Have an actual mlx5_sd instance in the core device, and fix the getter
accordingly. This allows SD stuff to flow, the feature becomes supported
only here.
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Reviewed-by: Gal Pressman <gal@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>

1) Each TX TLS device offloaded context has its own TIS object.  Extra work
is needed to get it working in a SD environment, where a stream can move
between different SQs (belonging to different mdevs).

2) Each RX TLS device offloaded context needs a DEK object from the DEK
pool.

Extra work is needed to get it working in a SD environment, as the DEK
pool currently falsely depends on TX cap, and is on the primary device
only.

Disallow this combination for now.
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Reviewed-by: Gal Pressman <gal@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>

Each queue counter object counts some events (in hardware) for the RQs
that are attached to it, like events of packet drops due to no receive
WQE (rx_out_of_buffer).

Each RQ can be attached to a queue counter only within the same vhca. To
still cover all RQs with these counters, we create multiple instances,
one per vhca.

The result that's shown to the user is now the sum of all instances.
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>

Implement driver support for the HW feature that allows RX steering of
one device to target other device's RQs.

In SD multi-mdev netdev mode, we set the secondaries into silent mode,
disconnecting them from the network. This feature is then used to steer
traffic from the primary to the secondaries.
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Reviewed-by: Gal Pressman <gal@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>

Distribute the channels between the different SD-devices to acheive
local numa node performance on multiple numas.

Each channel works against one specific mdev, creating all datapath
queues against it.

We distribute channels to mdevs in a round-robin policy.

Example for 2 mdevs and 6 channels:
+-------+---------+
| ch ix | mdev ix |
+-------+---------+
|   0   |    0    |
|   1   |    1    |
|   2   |    0    |
|   3   |    1    |
|   4   |    0    |
|   5   |    1    |
+-------+---------+

This round-robin distribution policy is preferred over another suggested
intuitive distribution, in which we first distribute one half of the
channels to mdev #0 and then the second half to mdev #1.

We prefer round-robin for a reason: it is less influenced by changes in
the number of channels. The mapping between channel index and mdev is
fixed, no matter how many channels the user configures. As the channel
stats are persistent to channels closure, changing the mapping every
single time would turn the accumulative stats less representing of the
channel's history.

Per-channel objects should stop using the primary mdev (priv->mdev)
directly, and instead move to using their own channel's mdev.
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Reviewed-by: Gal Pressman <gal@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>

Traffic queues will be created on all devices, including the
secondaries. Create the needed core layer resources for them as well.
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Reviewed-by: Gal Pressman <gal@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>

Integrate the SD library calls into the auxiliary_driver ops in
preparation for creating a single netdev for the multiple devices
belonging to the same SD group.

SD is still disabled at this stage. It is enabled by a downstream patch
when all needed parts are implemented.

The netdev is created only when the SD group, with all its participants,
are ready. It is later destroyed if any of the participating devices
drops.
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Reviewed-by: Gal Pressman <gal@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>

Print to kernel log when an SD group moves from/to ready state.
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>

Implement the needed SD steering adjustments for the primary and
secondaries.

While the SD multiple devices are used to avoid cross-numa memory, when
it comes to chip level all traffic goes only through the primary device.
The secondaries are forced to silent mode, to guarantee they are not
involved in any unexpected ingress/egress traffic.

In RX, secondary devices will not have steering objects. Traffic will be
steered from the primary device to the RQs of a secondary device using
advanced cross-vhca RX steering capabilities.

In TX, the primary creates a new TX flow table, which is aliased by the
secondaries.
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>

Use devcom to communicate between the different devices. Add a new
devcom component type for this.

Each device registers itself to the devcom component <SD, group ID>.
Once all devices of a component are registered, the component becomes
ready, and a primary device is elected.

