enum skb_drop_reason is more generic, we can adopt it instead.

Provide dev_kfree_skb_irq_reason() and dev_kfree_skb_any_reason().

This means drivers can use more precise drop reasons if they want to.
Signed-off-by: NEric Dumazet <edumazet@google.com>
Reviewed-by: NSimon Horman <simon.horman@corigine.com>
Reviewed-by: NYunsheng Lin <linyunsheng@huawei.com>
Link: https://lore.kernel.org/r/20230306204313.10492-1-edumazet@google.comSigned-off-by: NJakub Kicinski <kuba@kernel.org>

That really was meant to be a per netns attribute from the beginning.

The idea is that once proper isolation is in place in the main
namespace, additional demux in the child namespaces will be redundant.
Let's make child netns default rps mask empty by default.

To avoid bloating the netns with a possibly large cpumask, allocate
it on-demand during the first write operation.
Signed-off-by: NPaolo Abeni <pabeni@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Cited commit changed devlink to register its netdev notifier block on
the global netdev notifier chain instead of on the per network namespace
one.

However, when changing the network namespace of the devlink instance,
devlink still tries to unregister its notifier block from the chain of
the old namespace and register it on the chain of the new namespace.
This results in corruption of the notifier chains, as the same notifier
block is registered on two different chains: The global one and the per
network namespace one. In turn, this causes other problems such as the
inability to dismantle namespaces due to netdev reference count issues.

Fix by preventing devlink from moving its notifier block between
namespaces.

Reproducer:

 # echo "10 1" > /sys/bus/netdevsim/new_device
 # ip netns add test123
 # devlink dev reload netdevsim/netdevsim10 netns test123
 # ip netns del test123
 [   71.935619] unregister_netdevice: waiting for lo to become free. Usage count = 2
 [   71.938348] leaked reference.

Fixes: 565b4824 ("devlink: change port event netdev notifier from per-net to global")
Signed-off-by: NIdo Schimmel <idosch@nvidia.com>
Reviewed-by: NJiri Pirko <jiri@nvidia.com>
Reviewed-by: NJacob Keller <jacob.e.keller@intel.com>
Reviewed-by: NJakub Kicinski <kuba@kernel.org>
Link: https://lore.kernel.org/r/20230215073139.1360108-1-idosch@nvidia.comSigned-off-by: NPaolo Abeni <pabeni@redhat.com>

If RPS is enabled, this allows configuring a default rps
mask, which is effective since receive queue creation time.

A default RPS mask allows the system admin to ensure proper
isolation, avoiding races at network namespace or device
creation time.

The default RPS mask is initially empty, and can be
modified via a newly added sysctl entry.
Signed-off-by: NPaolo Abeni <pabeni@redhat.com>
Reviewed-by: NSimon Horman <simon.horman@corigine.com>
Reviewed-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NJakub Kicinski <kuba@kernel.org>

Add missing xdp_features field description in the struct net_device
documentation. This patch fix the following warning:

  [...]
  ./include/linux/netdevice.h:2375: warning: Function parameter or member 'xdp_features' not described in 'net_device'
  [...]

Fixes: d3d854fd ("netdev-genl: create a simple family for netdev stuff")
Signed-off-by: NLorenzo Bianconi <lorenzo@kernel.org>
Signed-off-by: NDaniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/7878544903d855b49e838c9d59f715bde0b5e63b.1675705948.git.lorenzo@kernel.org

Add a Netlink spec-compatible family for netdevs.
This is a very simple implementation without much
thought going into it.

It allows us to reap all the benefits of Netlink specs,
one can use the generic client to issue the commands:

  $ ./cli.py --spec netdev.yaml --dump dev_get
  [{'ifindex': 1, 'xdp-features': set()},
   {'ifindex': 2, 'xdp-features': {'basic', 'ndo-xmit', 'redirect'}},
   {'ifindex': 3, 'xdp-features': {'rx-sg'}}]

the generic python library does not have flags-by-name
support, yet, but we also don't have to carry strings
in the messages, as user space can get the names from
the spec.
Acked-by: NJesper Dangaard Brouer <brouer@redhat.com>
Co-developed-by: NLorenzo Bianconi <lorenzo@kernel.org>
Signed-off-by: NLorenzo Bianconi <lorenzo@kernel.org>
Co-developed-by: NKumar Kartikeya Dwivedi <memxor@gmail.com>
Signed-off-by: NKumar Kartikeya Dwivedi <memxor@gmail.com>
Co-developed-by: NMarek Majtyka <alardam@gmail.com>
Signed-off-by: NMarek Majtyka <alardam@gmail.com>
Signed-off-by: NJakub Kicinski <kuba@kernel.org>
Link: https://lore.kernel.org/r/327ad9c9868becbe1e601b580c962549c8cd81f2.1675245258.git.lorenzo@kernel.orgSigned-off-by: NAlexei Starovoitov <ast@kernel.org>

This patch introduces gso_ipv4_max_size and gro_ipv4_max_size
per device and adds netlink attributes for them, so that IPV4
BIG TCP can be guarded by a separate tunable in the next patch.

