Jakub Kicinski [Fri, 26 Nov 2021 04:05:34 +0000 (20:05 -0800)]
Merge branch 'net-ipa-gsi-channel-flow-control'
Alex Elder says:
====================
net: ipa: GSI channel flow control
Starting with IPA v4.2, endpoint DELAY mode (which prevents data
transfer on TX endpoints) does not work properly. To address this,
changes were made to allow underlying GSI channels to be put into
a "flow controlled" state, which achieves a similar objective.
The first patch in this series implements the flow controlled
channel state and the commands used to control it. It arranges
to use the new mechanism--instead of DELAY mode--for IPA v4.2+.
In IPA v4.11, the notion of GSI channel flow control was enhanced,
and implemented in a slightly different way. For the most part this
doesn't affect the way the IPA driver uses flow control, but the
second patch adds support for the newer mechanism.
====================
Alex Elder [Wed, 24 Nov 2021 19:44:16 +0000 (13:44 -0600)]
net: ipa: support enhanced channel flow control
IPA v4.2 introduced GSI channel flow control, used instead of IPA
endpoint DELAY mode to prevent a TX channel from injecting packets
into the IPA core. It used a new FLOW_CONTROLLED channel state
which could be entered using GSI generic commands.
IPA v4.11 extended the channel flow control model. Rather than
having a distinct FLOW_CONTROLLED channel state, each channel has a
"flow control" property that can be enabled or not--independent of
the channel state. The AP (or modem) can modify this property using
the same GSI generic commands as before.
The AP only uses channel flow control on modem TX channels, and only
when recovering from a modem crash. The AP has no way to discover
the state of a modem channel, so the fact that (starting with IPA
v4.11) flow control no longer uses a distinct channel state is
invisible to the AP. So enhanced flow control generally does not
change the way AP uses flow control.
There are a few small differences, however:
- There is a notion of "primary" or "secondary" flow control, and
when enabling or disabling flow control that must be specified
in a new field in the GSI generic command register. For now, we
always specify 0 (meaning "primary").
- When disabling flow control, it's possible a request will need
to be retried. We retry up to 5 times in this case.
- Another new generic command allows the current flow control
state to be queried. We do not use this.
Other than the need for retries, the code essentially works the same
way as before.
Signed-off-by: Alex Elder <elder@linaro.org> Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Alex Elder [Wed, 24 Nov 2021 19:44:15 +0000 (13:44 -0600)]
net: ipa: introduce channel flow control
One quirk for certain versions of IPA is that endpoint DELAY mode
does not work properly. IPA DELAY mode prevents any packets from
being delivered to the IPA core for processing on a TX endpoint.
The AP uses DELAY mode when the modem crashes, to prevent modem TX
endpoints from generating traffic during crash recovery. Without
this, there is a chance the hardware will stall during recovery from
a modem crash.
To achieve a similar effect, a GSI FLOW_CONTROLLED channel state
was created. A STARTED TX channel can be placed in FLOW_CONTROLLED
state, which prevents the transfer of any more packets. A channel
in FLOW_CONTROLLED state can be either returned to STARTED state, or
can be transitioned to STOPPED state.
Because this operates on GSI channels, two generic commands were
added to allow the AP to control this state for modem channels
(similar to the ALLOCATE and HALT channel commands).
Previously the code assumed this quirk only applied to IPA v4.2.
In fact, channel flow control (rather than endpoint DELAY mode)
should be used for all versions *starting* with IPA v4.2.
Signed-off-by: Alex Elder <elder@linaro.org> Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Jeremy Kerr [Thu, 25 Nov 2021 06:07:38 +0000 (14:07 +0800)]
mctp: serial: enforce fixed MTU
The current serial driver requires a maximum MTU of 68, and it doesn't
make sense to set a MTU below the MCTP-required baseline (of 68) either.
This change sets the min_mtu & max_mtu of the mctp netdev, essentially
disallowing changes. By using these instead of a ndo_change_mtu op, we
get the netlink extacks reported too.
Signed-off-by: Jeremy Kerr <jk@codeconstruct.com.au> Signed-off-by: Jakub Kicinski <kuba@kernel.org>
====================
net: ipa: small collected improvements
This series contains a somewhat unrelated set of changes, some
inspired by some recent work posted for back-port. For the most
part they're meant to improve the code without changing it's
functionality. Each basically stands on its own.
====================
Alex Elder [Wed, 24 Nov 2021 20:25:11 +0000 (14:25 -0600)]
net: ipa: rearrange GSI structure fields
The dummy net_device is a large field in the GSI structure, but it
is not at all interesting from the perspective of debugging. Move
it to the end of the GSI structure so the other fields are easier to
find in memory.
The channel and event ring arrays are also very large, so move them
near the end of the structure as well.
Swap the position of the result and completion fields to improve
structure packing.
Signed-off-by: Alex Elder <elder@linaro.org> Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Alex Elder [Wed, 24 Nov 2021 20:25:10 +0000 (14:25 -0600)]
net: ipa: GSI only needs one completion
A mutex ensures we never submit more than one GSI command of any
kind at once. This means the per-channel and per-event ring
completion structures provide no benefit. Instead, just use the
single (existing) GSI completion to signal the completion of GSI
commands of all types.
This makes gsi_evt_ring_init() a trivial function with no inverse,
so open-code it in its sole caller and get rid of the function.
Signed-off-by: Alex Elder <elder@linaro.org> Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Alex Elder [Wed, 24 Nov 2021 20:25:09 +0000 (14:25 -0600)]
net: ipa: skip SKB copy if no netdev
In ipa_endpoint_skb_copy(), a new socket buffer structure is
allocated so that some data can be copied into it. However, after
doing this, if the endpoint has a null netdev pointer, we just drop
free the socket buffer.
Instead, check endpoint->netdev pointer first, and just return early
if it's null. Also return early if the SKB allocation fails, to
avoid the deeper indentation in the normal path.