In principle, any of the devices can act as a primary, they are all
capable, and a random election would've worked. However, we aim to
achieve predictability and consistency, hence each group always choses
the same device, with the lowest PCI BUS number, as primary.
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>

Add implementation for querying the MPIR register for Socket-Direct
attributes, and instantiating a SD struct accordingly.
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>

Add Socket-Direct API with empty/minimal implementation.
We fill-in the implementation gradually in downstream patches.
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Reviewed-by: Gal Pressman <gal@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>

The sd_group field moved in the HW spec from the MPIR register
to the vport context.
Align the query accordingly.

Fixes: f5e95632 ("net/mlx5: Expose Management PCIe Index Register (MPIR)")
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>

The cited commit moved the code of mlx5e_create_tises() and changed the
loop to create TISes over MLX5_MAX_PORTS constant value, instead of
getting the correct lag ports supported by the device, which can cause
FW errors on devices with less than MLX5_MAX_PORTS ports.

Change that back to mlx5e_get_num_lag_ports(mdev).

Also IPoIB interfaces create there own TISes, they don't use the eth
TISes, pass a flag to indicate that.

Fixes: b25bd37c ("net/mlx5: Move TISes from priv to mdev HW resources")
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>

Victor Nogueira says:

====================
net: sched: Make tc-related drop reason more flexible for remaining qdiscs

This patch builds on Daniel's patch[1] to add initial support of tc drop
reason. The main goal is to distinguish between policy and error drops for
the remainder of the egress qdiscs (other than clsact).
The drop reason is set by cls_api and act_api in the tc skb cb in case
any error occurred in the data path.

Also add new skb drop reasons that are idiosyncratic to TC.

[1] https://lore.kernel.org/all/20231009092655.22025-1-daniel@iogearbox.net

Changes in V5:
- Drop "EXT_" from cookie error's drop reason name in doc

Changes in V4:
- Condense all the cookie drop reasons into one

Changes in V3:
- Removed duplicate assignment
- Rename function tc_skb_cb_drop_reason to tcf_get_drop_reason
- Move zone field upwards in struct tc_skb_cb to move hole to the end of
  the struct

Changes in V2:
- Dropped RFC tag
- Removed check for drop reason being overwritten by filter in cls_api.c
- Simplified logic and removed function tcf_init_drop_reason
====================
Signed-off-by: David S. Miller <davem@davemloft.net>

Continue expanding Daniel's patch by adding new skb drop reasons that
are idiosyncratic to TC.

More specifically:

- SKB_DROP_REASON_TC_COOKIE_ERROR: An error occurred whilst
  processing a tc ext cookie.

- SKB_DROP_REASON_TC_CHAIN_NOTFOUND: tc chain lookup failed.

- SKB_DROP_REASON_TC_RECLASSIFY_LOOP: tc exceeded max reclassify loop
  iterations
Signed-off-by: Victor Nogueira <victor@mojatatu.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

Incrementing on Daniel's patch[1], make tc-related drop reason more
flexible for remaining qdiscs - that is, all qdiscs aside from clsact.
In essence, the drop reason will be set by cls_api and act_api in case
any error occurred in the data path. With that, we can give the user more
detailed information so that they can distinguish between a policy drop
or an error drop.

[1] https://lore.kernel.org/all/20231009092655.22025-1-daniel@iogearbox.netSigned-off-by: Victor Nogueira <victor@mojatatu.com>
Acked-by: Daniel Borkmann <daniel@iogearbox.net>
Reviewed-by: Simon Horman <horms@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

Move drop_reason from struct tcf_result to skb cb - more specifically to
struct tc_skb_cb. With that, we'll be able to also set the drop reason for
the remaining qdiscs (aside from clsact) that do not have access to
tcf_result when time comes to set the skb drop reason.
Signed-off-by: Victor Nogueira <victor@mojatatu.com>
Acked-by: Daniel Borkmann <daniel@iogearbox.net>
Reviewed-by: Simon Horman <horms@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

This adds support for the LED's on most chip versions. Excluded are
the old non-PCIe versions and RTL8125. RTL8125 has a different LED
register layout, support for it will follow later.

LED's can be controlled from userspace using the netdev LED trigger.

Tested on RTL8168h.

