Move call to call_fib_entry_notifiers for new IPv4 routes to right
before the call to fib_insert_alias. At this point the only remaining
failure path is memory allocations in fib_insert_node. Handle that
very unlikely failure with a call to call_fib_entry_notifiers to
tell drivers about it.

At this point notifier handlers can decide the fate of the new route
with a clean path to delete the potential new entry if the notifier
returns non-0.
Signed-off-by: NDavid Ahern <dsa@cumulusnetworks.com>
Reviewed-by: NIdo Schimmel <idosch@mellanox.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Move call_fib_rule_notifiers up in fib_nl_newrule to the point right
before the rule is inserted into the list. At this point there are no
more failure paths within the core rule code, so if the notifier
does not fail then the rule will be inserted into the list.
Signed-off-by: NDavid Ahern <dsa@cumulusnetworks.com>
Reviewed-by: NIdo Schimmel <idosch@mellanox.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Notifier handlers use notifier_from_errno to convert any potential error
to an encoded format. As a consequence the other side, call_fib_notifier{s}
in this case, needs to use notifier_to_errno to return the error from
the handler back to its caller.
Signed-off-by: NDavid Ahern <dsa@cumulusnetworks.com>
Reviewed-by: NIdo Schimmel <idosch@mellanox.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Saeed Mahameed says:

====================
mlx5-updates-2018-03-27 (Misc updates & SQ recovery)

This series contains Misc updates and cleanups for mlx5e rx path
and SQ recovery feature for tx path.

From Tariq: (RX updates)
    - Disable Striding RQ when PCI devices, striding RQ limits the use
      of CQE compression feature, which is very critical for slow PCI
      devices performance, in this change we will prefer CQE compression
      over Striding RQ only on specific "slow"  PCIe links.
    - RX path cleanups
    - Private flag to enable/disable striding RQ

From Eran: (TX fast recovery)
    - TX timeout logic improvements, fast SQ recovery and TX error reporting
      if a HW error occurs while transmitting on a specific SQ, the driver will
      ignore such error and will wait for TX timeout to occur and reset all
      the rings. Instead, the current series improves the resiliency for such
      HW errors by detecting TX completions with errors, which will report them
      and perform a fast recover for the specific faulty SQ even before a TX
      timeout is detected.
====================
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Kirill Tkhai says:

====================
Introduce net_rwsem to protect net_namespace_list

The series introduces fine grained rw_semaphore, which will be used
instead of rtnl_lock() to protect net_namespace_list.

This improves scalability and allows to do non-exclusive sleepable
iteration for_each_net(), which is enough for most cases.

scripts/get_maintainer.pl gives enormous list of people, and I add
all to CC.

Note, that this patch is independent of "Close race between
{un, }register_netdevice_notifier and pernet_operations":
https://patchwork.ozlabs.org/project/netdev/list/?series=36495Signed-off-by: NKirill Tkhai <ktkhai@virtuozzo.com>
====================
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

rtnl_lock() doesn't protect net::ct::count,
and it's not needed for__nf_ct_unconfirmed_destroy()
and for nf_queue_nf_hook_drop().
Signed-off-by: NKirill Tkhai <ktkhai@virtuozzo.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Here we iterate for_each_net() and removes
vport from alive net to the exiting net.

ovs_net::dps are protected by ovs_mutex(),
and the others, who change it (ovs_dp_cmd_new(),
__dp_destroy()) also take it.
The same with datapath::ports list.

So, we remove rtnl_lock() here.
Signed-off-by: NKirill Tkhai <ktkhai@virtuozzo.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

rt_genid_bump_all() consists of ipv4 and ipv6 part.
ipv4 part is incrementing of net::ipv4::rt_genid,
and I see many places, where it's read without rtnl_lock().

ipv6 part calls __fib6_clean_all(), and it's also
called without rtnl_lock() in other places.

So, rtnl_lock() here was used to iterate net_namespace_list only,
and we can remove it.
Signed-off-by: NKirill Tkhai <ktkhai@virtuozzo.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This function iterates over net_namespace_list and flushes
the queue for every of them. What does this rtnl_lock()
protects?! Since we may add skbs to net::wext_nlevents
without rtnl_lock(), it does not protects us about queuers.

