Spectrum-1 and Spectrum-2 do not have a per-TC counter of number of packets
marked by ECN. The value reported currently is the total number of marked
packets. Showing this value at individual TC Qdiscs is misleading.

Move the counter to ethtool instead.
Signed-off-by: NPetr Machata <petrm@mellanox.com>
Signed-off-by: NIdo Schimmel <idosch@mellanox.com>
Signed-off-by: NJiri Pirko <jiri@mellanox.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Currently, only one SFN query is done from repetitive work at a time,
processing 64 entries. Another work iteration is scheduled in 100ms,
that means that the max rate of learned FDB entries is limited to 6400/s.
That is slow. Fix this by doing 2 optimizations:
1) Run 10 SFN queries at a time.
2) In case the SFN is not drained, schedule work with 0 delay to allow
   to continue processing rest of the records.

On a testing setup with 500K entries the time to process decreased
from 870secs to 10secs.
Signed-off-by: NJiri Pirko <jiri@mellanox.com>
Tested-by: NAlex Kushnarov <alexanderk@mellanox.com>
Signed-off-by: NIdo Schimmel <idosch@mellanox.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Return NULL instead of 0.
Signed-off-by: NSaeed Mahameed <saeedm@mellanox.com>
Reviewed-by: NMoshe Shemesh <moshe@mellanox.com>

drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c:191:13:
sparse: warning: incorrect type in assignment (different base types)
Signed-off-by: NSaeed Mahameed <saeedm@mellanox.com>
Reviewed-by: NMoshe Shemesh <moshe@mellanox.com>

Clang warns:

 In file included from
 ../drivers/net/ethernet/mellanox/mlx5/core/main.c:73:
 ../drivers/net/ethernet/mellanox/mlx5/core/diag/rsc_dump.h:4:9: warning:
 '__MLX5_RSC_DUMP_H' is used as a header guard here, followed by #define
 of a different macro [-Wheader-guard]
 #ifndef __MLX5_RSC_DUMP_H
         ^~~~~~~~~~~~~~~~~
 ../drivers/net/ethernet/mellanox/mlx5/core/diag/rsc_dump.h:5:9: note:
 '__MLX5_RSC_DUMP__H' is defined here; did you mean '__MLX5_RSC_DUMP_H'?
 #define __MLX5_RSC_DUMP__H
         ^~~~~~~~~~~~~~~~~~
         __MLX5_RSC_DUMP_H
 1 warning generated.

Make them match to get the intended behavior and remove the warning.

Fixes: 12206b17 ("net/mlx5: Add support for resource dump")
Link: https://github.com/ClangBuiltLinux/linux/issues/897Signed-off-by: NNathan Chancellor <natechancellor@gmail.com>
Reviewed-by: NJason Gunthorpe <jgg@mellanox.com>
Signed-off-by: NSaeed Mahameed <saeedm@mellanox.com>

We can avoid an indirect call per compressed completion wrapping the
completion handling call with the appropriate helper.
Signed-off-by: NTariq Toukan <tariqt@mellanox.com>
Reviewed-by: NMaxim Mikityanskiy <maximmi@mellanox.com>
Signed-off-by: NSaeed Mahameed <saeedm@mellanox.com>

We can avoid an indirect call per NAPI cycle wrapping the RX descriptors
posting call with the appropriate helper.
Signed-off-by: NTariq Toukan <tariqt@mellanox.com>
Reviewed-by: NMaxim Mikityanskiy <maximmi@mellanox.com>
Signed-off-by: NSaeed Mahameed <saeedm@mellanox.com>

The current steps that are performed when the trust state changes, if
the channels are active:

1. The trust state is changed in hardware.

2. The new inline mode is calculated.

3. If the new inline mode is different, the channels are recreated using
the new inline mode.

This approach has some issues:

1. There is a time gap between changing trust state in hardware and
starting sending enough inline headers (the latter happens after
recreation of channels). It leads to failed transmissions and error
CQEs.

2. If the new channels fail to open, we'll be left with the old ones,
but the hardware will be configured for the new trust state, so the
interval when we can see TX errors never ends.

This patch fixes the issues above by moving the trust state change into
the preactivate hook that runs during the recreation of the channels
when no channels are active, so it eliminates the gap of partially
applied configuration. If the inline mode doesn't change with the change
of the trust state, the channels won't be recreated, just like before
this patch.
Signed-off-by: NMaxim Mikityanskiy <maximmi@mellanox.com>
Reviewed-by: NTariq Toukan <tariqt@mellanox.com>
Signed-off-by: NSaeed Mahameed <saeedm@mellanox.com>

Sometimes the preactivate hook of mlx5e_safe_switch_channels needs more
parameters than just struct mlx5e_priv *. For such cases, a new
parameter (void *context) is added to preactivate hooks.