Note: The driver can't know which LED's are actually physically
wired. Therefore not every LED device may represent a physically
available LED.
Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Ido Schimmel says:

====================
Add MDB bulk deletion support

This patchset adds MDB bulk deletion support, allowing user space to
request the deletion of matching entries instead of dumping the entire
MDB and issuing a separate deletion request for each matching entry.
Support is added in both the bridge and VXLAN drivers in a similar
fashion to the existing FDB bulk deletion support.

The parameters according to which bulk deletion can be performed are
similar to the FDB ones, namely: Destination port, VLAN ID, state (e.g.,
"permanent"), routing protocol, source / destination VNI, destination IP
and UDP port. Flushing based on flags (e.g., "offload", "fast_leave",
"added_by_star_ex", "blocked") is not currently supported, but can be
added in the future, if a use case arises.

Patch #1 adds a new uAPI attribute to allow specifying the state mask
according to which bulk deletion will be performed, if any.

Patch #2 adds a new policy according to which bulk deletion requests
(with 'NLM_F_BULK' flag set) will be parsed.

Patches #3-#4 add a new NDO for MDB bulk deletion and invoke it from the
rtnetlink code when a bulk deletion request is made.

Patches #5-#6 implement the MDB bulk deletion NDO in the bridge and
VXLAN drivers, respectively.

Patch #7 allows user space to issue MDB bulk deletion requests by no
longer rejecting the 'NLM_F_BULK' flag when it is set in 'RTM_DELMDB'
requests.

Patches #8-#9 add selftests for both drivers, for both good and bad
flows.

iproute2 changes can be found here [1].

https://github.com/idosch/iproute2/tree/submit/mdb_flush_v1
====================
Signed-off-by: David S. Miller <davem@davemloft.net>

Add test cases to verify the behavior of the MDB bulk deletion
functionality in the VXLAN driver.
Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Acked-by: Petr Machata <petrm@nvidia.com>
Acked-by: Nikolay Aleksandrov <razor@blackwall.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

Add test cases to verify the behavior of the MDB bulk deletion
functionality in the bridge driver.
Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Acked-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: Hangbin Liu <liuhangbin@gmail.com>
Acked-by: Nikolay Aleksandrov <razor@blackwall.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

Now that both the common code as well as individual drivers support MDB
bulk deletion, allow user space to make such requests.
Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Reviewed-by: Petr Machata <petrm@nvidia.com>
Acked-by: Nikolay Aleksandrov <razor@blackwall.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

Implement MDB bulk deletion support in the VXLAN driver, allowing MDB
entries to be deleted in bulk according to provided parameters.
Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Reviewed-by: Petr Machata <petrm@nvidia.com>
Acked-by: Nikolay Aleksandrov <razor@blackwall.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

Implement MDB bulk deletion support in the bridge driver, allowing MDB
entries to be deleted in bulk according to provided parameters.
Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Reviewed-by: Petr Machata <petrm@nvidia.com>
Acked-by: Nikolay Aleksandrov <razor@blackwall.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

Invoke the new MDB bulk deletion device operation when the 'NLM_F_BULK'
flag is set in the netlink message header.
Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Reviewed-by: Petr Machata <petrm@nvidia.com>
Acked-by: Nikolay Aleksandrov <razor@blackwall.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

Add MDB net device operation that will be invoked by rtnetlink code in
response to received 'RTM_DELMDB' messages with the 'NLM_F_BULK' flag
set. Subsequent patches will implement the operation in the bridge and
VXLAN drivers.
Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Reviewed-by: Petr Machata <petrm@nvidia.com>
Acked-by: Nikolay Aleksandrov <razor@blackwall.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

For MDB bulk delete we will need to validate 'MDBA_SET_ENTRY'
differently compared to regular delete. Specifically, allow the ifindex
to be zero (in case not filtering on bridge port) and force the address
to be zero as bulk delete based on address is not supported.

Do that by introducing a new policy and choosing the correct policy
based on the presence of the 'NLM_F_BULK' flag in the netlink message
header. Use nlmsg_parse() for strict validation.
Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Reviewed-by: Petr Machata <petrm@nvidia.com>
Acked-by: Nikolay Aleksandrov <razor@blackwall.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

Currently, the 'state' field in 'struct br_port_msg' can be set to 1 if
the MDB entry is permanent or 0 if it is temporary. Additional states
might be added in the future.