It guarantees, two threads can't flush the queue in parallel,
that can change the order, but since skb can be queued
in any order, it doesn't matter, how many threads do this
in parallel. In case of several threads, this will be even
faster.

So, we can remove rtnl_lock() here, as it was used for
iteration over net_namespace_list only.
Signed-off-by: NKirill Tkhai <ktkhai@virtuozzo.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

rtnl_lock() is used everywhere, and contention is very high.
When someone wants to iterate over alive net namespaces,
he/she has no a possibility to do that without exclusive lock.
But the exclusive rtnl_lock() in such places is overkill,
and it just increases the contention. Yes, there is already
for_each_net_rcu() in kernel, but it requires rcu_read_lock(),
and this can't be sleepable. Also, sometimes it may be need
really prevent net_namespace_list growth, so for_each_net_rcu()
is not fit there.

This patch introduces new rw_semaphore, which will be used
instead of rtnl_mutex to protect net_namespace_list. It is
sleepable and allows not-exclusive iterations over net
namespaces list. It allows to stop using rtnl_lock()
in several places (what is made in next patches) and makes
less the time, we keep rtnl_mutex. Here we just add new lock,
while the explanation of we can remove rtnl_lock() there are
in next patches.

Fine grained locks generally are better, then one big lock,
so let's do that with net_namespace_list, while the situation
allows that.
Signed-off-by: NKirill Tkhai <ktkhai@virtuozzo.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Florian Fainelli says:

====================
net: bgmac: Couple of small bgmac changes

This patch series addresses two minor issues with the bgmac driver:

- provides the interface name through /proc/interrupts rather than "bgmac"
- makes sure the interrupts are masked during probe, in case the block was
  not properly reset
====================
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

We can have interrupts left enabled form e.g: the bootloader which used
the network device for network boot. Make sure we have those disabled as
early as possible to avoid spurious interrupts.
Signed-off-by: NFlorian Fainelli <f.fainelli@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

When the system contains several BGMAC adapters, it is nice to be able
to tell which one is which by looking at /proc/interrupts. Use the
network device name as a name to request_irq() with.
Signed-off-by: NFlorian Fainelli <f.fainelli@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

David Howells says:

====================
rxrpc: Tracing updates

Here are some patches that update tracing in AF_RXRPC and AFS:

 (1) Add a tracepoint for tracking resend events.

 (2) Use debug_ids in traces rather than pointers (as pointers are now hashed)
     and allow use of the same debug_id in AFS calls as in the corresponding
     AF_RXRPC calls.  This makes filtering the trace output much easier.

 (3) Add a tracepoint for tracking call completion.
====================
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Add support for the National Instruments XGE 1/10G network device.

It uses the EEPROM on the board via NVMEM.
Signed-off-by: NMoritz Fischer <mdf@kernel.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This adds bindings for the NI XGE 1G/10G network device.
Reviewed-by: NRob Herring <robh@kernel.org>
Signed-off-by: NMoritz Fischer <mdf@kernel.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Steffen Klassert says:

====================
pull request (net-next): ipsec-next 2018-03-29

1) Remove a redundant pointer initialization esp_input_set_header().
   From Colin Ian King.

2) Mark the xfrm kmem_caches as __ro_after_init.
   From Alexey Dobriyan.

3) Do the checksum for an ipsec offlad packet in software
   if the device does not advertise NETIF_F_HW_ESP_TX_CSUM.
   From Shannon Nelson.

4) Use booleans for true and false instead of integers
   in xfrm_policy_cache_flush().
   From Gustavo A. R. Silva

Please pull or let me know if there are problems.
====================
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

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: NEran Ben Elisha <eranbe@mellanox.com>
Signed-off-by: NSaeed 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: NEran Ben Elisha <eranbe@mellanox.com>
Signed-off-by: NSaeed Mahameed <saeedm@mellanox.com>

Move mlx5_ib dump error CQE implementation to mlx5 CQ header file in
order to use it in a downstream patch from mlx5e.

