian.org>
Reviewed-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Link: https://patch.msgid.link/20250415-nlmsg_v2-v1-7-a1c75d493fd7@debian.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

ipv4: Use nlmsg_payload in route file

Leverage the new nlmsg_payload() helper to avoid checking for message
size and then reading the nlmsg data.

Signed-off-by: Breno Leitao <leitao@debian.org>
Reviewed-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Link: https://patch.msgid.link/20250415-nlmsg_v2-v1-6-a1c75d493fd7@debian.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

ipv4: Use nlmsg_payload in fib_frontend file

Leverage the new nlmsg_payload() helper to avoid checking for message
size and then reading the nlmsg data.

Signed-off-by: Breno Leitao <leitao@debian.org>
Reviewed-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Link: https://patch.msgid.link/20250415-nlmsg_v2-v1-5-a1c75d493fd7@debian.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

ipv4: Use nlmsg_payload in devinet file

Leverage the new nlmsg_payload() helper to avoid checking for message
size and then reading the nlmsg data.

Signed-off-by: Breno Leitao <leitao@debian.org>
Reviewed-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Link: https://patch.msgid.link/20250415-nlmsg_v2-v1-4-a1c75d493fd7@debian.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

ipv6: Use nlmsg_payload in route file

Leverage the new nlmsg_payload() helper to avoid checking for message
size and then reading the nlmsg data.

Signed-off-by: Breno Leitao <leitao@debian.org>
Reviewed-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Link: https://patch.msgid.link/20250415-nlmsg_v2-v1-3-a1c75d493fd7@debian.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

ipv6: Use nlmsg_payload in addrconf file

Leverage the new nlmsg_payload() helper to avoid checking for message
size and then reading the nlmsg data.

Signed-off-by: Breno Leitao <leitao@debian.org>
Reviewed-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Link: https://patch.msgid.link/20250415-nlmsg_v2-v1-2-a1c75d493fd7@debian.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

ipv6: Use nlmsg_payload in addrlabel file

Leverage the new nlmsg_payload() helper to avoid checking for message
size and then reading the nlmsg data.

This changes function ip6addrlbl_valid_get_req() and
ip6addrlbl_valid_dump_req().

Signed-off-by: Breno Leitao <leitao@debian.org>
Reviewed-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Link: https://patch.msgid.link/20250415-nlmsg_v2-v1-1-a1c75d493fd7@debian.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

Merge branch 'net-stmmac-sti-cleanups'

Russell King says:

====================
net: stmmac: sti cleanups

Clean up the STI platform glue code.

- IS_PHY_IF_MODE_RGMII() is just a duplicate for
  phy_interface_mode_is_rgmii(), so use the generic version that we
  already have.

- add init/exit functions that call clk_prepare_enable(),
  sti_dwmac_set_mode() and clk_disable_unprepare() as appropriate,
  converting to devm_stmmac_pltfr_probe().

- the custom suspend/resume ops do basically what the generic ones
  do with init/exit functions populated, but also add runtime and
  noirq ops. Update STI to use the generic ops.
====================

Link: https://patch.msgid.link/Z_6Mfx_SrionoU-e@shell.armlinux.org.uk
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

net: stmmac: sti: convert to stmmac_pltfr_pm_ops

As we now have the plat_dat->init()/plat_dat->exit() populated which
have the required functionality on suspend/resume, we can now use
stmmac_pltfr_pm_ops which has methods that call these two functions.
Switch over to use this.

Doing so also fills in the runtime PM ops and _noirq variants as well.

Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Link: https://patch.msgid.link/E1u4jMo-000rCS-6f@rmk-PC.armlinux.org.uk
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

net: stmmac: sti: convert to devm_stmmac_pltfr_probe()

Convert sti to use the generic devm_stmmac_pltfr_probe() which will
call plat_dat->init()/plat_dat->exit() as appropriate, thus
simplifying the code.

Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Link: https://patch.msgid.link/E1u4jMj-000rCM-31@rmk-PC.armlinux.org.uk
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

net: stmmac: sti: use phy_interface_mode_is_rgmii()

Replace the custom IS_PHY_IF_MODE_RGMII() macro with our generic
phy_interface_mode_is_rgmii() inline function.

Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Link: https://patch.msgid.link/E1u4jMd-000rCG-VU@rmk-PC.armlinux.org.uk
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

docs: networking: clarify intended audience of netdevices.rst

The netdevices doc is dangerously broad. At least make it clear
that it's intended for developers, not for users.

Acked-by: Stanislav Fomichev <sdf@fomichev.me>
Link: https://patch.msgid.link/20250415172653.811147-1-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

net: phy: mediatek: init val in .phy_led_polarity_set for AN7581

Fix smatch warning for uninitialised val in .phy_led_polarity_set for
AN7581 driver.

Correctly init to 0 to set polarity high by default.

Reported-by: Simon Horman <horms@kernel.org>
Fixes: 6a325aed130b ("net: phy: mediatek: add Airoha PHY ID to SoC driver")
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
Link: https://patch.msgid.link/20250415105313.3409-1-ansuelsmth@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

net: stmmac: sun8i: use stmmac_pltfr_probe()

Using stmmac_pltfr_probe() simplifies the probe function. This will not
only call plat_dat->init (sun8i_dwmac_init), but also plat_dat->exit
(sun8i_dwmac_exit) appropriately if stmmac_dvr_probe() fails. This
results in an overall simplification of the glue driver.

Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Reviewed-by: Maxime Chevallier <maxime.chevallier@bootlin.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Tested-by: Corentin Labbe <clabbe.montjoie@gmail.com>
Link: https://patch.msgid.link/E1u4dKb-000dV7-3B@rmk-PC.armlinux.org.uk
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

net: stmmac: intel: remove unnecessary setting max_speed

Phylink will already limit the MAC speed according to the interface,
so if 2500BASE-X is selected, the maximum speed will be 2.5G.
Similarly, if SGMII is selected, the maximum speed will be 1G.
It is, therefore, not necessary to set a speed limit. Remove setting
plat_dat->max_speed from this glue driver.

Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Reviewed-by: Maxime Chevallier <maxime.chevallier@bootlin.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Link: https://patch.msgid.link/E1u4dIh-000dT5-Kt@rmk-PC.armlinux.org.uk
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

eth: bnxt: add support rx side device memory TCP

Currently, bnxt_en driver satisfies the requirements of the Device
memory TCP, which is HDS.
So, it implements rx-side Device memory TCP for bnxt_en driver.
It requires only converting the page API to netmem API.
`struct page` of agg rings are changed to `netmem_ref netmem` and
corresponding functions are changed to a variant of netmem API.

It also passes PP_FLAG_ALLOW_UNREADABLE_NETMEM flag to a parameter of
page_pool.
The netmem will be activated only when a user requests devmem TCP.

When netmem is activated, received data is unreadable and netmem is
disabled, received data is readable.
But drivers don't need to handle both cases because netmem core API will
handle it properly.
So, using proper netmem API is enough for drivers.

Device memory TCP can be tested with
tools/testing/selftests/drivers/net/hw/ncdevmem.
This is tested with BCM57504-N425G and firmware version 232.0.155.8/pkg
232.1.132.8.