Some of the existing normal functions are currently used as preactivate
callbacks. To avoid adding an extra unused parameter, they are wrapped
in an automatic way using the MLX5E_DEFINE_PREACTIVATE_WRAPPER_CTX
macro.
Signed-off-by: NMaxim Mikityanskiy <maximmi@mellanox.com>
Reviewed-by: NTariq Toukan <tariqt@mellanox.com>
Signed-off-by: NSaeed Mahameed <saeedm@mellanox.com>

Currently mlx5e_switch_priv_channels expects that the preactivate hook
doesn't fail, however, it can fail, because it may set hardware
parameters. This commit addresses this issue and provides a way to
recover from failures of the preactivate hook: the old channels are not
closed until the point where nothing can fail anymore, so in case
preactivate fails, the driver can roll back the old channels and
activate them again.
Signed-off-by: NMaxim Mikityanskiy <maximmi@mellanox.com>
Reviewed-by: NTariq Toukan <tariqt@mellanox.com>
Signed-off-by: NSaeed Mahameed <saeedm@mellanox.com>

The number of queues is now updated by mlx5e_update_netdev_queues in a
centralized way, when no channels are active. Remove an extra occurrence
of netif_set_real_num_tx_queues to prepare it for the next commit.
Signed-off-by: NMaxim Mikityanskiy <maximmi@mellanox.com>
Reviewed-by: NTariq Toukan <tariqt@mellanox.com>
Signed-off-by: NSaeed Mahameed <saeedm@mellanox.com>

Currently, mlx5e notifies the kernel about the number of queues and sets
the default XPS cpumasks when channels are activated. This
implementation has several corner cases, in which the kernel may not be
updated on time, or XPS cpumasks may be reset when not directly touched
by the user.

This commit fixes these corner cases to match the following expected
behavior:

1. The number of queues always corresponds to the number of channels
configured.

2. XPS cpumasks are set to driver's defaults on netdev attach.

3. XPS cpumasks set by user are not reset, unless the number of channels
changes. If the number of channels changes, they are reset to driver's
defaults. (In general case, when the number of channels increases or
decreases, it's not possible to guess how to convert the current XPS
cpumasks to work with the new number of channels, so we let the user
reconfigure it if they change the number of channels.)

XPS cpumasks are no longer stored per channel. Only one temporary
cpumask is used. The old stored cpumasks didn't reflect the user's
changes and were not used after applying them.

A scratchpad area is added to struct mlx5e_priv. As cpumask_var_t
requires allocation, and the preactivate hook can't fail, we need to
preallocate the temporary cpumask in advance. It's stored in the
scratchpad.

Fixes: 149e566f ("net/mlx5e: Expand XPS cpumask to cover all online cpus")
Signed-off-by: NMaxim Mikityanskiy <maximmi@mellanox.com>
Reviewed-by: NTariq Toukan <tariqt@mellanox.com>
Signed-off-by: NSaeed Mahameed <saeedm@mellanox.com>

mlx5e_ethtool_set_channels updates the indirection table before
switching to the new channels. If the switch fails, the indirection
table is new, but the channels are old, which is wrong. Fix it by using
the preactivate hook of mlx5e_safe_switch_channels to update the
indirection table at the stage when nothing can fail anymore.

As the code that updates the indirection table is now encapsulated into
a new function, use that function in the attach flow when the driver has
to reduce the number of channels, and prepare the code for the next
commit.

Fixes: 85082dba ("net/mlx5e: Correctly handle RSS indirection table when changing number of channels")
Signed-off-by: NMaxim Mikityanskiy <maximmi@mellanox.com>
Reviewed-by: NTariq Toukan <tariqt@mellanox.com>
Signed-off-by: NSaeed Mahameed <saeedm@mellanox.com>

mlx5e_safe_switch_channels accepts a callback to be called before
activating new channels. It is intended to configure some hardware
parameters in cases where channels are recreated because some
configuration has changed.