Signed-off-by: Alex Elder <elder@linaro.org> Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Alex Elder [Wed, 24 Nov 2021 20:25:08 +0000 (14:25 -0600)]
net: ipa: explicitly disable HOLB drop during setup
During setup, ipa_endpoint_program() programs each endpoint with
various configuration parameters. One of those registers defines
whether to drop packets when a head-of-line blocking condition is
detected on an RX endpoint. We currently assume this is disabled;
instead, explicitly set it to be disabled.
Signed-off-by: Alex Elder <elder@linaro.org> Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Alex Elder [Wed, 24 Nov 2021 20:25:07 +0000 (14:25 -0600)]
net: ipa: rework how HOL_BLOCK handling is specified
The head-of-line block (HOLB) drop timer is only meaningful when
dropping packets due to blocking is enabled. Given that, redefine
the interface so the timer is specified when enabling HOLB drop, and
use a different function when disabling.
To enable and disable HOLB drop, these functions will now be used:
ipa_endpoint_init_hol_block_enable(endpoint, milliseconds)
ipa_endpoint_init_hol_block_disable(endpoint)
The existing ipa_endpoint_init_hol_block_enable() becomes a helper
function, renamed ipa_endpoint_init_hol_block_en(), and used with
ipa_endpoint_init_hol_block_timer() to enable HOLB block on an
endpoint.
Signed-off-by: Alex Elder <elder@linaro.org> Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Alex Elder [Wed, 24 Nov 2021 20:25:06 +0000 (14:25 -0600)]
net: ipa: zero unused portions of filter table memory
Not all filter table entries are used. Only certain endpoints
support filtering, and the table begins with a bitmap indicating
which endpoints use the "slots" that follow for filter rules.
Currently, unused filter table entries are not initialized.
Instead, zero-fill the entire unused portion of the filter table
memory regions, to make it more obvious that memory is unused (and
not subsequently modified).
This is not strictly necessary, but the result is reassuring when
looking at filter table memory.
Signed-off-by: Alex Elder <elder@linaro.org> Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Alex Elder [Wed, 24 Nov 2021 20:25:05 +0000 (14:25 -0600)]
net: ipa: kill ipa_modem_init()
A recent commit made disabling the SMP2P "setup ready" interrupt
unrelated to ipa_modem_stop(). Given that, it seems fitting to get
rid of ipa_modem_init() and ipa_modem_exit() (which are trivial
wrapper functions), and call ipa_smp2p_init() and ipa_smp2p_exit()
directly instead.
Signed-off-by: Alex Elder <elder@linaro.org> Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Vladimir Oltean [Thu, 25 Nov 2021 12:58:08 +0000 (14:58 +0200)]
net: dsa: felix: enable cut-through forwarding between ports by default
The VSC9959 switch embedded within NXP LS1028A (and that version of
Ocelot switches only) supports cut-through forwarding - meaning it can
start the process of looking up the destination ports for a packet, and
forward towards those ports, before the entire packet has been received
(as opposed to the store-and-forward mode).
The up side is having lower forwarding latency for large packets. The
down side is that frames with FCS errors are forwarded instead of being
dropped. However, erroneous frames do not result in incorrect updates of
the FDB or incorrect policer updates, since these processes are deferred
inside the switch to the end of frame. Since the switch starts the
cut-through forwarding process after all packet headers (including IP,
if any) have been processed, packets with large headers and small
payload do not see the benefit of lower forwarding latency.
There are two cases that need special attention.
The first is when a packet is multicast (or flooded) to multiple
destinations, one of which doesn't have cut-through forwarding enabled.
The switch deals with this automatically by disabling cut-through
forwarding for the frame towards all destination ports.
The second is when a packet is forwarded from a port of lower link speed
towards a port of higher link speed. This is not handled by the hardware
and needs software intervention.
Since we practically need to update the cut-through forwarding domain
from paths that aren't serialized by the rtnl_mutex (phylink
mac_link_down/mac_link_up ops), this means we need to serialize physical
link events with user space updates of bonding/bridging domains.
Enabling cut-through forwarding is done per {egress port, traffic class}.
I don't see any reason why this would be a configurable option as long
as it works without issues, and there doesn't appear to be any user
space configuration tool to toggle this on/off, so this patch enables
cut-through forwarding on all eligible ports and traffic classes.
Vladimir Oltean [Thu, 25 Nov 2021 12:58:07 +0000 (14:58 +0200)]
net: ocelot: remove "bridge" argument from ocelot_get_bridge_fwd_mask
The only called takes ocelot_port->bridge and passes it as the "bridge"
argument to this function, which then compares it with
ocelot_port->bridge. This is not useful.
Instead, we would like this function to return 0 if ocelot_port->bridge
is not present, which is what this patch does.
Ong Boon Leong [Wed, 24 Nov 2021 11:40:19 +0000 (19:40 +0800)]
net: stmmac: perserve TX and RX coalesce value during XDP setup
When XDP program is loaded, it is desirable that the previous TX and RX
coalesce values are not re-inited to its default value. This prevents
unnecessary re-configurig the coalesce values that were working fine
before.
Fixes: ac746c8520d9 ("net: stmmac: enhance XDP ZC driver level switching performance") Signed-off-by: Ong Boon Leong <boon.leong.ong@intel.com> Tested-by: Kurt Kanzenbach <kurt@linutronix.de> Link: https://lore.kernel.org/r/20211124114019.3949125-1-boon.leong.ong@intel.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Yang Yingliang [Wed, 24 Nov 2021 08:40:48 +0000 (16:40 +0800)]
tsnep: Add missing of_node_put() in tsnep_mdio_init()
The node pointer is returned by of_get_child_by_name() with
refcount incremented in tsnep_mdio_init(). Calling of_node_put()
to aovid the refcount leak in tsnep_mdio_init().
Heiner Kallweit [Wed, 24 Nov 2021 20:44:40 +0000 (21:44 +0100)]
r8169: disable detection of chip version 60
It seems only XID 609 made it to the mass market. Therefore let's
disable detection of the other RTL8125a XID's. If nobody complains
we can remove support for RTL_GIGA_MAC_VER_60 later.