Reviewed-by: Mina Almasry <almasrymina@google.com>
Tested-by: David Wei <dw@davidwei.uk>
Signed-off-by: Taehee Yoo <ap420073@gmail.com>
Link: https://patch.msgid.link/20250415052458.1260575-1-ap420073@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

net: lan743x: Allocate rings outside ZONE_DMA

The driver allocates ring elements using GFP_DMA flags. There is
no dependency from LAN743x hardware on memory allocation should be
in DMA_ZONE. Hence modifying the flags to use only GFP_ATOMIC. This
is consistent with other callers of lan743x_rx_init_ring_element().

Reported-by: Zhang, Liyin(CN) <Liyin.Zhang.CN@windriver.com>
Signed-off-by: Thangaraj Samynathan <thangaraj.s@microchip.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://patch.msgid.link/20250415044509.6695-1-thangaraj.s@microchip.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

Merge branch 'net-ethernet-ti-am65-cpsw-fix-mac-address-fetching'

Michael Walle says:

====================
net: ethernet: ti: am65-cpsw: Fix MAC address fetching

MAC addresses can be fetched from a NVMEM device. of_get_mac_address()
will return EPROBE_DEFER if that device is not available yet. That
isn't handled correctly by the driver and it will always fall back
to either a random MAC address or it's own "fetch by fuse" method.

Also, if the ethernet (sub)node has a link to the nvmem device,
it will fail to create a device link as the fwnode parameter isn't
populated. That's fixed in the first patch.
====================

Link: https://patch.msgid.link/20250414084336.4017237-1-mwalle@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

net: ethernet: ti: am65-cpsw: handle -EPROBE_DEFER

of_get_mac_address() might fetch the MAC address from NVMEM and that
driver might not have been loaded. In that case, -EPROBE_DEFER is
returned. Right now, this will trigger an immediate fallback to
am65_cpsw_am654_get_efuse_macid() possibly resulting in a random MAC
address although the MAC address is stored in the referenced NVMEM.

Fix it by handling the -EPROBE_DEFER return code correctly. This also
means that the creation of the MDIO device has to be moved to a later
stage as -EPROBE_DEFER must not be returned after child devices are
created.

Signed-off-by: Michael Walle <mwalle@kernel.org>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Link: https://patch.msgid.link/20250414084336.4017237-3-mwalle@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

net: ethernet: ti: am65-cpsw: set fwnode for ports

fwnode needs to be set for a device for fw_devlink to be able to
track/enforce its dependencies correctly. Without this, you'll see error
messages like this when the supplier has probed and tries to make sure
all its fwnode consumers are linked to it using device links:

am65-cpsw-nuss 8000000.ethernet: Failed to create device link (0x180) with supplier ..

Reviewed-by: Saravana Kannan <saravanak@google.com>
Signed-off-by: Michael Walle <mwalle@kernel.org>
Link: https://patch.msgid.link/20250414084336.4017237-2-mwalle@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

Merge branch 'net-ptp-driver-opt-in-for-supported-ptp-ioctl-flags'

Jacob Keller says:

====================
net: ptp: driver opt-in for supported PTP ioctl flags

Both the PTP_EXTTS_REQUEST(2) and PTP_PEROUT_REQUEST(2) ioctls take flags
from userspace to modify their behavior. Drivers are supposed to check
these flags, rejecting requests for flags they do not support.

Many drivers today do not check these flags, despite many attempts to
squash individual drivers as these mistakes are discovered. Additionally,
any new flags added can require updating every driver if their validation
checks are poorly implemented.

It is clear that driver authors will not reliably check for unsupported
flags. The root of the issue is that drivers must essentially opt out of
every flag, rather than opt in to the ones they support.

Instead, lets introduce .supported_perout_flags and .supported_extts_flags
to the ptp_clock_info structure. This is a pattern taken from several
ethtool ioctls which enabled validation to move out of the drivers and into
the shared ioctl handlers. This pattern has worked quite well and makes it
much more difficult for drivers to accidentally accept flags they do not
support.

With this approach, drivers which do not set the supported fields will have
the core automatically reject any request which has flags. Drivers must opt
in to each flag they support by adding it to the list, with the sole
exception being the PTP_ENABLE_FEATURE flag of the PTP_EXTTS_REQUEST ioctl
since it is entirely handled by the ptp_chardev.c file.

This change will ensure that all current and future drivers are safe for
extension when we need to extend these ioctls.

I opted to keep all the driver changes into one patch per ioctl type. The
changes are relatively small and straight forward. Splitting it per-driver
would make the series large, and also break flags between the introduction
of the supported field and setting it in each driver.

The non-Intel drivers are compile-tested only, and I would appreciate
confirmation and testing from their respective maintainers. (It is also
likely that I missed some of the driver authors especially for drivers
which didn't make any checks at all and do not set either of the supported
flags yet)

v1: https://lore.kernel.org/20250408-jk-supported-perout-flags-v1-0-d2f8e3df64f3@intel.com
====================

Link: https://patch.msgid.link/20250414-jk-supported-perout-flags-v2-0-f6b17d15475c@intel.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

net: ptp: introduce .supported_perout_flags to ptp_clock_info

The PTP_PEROUT_REQUEST2 ioctl has gained support for flags specifying
specific output behavior including PTP_PEROUT_ONE_SHOT,
PTP_PEROUT_DUTY_CYCLE, PTP_PEROUT_PHASE.

Driver authors are notorious for not checking the flags of the request.
This results in misinterpreting the request, generating an output signal
that does not match the requested value. It is anticipated that even more
flags will be added in the future, resulting in even more broken requests.

Expecting these issues to be caught during review or playing whack-a-mole
after the fact is not a great solution.

Instead, introduce the supported_perout_flags field in the ptp_clock_info
structure. Update the core character device logic to explicitly reject any
request which has a flag not on this list.

This ensures that drivers must 'opt in' to the flags they support. Drivers
which don't set the .supported_perout_flags field will not need to check
that unsupported flags aren't passed, as the core takes care of this.

Update the drivers which do support flags to set this new field.

Note the following driver files set n_per_out to a non-zero value but did
not check the flags at all:

 • drivers/ptp/ptp_clockmatrix.c
 • drivers/ptp/ptp_idt82p33.c
 • drivers/ptp/ptp_fc3.c
 • drivers/net/ethernet/ti/am65-cpts.c
 • drivers/net/ethernet/aquantia/atlantic/aq_ptp.c
 • drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.c
 • drivers/net/dsa/sja1105/sja1105_ptp.c
 • drivers/net/ethernet/freescale/dpaa2/dpaa2-ptp.c
 • drivers/net/ethernet/mscc/ocelot_vsc7514.c
 • drivers/net/ethernet/intel/i40e/i40e_ptp.c

Reviewed-by: Vadim Fedorenko <vadim.fedorenko@linux.dev>
Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Reviewed-by: Kory Maincent <kory.maincent@bootlin.com>
Link: https://patch.msgid.link/20250414-jk-supported-perout-flags-v2-2-f6b17d15475c@intel.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

net: ptp: introduce .supported_extts_flags to ptp_clock_info

The PTP_EXTTS_REQUEST(2) ioctl has a flags field which specifies how the
external timestamp request should behave. This includes which edge of the
signal to timestamp, as well as a specialized "offset" mode. It is expected
that more flags will be added in the future.