To not break the old application using "gso/gro_max_size" for
IPv4 GSO packets, this patch updates "gso/gro_ipv4_max_size"
in netif_set_gso/gro_max_size() if the new size isn't greater
than GSO_LEGACY_MAX_SIZE, so that nothing will change even if
userspace doesn't realize the new netlink attributes.
Signed-off-by: NXin Long <lucien.xin@gmail.com>
Reviewed-by: NDavid Ahern <dsahern@kernel.org>
Reviewed-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NJakub Kicinski <kuba@kernel.org>

Almost all validation logic is in the drivers, but they are
missing reliable way to convey failure reason to userspace
applications.

Let's use extack to return this information to users.
Signed-off-by: NLeon Romanovsky <leonro@nvidia.com>
Signed-off-by: NJakub Kicinski <kuba@kernel.org>

Almost all validation logic is in the drivers, but they are
missing reliable way to convey failure reason to userspace
applications.

Let's use extack to return this information to users.
Signed-off-by: NLeon Romanovsky <leonro@nvidia.com>
Signed-off-by: NJakub Kicinski <kuba@kernel.org>

Define a new kfunc set (xdp_metadata_kfunc_ids) which implements all possible
XDP metatada kfuncs. Not all devices have to implement them. If kfunc is not
supported by the target device, the default implementation is called instead.
The verifier, at load time, replaces a call to the generic kfunc with a call
to the per-device one. Per-device kfunc pointers are stored in separate
struct xdp_metadata_ops.

Cc: John Fastabend <john.fastabend@gmail.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Martin KaFai Lau <martin.lau@linux.dev>
Cc: Jakub Kicinski <kuba@kernel.org>
Cc: Willem de Bruijn <willemb@google.com>
Cc: Jesper Dangaard Brouer <brouer@redhat.com>
Cc: Anatoly Burakov <anatoly.burakov@intel.com>
Cc: Alexander Lobakin <alexandr.lobakin@intel.com>
Cc: Magnus Karlsson <magnus.karlsson@gmail.com>
Cc: Maryam Tahhan <mtahhan@redhat.com>
Cc: xdp-hints@xdp-project.net
Cc: netdev@vger.kernel.org
Signed-off-by: NStanislav Fomichev <sdf@google.com>
Link: https://lore.kernel.org/r/20230119221536.3349901-8-sdf@google.comSigned-off-by: NMartin KaFai Lau <martin.lau@kernel.org>

Currently, in bonding it reused the IFF_SLAVE flag and checked it
in ipv6 addrconf to prevent ipv6 addrconf.

However, it is not a proper flag to use for no ipv6 addrconf, for
bonding it has to move IFF_SLAVE flag setting ahead of dev_open()
in bond_enslave(). Also, IFF_MASTER/SLAVE are historical flags
used in bonding and eql, as Jiri mentioned, the new devices like
Team, Failover do not use this flag.

So as Jiri suggested, this patch adds IFF_NO_ADDRCONF in priv_flags
of the device to indicate no ipv6 addconf, and uses it in bonding
and moves IFF_SLAVE flag setting back to its original place.
Signed-off-by: NXin Long <lucien.xin@gmail.com>
Signed-off-by: NJakub Kicinski <kuba@kernel.org>

Both in RX and TX, the traffic that performs IPsec packet offload
transformation is accounted by HW. It is needed to properly handle
hard limits that require to drop the packet.

It means that XFRM core needs to update internal counters with the one
that accounted by the HW, so new callbacks are introduced in this patch.

In case of soft or hard limit is occurred, the driver should call to
xfrm_state_check_expire() that will perform key rekeying exactly as
done by XFRM core.
Signed-off-by: NLeon Romanovsky <leonro@nvidia.com>
Signed-off-by: NSteffen Klassert <steffen.klassert@secunet.com>

Extend netlink interface to add and delete XFRM policy from the device.
This functionality is a first step to implement packet IPsec offload solution.
Signed-off-by: NRaed Salem <raeds@nvidia.com>
Signed-off-by: NLeon Romanovsky <leonro@nvidia.com>
Signed-off-by: NSteffen Klassert <steffen.klassert@secunet.com>

Add a helper for drivers wanting to set SW IRQ coalescing
by default. The related sysfs attributes can be used to
override the default values.