Maciej Żenczykowski [Tue, 23 Nov 2021 20:37:15 +0000 (12:37 -0800)]
net: allow SO_MARK with CAP_NET_RAW
A CAP_NET_RAW capable process can already spoof (on transmit) anything
it desires via raw packet sockets... There is no good reason to not
allow it to also be able to play routing tricks on packets from its
own normal sockets.
There is a desire to be able to use SO_MARK for routing table selection
(via ip rule fwmark) from within a user process without having to run
it as root. Granting it CAP_NET_RAW is much less dangerous than
CAP_NET_ADMIN (CAP_NET_RAW doesn't permit persistent state change,
while CAP_NET_ADMIN does - by for example allowing the reconfiguration
of the routing tables and/or bringing up/down devices).
Let's keep CAP_NET_ADMIN for persistent state changes,
while using CAP_NET_RAW for non-configuration related stuff.
Maciej Żenczykowski [Tue, 23 Nov 2021 20:37:02 +0000 (12:37 -0800)]
net: allow CAP_NET_RAW to setsockopt SO_PRIORITY
CAP_NET_ADMIN is and should continue to be about configuring the
system as a whole, not about configuring per-socket or per-packet
parameters.
Sending and receiving raw packets is what CAP_NET_RAW is all about.
It can already send packets with any VLAN tag, and any IPv4 TOS
mark, and any IPv6 TCLASS mark, simply by virtue of building
such a raw packet. Not to mention using any protocol and source/
/destination ip address/port tuple.
These are the fields that networking gear uses to prioritize packets.
Hence, a CAP_NET_RAW process is already capable of affecting traffic
prioritization after it hits the wire. This change makes it capable
of affecting traffic prioritization even in the host at the nic and
before that in the queueing disciplines (provided skb->priority is
actually being used for prioritization, and not the TOS/TCLASS field)
Hence it makes sense to allow a CAP_NET_RAW process to set the
priority of sockets and thus packets it sends.
Rahul Lakkireddy [Tue, 23 Nov 2021 15:47:17 +0000 (21:17 +0530)]
cxgb4: allow reading unrecognized port module eeprom
Even if firmware fails to recognize the plugged-in port module type,
allow reading port module EEPROM anyway. This helps in obtaining
necessary diagnostics information for debugging and analysis.
Ido Schimmel [Wed, 24 Nov 2021 10:11:22 +0000 (12:11 +0200)]
net: bridge: Allow base 16 inputs in sysfs
Cited commit converted simple_strtoul() to kstrtoul() as suggested by
the former's documentation. However, it also forced all the inputs to be
decimal resulting in user space breakage.
Fix by setting the base to '0' so that the base is automatically
detected.
Before:
# ip link add name br0 type bridge vlan_filtering 1
# echo "0x88a8" > /sys/class/net/br0/bridge/vlan_protocol
bash: echo: write error: Invalid argument
After:
# ip link add name br0 type bridge vlan_filtering 1
# echo "0x88a8" > /sys/class/net/br0/bridge/vlan_protocol
# echo $?
0
Fixes: 520fbdf7fb19 ("net/bridge: replace simple_strtoul to kstrtol") Signed-off-by: Ido Schimmel <idosch@nvidia.com> Acked-by: Nikolay Aleksandrov <nikolay@nvidia.com> Link: https://lore.kernel.org/r/20211124101122.3321496-1-idosch@idosch.org Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Yufeng Mo [Wed, 24 Nov 2021 01:06:54 +0000 (09:06 +0800)]
net: hns3: add dql info when tx timeout
When tx timeout occurs, the info of dql maybe helpful, so print
these info to hns3_get_tx_timeo_queue_info().
Signed-off-by: Yufeng Mo <moyufeng@huawei.com> Signed-off-by: Guangbin Huang <huangguangbin2@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Jie Wang [Wed, 24 Nov 2021 01:06:53 +0000 (09:06 +0800)]
net: hns3: debugfs add drop packet statistics of multicast and broadcast for igu
Currently, there is no way to get drop packet number of multicast and
broadcast in IGU hardware module, it is not convenient to find problem
when multicast packet or broadcast packet is dropped in IGU, so this
patch adds statistics for them in debugfs.
Signed-off-by: Jie Wang <wangjie125@huawei.com> Signed-off-by: Guangbin Huang <huangguangbin2@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Yufeng Mo [Wed, 24 Nov 2021 01:06:52 +0000 (09:06 +0800)]
net: hns3: format the output of the MAC address
Printing the whole MAC addresse may bring security risks. Therefore,
the MAC address is partially encrypted to improve security.
Signed-off-by: Yufeng Mo <moyufeng@huawei.com> Signed-off-by: Guangbin Huang <huangguangbin2@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Yufeng Mo [Wed, 24 Nov 2021 01:06:51 +0000 (09:06 +0800)]
net: hns3: add log for workqueue scheduled late
When the mbx or reset message arrives, the driver is informed
through an interrupt. This task can be processed only after
the workqueue is scheduled. In some cases, this workqueue
scheduling takes a long time. As a result, the mbx or reset
service task cannot be processed in time. So add some warning
message to improve debugging efficiency for this case.
Signed-off-by: Yufeng Mo <moyufeng@huawei.com> Signed-off-by: Guangbin Huang <huangguangbin2@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Reported-by: Abaci Robot <abaci@linux.alibaba.com> Signed-off-by: Jiapeng Chong <jiapeng.chong@linux.alibaba.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Kuniyuki Iwashima [Mon, 22 Nov 2021 10:16:21 +0000 (19:16 +0900)]
dccp/tcp: Remove an unused argument in inet_csk_listen_start().
The commit 1295e2cf3065 ("inet: minor optimization for backlog setting in
listen(2)") added change so that sk_max_ack_backlog is initialised earlier
in inet_dccp_listen() and inet_listen(). Since then, we no longer use
backlog in inet_csk_listen_start(), so let's remove it.
Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.co.jp> Acked-by: Yafang Shao <laoar.shao@gmail.com> Reviewed-by: Richard Sailer <richard_siegfried@systemli.org> Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Kurt Kanzenbach [Mon, 22 Nov 2021 11:19:31 +0000 (12:19 +0100)]
net: stmmac: Calculate CDC error only once
The clock domain crossing error (CDC) is calculated at every fetch of Tx or Rx
timestamps. It includes a division. Especially on arm32 based systems it is
expensive. It also requires two conditionals in the hotpath.
Add a compensation value cache to struct plat_stmmacenet_data and subtract it
unconditionally in the RX/TX functions which spares the conditionals.
The value is initialized to 0 and if supported calculated in the PTP
initialization code.
Jakub Kicinski [Tue, 23 Nov 2021 01:24:47 +0000 (17:24 -0800)]
net: remove .ndo_change_proto_down
.ndo_change_proto_down was added seemingly to enable out-of-tree
implementations. Over 2.5yrs later we still have no real users
upstream. Hardwire the generic implementation for now, we can
revert once real users materialize. (rocker is a test vehicle,
not a user.)
We need to drop the optimization on the sysfs side, because
unlike ndos priv_flags will be changed at runtime, so we'd
need READ_ONCE/WRITE_ONCE everywhere..
Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Tue, 23 Nov 2021 12:17:24 +0000 (12:17 +0000)]
Merge branch '100GbE' of git://git.kernel.org/pub/scm/linux/kernel/git/tnguy/next-queue
Tony Nguyen says:
====================
100GbE Intel Wired LAN Driver Updates 2021-11-22
Shiraz Saleem says:
Currently E800 devices come up as RoCEv2 devices by default.
This series add supports for users to configure iWARP or RoCEv2 functionality
per PCI function. devlink parameters is used to realize this and is keyed
off similar work in [1].
David S. Miller [Tue, 23 Nov 2021 12:14:49 +0000 (12:14 +0000)]
Merge branch 'mvpp2-5gbase-r-support'
Marek Behún says:
====================
Add 5gbase-r support for mvpp2
this adds support for 5gbase-r for mvpp2 driver. Current versions of
TF-A firmware support changing the PHY to 5gbase-r via SMC calls, at
least on Macchiatobin.
Tested on Macchiatobin.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Gerhard Engleder [Mon, 22 Nov 2021 20:32:25 +0000 (21:32 +0100)]
tsnep: Fix set MAC address
Commit 4dfb9982644b ("tsn: Fix build.") fixed compilation with const
dev_addr. In tsnep_netdev_set_mac_address() the call of ether_addr_copy()
was replaced with dev_set_mac_address(), which calls
ndo_set_mac_address(). This results in an endless recursive loop because
ndo_set_mac_address is set to tsnep_netdev_set_mac_address.
Call eth_hw_addr_set() instead of dev_set_mac_address() in
ndo_set_mac_address()/tsnep_netdev_set_mac_address() to copy the address
as intended.
Fixes: 4dfb9982644b ("tsn: Fix build.") Signed-off-by: Gerhard Engleder <gerhard@engleder-embedded.com> Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Tue, 23 Nov 2021 11:53:16 +0000 (11:53 +0000)]
Merge branch 'qca8k-mirror-and-lag-support'
Ansuel Smith says:
====================
Add mirror and LAG support to qca8k
With the continue of adding 'Multiple feature to qca8k'
The switch supports mirror mode and LAG.
In mirror mode a port is set as mirror and other port are configured
to both igress or egress mode. With no port configured for mirror,
the mirror port is disabled and reverted to normal port.
The switch supports max 4 LAG with 4 different member max.
Current supported mode is Hash mode in both L2 or L2+3 mode.
There is a problematic implementation for the hash mode where
with multiple LAG configured, someone has to remove them to
change the hash mode as it's global.
When a member of the LAG is disconnected, the traffic is redirected
to the other port.
Some warning are present from checkpatch but can't really be fixed
as it would result in making the regs less readable.
(They really did their best with the LAG reg logic and complexity)
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Ansuel Smith [Tue, 23 Nov 2021 02:59:11 +0000 (03:59 +0100)]
net: dsa: qca8k: add LAG support
Add LAG support to this switch. In Documentation this is described as
trunk mode. A max of 4 LAGs are supported and each can support up to 4
port. The current tx mode supported is Hash mode with both L2 and L2+3
mode.
When no port are present in the trunk, the trunk is disabled in the
switch.
When a port is disconnected, the traffic is redirected to the other
available port.
The hash mode is global and each LAG require to have the same hash mode
set. To change the hash mode when multiple LAG are configured, it's
required to remove each LAG and set the desired hash mode to the last.
An error is printed when it's asked to set a not supported hadh mode.
Signed-off-by: Ansuel Smith <ansuelsmth@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Ansuel Smith [Tue, 23 Nov 2021 02:59:10 +0000 (03:59 +0100)]
net: dsa: qca8k: add support for mirror mode
The switch supports mirror mode. Only one port can set as mirror port and
every other port can set to both ingress and egress mode. The mirror
port is disabled and reverted to normal operation once every port is
removed from sending packet to it.
Signed-off-by: Ansuel Smith <ansuelsmth@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Jeremy Kerr [Tue, 23 Nov 2021 07:50:46 +0000 (15:50 +0800)]
mctp: Add MCTP-over-serial transport binding
This change adds a MCTP Serial transport binding, as defined by DMTF
specificiation DSP0253 - "MCTP Serial Transport Binding". This is
implemented as a new serial line discipline, and can be attached to
arbitrary tty devices.
From the Kconfig description:
This driver provides an MCTP-over-serial interface, through a
serial line-discipline, as defined by DMTF specification "DSP0253 -
MCTP Serial Transport Binding". By attaching the ldisc to a serial
device, we get a new net device to transport MCTP packets.
This allows communication with external MCTP endpoints which use
serial as their transport. It can also be used as an easy way to
provide MCTP connectivity between virtual machines, by forwarding
data between simple virtual serial devices.
