Our implementation has restriction on stack pointers for function
calls. Move the common checks into a helper for reuse. The state
has to be encapsulated into a structure to support parameters
other than BPF_REG_2.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: Quentin Monnet <quentin.monnet@netronome.com>
Reviewed-by: Jiong Wang <jiong.wang@netronome.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
This patch is the front end of this optimisation, it detects and marks
those packet reads that could be cached. Then the optimisation "backend"
will be activated automatically.
Signed-off-by: Jiong Wang <jiong.wang@netronome.com>
Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
This patch add the support for unaligned read offset, i.e. the read offset
to the start of packet cache area is not aligned to REG_WIDTH. In this
case, the read area might across maximum three transfer-in registers.
Signed-off-by: Jiong Wang <jiong.wang@netronome.com>
Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
This patch assumes there is a packet data cache, and would try to read
packet data from the cache instead of from memory.
This patch only implements the optimisation "backend", it doesn't build
the packet data cache, so this optimisation is not enabled.
This patch has only enabled aligned packet data read, i.e. when the read
offset to the start of cache is REG_WIDTH aligned.
Signed-off-by: Jiong Wang <jiong.wang@netronome.com>
Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
An error TX completion (CQE) which arrived on a specific SQ indicates
that this SQ got moved by the hardware to error state, which means all
pending and incoming TX requests are dropped or will be dropped and no
further "Good" CQEs will be generated for that SQ.
Before this patch TX completions (CQEs) were not monitored and were
handled as a regular CQE. This caused the SQ to stay in an error state,
making it useless for xmiting new packets.
Mitigation plan:
In case of an error completion, schedule a recovery work which would do
the following:
- Mark the TXQ as DRV_XOFF to disable new packets to arrive from the
stack
- NAPI to flush all pending SQ WQEs (via flush_in_error_en bit) to
release SW and HW resources(SKB, DMA, etc) and have the SQ and CQ
consumer/producer indices synced.
- Modify the SQ state ERR -> RST -> RDY (restart the SQ).
- Reactivate the SQ and reset SQ cc and pc
If we identify two consecutive requests for SQ recover in less than
500 msecs, drop the recover request to avoid CPU overload, as this
scenario most likely happened due to a severe repeated bug.
In addition, add SQ recover SW counter to monitor successful recoveries.
Signed-off-by: Eran Ben Elisha <eranbe@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Monitor and dump xmit error completions. In addition, add err_cqe
counter to track the number of error completion per send queue.
Signed-off-by: Eran Ben Elisha <eranbe@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Move query SQ state function from mlx5_ib to mlx5_core in order to
have it in shared code.
It will be used in a downstream patch from mlx5e.
Signed-off-by: Eran Ben Elisha <eranbe@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Driver callback for handling TX timeout should access some internal
resources (SQ, CQ) in order to decide if the tx timeout work should be
scheduled. These resources might be unavailable if channels are closed
in parallel (ifdown for example).
The state lock is the mechanism to protect from such races.
Move all TX timeout logic to be in the work under a state lock.
In addition, Move the work from the global WQ to mlx5e WQ to make sure
this work is flushed when device is detached..
Also, move the mlx5e_tx_timeout_work code to be next to the TX timeout
NDO for better code locality.
Fixes: 3947ca1859 ("net/mlx5e: Implement ndo_tx_timeout callback")
Signed-off-by: Eran Ben Elisha <eranbe@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Commit 58d522912a ("net/mlx5e: Support TX packet copy into WQE")
introduced the max inline WQE as an ethtool tunable. One commit later,
that functionality was made dependent on BlueFlame.
Commit 6982ab6097 ("net/mlx5e: Xmit, no write combining") removed
BlueFlame support, and with it the max inline WQE.
This patch cleans up the leftovers from the removed feature.
Signed-off-by: Gal Pressman <galp@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Add a control private flag in ethtool to enable/disable
Striding RQ feature.
Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Do not implicit a call to mlx5e_init_rq_type_params() upon every
change in RQ type. It should be called only on channels creation.