In a similar fashion to 'NDA_NDM_STATE_MASK', add an MDB state mask uAPI
attribute that will allow the upcoming bulk deletion API to bulk delete
MDB entries with a certain state or any state.
Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Reviewed-by: Petr Machata <petrm@nvidia.com>
Acked-by: Nikolay Aleksandrov <razor@blackwall.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

Maintain Consistent Formatting: Insert Space after #include
Suggested-by: Jakub Kicinski <kuba@kernel.org>
Reviewed-by: Subbaraya Sundeep <sbhatta@marvell.com>
Signed-off-by: Wang Jinchao <wangjinchao@xfusion.com>
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Daniel Borkmann says:

====================
pull-request: bpf-next 2023-12-19

Hi David, hi Jakub, hi Paolo, hi Eric,

The following pull-request contains BPF updates for your *net-next* tree.

We've added 2 non-merge commits during the last 1 day(s) which contain
a total of 40 files changed, 642 insertions(+), 2926 deletions(-).

The main changes are:

1) Revert all of BPF token-related patches for now as per list discussion [0],
   from Andrii Nakryiko.

   [0] https://lore.kernel.org/bpf/CAHk-=wg7JuFYwGy=GOMbRCtOL+jwSQsdUaBsRWkDVYbxipbM5A@mail.gmail.com

2) Fix a syzbot-reported use-after-free read in nla_find() triggered from
   bpf_skb_get_nlattr_nest() helper, from Jakub Kicinski.

bpf-next-for-netdev

* tag 'for-netdev' of https://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf-next:
  Revert BPF token-related functionality
  bpf: Use nla_ok() instead of checking nla_len directly
====================

Link: https://lore.kernel.org/r/20231219170359.11035-1-daniel@iogearbox.netSigned-off-by: Paolo Abeni <pabeni@redhat.com>

This patch includes the following revert (one  conflicting BPF FS
patch and three token patch sets, represented by merge commits):
  - revert 0f5d5454 "Merge branch 'bpf-fs-mount-options-parsing-follow-ups'";
  - revert 750e7857 "bpf: Support uid and gid when mounting bpffs";
  - revert 73376328 "Merge branch 'bpf-token-support-in-libbpf-s-bpf-object'";
  - revert c35919dc "Merge branch 'bpf-token-and-bpf-fs-based-delegation'".

Link: https://lore.kernel.org/bpf/CAHk-=wg7JuFYwGy=GOMbRCtOL+jwSQsdUaBsRWkDVYbxipbM5A@mail.gmail.comSigned-off-by: Andrii Nakryiko <andrii@kernel.org>

Jiri Pirko says:

====================
devlink: introduce notifications filtering

From: Jiri Pirko <jiri@nvidia.com>

Currently the user listening on a socket for devlink notifications
gets always all messages for all existing devlink instances and objects,
even if he is interested only in one of those. That may cause
unnecessary overhead on setups with thousands of instances present.

User is currently able to narrow down the devlink objects replies
to dump commands by specifying select attributes.

Allow similar approach for notifications providing user a new
notify-filter-set command to select attributes with values
the notification message has to match. In that case, it is delivered
to the socket.

Note that the filtering is done per-socket, so multiple users may
specify different selection of attributes with values.

This patchset initially introduces support for following attributes:
DEVLINK_ATTR_BUS_NAME
DEVLINK_ATTR_DEV_NAME
DEVLINK_ATTR_PORT_INDEX

Patches #1 - #4 are preparations in devlink code, patch #3 is
                an optimization done on the way.
Patches #5 - #7 are preparations in netlink and generic netlink code.
Patch #8 is the main one in this set implementing of
         the notify-filter-set command and the actual
         per-socket filtering.
Patch #9 extends the infrastructure allowing to filter according
         to a port index.