Say y here if you need to connect to MCTP endpoints over serial. To
compile as a module, use m; the module will be called mctp-serial.
Once the N_MCTP line discipline is set [using ioctl(TCIOSETD)], we get a
new netdev suitable for MCTP communication.
The 'mctp' utility[1] provides a simple wrapper for this ioctl, using
'link serial <device>':
# mctp link serial /dev/ttyS0 &
# mctp link
dev lo index 1 address 0x00:00:00:00:00:00 net 1 mtu 65536 up
dev mctpserial0 index 5 address 0x(no-addr) net 1 mtu 68 down
[1]: https://github.com/CodeConstruct/mctp
Signed-off-by: Jeremy Kerr <jk@codeconstruct.com.au> Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Tue, 23 Nov 2021 11:44:31 +0000 (11:44 +0000)]
Merge branch 'mlxsw-updates'
Ido Schimmel says:
====================
mlxsw: Various updates
Patch #1 removes deadcode reported by Coverity.
Patch #2 adds a shutdown method in the PCI driver to ensure the kexeced
kernel starts working with a device that is in a sane state.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Danielle Ratson [Tue, 23 Nov 2021 07:54:47 +0000 (09:54 +0200)]
mlxsw: pci: Add shutdown method in PCI driver
On an arm64 platform with the Spectrum ASIC, after loading and executing
a new kernel via kexec, the following trace [1] is observed. This seems
to be caused by the fact that the device is not properly shutdown before
executing the new kernel.
Fix this by implementing a shutdown method which mirrors the remove
method, as recommended by the kexec maintainer [2][3].
Cc: Eric Biederman <ebiederm@xmission.com> Signed-off-by: Danielle Ratson <danieller@nvidia.com> Signed-off-by: Ido Schimmel <idosch@nvidia.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Danielle Ratson [Tue, 23 Nov 2021 07:54:46 +0000 (09:54 +0200)]
mlxsw: spectrum_router: Remove deadcode in mlxsw_sp_rif_mac_profile_find
The function idr_for_each_entry() already checks that the next entry in
the IDR is not NULL.
Therefore, checking that again in every iteration leads to deadcode.
Remove the unnecessary check in order to avoid that.
Addresses-Coverity: ("Logically dead code") Signed-off-by: Danielle Ratson <danieller@nvidia.com> Signed-off-by: Ido Schimmel <idosch@nvidia.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Shiraz Saleem [Mon, 18 Oct 2021 23:16:02 +0000 (18:16 -0500)]
net/ice: Add support for enable_iwarp and enable_roce devlink param
Allow support for 'enable_iwarp' and 'enable_roce' devlink params to turn
on/off iWARP or RoCE protocol support for E800 devices.
For example, a user can turn on iWARP functionality with,
devlink dev param set pci/0000:07:00.0 name enable_iwarp value true cmode runtime
This add an iWARP auxiliary rdma device, ice.iwarp.<>, under this PF.
A user request to enable both iWARP and RoCE under the same PF is rejected
since this device does not support both protocols simultaneously on the
same port.
David S. Miller [Mon, 22 Nov 2021 15:35:17 +0000 (15:35 +0000)]
Merge branch 'qca8k-next'
Ansuel Smith says:
====================
Multiple cleanup and feature for qca8k
This is a reduced version of the old massive series.
Refer to the changelog to know what is removed from this.
We clean and convert the driver to GENMASK FIELD_PREP to clean multiple
use of various naming scheme. (example we have a mix of _MASK, _S _M,
and various name) The idea is to ""simplify"" and remove some shift and
data handling by using FIELD API.
The patch contains various checkpatch warning and are ignored to not
create more mess in the header file. (fixing the too long line warning
would results in regs definition less readable)
We conver the driver to regmap API as ipq40xx SoC is based on the same
reg structure and we need to generilize the read/write access to split
the driver to commond and specific code.
The conversion to regmap is NOT done for the read/write/rmw operation,
the function are reworked to use the regmap helper instead.
This is done to keep the patch delta low but will come sooner or later
when the code will be split.
Any new feature will directly use the regmap helper and the reg
set/clear and the busy wait function are migrated to regmap helper as
the use of these function is low.
We also add a minor patch for MIB counter as qca8337 is missing 2 extra
counter, support for mdb and ageing settings.
v3:
- Try to reduce regmap conversion patch
v2:
- Move regmap init to sw_probe instead of moving switch id check.
- Removed LAGs, mirror extra features will come later in another
smaller series.
- Squash 2 ageing patch
- Add more description about the regmap patch
- Rework mdb patch to do operation under the same lock
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Ansuel Smith [Mon, 22 Nov 2021 15:23:48 +0000 (16:23 +0100)]
net: dsa: qca8k: add support for mdb_add/del
Add support for mdb add/del function. The ARL table is used to insert
the rule. The rule will be searched, deleted and reinserted with the
port mask updated. The function will check if the rule has to be updated
or insert directly with no deletion of the old rule.
If every port is removed from the port mask, the rule is removed.
The rule is set STATIC in the ARL table (aka it doesn't age) to not be
flushed by fast age function.
Signed-off-by: Ansuel Smith <ansuelsmth@gmail.com> Reviewed-by: Vladimir Oltean <olteanv@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Ansuel Smith [Mon, 22 Nov 2021 15:23:47 +0000 (16:23 +0100)]
net: dsa: qca8k: add set_ageing_time support
qca8k support setting ageing time in step of 7s. Add support for it and
set the max value accepted of 7645m.
Documentation talks about support for 10000m but that values doesn't
make sense as the value doesn't match the max value in the reg.
Signed-off-by: Ansuel Smith <ansuelsmth@gmail.com> Reviewed-by: Vladimir Oltean <olteanv@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Ansuel Smith [Mon, 22 Nov 2021 15:23:46 +0000 (16:23 +0100)]
net: dsa: qca8k: add support for port fast aging
The switch supports fast aging by flushing any rule in the ARL
table for a specific port.