Driver authors routinely do not check the flags, often accepting requests
with flags which they do not support. Even drivers which do check flags may
not be future-proofed to reject flags not yet defined. Thus, any future
flag additions often require manually updating drivers to reject these
flags.

This approach of hoping we catch flag checks during review, or playing
whack-a-mole after the fact is the wrong approach.

Introduce the "supported_extts_flags" field to the ptp_clock_info
structure. This field defines the set of flags the device actually
supports.

Update the core character device logic to check this field and reject
unsupported requests. Getting this right is somewhat tricky. First, to
avoid unnecessary repetition and make basic functionality work when
.supported_extts_flags is 0, the core always accepts the PTP_ENABLE_FEATURE
flag. This flag is used to set the 'on' parameter to the .enable function
and is thus always 'supported' by all drivers.

For backwards compatibility, the PTP_RISING_EDGE and PTP_FALLING_EDGE flags
are merely "hints" when using the old PTP_EXTTS_REQUEST ioctl, and are not
expected to be enforced. If the user issues PTP_EXTTS_REQUEST2, the
PTP_STRICT_FLAGS flag is added which is supposed to inform the driver to
strictly validate the flags and reject unsupported requests. To handle
this, first check if the driver reports PTP_STRICT_FLAGS support. If it
does not, then always allow the PTP_RISING_EDGE and PTP_FALLING_EDGE flags.
This keeps backwards compatibility with the original PTP_EXTTS_REQUEST
ioctl where these flags are not guaranteed to be honored.

This way, drivers which do not set the supported_extts_flags will continue
to accept requests for the original PTP_EXTTS_REQUEST ioctl. The core will
automatically reject requests with new flags, and correctly reject requests
with PTP_STRICT_FLAGS, where the driver is supposed to strictly validate
the flags.

Update the various drivers, refactoring their validation logic into the
.supported_extts_flags field. For consistency and readability,
PTP_ENABLE_FEATURE is not set in the supported flags list, and
PTP_EXTTS_EDGES is expanded to PTP_RISING_EDGE | PTP_FALLING_EDGE in all
cases.

Note the following driver files set n_ext_ts to a non-zero value but did
not check flags at all:

 • drivers/net/ethernet/freescale/dpaa2/dpaa2-ptp.c
 • drivers/net/ethernet/freescale/enetc/enetc_ptp.c
 • drivers/net/ethernet/intel/i40e/i40e_ptp.c
 • drivers/net/ethernet/marvell/octeontx2/nic/otx2_ptp.c
 • drivers/net/ethernet/renesas/ravb_ptp.c
 • drivers/net/ethernet/renesas/rtsn.c
 • drivers/net/ethernet/renesas/rtsn.h
 • drivers/net/ethernet/ti/am65-cpts.c
 • drivers/net/ethernet/ti/cpts.h
 • drivers/net/ethernet/ti/icssg/icss_iep.c
 • drivers/net/ethernet/xscale/ptp_ixp46x.c
 • drivers/net/phy/bcm-phy-ptp.c
 • drivers/ptp/ptp_ocp.c
 • drivers/ptp/ptp_pch.c
 • drivers/ptp/ptp_qoriq.c

These drivers behavior does change slightly: they will now reject the
PTP_EXTTS_REQUEST2 ioctl, because they do not strictly validate their
flags. This also makes them no longer incorrectly accept PTP_EXT_OFFSET.

Also note that the renesas ravb driver does not support PTP_STRICT_FLAGS.
We could leave the .supported_extts_flags as 0, but I added the
PTP_RISING_EDGE | PTP_FALLING_EDGE since the driver previously manually
validated these flags. This is equivalent to 0 because the core will allow
these flags regardless unless PTP_STRICT_FLAGS is also set.

Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Reviewed-by: Kory Maincent <kory.maincent@bootlin.com>
Link: https://patch.msgid.link/20250414-jk-supported-perout-flags-v2-1-f6b17d15475c@intel.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

net: bridge: locally receive all multicast packets if IFF_ALLMULTI is set

If multicast snooping is enabled, multicast packets may not always end up
on the local bridge interface, if the host is not a member of the multicast
group. Similar to how IFF_PROMISC allows all packets to be received
locally, let IFF_ALLMULTI allow all multicast packets to be received.

OpenWrt uses a user space daemon for DHCPv6/RA/NDP handling, and in relay
mode it sets the ALLMULTI flag in order to receive all relevant queries on
the network.

This works for normal network interfaces and non-snooping bridges, but not
snooping bridges (unless multicast routing is enabled).

Reported-by: Felix Fietkau <nbd@nbd.name>
Closes: https://github.com/openwrt/openwrt/issues/15857#issuecomment-2662851243
Signed-off-by: Shengyu Qu <wiagn233@outlook.com>
Reviewed-by: Ido Schimmel <idosch@nvidia.com>
Acked-by: Nikolay Aleksandrov <razor@blackwall.org>
Link: https://patch.msgid.link/OSZPR01MB8434308370ACAFA90A22980798B32@OSZPR01MB8434.jpnprd01.prod.outlook.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

net: phy: remove redundant dependency on NETDEVICES for PHYLINK and PHYLIB

drivers/net/phy/Kconfig is included from drivers/net/Kconfig in an
"if NETDEVICES" section. Therefore we don't have to duplicate the
dependency here. And if e.g. PHYLINK is selected somewhere, then the
dependency is ignored anyway (see note in Kconfig help).

Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Reviewed-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Link: https://patch.msgid.link/085892cd-aa11-4c22-bf8a-574a5c6dcd7c@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

tc: Return an error if filters try to attach too many actions

While developing the fix for the buffer sizing issue in [0], I noticed
that the kernel will happily accept a long list of actions for a filter,
and then just silently truncate that list down to a maximum of 32
actions.

That seems less than ideal, so this patch changes the action parsing to
return an error message and refuse to create the filter in this case.
This results in an error like:

 # ip link add type veth
 # tc qdisc replace dev veth0 root handle 1: fq_codel
 # tc -echo filter add dev veth0 parent 1: u32 match u32 0 0 $(for i in $(seq 33); do echo action pedit munge ip dport set 22; done)
Error: Only 32 actions supported per filter.
We have an error talking to the kernel

Instead of just creating a filter with 32 actions and dropping the last
one.

This is obviously a change in UAPI. But seeing as creating more than 32
filters has never actually *worked*, it seems that returning an explicit
error is better, and any use cases that get broken by this were already
broken just in more subtle ways.

[0] https://lore.kernel.org/r/20250407105542.16601-1-toke@redhat.com

Signed-off-by: Toke Høiland-Jørgensen <toke@redhat.com>
Link: https://patch.msgid.link/20250409145523.164506-1-toke@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

octeon_ep_vf: Remove octep_vf_wq

commit cb7dd712189f ("octeon_ep_vf: Add driver framework and device
initialization") added octep_vf_wq but it has never been used. Remove it.