Follow Jakub's suggestion and put this functionality into
net core so that drivers wanting to use software interrupt
coalescing per default don't have to open-code it.

Note that this function needs to be called before the
netdevice is registered.
Suggested-by: NJakub Kicinski <kuba@kernel.org>
Signed-off-by: NHeiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Long standing KCSAN issues are caused by data-race around
some dev->stats changes.

Most performance critical paths already use per-cpu
variables, or per-queue ones.

It is reasonable (and more correct) to use atomic operations
for the slow paths.

This patch adds an union for each field of net_device_stats,
so that we can convert paths that are not yet protected
by a spinlock or a mutex.

netdev_stats_to_stats64() no longer has an #if BITS_PER_LONG==64

Note that the memcpy() we were using on 64bit arches
had no provision to avoid load-tearing,
while atomic_long_read() is providing the needed protection
at no cost.
Signed-off-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Move the received_rps counter value next to the other RPS-related members
in softnet_data. This closes two four-byte holes in the structure, making
room for another pointer in the first two cache lines without bumping the
xmit struct to its own line.
Acked-by: NSong Liu <song@kernel.org>
Reviewed-by: NStanislav Fomichev <sdf@google.com>
Signed-off-by: NToke Høiland-Jørgensen <toke@redhat.com>
Link: https://lore.kernel.org/r/20221108140601.149971-2-toke@redhat.comSigned-off-by: NAlexei Starovoitov <ast@kernel.org>

Currently, net_dev() netdev notifier variant follows the netdev with
per-net notifier from namespace to namespace. This is implemented
by move_netdevice_notifiers_dev_net() helper.

For devlink it is needed to re-register per-net notifier during
devlink reload. Introduce a new helper called
move_netdevice_notifier_net() and share the unregister/register code
with existing move_netdevice_notifiers_dev_net() helper.
Signed-off-by: NJiri Pirko <jiri@nvidia.com>
Reviewed-by: NIdo Schimmel <idosch@nvidia.com>
Signed-off-by: NJakub Kicinski <kuba@kernel.org>

Allow a network interface to be renamed when the interface
is up.

As described in the netconsole documentation [1], when netconsole is
used as a built-in, it will bring up the specified interface as soon as
possible. As a result, user space will not be able to rename the
interface since the kernel disallows renaming of interfaces that are
administratively up unless the 'IFF_LIVE_RENAME_OK' private flag was set
by the kernel.

The original solution [2] to this problem was to add a new parameter to
the netconsole configuration parameters that allows renaming of
the interface used by netconsole while it is administratively up.
However, during the discussion that followed, it became apparent that we
have no reason to keep the current restriction and instead we should
allow user space to rename interfaces regardless of their administrative
state:

1. The restriction was put in place over 20 years ago when renaming was
only possible via IOCTL and before rtnetlink started notifying user
space about such changes like it does today.

2. The 'IFF_LIVE_RENAME_OK' flag was added over 3 years ago in version
5.2 and no regressions were reported.

3. In-kernel listeners to 'NETDEV_CHANGENAME' do not seem to care about
the administrative state of interface.

Therefore, allow user space to rename running interfaces by removing the
restriction and the associated 'IFF_LIVE_RENAME_OK' flag. Help in
possible triage by emitting a message to the kernel log that an
interface was renamed while UP.

[1] https://www.kernel.org/doc/Documentation/networking/netconsole.rst
[2] https://lore.kernel.org/netdev/20221102002420.2613004-1-andy.ren@getcruise.com/Signed-off-by: NAndy Ren <andy.ren@getcruise.com>
Reviewed-by: NIdo Schimmel <idosch@nvidia.com>
Reviewed-by: NDavid Ahern <dsahern@kernel.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Remove ndo_get_devlink_port which is no longer used alongside with the
implementations in drivers.
Signed-off-by: NJiri Pirko <jiri@nvidia.com>
Signed-off-by: NJakub Kicinski <kuba@kernel.org>

Currently, ethernet drivers are using devlink_port_type_eth_set() and
devlink_port_type_clear() to set devlink port type and link to related
netdev.

Instead of calling them directly, let the driver use
SET_NETDEV_DEVLINK_PORT macro to assign devlink_port pointer and let
devlink to track it. Note the devlink port pointer is static during
the time netdevice is registered.