Fixes: 2fc4bfb725 ("net/mlx5e: Dynamic RQ type infrastructure")
Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
It can be derived from other params, calculate it
via the dedicated function when needed.
Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Introduce functions to calculate them when needed.
They can be derived from other params.
This will simplify transition between RQ configurations.
In general, any parameter that is not explicitly set
or controlled, but derived from other parameters,
should not have a control-path field itself, but a
getter function.
Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
In copying skb header to skb->data, replace the call to
skb_copy_to_linear_data_offset() with a zero offset with
the call to the no-offset function skb_copy_to_linear_data().
Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Pass the base dma address and offset to dma_sync_single_range_for_cpu(),
instead of doing the pre-calculation.
Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Clean it up as it's not in use.
Fixes: d9d9f156f3 ("net/mlx5e: Expand WQE stride when CQE compression is enabled")
Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
We turn the feature off for servers with PCI BW bounded
by a threshold (16G) and lower than MAX LINK BW.
This improves the effectiveness of CQE compression feature,
that is defaulted to ON for the same case.
Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Get the link/pci speed query and logic into a single function.
Unify the heuristics and use a single PCI threshold (16G) for all.
Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
The FLOW_RSS flag was causing us to insert UDP filters when TCP was wanted.
Fixes: 42356d9a13 ("sfc: support RSS spreading of ethtool ntuple filters")
Signed-off-by: Edward Cree <ecree@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Otherwise races are possible between ethtool ops and
efx_ef10_rx_restore_rss_contexts().
Also, don't try to perform the restore on every reset, only after an MC
reboot, otherwise we'll leak RSS contexts on the NIC.
Fixes: 42356d9a13 ("sfc: support RSS spreading of ethtool ntuple filters")
Signed-off-by: Edward Cree <ecree@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
If some other operation gets the MCDI lock ahead of us and performs an MC
reboot, then our attempt to insert the filter will fail with EINVAL,
because the destination VI (spec->dmaq_id, MC_CMD_FILTER_OP_IN_RX_QUEUE) does
not exist. But the caller's request (which might e.g. be an ethtool ntuple
request from userland) isn't invalid, it just got unlucky; so return EAGAIN.
Signed-off-by: Edward Cree <ecree@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
With this change, the spinlock efx->filter_lock is no longer used and is
thus removed.
Signed-off-by: Edward Cree <ecree@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
efx->filter_lock remains in place for use on farch, but EF10 now ignores it.
EFX_EF10_FILTER_FLAG_BUSY is no longer needed, hence it is removed.
Signed-off-by: Edward Cree <ecree@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Instead of having an efx->type->filter_rfs_insert() method, just use
workitems with a worker function that calls efx->type->filter_insert().
The only user of this is efx_filter_rfs(), which now queues a call to
efx_filter_rfs_work().
Similarly, efx_filter_rfs_expire() is now a worker function called on a
new channel->filter_work work_struct, so the method
efx->type->filter_rfs_expire_one() is no longer called in atomic context.
We also add a new mutex efx->rps_mutex to protect the RPS state (efx->
rps_expire_channel, efx->rps_expire_index, and channel->rps_flow_id) so
that the taking of efx->filter_lock can be moved to
efx->type->filter_rfs_expire_one().
Thus, all filter table functions are now called in a sleepable context,
allowing them to use sleeping locks in a future patch.
Signed-off-by: Edward Cree <ecree@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Use relaxed I/O on the hot path. This achieves significant performance
improvements. On a 10G link, this makes a basic iperf TCP test go from
an average of 4.5 Gbits/sec to about 9.40 Gbits/sec.
Signed-off-by: Yan Markman <ymarkman@marvell.com>
[Maxime: Commit message, cosmetic changes]
Signed-off-by: Maxime Chevallier <maxime.chevallier@bootlin.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Since commit c5ad119fb6
("net: sched: pfifo_fast use skb_array") driver is exposed
to an issue where it is hitting NULL skbs while handling TX
completions. Driver uses mmiowb() to flush the writes to the
doorbell bar which is a write-combined bar, however on x86
mmiowb() does not flush the write combined buffer.