Reported-by: Dr. David Alan Gilbert <linux@treblig.org>
Closes: https://lore.kernel.org/netdev/Z70bEoTKyeBau52q@gallifrey/
Signed-off-by: Simon Horman <horms@kernel.org>
Reviewed-by: Dr. David Alan Gilbert <linux@treblig.org>
Link: https://patch.msgid.link/20250414-octeon-wq-v1-1-23700e4bd208@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

Merge branch 'net-stmmac-ingenic-cleanups'

Russell King says:

====================
net: stmmac: ingenic: cleanups

Another series for another stmmac glue platform.

Convert Ingenic to use the stmmac platform PM ops and the
devm_stmmac_pltfr_probe() helper.
====================

Link: https://patch.msgid.link/Z_0u9pA0Ziop-BuU@shell.armlinux.org.uk
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

net: stmmac: ingenic: convert to devm_stmmac_pltfr_probe()

As Ingenic now uses the stmmac platform PM ops, convert it to use
devm_stmmac_pltfr_probe() which will call the plat_dat->init() method
before stmmac_drv_probe() and appropriately cleaning up via the
->exit() method, thus simplifying the code. Using the devm_*()
variant also allows removal of the explicit call to
stmmac_pltfr_remove().

Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Link: https://patch.msgid.link/E1u4M5S-000YGJ-9K@rmk-PC.armlinux.org.uk
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

net: stmmac: ingenic: convert to stmmac_pltfr_pm_ops

Convert the Ingenic glue driver to use the generic stmmac platform
power management operations.

In order to do this, we need to make ingenic_mac_init() arguments
compatible with plat_dat->init() by adding a plat_dat member to struct
ingenic_mac. This allows the custom suspend/resume operations to be
removed, and the PM ops pointer replaced with stmmac_pltfr_pm_ops.

This will adds runtime PM and noirq suspend/resume ops to this driver.

Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Link: https://patch.msgid.link/E1u4M5N-000YGD-5i@rmk-PC.armlinux.org.uk
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

net: stmmac: imx: use stmmac_pltfr_probe()

Using stmmac_pltfr_probe() simplifies the probe function. This will not
only call plat_dat->init (imx_dwmac_init), but also plat_dat->exit
(imx_dwmac_exit) appropriately if stmmac_dvr_probe() fails. This
results in an overall simplification of the glue driver.

Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Link: https://patch.msgid.link/E1u4Flp-000XlM-Tb@rmk-PC.armlinux.org.uk
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

Merge branch 'net-stmmac-anarion-cleanups'

Russell King says:

====================
net: stmmac: anarion: cleanups

A series of cleanups to the anarion glue driver.

Clean up anarion_config_dt() error handling, printing a human readable
error rather than the numeric errno, and use ERR_CAST().

Using a switch statement with incorrect "fallthrough;" for RGMII vs
non-RGMII is unnecessary when we have phy_interface_mode_is_rgmii().
Convert to use the helper.

Use stmmac_pltfr_probe() rahter than open-coding the call to the
init function (which stmmac_pltfr_probe() will do for us.)

Finally, convert to use devm_stmmac_pltfr_probe() which allows the
removal of the .remove initialiser in the driver structure.

Not tested on hardware.
====================

Link: https://patch.msgid.link/Z_zP9BvZlqeq3Ssl@shell.armlinux.org.uk
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

net: stmmac: anarion: use devm_stmmac_pltfr_probe()

Convert anarion to use devm_stmmac_pltfr_probe() which allows the
removal of an explicit call to stmmac_pltfr_remove().

Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Link: https://patch.msgid.link/E1u4Flf-000XjS-Fi@rmk-PC.armlinux.org.uk
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

net: stmmac: anarion: use stmmac_pltfr_probe()

Rather than open-coding the call to anarion_gmac_init() and then
stmmac_dvr_probe(), omitting the cleanup of calling
anarion_gmac_exit(), use stmmac_pltfr_probe() which will handle this
for us.

Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Link: https://patch.msgid.link/E1u4Fla-000XjM-Bw@rmk-PC.armlinux.org.uk
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

net: stmmac: anarion: clean up interface parsing

anarion_config_dt() used a switch statement to check for the RGMII
modes, complete with an unnecessary "fallthrough", and also printed
the numerical value of the PHY interface mode on error. Clean this
up using the phy_interface_mode_is_rgmii() helper, and print the
English version of the PHY interface mode on error.

Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Link: https://patch.msgid.link/E1u4FlV-000XjG-83@rmk-PC.armlinux.org.uk
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

net: stmmac: anarion: clean up anarion_config_dt() error handling

When enabled, print a user friendly description of the error when
failing to ioremap() the control resource, and use ERR_CAST() when
propagating the error. This allows us to get rid of the "err" local
variable in anarion_config_dt().

Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Link: https://patch.msgid.link/E1u4FlQ-000XjA-2V@rmk-PC.armlinux.org.uk
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

Merge branch 'net-introduce-nlmsg_payload-helper'

Breno Leitao says:

====================
net: Introduce nlmsg_payload helper

In the current codebase, there are multiple instances where the
structure size is checked before assigning it to a Netlink message. This
check is crucial for ensuring that the structure is correctly mapped
onto the Netlink message, providing a layer of security.

To streamline this process, Jakub Kicinski suggested creating a helper
function, `nlmsg_payload`, which verifies if the structure fits within
the message. If it does, the function returns the data; otherwise, it
returns NULL. This approach simplifies the code and reduces redundancy.

This patchset introduces the `nlmsg_payload` helper and updates several
parts of the code to use it. Further updates will follow in subsequent
patchsets.

v1: https://lore.kernel.org/20250411-nlmsg-v1-0-ddd4e065cb15@debian.org
====================

Link: https://patch.msgid.link/20250414-nlmsg-v2-0-3d90cb42c6af@debian.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

net: fib_rules: Use nlmsg_payload in fib_{new,del}rule()

Leverage the new nlmsg_payload() helper to avoid checking for message
size and then reading the nlmsg data.

Suggested-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Signed-off-by: Breno Leitao <leitao@debian.org>
Reviewed-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Link: https://patch.msgid.link/20250414-nlmsg-v2-10-3d90cb42c6af@debian.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

net: fib_rules: Use nlmsg_payload in fib_valid_dumprule_req

Leverage the new nlmsg_payload() helper to avoid checking for message
size and then reading the nlmsg data.

Signed-off-by: Breno Leitao <leitao@debian.org>
Reviewed-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Link: https://patch.msgid.link/20250414-nlmsg-v2-9-3d90cb42c6af@debian.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

mpls: Use nlmsg_payload in mpls_valid_getroute_req

Leverage the new nlmsg_payload() helper to avoid checking for message
size and then reading the nlmsg data.

Signed-off-by: Breno Leitao <leitao@debian.org>
Reviewed-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Link: https://patch.msgid.link/20250414-nlmsg-v2-8-3d90cb42c6af@debian.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

ipv6: Use nlmsg_payload in inet6_rtm_valid_getaddr_req

Leverage the new nlmsg_payload() helper to avoid checking for message
size and then reading the nlmsg data.

Signed-off-by: Breno Leitao <leitao@debian.org>
Reviewed-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Link: https://patch.msgid.link/20250414-nlmsg-v2-7-3d90cb42c6af@debian.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

ipv6: Use nlmsg_payload in inet6_valid_dump_ifaddr_req

Leverage the new nlmsg_payload() helper to avoid checking for message
size and then reading the nlmsg data.