Example:
$ devlink mon port pci/0000:08:00.0/32768
[port,new] pci/0000:08:00.0/32768: type notset flavour pcisf controller 0 pfnum 0 sfnum 107 splittable false
  function:
    hw_addr 00:00:00:00:00:00 state inactive opstate detached roce enable
[port,new] pci/0000:08:00.0/32768: type eth flavour pcisf controller 0 pfnum 0 sfnum 107 splittable false
  function:
    hw_addr 00:00:00:00:00:00 state inactive opstate detached roce enable
[port,new] pci/0000:08:00.0/32768: type eth netdev eth3 flavour pcisf controller 0 pfnum 0 sfnum 107 splittable false
  function:
    hw_addr 00:00:00:00:00:00 state inactive opstate detached roce enable
[port,new] pci/0000:08:00.0/32768: type eth netdev eth3 flavour pcisf controller 0 pfnum 0 sfnum 107 splittable false
  function:
    hw_addr 00:00:00:00:00:00 state inactive opstate detached roce enable
[port,new] pci/0000:08:00.0/32768: type eth flavour pcisf controller 0 pfnum 0 sfnum 107 splittable false
  function:
    hw_addr 00:00:00:00:00:00 state inactive opstate detached roce enable
[port,new] pci/0000:08:00.0/32768: type notset flavour pcisf controller 0 pfnum 0 sfnum 107 splittable false
  function:
    hw_addr 00:00:00:00:00:00 state inactive opstate detached roce enable
[port,del] pci/0000:08:00.0/32768: type notset flavour pcisf controller 0 pfnum 0 sfnum 107 splittable false
  function:
    hw_addr 00:00:00:00:00:00 state inactive opstate detached roce enable
====================

Link: https://lore.kernel.org/r/20231216123001.1293639-1-jiri@resnulli.usSigned-off-by: Paolo Abeni <pabeni@redhat.com>

Expose the previously introduced notification multicast messages
filtering infrastructure and allow the user to select messages using
port index.
Signed-off-by: Jiri Pirko <jiri@nvidia.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>

Currently the user listening on a socket for devlink notifications
gets always all messages for all existing instances, even if he is
interested only in one of those. That may cause unnecessary overhead
on setups with thousands of instances present.

User is currently able to narrow down the devlink objects replies
to dump commands by specifying select attributes.

Allow similar approach for notifications. Introduce a new devlink
NOTIFY_FILTER_SET which the user passes the select attributes. Store
these per-socket and use them for filtering messages
during multicast send.
Signed-off-by: Jiri Pirko <jiri@nvidia.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>

Currently it is possible for netlink kernel user to pass custom
filter function to broadcast send function netlink_broadcast_filtered().
However, this is not exposed to multicast send and to generic
netlink users.

Extend the api and introduce a netlink helper nlmsg_multicast_filtered()
and a generic netlink helper genlmsg_multicast_netns_filtered()
to allow generic netlink families to specify filter function
while sending multicast messages.
Signed-off-by: Jiri Pirko <jiri@nvidia.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>

Make the code using filter function a bit nicer by consolidating the
filter function arguments using typedef.
Suggested-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: Jiri Pirko <jiri@nvidia.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>

Introduce an xarray for Generic netlink family to store per-socket
private. Initialize this xarray only if family uses per-socket privs.

Introduce genl_sk_priv_get() to get the socket priv pointer for a family
and initialize it in case it does not exist.
Introduce __genl_sk_priv_get() to obtain socket priv pointer for a
family under RCU read lock.

Allow family to specify the priv size, init() and destroy() callbacks.
Signed-off-by: Jiri Pirko <jiri@nvidia.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>

Introduce a helper devlink_nl_notify_send() so each object notification
function does not have to call genlmsg_multicast_netns() with the same
arguments.
Signed-off-by: Jiri Pirko <jiri@nvidia.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>

Introduce devlink_nl_notify_need() helper and using it to check at the
beginning of notification functions to avoid overhead of composing
notification messages in case nobody listens.
Signed-off-by: Jiri Pirko <jiri@nvidia.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>