Signed-off-by: Ansuel Smith <ansuelsmth@gmail.com> Reviewed-by: Vladimir Oltean <olteanv@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Ansuel Smith [Mon, 22 Nov 2021 15:23:45 +0000 (16:23 +0100)]
net: dsa: qca8k: add additional MIB counter and make it dynamic
We are currently missing 2 additionals MIB counter present in QCA833x
switch.
QC832x switch have 39 MIB counter and QCA833X have 41 MIB counter.
Add the additional MIB counter and rework the MIB function to print the
correct supported counter from the match_data struct.
Signed-off-by: Ansuel Smith <ansuelsmth@gmail.com> Reviewed-by: Vladimir Oltean <olteanv@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Ansuel Smith [Mon, 22 Nov 2021 15:23:44 +0000 (16:23 +0100)]
net: dsa: qca8k: initial conversion to regmap helper
Convert any qca8k set/clear/pool to regmap helper and add
missing config to regmap_config struct.
Read/write/rmw operation are reworked to use the regmap helper
internally to keep the delta of this patch low. These additional
function will then be dropped when the code split will be proposed.
Ipq40xx SoC have the internal switch based on the qca8k regmap but use
mmio for read/write/rmw operation instead of mdio.
In preparation for the support of this internal switch, convert the
driver to regmap API to later split the driver to common and specific
code. The overhead introduced by the use of regamp API is marginal as the
internal mdio will bypass it by using its direct access and regmap will be
used only by configuration functions or fdb access.
Signed-off-by: Ansuel Smith <ansuelsmth@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Ansuel Smith [Mon, 22 Nov 2021 15:23:43 +0000 (16:23 +0100)]
net: dsa: qca8k: move regmap init in probe and set it mandatory
In preparation for regmap conversion, move regmap init in the probe
function and make it mandatory as any read/write/rmw operation will be
converted to regmap API.
Signed-off-by: Ansuel Smith <ansuelsmth@gmail.com> Reviewed-by: Vladimir Oltean <olteanv@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Ansuel Smith [Mon, 22 Nov 2021 15:23:42 +0000 (16:23 +0100)]
net: dsa: qca8k: remove extra mutex_init in qca8k_setup
Mutex is already init in sw_probe. Remove the extra init in qca8k_setup.
Signed-off-by: Ansuel Smith <ansuelsmth@gmail.com> Reviewed-by: Vladimir Oltean <olteanv@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Ansuel Smith [Mon, 22 Nov 2021 15:23:41 +0000 (16:23 +0100)]
net: dsa: qca8k: convert to GENMASK/FIELD_PREP/FIELD_GET
Convert and try to standardize bit fields using
GENMASK/FIELD_PREP/FIELD_GET macros. Rework some logic to support the
standard macro and tidy things up. No functional change intended.
Signed-off-by: Ansuel Smith <ansuelsmth@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Ansuel Smith [Mon, 22 Nov 2021 15:23:40 +0000 (16:23 +0100)]
net: dsa: qca8k: remove redundant check in parse_port_config
The very next check for port 0 and 6 already makes sure we don't go out
of bounds with the ports_config delay table.
Remove the redundant check.
Reported-by: kernel test robot <lkp@intel.com> Reported-by: Dan Carpenter <dan.carpenter@oracle.com> Signed-off-by: Ansuel Smith <ansuelsmth@gmail.com> Reviewed-by: Vladimir Oltean <olteanv@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Mon, 22 Nov 2021 15:13:54 +0000 (15:13 +0000)]
Merge branch 'skbuff-struct-group'
Kees Cook says:
====================
skbuff: Switch structure bounds to struct_group()
This is a pair of patches to add struct_group() to struct sk_buff. The
first is needed to work around sparse-specific complaints, and is new
for v2. The second patch is the same as originally sent as v1.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Kees Cook [Sun, 21 Nov 2021 00:31:49 +0000 (16:31 -0800)]
skbuff: Switch structure bounds to struct_group()
In preparation for FORTIFY_SOURCE performing compile-time and run-time
field bounds checking for memcpy(), memmove(), and memset(), avoid
intentionally writing across neighboring fields.
Replace the existing empty member position markers "headers_start" and
"headers_end" with a struct_group(). This will allow memcpy() and sizeof()
to more easily reason about sizes, and improve readability.
"pahole" shows no size nor member offset changes to struct sk_buff.
"objdump -d" shows no object code changes (outside of WARNs affected by
source line number changes).
Signed-off-by: Kees Cook <keescook@chromium.org> Reviewed-by: Gustavo A. R. Silva <gustavoars@kernel.org> Reviewed-by: Jason A. Donenfeld <Jason@zx2c4.com> # drivers/net/wireguard/* Link: https://lore.kernel.org/lkml/20210728035006.GD35706@embeddedor Signed-off-by: David S. Miller <davem@davemloft.net>
Kees Cook [Sun, 21 Nov 2021 00:31:48 +0000 (16:31 -0800)]
skbuff: Move conditional preprocessor directives out of struct sk_buff
In preparation for using the struct_group() macro in struct sk_buff,
move the conditional preprocessor directives out of the region of struct
sk_buff that will be enclosed by struct_group(). While GCC and Clang are
happy with conditional preprocessor directives here, sparse is not, even
under -Wno-directive-within-macro[1], as would be seen under a C=1 build:
net/core/filter.c: note: in included file (through include/linux/netlink.h, include/linux/sock_diag.h):
./include/linux/skbuff.h:820:1: warning: directive in macro's argument list
./include/linux/skbuff.h:822:1: warning: directive in macro's argument list
./include/linux/skbuff.h:846:1: warning: directive in macro's argument list
./include/linux/skbuff.h:848:1: warning: directive in macro's argument list
Additionally remove empty macro argument definitions and usage.
Antoine Tenart [Mon, 22 Nov 2021 14:24:56 +0000 (15:24 +0100)]
sections: global data can be in .bss
When checking an address is located in a global data section also check
for the .bss section as global variables initialized to 0 can be in
there (-fzero-initialized-in-bss).