Signed-off-by: Breno Leitao <leitao@debian.org>
Reviewed-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Link: https://patch.msgid.link/20250414-nlmsg-v2-6-3d90cb42c6af@debian.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

mpls: Use nlmsg_payload in mpls_valid_fib_dump_req

Leverage the new nlmsg_payload() helper to avoid checking for message
size and then reading the nlmsg data.

Signed-off-by: Breno Leitao <leitao@debian.org>
Reviewed-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Link: https://patch.msgid.link/20250414-nlmsg-v2-5-3d90cb42c6af@debian.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

rtnetlink: Use nlmsg_payload in valid_fdb_dump_strict

Leverage the new nlmsg_payload() helper to avoid checking for message
size and then reading the nlmsg data.

Signed-off-by: Breno Leitao <leitao@debian.org>
Reviewed-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Link: https://patch.msgid.link/20250414-nlmsg-v2-4-3d90cb42c6af@debian.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

neighbour: Use nlmsg_payload in neigh_valid_get_req

Update neigh_valid_get_req function to utilize the new nlmsg_payload()
helper function.

This change improves code clarity and safety by ensuring that the
Netlink message payload is properly validated before accessing its data.

Signed-off-by: Breno Leitao <leitao@debian.org>
Reviewed-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Link: https://patch.msgid.link/20250414-nlmsg-v2-3-3d90cb42c6af@debian.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

neighbour: Use nlmsg_payload in neightbl_valid_dump_info

Update neightbl_valid_dump_info function to utilize the new
nlmsg_payload() helper function.

This change improves code clarity and safety by ensuring that the
Netlink message payload is properly validated before accessing its data.

Signed-off-by: Breno Leitao <leitao@debian.org>
Reviewed-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Link: https://patch.msgid.link/20250414-nlmsg-v2-2-3d90cb42c6af@debian.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

netlink: Introduce nlmsg_payload helper

Create a new helper function, nlmsg_payload(), to simplify checking and
retrieving Netlink message payloads.

This reduces boilerplate code for users who need to verify the message
length before accessing its data.

Suggested-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Breno Leitao <leitao@debian.org>
Reviewed-by: Jakub Kicinski <kuba@kernel.org>
Reviewed-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Link: https://patch.msgid.link/20250414-nlmsg-v2-1-3d90cb42c6af@debian.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

Merge branch 'qed-deadcoding'

Dr. David Alan Gilbert says:

====================
qed deadcoding

  This is a set of deadcode removals for the qed ethernet
device.  I've tried to avoid removing anything that
are trivial firmware wrappers.

  One odd one I've not removed is qed_bw_update(),
it doesn't seem to be called but looks like the only
caller of the bw_update(..) method which qedf does
define.  Perhaps qed_bw_update is supposed to be called
somewhere?
====================

Link: https://patch.msgid.link/20250414005247.341243-1-linux@treblig.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

qed: Remove unused qed_db_recovery_dp

qed_db_recovery_dp() was added in 2018 as part of
commit 36907cd5cd72 ("qed: Add doorbell overflow recovery mechanism")
but has remained unused.

Remove it.

Signed-off-by: Dr. David Alan Gilbert <linux@treblig.org>
Link: https://patch.msgid.link/20250414005247.341243-6-linux@treblig.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

qed: Remove unused qed_print_mcp_trace_*

While most of the trace code is reachable by other routes
(I think mostly via the qed_features_lookup[] array), there
are a couple of unused wrappers.

qed_print_mcp_trace_line() and qed_print_mcp_trace_results_cont()
were added in 2018 as part of
commit a3f723079df8 ("qed*: Utilize FW 8.37.7.0")
but have remained unused.

Remove them.

Signed-off-by: Dr. David Alan Gilbert <linux@treblig.org>
Link: https://patch.msgid.link/20250414005247.341243-5-linux@treblig.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

qed: Remove unused qed_ptt_invalidate

qed_ptt_invalidate() was added in 2015 as part of
commit fe56b9e6a8d9 ("qed: Add module with basic common support")
but has remained unused.

Remove it.

Signed-off-by: Dr. David Alan Gilbert <linux@treblig.org>
Link: https://patch.msgid.link/20250414005247.341243-4-linux@treblig.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

qed: Remove unused qed_calc_*_ctx_validation functions

qed_calc_session_ctx_validation() and qed_calc_task_ctx_validation()
were added as part of 2017's
commit da09091732ae ("qed*: Utilize FW 8.33.1.0")

but have remained unused.

Remove them.

This leaves; con_region_offsets[], task_region_offsets[],
cdu_crc8_table and qed_calc_cdu_validation_byte() unused.

Remove them.

Signed-off-by: Dr. David Alan Gilbert <linux@treblig.org>
Link: https://patch.msgid.link/20250414005247.341243-3-linux@treblig.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

qed: Remove unused qed_memset_*ctx functions

qed_memset_session_ctx() and qed_memset_task_ctx() were added in 2017
as part of
commit da09091732ae ("qed*: Utilize FW 8.33.1.0")

but have not been used.

Remove them.

Signed-off-by: Dr. David Alan Gilbert <linux@treblig.org>
Link: https://patch.msgid.link/20250414005247.341243-2-linux@treblig.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

net: phy: remove device_phy_find_device

AFAICS this function has never had a user.

Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Link: https://patch.msgid.link/ab7b8094-2eea-4e82-a047-fd60117f220b@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

Merge branch 'mptcp-various-small-and-unrelated-improvements'

Matthieu Baerts says:

====================
mptcp: various small and unrelated improvements

Here are various unrelated patches:

- Patch 1: sched: remove unused structure.

- Patch 2: sched: split the validation part, a preparation for later.

- Patch 3: pm: clarify code, not to think there is a possible UaF.
  Note: a previous version has already been sent individually to Netdev.

- Patch 4: subflow: simplify subflow_hmac_valid by passing subflow_req.

- Patch 5: mib: add counter for MPJoin rejected by the PM.

- Patch 6: selftests: validate this new MPJoinRejected counter.

- Patch 7: selftests: define nlh variable only where needed.

- Patch 8: selftests: show how to use IPPROTO_MPTCP with getaddrinfo.
  Note: a previous version has already been sent individually to Netdev.

v1: https://lore.kernel.org/20250411-net-next-mptcp-sched-mib-sft-misc-v1-0-85ac8c6654c3@kernel.org
====================

Link: https://patch.msgid.link/20250413-net-next-mptcp-sched-mib-sft-misc-v2-0-0f83a4350150@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

selftests: mptcp: use IPPROTO_MPTCP for getaddrinfo

mptcp_connect.c is a startup tutorial of MPTCP programming, however
there is a lack of ai_protocol(IPPROTO_MPTCP) usage. Add comment for
getaddrinfo MPTCP support.

This patch first uses IPPROTO_MPTCP to get addrinfo, and if glibc
version is too old, it falls back to using IPPROTO_TCP.