This patch fixes this problem by replacing mmiowb() with wmb()
after the write combined doorbell write so that writes are
flushed and synchronized from more than one processor.
V1->V2:
-------
This patch was marked as "superseded" in patchwork.
(Not really sure for what reason).Resending it as v2.
Signed-off-by: Ariel Elior <ariel.elior@cavium.com>
Signed-off-by: Manish Chopra <manish.chopra@cavium.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
mlx4_delete_all_resources_for_slave in resource tracker should free all
memory allocated for a slave.
While releasing memory of fs_rule, it misses releasing memory of
fs_rule->mirr_mbox.
Fixes: 78efed2751 ('net/mlx4_core: Support mirroring VF DMFS rules on both ports')
Signed-off-by: Moshe Shemesh <moshe@mellanox.com>
Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Global pause and PFC configuration should be mutually exclusive (i.e. only
one of them at most can be set). However, once PFC was turned off,
driver automatically turned Global pause on. This is a bug.
Fix the driver behaviour to turn off PFC/Global once the user turned the
other on.
This also fixed a weird behaviour that at a current time, the profile
had both PFC and global pause configuration turned on, which is
Hardware-wise impossible and caused returning false positive indication
to query tools.
In addition, fix error code when setting global pause or PFC to change
metadata only upon successful change.
Also, removed useless debug print.
Fixes: af7d518526 ("net/mlx4_en: Add DCB PFC support through CEE netlink commands")
Fixes: c27a02cd94 ("mlx4_en: Add driver for Mellanox ConnectX 10GbE NIC")
Signed-off-by: Eran Ben Elisha <eranbe@mellanox.com>
Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Saeed Mahameed says:
====================
Mellanox, mlx5 fixes 2018-03-23
The following series includes fixes for mlx5 netdev and eswitch.
v1->v2:
- Fixed commit message quotation marks in patch #7
For -stable v4.12
('net/mlx5e: Avoid using the ipv6 stub in the TC offload neigh update path')
('net/mlx5e: Fix traffic being dropped on VF representor')
For -stable v4.13
('net/mlx5e: Fix memory usage issues in offloading TC flows')
('net/mlx5e: Verify coalescing parameters in range')
For -stable v4.14
('net/mlx5e: Don't override vport admin link state in switchdev mode')
For -stable v4.15
('108b2b6d5c02 net/mlx5e: Sync netdev vxlan ports at open')
Please pull and let me know if there's any problem.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Saeed Mahameed says:
====================
mlx5-updates-2018-03-22 (Misc updates)
This series includes misc updates for mlx5 core and netdev dirver,
Highlights:
From Inbar, three patches to add support for PFC stall prevention
statistics and enable/disable through new ethtool tunable, as requested
from previous submission.
From Moshe, four patches, added more drop counters:
- drop counter for netdev steering miss
- drop counter for when VF logical link is down
- drop counter for when netdev logical link is down.
From Or, three patches to support vlan push/pop offload via tc HW action,
for newer HW (Connectx-5 and onward) via HW steering flow actions rather
than the emulated path for the older HW brands.
And five more misc small trivial patches.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Napi is checking Tx queue status and waking the Tx queue if required.
Same operation is being done while freeing every Tx buffer.
So removed the duplicate operation of checking Tx queue status from the Tx
buffer free functions.
Signed-off-by: Intiyaz Basha <intiyaz.basha@cavium.com>
Signed-off-by: Felix Manlunas <felix.manlunas@cavium.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
pci_set_drvdata() is called only after registering the net_device,
therefore we could run into a NPE if one of the functions using
driver_data is called before it's set.
Fix this by calling pci_set_drvdata() before registering the
net_device.
This fix is a candidate for stable. As far as I can see the
bug has been there in kernel version 3.2 already, therefore
I can't provide a reference which commit is fixed by it.
The fix may need small adjustments per kernel version because
due to other changes the label which is jumped to if
register_netdev() fails has changed over time.