This was found when looking at ensure_safe_net_sysctl which was failing
to detect non-init sysctl pointing to a global data section when the
data was in the .bss section.
Signed-off-by: Antoine Tenart <atenart@kernel.org> Acked-by: Steven Rostedt (VMware) <rostedt@goodmis.org> Acked-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: David S. Miller <davem@davemloft.net>
Christophe JAILLET [Sun, 21 Nov 2021 21:56:39 +0000 (22:56 +0100)]
hv_netvsc: Use bitmap_zalloc() when applicable
'send_section_map' is a bitmap. So use 'bitmap_zalloc()' to simplify code,
improve the semantic and avoid some open-coded arithmetic in allocator
arguments.
Also change the corresponding 'kfree()' into 'bitmap_free()' to keep
consistency.
While at it, change an '== NULL' test into a '!'.
Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr> Signed-off-by: David S. Miller <davem@davemloft.net>
Christophe JAILLET [Sun, 21 Nov 2021 15:32:04 +0000 (16:32 +0100)]
rds: Fix a typo in a comment
s/cold/could/
Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr> Acked-By: Devesh Sharma <devesh.s.sharma@oracle.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Jakub Kicinski [Sat, 20 Nov 2021 17:15:10 +0000 (09:15 -0800)]
pcmcia: hide the MAC address helpers if !NET
pcmcia_get_mac_from_cis is only called from networking and
recent changes made it call dev_addr_mod() which is itself
only defined if NET.
Reported-by: kernel test robot <lkp@intel.com> Fixes: adeef3e32146 ("net: constify netdev->dev_addr") Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: David S. Miller <davem@davemloft.net>
M Chetan Kumar [Sat, 20 Nov 2021 16:21:55 +0000 (21:51 +0530)]
net: wwan: iosm: device trace collection using relayfs
This patch brings in support for device trace collection.
It implements relayfs interface for pushing device trace
from kernel space to user space.
Driver gets the debugfs base directory associated to WWAN
Device and creates trace_control and trace debugfs for
device tracing. Both trace_control & trace debugfs are
created under /sys/kernel/debug/wwan/wwan0/.
In order to collect device trace on trace0 interface, user
need to write 1 to trace_ctl interface.
Signed-off-by: M Chetan Kumar <m.chetan.kumar@linux.intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
M Chetan Kumar [Sat, 20 Nov 2021 16:21:54 +0000 (21:51 +0530)]
net: wwan: common debugfs base dir for wwan device
This patch set brings in a common debugfs base directory
i.e. /sys/kernel/debugfs/wwan/ in WWAN Subsystem for a
WWAN device instance. So that it avoids driver polluting
debugfs root with unrelated directories & possible name
collusion.
Having a common debugfs base directory for WWAN drivers
eases user to match control devices with debugfs entries.
WWAN Subsystem creates dentry (/sys/kernel/debugfs/wwan)
on module load & removes dentry on module unload.
When driver registers a new wwan device, dentry (wwanX)
is created for WWAN device instance & on driver unregister
dentry is removed.
New API is introduced to return the wwan device instance
dentry so that driver can create debugfs entries under it.
Signed-off-by: M Chetan Kumar <m.chetan.kumar@linux.intel.com> Reviewed-by: Loic Poulain <loic.poulain@linaro.org> Signed-off-by: David S. Miller <davem@davemloft.net>
Jakub Kicinski [Sat, 20 Nov 2021 15:31:19 +0000 (07:31 -0800)]
octeon: constify netdev->dev_addr
Argument of a helper is missing a const.
Reported-by: kernel test robot <lkp@intel.com> Fixes: adeef3e32146 ("net: constify netdev->dev_addr") Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Mon, 22 Nov 2021 13:19:04 +0000 (13:19 +0000)]
Merge branch 'tsn-endpoint-driver'
Gerhard Engleder says:
====================
TSN endpoint Ethernet MAC driver
This series adds a driver for my FPGA based TSN endpoint Ethernet MAC.
It also includes the device tree binding.
The device is designed as Ethernet MAC for TSN networks. It will be used
in PLCs with real-time requirements up to isochronous communication with
protocols like OPC UA Pub/Sub.
v3:
- set MAC mode based on PHY information (Andrew Lunn)
- remove/postpone loopback mode interface (Andrew Lunn)
- add suppress_preamble node support (Andrew Lunn)
- add mdio timeout (Andrew Lunn)
- no need to call phy_start_aneg (Andrew Lunn)
- remove unreachable code (Andrew Lunn)
- move 'struct napi_struct' closer to queues (Vinicius Costa Gomes)
- remove unused variable (kernel test robot)
- switch from mdio interrupt to polling
- mdio register without PHY address flag
- thread safe interrupt enable register
- add PTP_1588_CLOCK_OPTIONAL dependency to Kconfig
- introduce dmadev for DMA allocation
- mdiobus for platforms without device tree
- prepare MAC address support for platforms without device tree
- add missing interrupt disable to probe error path
v2:
- add C45 check (Andrew Lunn)
- forward phy_connect_direct() return value (Andrew Lunn)
- use phy_remove_link_mode() (Andrew Lunn)
- do not touch PHY directly, use PHY subsystem (Andrew Lunn)
- remove management data lock (Andrew Lunn)
- use phy_loopback (Andrew Lunn)
- remove GMII2RGMII handling, use xgmiitorgmii (Andrew Lunn)
- remove char device for direct TX/RX queue access (Andrew Lunn)
- mdio node for mdiobus (Rob Herring)
- simplify compatible node (Rob Herring)
- limit number of items of reg and interrupts nodes (Rob Herring)
- restrict phy-connection-type node (Rob Herring)
- reference to mdio.yaml under mdio node (Rob Herring)
- remove device tree (Michal Simek)
- fix %llx warning (kernel test robot)
- fix unused tmp variable warning (kernel test robot)
- add missing of_node_put() for of_parse_phandle()
- use devm_mdiobus_alloc()
- simplify mdiobus read/write
- reduce required nodes
- ethtool priv flags interface for loopback
- add missing static for some functions
- remove obsolete hardware defines
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Gerhard Engleder [Fri, 19 Nov 2021 22:58:25 +0000 (23:58 +0100)]
dt-bindings: net: Add tsnep Ethernet controller
The TSN endpoint Ethernet MAC is a FPGA based network device for
real-time communication.