Co-developed-by: Geliang Tang <geliang@kernel.org>
Signed-off-by: Geliang Tang <geliang@kernel.org>
Signed-off-by: zhenwei pi <pizhenwei@bytedance.com>
Reviewed-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Link: https://patch.msgid.link/20250413-net-next-mptcp-sched-mib-sft-misc-v2-8-0f83a4350150@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

selftests: mptcp: diag: drop nlh parameter of recv_nlmsg

It's strange that 'nlh' variable is set to NULL in get_mptcpinfo() and then
this NULL pointer is passed to recv_nlmsg(). In fact, this variable should
be defined in recv_nlmsg(), not get_mptcpinfo().

So this patch drops this useless 'nlh' parameter of recv_nlmsg() and define
'nlh' variable in recv_nlmsg().

Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
Reviewed-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Link: https://patch.msgid.link/20250413-net-next-mptcp-sched-mib-sft-misc-v2-7-0f83a4350150@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

selftests: mptcp: validate MPJoinRejected counter

The parent commit adds this new counter, incremented when receiving a
connection request, if the PM didn't allow the creation of new subflows.

Most of the time, it is then kept at 0, except when the PM limits cause
the receiver side to reject new MPJoin connections. This is the case in
the following tests:

 - single subflow, limited by server
 - multiple subflows, limited by server
 - subflows limited by server w cookies
 - userspace pm type rejects join
 - userspace pm type prevents mp_prio

Simply set join_syn_rej=1 when checking the MPJoin counters for these
tests.

Reviewed-by: Geliang Tang <geliang@kernel.org>
Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Link: https://patch.msgid.link/20250413-net-next-mptcp-sched-mib-sft-misc-v2-6-0f83a4350150@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

mptcp: add MPJoinRejected MIB counter

This counter is useful to understand why some paths are rejected, and
not created as expected.

It is incremented when receiving a connection request, if the PM didn't
allow the creation of new subflows.

Reviewed-by: Geliang Tang <geliang@kernel.org>
Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Link: https://patch.msgid.link/20250413-net-next-mptcp-sched-mib-sft-misc-v2-5-0f83a4350150@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

mptcp: pass right struct to subflow_hmac_valid

subflow_hmac_valid() needs to access the MPTCP socket and the subflow
request, but not the request sock that is passed in argument.

Instead, the subflow request can be directly passed to avoid getting it
via an additional cast.

Reviewed-by: Geliang Tang <geliang@kernel.org>
Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Link: https://patch.msgid.link/20250413-net-next-mptcp-sched-mib-sft-misc-v2-4-0f83a4350150@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

mptcp: pm: Return local variable instead of freed pointer

Commit e4c28e3d5c090 ("mptcp: pm: move generic PM helpers to pm.c")
removed an unnecessary if-check, which resulted in returning a freed
pointer.

This still works due to the implicit boolean conversion when returning
the freed pointer from mptcp_remove_anno_list_by_saddr(), but it can be
confusing and potentially error-prone. To improve clarity, add a local
variable to explicitly return a boolean value instead.

Signed-off-by: Thorsten Blum <thorsten.blum@linux.dev>
Reviewed-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Link: https://patch.msgid.link/20250413-net-next-mptcp-sched-mib-sft-misc-v2-3-0f83a4350150@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

mptcp: sched: split validation part

A new interface .validate has been added in struct bpf_struct_ops
recently. This patch prepares a future struct_ops support by
implementing it as a new helper mptcp_validate_scheduler() for struct
mptcp_sched_ops.

In this helper, check whether the required ops "get_subflow" of struct
mptcp_sched_ops has been implemented.

Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
Reviewed-by: Mat Martineau <martineau@kernel.org>
Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Link: https://patch.msgid.link/20250413-net-next-mptcp-sched-mib-sft-misc-v2-2-0f83a4350150@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

mptcp: sched: remove mptcp_sched_data

This is a follow-up of commit b68b106b0f15 ("mptcp: sched: reduce size
for unused data"), now removing the mptcp_sched_data structure.

Now is a good time to do that, because the previously mentioned WIP work
has been updated, no longer depending on this structure.

Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Link: https://patch.msgid.link/20250413-net-next-mptcp-sched-mib-sft-misc-v2-1-0f83a4350150@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

net: txgbe: Update module description

Because of the addition of support for 25G/40G devices, update the module
description.

Signed-off-by: Jiawen Wu <jiawenwu@trustnetic.com>
Link: https://patch.msgid.link/20250414022421.375101-1-jiawenwu@trustnetic.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

ixgbe: add support for FW rollback mode

The driver should detect whether the device entered FW rollback
mode and then notify user with the dedicated message including
FW and NVM versions.

Even if the driver detected rollback mode, this should not result
in an probe error and the normal flow proceeds.

FW tries to rollback to "old" operational FW located in the
inactive NVM bank in cases when newly loaded FW exhibits faulty
behavior. If something goes wrong during boot the FW may switch
into rollback mode in an attempt to avoid recovery mode and stay
operational. After rollback is successful, the banks are swapped,
and the "rollback" bank becomes the active bank for the next reset.

Reviewed-by: Mateusz Polchlopek <mateusz.polchlopek@intel.com>
Signed-off-by: Andrii Staikov <andrii.staikov@intel.com>
Signed-off-by: Jedrzej Jagielski <jedrzej.jagielski@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>

ixgbe: add E610 implementation of FW recovery mode

Add E610 implementation of fw_recovery_mode MAC operation.

In case of E610 information about recovery mode is obtained
from FW_MODES field in IXGBE_GL_MNG_FWSM register (0x000B6134).

Introduce recovery specific probing flow and init only
vital features.

User should be able to perform NVM update using devlink
once FW error is detected in order to load a healthy img.

Reviewed-by: Mateusz Polchlopek <mateusz.polchlopek@intel.com>
Co-developed-by: Stefan Wegrzyn <stefan.wegrzyn@intel.com>
Signed-off-by: Stefan Wegrzyn <stefan.wegrzyn@intel.com>
Signed-off-by: Jedrzej Jagielski <jedrzej.jagielski@intel.com>
Tested-by: Bharath R <bharath.r@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>

ixgbe: add FW API version check

Add E610 specific function checking whether the FW API version
is compatible with the driver expectations.

The major API version should be less than or equal to the expected
API version. If not the driver won't be fully operational.

Check the minor version, and if it is more than two versions lesser
or greater than the expected version, print a message indicating
that the NVM or driver should be updated respectively.

Reviewed-by: Mateusz Polchlopek <mateusz.polchlopek@intel.com>
Co-developed-by: Piotr Kwapulinski <piotr.kwapulinski@intel.com>
Signed-off-by: Piotr Kwapulinski <piotr.kwapulinski@intel.com>
Signed-off-by: Jedrzej Jagielski <jedrzej.jagielski@intel.com>
Tested-by: Bharath R <bharath.r@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>

ixgbe: add support for devlink reload

The E610 adapters contain an embedded chip with firmware which can be
updated using devlink flash. The firmware which runs on this chip is
referred to as the Embedded Management Processor firmware (EMP
firmware).

Activating the new firmware image currently requires that the system be
rebooted. This is not ideal as rebooting the system can cause unwanted
downtime.

The EMP firmware itself can be reloaded by issuing a special update
to the device called an Embedded Management Processor reset (EMP
reset). This reset causes the device to reset and reload the EMP
firmware.