Recently, this callback has started being used to update the driver's
internal MLX5E_STATE_XDP_OPEN flag, and the following patches also
intend to use this callback for software preparations. This patch
renames the hw_modify callback to preactivate, so that the name fits
better.
Signed-off-by: NMaxim Mikityanskiy <maximmi@mellanox.com>
Reviewed-by: NTariq Toukan <tariqt@mellanox.com>
Signed-off-by: NSaeed Mahameed <saeedm@mellanox.com>

As a preparation for one of the following commits, create a function to
encapsulate the code that notifies the kernel about the new amount of
RX and TX queues. The code will be called multiple times in the next
commit.
Signed-off-by: NMaxim Mikityanskiy <maximmi@mellanox.com>
Reviewed-by: NTariq Toukan <tariqt@mellanox.com>
Signed-off-by: NSaeed Mahameed <saeedm@mellanox.com>

The LRO boolean state in params->lro_en must not be set in case
the NIC is not capable.
Enforce this check and remove the TODO comment.
Signed-off-by: NTariq Toukan <tariqt@mellanox.com>
Signed-off-by: NSaeed Mahameed <saeedm@mellanox.com>

We shall always extract channel index out of the txq, regardless
of the relation between txq_ix and num channels. The extraction is
always valid, as if txq is smaller than number of channels,
txq_ix == priv->txq2sq[txq_ix]->ch_ix.

By doing so, we can remove an if clause from the select queue method,
and have one flow for all packets.
Signed-off-by: NEran Ben Elisha <eranbe@mellanox.com>
Reviewed-by: NTariq Toukan <tariqt@mellanox.com>
Signed-off-by: NSaeed Mahameed <saeedm@mellanox.com>

Use the cookie index received along with the packet to lookup original
flow_offload cookie binary and pass it down to devlink_trap_report().
Add "fa_cookie" metadata to the ACL trap.
Signed-off-by: NJiri Pirko <jiri@mellanox.com>
Signed-off-by: NIdo Schimmel <idosch@mellanox.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

In case the received packet comes in due to one of ACL discard traps,
take the user_def_val_orig_pkt_len field from CQE and store it
in skb->cb as ACL cookie index.
Signed-off-by: NJiri Pirko <jiri@mellanox.com>
Signed-off-by: NIdo Schimmel <idosch@mellanox.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Track cookies coming down to driver by flow_offload.
Assign a cookie_index to each unique cookie binary. Use previously
defined "Trap with userdef" flex action to ask HW to pass cookie_index
alongside with the dropped packets.
Signed-off-by: NJiri Pirko <jiri@mellanox.com>
Signed-off-by: NIdo Schimmel <idosch@mellanox.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Expose "Trap action with userdef". It is the same as already
defined "Trap action" with a difference that it would ask the policy
engine to pass arbitrary value (userdef) alongside with received packets.
This would be later on used to carry cookie index.
Signed-off-by: NJiri Pirko <jiri@mellanox.com>
Signed-off-by: NIdo Schimmel <idosch@mellanox.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Add cookie argument to devlink_trap_report() allowing driver to pass in
the user cookie. Pass on the cookie down to drop monitor code.
Signed-off-by: NJiri Pirko <jiri@mellanox.com>
Signed-off-by: NIdo Schimmel <idosch@mellanox.com>
Reviewed-by: NJakub Kicinski <kuba@kernel.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Add the trap group used to report ACL drops. Setup the trap IDs for
ingress/egress flow action drop. Register the two packet traps
associated with ACL trap group with devlink during driver
initialization. As these are "source traps", set the disabled
trap group to be the dummy, discarding as many packets in HW
as possible.
Signed-off-by: NJiri Pirko <jiri@mellanox.com>
Signed-off-by: NIdo Schimmel <idosch@mellanox.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

For "source traps" it is not possible to change HPKT action to discard.
But there is still need to disallow packets arriving to CPU as much as
possible. Handle this by introduction of a "dummy group". It has a
"thin" policer, which passes as less packets to CPU as possible. The
rest is going to be discarded there. The "dummy group" is to be used
later on by ACL trap (which is a "source trap").
Signed-off-by: NJiri Pirko <jiri@mellanox.com>
Signed-off-by: NIdo Schimmel <idosch@mellanox.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Extend the mlxsw_listener struct to contain trap group for disabled
traps too. Rename the original "trap_group" item to "en_trap_group" as
it represents enabled state. Let both groups be the same for MLXSW_RXL
however extend MLXSW_RXL_DIS to register separate groups for enable and
disable.
Signed-off-by: NJiri Pirko <jiri@mellanox.com>
Signed-off-by: NIdo Schimmel <idosch@mellanox.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