In addition, use print_hex_dump instead of manual dumping of the buffer.
Signed-off-by: NEran Ben Elisha <eranbe@mellanox.com>
Signed-off-by: NSaeed 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: NEran Ben Elisha <eranbe@mellanox.com>
Signed-off-by: NSaeed 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: 3947ca18 ("net/mlx5e: Implement ndo_tx_timeout callback")
Signed-off-by: NEran Ben Elisha <eranbe@mellanox.com>
Signed-off-by: NSaeed Mahameed <saeedm@mellanox.com>

Commit 58d52291 ("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 6982ab60 ("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: NGal Pressman <galp@mellanox.com>
Signed-off-by: NSaeed Mahameed <saeedm@mellanox.com>

Add a control private flag in ethtool to enable/disable
Striding RQ feature.
Signed-off-by: NTariq Toukan <tariqt@mellanox.com>
Signed-off-by: NSaeed 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: 2fc4bfb7 ("net/mlx5e: Dynamic RQ type infrastructure")
Signed-off-by: NTariq Toukan <tariqt@mellanox.com>
Signed-off-by: NSaeed Mahameed <saeedm@mellanox.com>

It can be derived from other params, calculate it
via the dedicated function when needed.
Signed-off-by: NTariq Toukan <tariqt@mellanox.com>
Signed-off-by: NSaeed 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: NTariq Toukan <tariqt@mellanox.com>
Signed-off-by: NSaeed 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: NTariq Toukan <tariqt@mellanox.com>
Signed-off-by: NSaeed 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: NTariq Toukan <tariqt@mellanox.com>
Signed-off-by: NSaeed Mahameed <saeedm@mellanox.com>

Clean it up as it's not in use.

Fixes: d9d9f156 ("net/mlx5e: Expand WQE stride when CQE compression is enabled")
Signed-off-by: NTariq Toukan <tariqt@mellanox.com>
Signed-off-by: NSaeed 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: NTariq Toukan <tariqt@mellanox.com>
Signed-off-by: NSaeed 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: NTariq Toukan <tariqt@mellanox.com>
Signed-off-by: NSaeed Mahameed <saeedm@mellanox.com>

Add a tracepoint to track rxrpc calls moving into the completed state and
to log the completion type and the recorded error value and abort code.
Signed-off-by: NDavid Howells <dhowells@redhat.com>

In rxrpc and afs, use the debug_ids that are monotonically allocated to
various objects as they're allocated rather than pointers as kernel
pointers are now hashed making them less useful.  Further, the debug ids
aren't reused anywhere nearly as quickly.

In addition, allow kernel services that use rxrpc, such as afs, to take
numbers from the rxrpc counter, assign them to their own call struct and
pass them in to rxrpc for both client and service calls so that the trace
lines for each will have the same ID tag.
Signed-off-by: NDavid Howells <dhowells@redhat.com>

Add a tracepoint to trace packet resend events and to dump the Tx
annotation buffer for added illumination.
Signed-off-by: NDavid Howells <dhowells@rdhat.com>

Edward Cree says:

====================
sfc: rework locking around filter management

The use of a spinlock to protect filter state combined with the need for a
 sleeping operation (MCDI) to apply that state to the NIC (on EF10) led to
 unfixable race conditions, around the handling of filter restoration after
 an MC reboot.
So, this patch series removes the requirement to be able to modify the SW
 filter table from atomic context, by using a workqueue to request
 asynchronous filter operations (which are needed for ARFS).  Then, the
 filter table locks are changed to mutexes, replacing the dance of spinlocks
 and 'busy' flags.  Also, a mutex is added to protect the RSS context state,
 since otherwise a similar race is possible around restoring that after an
 MC reboot.  While we're at it, fix a couple of other related bugs.
====================
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The FLOW_RSS flag was causing us to insert UDP filters when TCP was wanted.

Fixes: 42356d9a ("sfc: support RSS spreading of ethtool ntuple filters")
Signed-off-by: NEdward Cree <ecree@solarflare.com>
Signed-off-by: NDavid 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: 42356d9a ("sfc: support RSS spreading of ethtool ntuple filters")
Signed-off-by: NEdward Cree <ecree@solarflare.com>
Signed-off-by: NDavid 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: NEdward Cree <ecree@solarflare.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

With this change, the spinlock efx->filter_lock is no longer used and is
 thus removed.
Signed-off-by: NEdward Cree <ecree@solarflare.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>