Implement support for devlink reload with the "fw_activate" flag. This
allows user space to request the firmware be activated immediately.

Reviewed-by: Mateusz Polchlopek <mateusz.polchlopek@intel.com>
Tested-by: Bharath R <bharath.r@intel.com>
Co-developed-by: Slawomir Mrozowicz <slawomirx.mrozowicz@intel.com>
Signed-off-by: Slawomir Mrozowicz <slawomirx.mrozowicz@intel.com>
Co-developed-by: Piotr Kwapulinski <piotr.kwapulinski@intel.com>
Signed-off-by: Piotr Kwapulinski <piotr.kwapulinski@intel.com>
Co-developed-by: Stefan Wegrzyn <stefan.wegrzyn@intel.com>
Signed-off-by: Stefan Wegrzyn <stefan.wegrzyn@intel.com>
Signed-off-by: Jedrzej Jagielski <jedrzej.jagielski@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>

ixgbe: add device flash update via devlink

Use the pldmfw library to implement device flash update for
the Intel ixgbe networking device driver specifically for E610 devices.
This support uses the devlink flash update interface.

Using the pldmfw library, the provided firmware file will be scanned for
the three major components, "fw.undi" for the Option ROM, "fw.mgmt" for
the main NVM module containing the primary device firmware, and
"fw.netlist" containing the netlist module.

The flash is separated into two banks, the active bank containing the
running firmware, and the inactive bank which we use for update. Each
module is updated in a staged process. First, the inactive bank is
erased, preparing the device for update. Second, the contents of the
component are copied to the inactive portion of the flash. After all
components are updated, the driver signals the device to switch the
active bank during the next EMP reset.

With this implementation, basic flash update for the E610 hardware is
supported.

Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
Tested-by: Bharath R <bharath.r@intel.com>
Co-developed-by: Slawomir Mrozowicz <slawomirx.mrozowicz@intel.com>
Signed-off-by: Slawomir Mrozowicz <slawomirx.mrozowicz@intel.com>
Co-developed-by: Piotr Kwapulinski <piotr.kwapulinski@intel.com>
Signed-off-by: Piotr Kwapulinski <piotr.kwapulinski@intel.com>
Co-developed-by: Stefan Wegrzyn <stefan.wegrzyn@intel.com>
Signed-off-by: Stefan Wegrzyn <stefan.wegrzyn@intel.com>
Signed-off-by: Jedrzej Jagielski <jedrzej.jagielski@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>

ixgbe: extend .info_get() with stored versions

Add functions reading inactive versions from the inactive flash
bank.

Print stored versions for the content present in the inactive bank.
If there's pending update the versions reflect the ones which
are going to be loaded after reload. If there's no pending update
both running and stored are the same, which means there won't
be any NVM change on reload.

Co-developed-by: Slawomir Mrozowicz <slawomirx.mrozowicz@intel.com>
Signed-off-by: Slawomir Mrozowicz <slawomirx.mrozowicz@intel.com>
Co-developed-by: Piotr Kwapulinski <piotr.kwapulinski@intel.com>
Signed-off-by: Piotr Kwapulinski <piotr.kwapulinski@intel.com>
Signed-off-by: Jedrzej Jagielski <jedrzej.jagielski@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>

ixgbe: add E610 functions getting PBA and FW ver info

Introduce 2 E610 specific callbacks implementations:
-ixgbe_start_hw_e610() which expands the regular .start_hw callback with
getting FW version information
-ixgbe_read_pba_string_e610() which gets Product Board Assembly string

Extend EEPROM ops with new .read_pba_string in order to distinguish
generic one and the E610 one.

Reviewed-by: Mateusz Polchlopek <mateusz.polchlopek@intel.com>
Tested-by: Bharath R <bharath.r@intel.com>
Co-developed-by: Stefan Wegrzyn <stefan.wegrzyn@intel.com>
Signed-off-by: Stefan Wegrzyn <stefan.wegrzyn@intel.com>
Signed-off-by: Jedrzej Jagielski <jedrzej.jagielski@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>

ixgbe: add .info_get extension specific for E610 devices

E610 devices give possibility to show more detailed info than the previous
boards.
Extend reporting NVM info with following pieces:
 fw.mgmt.api -> version number of the API
 fw.mgmt.build -> identifier of the source for the FW
 fw.mgmt.srev -> number defining FW's security revision
 fw.psid.api -> version defining the format of the flash contents
 fw.undi.srev -> number defining OROM's security revision
 fw.netlist -> version of the netlist module
 fw.netlist.build -> first 4 bytes of the netlist hash

Co-developed-by: Slawomir Mrozowicz <slawomirx.mrozowicz@intel.com>
Signed-off-by: Slawomir Mrozowicz <slawomirx.mrozowicz@intel.com>
Co-developed-by: Piotr Kwapulinski <piotr.kwapulinski@intel.com>
Signed-off-by: Piotr Kwapulinski <piotr.kwapulinski@intel.com>
Signed-off-by: Jedrzej Jagielski <jedrzej.jagielski@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>

ixgbe: read the netlist version information

Add functions reading the netlist version info and use them
as a part of the setting NVM info procedure.

Reviewed-by: Mateusz Polchlopek <mateusz.polchlopek@intel.com>
Tested-by: Bharath R <bharath.r@intel.com>
Signed-off-by: Slawomir Mrozowicz <slawomirx.mrozowicz@intel.com>
Co-developed-by: Piotr Kwapulinski <piotr.kwapulinski@intel.com>
Signed-off-by: Piotr Kwapulinski <piotr.kwapulinski@intel.com>
Signed-off-by: Jedrzej Jagielski <jedrzej.jagielski@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>

ixgbe: read the OROM version information

Add functions reading the OROM version info and use them
as a part of the setting NVM info procedure.

Reviewed-by: Mateusz Polchlopek <mateusz.polchlopek@intel.com>
Reviewed-by: Przemek Kitszel <przemyslaw.kitszel@intel.com>
Tested-by: Bharath R <bharath.r@intel.com>
Signed-off-by: Slawomir Mrozowicz <slawomirx.mrozowicz@intel.com>
Co-developed-by: Piotr Kwapulinski <piotr.kwapulinski@intel.com>
Signed-off-by: Piotr Kwapulinski <piotr.kwapulinski@intel.com>
Signed-off-by: Jedrzej Jagielski <jedrzej.jagielski@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>

ixgbe: add E610 functions for acquiring flash data

Read NVM related info from the flash.

Add several helper functions used to access the flash data,
find memory banks, calculate offsets, calculate the flash size.