It is integrated as normal Ethernet controller with
ethernet-controller.yaml and mdio.yaml.
Signed-off-by: Gerhard Engleder <gerhard@engleder-embedded.com> Reviewed-by: Rob Herring <robh@kernel.org> Reviewed-by: Andrew Lunn <andrew@lunn.ch> Signed-off-by: David S. Miller <davem@davemloft.net>
Gerhard Engleder [Fri, 19 Nov 2021 22:58:24 +0000 (23:58 +0100)]
dt-bindings: Add vendor prefix for Engleder
Engleder develops FPGA based controllers for real-time communication.
Signed-off-by: Gerhard Engleder <gerhard@engleder-embedded.com> Acked-by: Rob Herring <robh@kernel.org> Reviewed-by: Andrew Lunn <andrew@lunn.ch> Signed-off-by: David S. Miller <davem@davemloft.net>
Russell King (Oracle) [Fri, 19 Nov 2021 16:28:06 +0000 (16:28 +0000)]
net: phylink: handle NA interface mode in phylink_fwnode_phy_connect()
Commit 4904b6ea1f9db ("net: phy: phylink: Use PHY device interface if
N/A") introduced handling for the phy interface mode where this is not
known at phylink creation time. This was never added to the OF/fwnode
paths, but is necessary when the phy is present in DT, but the phy-mode
is not specified.
Add this handling.
Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk> Acked-by: Florian Fainelli <f.fainelli@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Sean Anderson [Fri, 19 Nov 2021 15:58:09 +0000 (10:58 -0500)]
net: phylink: Add helpers for c22 registers without MDIO
Some devices expose memory-mapped c22-compliant PHYs. Because these
devices do not have an MDIO bus, we cannot use the existing helpers.
Refactor the existing helpers to allow supplying the values for c22
registers directly, instead of using MDIO to access them. Only get_state
and set_advertisement are converted, since they contain the most complex
logic. Because set_advertisement is never actually used outside
phylink_mii_c22_pcs_config, move the MDIO-writing part into that
function. Because some modes do not need the advertisement register set
at all, we use -EINVAL for this purpose.
Additionally, a new function phylink_pcs_enable_an is provided to
determine whether to enable autonegotiation.
Signed-off-by: Sean Anderson <sean.anderson@seco.com> Reviewed-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk> Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Mon, 22 Nov 2021 12:31:50 +0000 (12:31 +0000)]
Merge branch 'ethtool-copybreak'
Guangbin Huang says:
====================
ethtool: add support to set/get tx copybreak buf size and rx buf len
This series add support to set/get tx copybreak buf size and rx buf len via
ethtool and hns3 driver implements them.
Tx copybreak buf size is used for tx copybreak feature which for small size
packet or frag. Use ethtool --get-tunable command to get it, and ethtool
--set-tunable command to set it, examples are as follow:
1. set tx spare buf size to 102400:
$ ethtool --set-tunable eth1 tx-buf-size 102400
Rx buf len is buffer length of each rx BD. Use ethtool -g command to get
it, and ethtool -G command to set it, examples are as follow:
1. set rx buf len to 4096
$ ethtool -G eth1 rx-buf-len 4096
2. get rx buf len
$ ethtool -g eth1
...
RX Buf Len: 4096
Change log:
V5 -> V6
1.Fix compile error for divers/s390.
V4 -> V5
1.Change struct ethtool_ringparam_ext to kernel_ethtool_ringparam.
2.change "__u32 rx_buf_len" to "u32 rx_buf_len".
V3 -> V4
1.Fix a few allmodconfig compile warning.
2.Add more '=' synbol to ethtool-netlink.rst to refine format.
3.Move definement of struct ethtool_ringparam_ext to include/linux/ethtool.h.
4.Move related modify of rings_fill_reply() from patch 4/6 to patch 3/6.
V2 -> V3
1.Remove documentation for tx copybreak buf size, there is description for
it in userspace ethtool.
2.Move extending parameters for get/set_ringparam function from patch3/6
to patch 4/6.
V1 -> V2
1.Add documentation for rx buf len and tx copybreak buf size.
2.Extend structure ringparam_ext for extenal ring params.
3.Change type of ETHTOOL_A_RINGS_RX_BUF_LEN from NLA_U32 to
NLA_POLICY_MIN(NLA_U32, 1).
4.Add supported_ring_params in ethtool_ops to indicate if support external
params.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Hao Chen [Thu, 18 Nov 2021 12:12:45 +0000 (20:12 +0800)]
net: hns3: remove the way to set tx spare buf via module parameter
The way to set tx spare buf via module parameter is not such
convenient as the way to set it via ethtool.
So,remove the way to set tx spare buf via module parameter.
Signed-off-by: Hao Chen <chenhao288@hisilicon.com> Signed-off-by: Guangbin Huang <huangguangbin2@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Hao Chen [Thu, 18 Nov 2021 12:12:44 +0000 (20:12 +0800)]
net: hns3: add support to set/get rx buf len via ethtool for hns3 driver
Rx buf len is for rx BD buffer size, support setting it via ethtool -G
parameter and getting it via ethtool -g parameter.
Signed-off-by: Hao Chen <chenhao288@hisilicon.com> Signed-off-by: Guangbin Huang <huangguangbin2@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Hao Chen [Thu, 18 Nov 2021 12:12:43 +0000 (20:12 +0800)]
ethtool: extend ringparam setting/getting API with rx_buf_len
Add two new parameters kernel_ringparam and extack for
.get_ringparam and .set_ringparam to extend more ring params
through netlink.
Signed-off-by: Hao Chen <chenhao288@hisilicon.com> Signed-off-by: Guangbin Huang <huangguangbin2@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
projects / users / jedix / linux-maple.git / log