For source traps, the "thin policer" is going to be used in order
to reduce the amount of trapped packets to minimum. However, there
will be still small number of packets coming in that need to be dropped
in the driver. Allow to enable/disable rx_listener related to specific
trap in order to prevent unwanted packets to go up the stack.
Signed-off-by: NJiri Pirko <jiri@mellanox.com>
Signed-off-by: NIdo Schimmel <idosch@mellanox.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Introduce a new set of traps:
DISCARD_INGRESS_ACL and DISCARD_EGRESS_ACL
Set the trap_action from NOP to TRAP which causes the packets dropped
by the TRAP action to be trapped under new trap IDs, depending on the
ingress/egress binding point.
Signed-off-by: NJiri Pirko <jiri@mellanox.com>
Signed-off-by: NIdo Schimmel <idosch@mellanox.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The ACL flex action will have to know if it is in ingress or egress,
so it can use correct trap ID. Pass the ingress indication down to it.
Signed-off-by: NJiri Pirko <jiri@mellanox.com>
Signed-off-by: NIdo Schimmel <idosch@mellanox.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Action drop is going to be tracked by two separate traps, one for
ingress and one for egress. Prepare for it and disallow the possibility
to have drop action in blocks which are bound to both ingress and
egress.
Signed-off-by: NJiri Pirko <jiri@mellanox.com>
Signed-off-by: NIdo Schimmel <idosch@mellanox.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Count the number of ingress and egress block bindings. Use the egress
counter in "is_egress_bound" helper. Add couple of helpers to check
ingress and mixed bound.
Signed-off-by: NJiri Pirko <jiri@mellanox.com>
Signed-off-by: NIdo Schimmel <idosch@mellanox.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Rename function mlxsw_core_trap_action_set() to
mlxsw_core_trap_state_set() and pass bool enabled instead of action.
Figure out the action according to the enabled state there.
Signed-off-by: NJiri Pirko <jiri@mellanox.com>
Signed-off-by: NIdo Schimmel <idosch@mellanox.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The listener fields en_action and dis_action now contain the actions to
be used for TRAP and DROP devlink trap actions. Use them directly
instead of the hard-coded values.
Signed-off-by: NJiri Pirko <jiri@mellanox.com>
Signed-off-by: NIdo Schimmel <idosch@mellanox.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Introduce a new macro MLXSW_RXL_DIS that allows to register listeners
as disabled. That allows that from now on, the "action" can be
understood always as "enabled action" and "unreg_action" as "disabled
action". Rename them and treat them accordingly.

Use the new macro for defining drops in spectrum_trap.c.
Signed-off-by: NJiri Pirko <jiri@mellanox.com>
Signed-off-by: NIdo Schimmel <idosch@mellanox.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Currently it does not really matter if it is set to DISCARD or
SET_FW_DEFAULT because it is set only during unregister of the listener.
The unreg_action is going to be used for disabling the listener too, so
change to SET_FW_DEFAULT and ensure the HW is going to behave correctly.
Signed-off-by: NJiri Pirko <jiri@mellanox.com>
Signed-off-by: NIdo Schimmel <idosch@mellanox.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

There is a spelling mistake in a literal string. Fix it.
Signed-off-by: NColin Ian King <colin.king@canonical.com>
Acked-by: NJiri Pirko <jiri@mellanox.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Since commit f3a52c61 ("mlxsw: pci: Utilize MRSR register to perform
FW reset") the driver no longer issues a reset via the PCI BAR, so the
offset of the reset bit is unused. Remove it.
Signed-off-by: NIdo Schimmel <idosch@mellanox.com>
Acked-by: NJiri Pirko <jiri@mellanox.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

During packet receive, only the first matching RX listener
in rx_listener_list is going to get the packet. So there is no
meaning in registering two equal listeners with different privs.
Remove priv from equality comparison and disable possibility
of doing it.
Signed-off-by: NJiri Pirko <jiri@mellanox.com>
Signed-off-by: NIdo Schimmel <idosch@mellanox.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

There are couple of places where block pointer as a function argument
can be const. So make those const.
Signed-off-by: NJiri Pirko <jiri@mellanox.com>
Signed-off-by: NIdo Schimmel <idosch@mellanox.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The global arrays are treated as const, they should be const, so make
them const.
Signed-off-by: NJiri Pirko <jiri@mellanox.com>
Signed-off-by: NIdo Schimmel <idosch@mellanox.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

No need to initialize to false, so remove it.
Signed-off-by: NJiri Pirko <jiri@mellanox.com>
Signed-off-by: NIdo Schimmel <idosch@mellanox.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>