Reviewed-by: Przemek Kitszel <przemyslaw.kitszel@intel.com>
Reviewed-by: Mateusz Polchlopek <mateusz.polchlopek@intel.com>
Tested-by: Bharath R <bharath.r@intel.com>
Signed-off-by: Slawomir Mrozowicz <slawomirx.mrozowicz@intel.com>
Co-developed-by: Piotr Kwapulinski <piotr.kwapulinski@intel.com>
Signed-off-by: Piotr Kwapulinski <piotr.kwapulinski@intel.com>
Co-developed-by: Jedrzej Jagielski <jedrzej.jagielski@intel.com>
Signed-off-by: Jedrzej Jagielski <jedrzej.jagielski@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>

ixgbe: add handler for devlink .info_get()

Provide devlink .info_get() callback implementation to allow the
driver to report detailed version information. The following info
is reported:

 "serial_number" -> The PCI DSN of the adapter
 "fw.bundle_id" -> Unique identifier for the combined flash image
 "fw.undi" -> Version of the Option ROM containing the UEFI driver
 "board.id" -> The PBA ID string

Reviewed-by: Mateusz Polchlopek <mateusz.polchlopek@intel.com>
Tested-by: Bharath R <bharath.r@intel.com>
Signed-off-by: Jedrzej Jagielski <jedrzej.jagielski@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>

ixgbe: add initial devlink support

Add an initial support for devlink interface to ixgbe driver.

Similarly to i40e driver the implementation doesn't enable
devlink to manage device-wide configuration. Devlink instance
is created for each physical function of PCIe device.

Create separate directory for devlink related ixgbe files
and use naming scheme similar to the one used in the ice driver.

Add a stub for Documentation, to be extended by further patches.

Change struct ixgbe_adapter allocation to be done by devlink (Przemek),
as suggested by Jiri.

Reviewed-by: Mateusz Polchlopek <mateusz.polchlopek@intel.com>
Co-developed-by: Przemek Kitszel <przemyslaw.kitszel@intel.com>
Signed-off-by: Przemek Kitszel <przemyslaw.kitszel@intel.com>
Tested-by: Bharath R <bharath.r@intel.com>
Signed-off-by: Jedrzej Jagielski <jedrzej.jagielski@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>

ixgbe: wrap netdev_priv() usage

Wrap use of netdev_priv() in order to change the allocator of the device
private structure from alloc_etherdev_mq() to the devlink in next commit.

All but one netdev_priv() calls in the whole driver are replaced, the
remaining one is called on MACVLAN (so not ixgbe) device.

Signed-off-by: Przemek Kitszel <przemyslaw.kitszel@intel.com>
Tested-by: Bharath R <bharath.r@intel.com>
Signed-off-by: Jedrzej Jagielski <jedrzej.jagielski@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>

devlink: add value check to devlink_info_version_put()

Prevent from proceeding if there's nothing to print.

Suggested-by: Przemek Kitszel <przemyslaw.kitszel@intel.com>
Reviewed-by: Jiri Pirko <jiri@nvidia.com>
Reviewed-by: Kalesh AP <kalesh-anakkur.purayil@broadcom.com>
Tested-by: Bharath R <bharath.r@intel.com>
Signed-off-by: Jedrzej Jagielski <jedrzej.jagielski@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>

net: ncsi: Fix GCPS 64-bit member variables

Correct Get Controller Packet Statistics (GCPS) 64-bit wide member
variables, as per DSP0222 v1.0.0 and forward specs. The Driver currently
collects these stats, but they are yet to be exposed to the user.
Therefore, no user impact.

Statistics fixes:
Total Bytes Received (byte range 28..35)
Total Bytes Transmitted (byte range 36..43)
Total Unicast Packets Received (byte range 44..51)
Total Multicast Packets Received (byte range 52..59)
Total Broadcast Packets Received (byte range 60..67)
Total Unicast Packets Transmitted (byte range 68..75)
Total Multicast Packets Transmitted (byte range 76..83)
Total Broadcast Packets Transmitted (byte range 84..91)
Valid Bytes Received (byte range 204..11)

Signed-off-by: Hari Kalavakunta <kalavakunta.hari.prasad@gmail.com>
Reviewed-by: Paul Fertser <fercerpav@gmail.com>
Link: https://patch.msgid.link/20250410012309.1343-1-kalavakunta.hari.prasad@gmail.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>

tipc: Removing deprecated strncpy()

This patch suggests the replacement of strncpy with strscpy
as per Documentation/process/deprecated.
The strncpy() fails to guarantee NULL termination,
The function adds zero pads which isn't really convenient for short strings
as it may cause performance issues.

strscpy() is a preferred replacement because
it overcomes the limitations of strncpy mentioned above.

Compile Tested

Signed-off-by: Kevin Paul Reddy Janagari <kevinpaul468@gmail.com>
Reviewed-by: Tung Nguyen <tung.quang.nguyen@est.tech>
Tested-by: Tung Nguyen <tung.quang.nguyen@est.tech>
Link: https://patch.msgid.link/20250411085010.6249-1-kevinpaul468@gmail.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>

Merge branch 'net-dsa-mt7530-modernize-mib-handling-fix'

Christian Marangi says:

====================
net: dsa: mt7530: modernize MIB handling + fix

This small series modernize MIB handling for MT7530 and also
implement .get_stats64.

It was reported that kernel and Switch MIB desync in scenario where
a packet is forwarded from a port to another. In such case, the
forwarding is offloaded and the kernel is not aware of the
transmitted packet. To handle this, read the counter directly
from Switch registers.
====================

Link: https://patch.msgid.link/20250410163022.3695-1-ansuelsmth@gmail.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>

net: dsa: mt7530: implement .get_stats64

It was reported that the internally calculated counter might differ from
the real one from the Switch MIB. This can happen if the switch directly
forward packets between the ports or offload small packets like ARP
request. In such case, the kernel counter will desync compared to the
real one transmitted and received by the Switch.

To correctly provide the real info to the kernel, implement .get_stats64
that will directly read the current MIB counter from the switch
register.

Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
Link: https://patch.msgid.link/20250410163022.3695-7-ansuelsmth@gmail.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>

net: dsa: mt7530: move remaining MIB counter to define

For consistency with the other MIB counter, move also the remaining MIB
counter to define and update the custom MIB table.

Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
Link: https://patch.msgid.link/20250410163022.3695-6-ansuelsmth@gmail.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>

net: dsa: mt7530: move pkt stats and err MIB counter to eth_mac stats API

Drop custom handling of TX/RX packet stats and error MIB counter and handle
them in the standard .get_eth_mac_stats API

The MIB entry are dropped from the custom MIB table and converted to
a define providing only the MIB offset.

Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
Link: https://patch.msgid.link/20250410163022.3695-5-ansuelsmth@gmail.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>

net: dsa: mt7530: move pause MIB counter to eth_ctrl stats API

Drop custom handling of TX/RX pause frame MIB counter and handle
them in the standard .get_eth_ctrl_stats API

The MIB entry are dropped from the custom MIB table and converted to
a define providing only the MIB offset.

Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
Link: https://patch.msgid.link/20250410163022.3695-4-ansuelsmth@gmail.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>

net: dsa: mt7530: move pkt size and rx err MIB counter to rmon stats API

Drop custom handling of packet size and RX error MIB counter and handle
them in the standard .get_rmon_stats API

The MIB entry are dropped from the custom MIB table and converted to
a define providing only the MIB offset.

Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
Link: https://patch.msgid.link/20250410163022.3695-3-ansuelsmth@gmail.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>

net: dsa: mt7530: generalize read port stats logic

In preparation for migration to use of standard MIB API, generalize the
read port stats logic to a dedicated function.

This will permit to manually provide the offset and size of the MIB
counter to directly access specific counter.

Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
Link: https://patch.msgid.link/20250410163022.3695-2-ansuelsmth@gmail.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>