Reported-by: David Miller <davem@davemloft.net>
Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Fixes the following sparse warning:
drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_utils.c:508:5: warning:
symbol 'hw_atl_utils_mpi_set_speed' was not declared. Should it be static?
Signed-off-by: Wei Yongjun <weiyongjun1@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Some helper functions that search for given entries in the TCAM filter
on PPv2 controller make use of dynamically alloced temporary variables,
allocated with GFP_KERNEL. These functions can be called in atomic
context, and dynamic alloc is not really needed in these cases anyways.
This commit gets rid of dynamic allocs and use stack allocation in the
following functions, and where they're used :
- mvpp2_prs_flow_find
- mvpp2_prs_vlan_find
- mvpp2_prs_double_vlan_find
- mvpp2_prs_mac_da_range_find
For all these functions, instead of returning an temporary object
representing the TCAM entry, we simply return the TCAM id that matches
the requested entry.
Signed-off-by: Maxime Chevallier <maxime.chevallier@bootlin.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The mvpp2_prs_hw_read function uses the 'index' field of the struct
mvpp2_prs_entry to initialize the rest of the fields. This makes it
unclear from a caller's perspective, who needs to manipulate a struct
that is not entirely initialized.
This commit makes it an init function for prs_entry, by passing it the
index as a parameter. The function now zeroes the entry, and sets the
index field before doing all other init from HW.
The function is renamed 'mvpp2_prs_init_from_hw' to make that clear.
Signed-off-by: Maxime Chevallier <maxime.chevallier@bootlin.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Jeff Kirsher says:
====================
40GbE Intel Wired LAN Driver Updates 2018-03-26
This series contains updates to i40e only.
Jake provides several patches which remove the need for cmpxchg64(),
starting with moving I40E_FLAG_[UDP]_FILTER_SYNC from pf->flags to pf->state
since they are modified during run time possibly when the RTNL lock is not
held so they should be a state bits and not flags. Moved additional
"flags" which should be state fields, into pf->state. Ensure we hold
the RTNL lock for the entire sequence of preparing for reset and when
resuming, which will protect the flags related to interrupt scheme under
RTNL lock so that their modification is properly threaded. Finally,
cleanup the use of cmpxchg64() since it is no longer needed. Cleaned up
the holes in the feature flags created my moving some flags to the state
field.
Björn Töpel adds XDP_REDIRECT support as well as tweaking the page
counting for XDP_REDIRECT so that it will function properly.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Jeff Kirsher says:
====================
100GbE Intel Wired LAN Driver Updates 2018-03-26
This patch series adds the ice driver, which will support the Intel(R)
E800 Series of network devices.
This is the first phase in the release of this driver where we implement
basic transmit and receive. The idea behind the multi-phase release is to
aid in code review as well as testing. Subsequent phases will implement
advanced features (like SR-IOV, tunnelling, flow director, QoS, etc.) that
build upon the previous phase(s). Each phase will be submitted as a patch
series.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
The driver now acts upon the XDP_REDIRECT return action. Two new ndos
are implemented, ndo_xdp_xmit and ndo_xdp_flush.
XDP_REDIRECT action enables XDP program to redirect frames to other
netdevs.
Signed-off-by: Björn Töpel <bjorn.topel@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
This commit tweaks the page counting for XDP_REDIRECT to function
properly. XDP_REDIRECT support will be added in a future commit.
The current page counting scheme assumes that the reference count
cannot decrease until the received frame is sent to the upper layers
of the networking stack. This assumption does not hold for the
XDP_REDIRECT action, since a page (pointed out by xdp_buff) can have
its reference count decreased via the xdp_do_redirect call.
To work around that, we now start off by a large page count and then
don't allow a refcount less than two.
Signed-off-by: Björn Töpel <bjorn.topel@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Remove the gaps created by the recent refactor of various feature flags
that have moved to the state field. Use only a u32 now that we have
fewer than 32 flags in the field.
Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
When mlx5_core is loaded it is expected to sync ports
with all vxlan devices so it can support vxlan encap/decap.
This is done via udp_tunnel_get_rx_info(). Currently this
call is set in mlx5e_nic_enable() and if the netdev is not in
NETREG_REGISTERED state it will not be called.
Normally on load the netdev state is not NETREG_REGISTERED
so udp_tunnel_get_rx_info() will not be called.
Moving udp_tunnel_get_rx_info() to mlx5e_open() so
it will be called on netdev UP event and allow encap/decap.
Fixes: 610e89e05c ("net/mlx5e: Don't sync netdev state when not registered")
Signed-off-by: Shahar Klein <shahark@mellanox.com>
Reviewed-by: Roi Dayan <roid@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Currently we use the global ipv6_stub var to access the ipv6 global
nd table. This practice gets us to troubles when the stub is only partially
set e.g when ipv6 is loaded under the disabled policy. In this case, as of commit
343d60aada ("ipv6: change ipv6_stub_impl.ipv6_dst_lookup to take net argument")
the stub is not null, but stub->nd_tbl is and we crash.
As we can access the ipv6 nd_tbl directly, the fix is just to avoid the
reference through the stub. There is one place in the code where we
issue ipv6 route lookup and keep doing it through the stub, but that
mentioned commit makes sure we get -EAFNOSUPPORT from the stack.
Fixes: 232c001398 ("net/mlx5e: Add support to neighbour update flow")
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Reviewed-by: Aviv Heller <avivh@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
For NIC flows, the parsed attributes are not freed when we exit
successfully from mlx5e_configure_flower().
There is possible double free for eswitch flows. If error is returned
from rhashtable_insert_fast(), the parse attrs will be freed in
mlx5e_tc_del_flow(), but they will be freed again before exiting
mlx5e_configure_flower().
To fix both issues we do the following:
(1) change the condition that determines if to issue the free call to
check if this flow is NIC flow, or it does not have encap action.
(2) reorder the code such that that the check and free calls are done
before we attempt to add into the hash table.
Fixes: 232c001398 ('net/mlx5e: Add support to neighbour update flow')
Signed-off-by: Jianbo Liu <jianbol@mellanox.com>
Reviewed-by: Or Gerlitz <ogerlitz@mellanox.com>
Reviewed-by: Roi Dayan <roid@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Increase representor netdev RQ size to avoid dropped packets.
The current size (two) is just too small to keep up with
conventional slow path traffic patterns.
Also match the SQ size to the RQ size.
Fixes: cb67b83292 ("net/mlx5e: Introduce SRIOV VF representors")
Signed-off-by: Roi Dayan <roid@mellanox.com>
Reviewed-by: Paul Blakey <paulb@mellanox.com>
Reviewed-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Add check of coalescing parameters received through ethtool are within
range of values supported by the HW.
Driver gets the coalescing rx/tx-usecs and rx/tx-frames as set by the
users through ethtool. The ethtool support up to 32 bit value for each.
However, mlx5 modify cq limits the coalescing time parameter to 12 bit
and coalescing frames parameters to 16 bits.
Return out of range error if user tries to set these parameters to
higher values.
Fixes: f62b8bb8f2 ('net/mlx5: Extend mlx5_core to support ConnectX-4 Ethernet functionality')
Signed-off-by: Moshe Shemesh <moshe@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Add dependancy for switchdev to be congfigured as any user-space control
plane SW is expected to use the HW switchdev ID to locate the representors
related to VFs of a certain PF and apply SW/offloaded switching on them.
Fixes: e80541ecab ('net/mlx5: Add CONFIG_MLX5_ESWITCH Kconfig')
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Reviewed-by: Mark Bloch <markb@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
SQs are 32 and not 16 bits, hence it's wrong to use only 16 bits to
store the sq number for which are going to set steering rule, fix that.
Fixes: cb67b83292 ('net/mlx5e: Introduce SRIOV VF representors')
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Reviewed-by: Mark Bloch <markb@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