In devlink code, use per-namespace netdev notifier to track
the netdevices with devlink_port assigned and change the internal
devlink_port type and related type pointer accordingly.
Signed-off-by: NJiri Pirko <jiri@nvidia.com>
Signed-off-by: NJakub Kicinski <kuba@kernel.org>

This patch pass netlink message header and portid to rtnl_configure_link()
All the functions in this call chain need to add the parameters so we can
use them in the last call rtnl_notify(), and notify the userspace about
the new link info if NLM_F_ECHO flag is set.

- rtnl_configure_link()
  - __dev_notify_flags()
    - rtmsg_ifinfo()
      - rtmsg_ifinfo_event()
        - rtmsg_ifinfo_build_skb()
        - rtmsg_ifinfo_send()
	  - rtnl_notify()

Also move __dev_notify_flags() declaration to net/core/dev.h, as Jakub
suggested.
Signed-off-by: NHangbin Liu <liuhangbin@gmail.com>
Reviewed-by: NGuillaume Nault <gnault@redhat.com>
Signed-off-by: NJakub Kicinski <kuba@kernel.org>

Currently, use dev->reg_state == NETREG_UNREGISTERING to check the status
which is NETREG_UNREGISTERING, rather than using netdev_unregistering.
Also, A helper function which is netdev_unregistering on nedevice.h is no
longer used. Thus, netdev_unregistering removes from netdevice.h.
Signed-off-by: NJuhee Kang <claudiajkang@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This reverts commit 854701ba.

We have more violations around, which leads to:

  WARNING: CPU: 2 PID: 1 at include/linux/cpumask.h:110 __netif_set_xps_queue+0x14e/0x770

Let's back this out and retry with a larger clean up in -next.

Fixes: 854701ba ("net: fix cpu_max_bits_warn() usage in netif_attrmask_next{,_and}")
Link: https://lore.kernel.org/all/20221014030459.3272206-2-guoren@kernel.org/Signed-off-by: NJakub Kicinski <kuba@kernel.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The functions require to be passed with a cpu index prior to one that is
the first to start search, so the valid input range is [-1, nr_cpu_ids-1).
However, the code checks against [-1, nr_cpu_ids).
Acked-by: NJakub Kicinski <kuba@kernel.org>
Signed-off-by: NYury Norov <yury.norov@gmail.com>

When adding optional new features to Qdisc offloads, existing drivers
must reject the new configuration until they are coded up to act on it.

Since modifying all drivers in lockstep with the changes in the Qdisc
can create problems of its own, it would be nice if there existed an
automatic opt-in mechanism for offloading optional features.

Jakub proposes that we multiplex one more kind of call through
ndo_setup_tc(): one where the driver populates a Qdisc-specific
capability structure.

First user will be taprio in further changes. Here we are introducing
the definitions for the base functionality.

Link: https://patchwork.kernel.org/project/netdevbpf/patch/20220923163310.3192733-3-vladimir.oltean@nxp.com/Suggested-by: NJakub Kicinski <kuba@kernel.org>
Co-developed-by: NJakub Kicinski <kuba@kernel.org>
Signed-off-by: NVladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: NJakub Kicinski <kuba@kernel.org>

After commit 3226b158 ("net: avoid 32 x truesize under-estimation
for tiny skbs") we are observing 10-20% regressions in performance
tests with small packets. The perf trace points to high pressure on
the slab allocator.

This change tries to improve the allocation schema for small packets
using an idea originally suggested by Eric: a new per CPU page frag is
introduced and used in __napi_alloc_skb to cope with small allocation
requests.

To ensure that the above does not lead to excessive truesize
underestimation, the frag size for small allocation is inflated to 1K
and all the above is restricted to build with 4K page size.

Note that we need to update accordingly the run-time check introduced
with commit fd9ea57f ("net: add napi_get_frags_check() helper").

Alex suggested a smart page refcount schema to reduce the number
of atomic operations and deal properly with pfmemalloc pages.

Under small packet UDP flood, I measure a 15% peak tput increases.
Suggested-by: NEric Dumazet <eric.dumazet@gmail.com>
Suggested-by: NAlexander H Duyck <alexanderduyck@fb.com>
Signed-off-by: NPaolo Abeni <pabeni@redhat.com>
Reviewed-by: NEric Dumazet <edumazet@google.com>
Reviewed-by: NAlexander Duyck <alexanderduyck@fb.com>
Link: https://lore.kernel.org/r/6b6f65957c59f86a353fc09a5127e83a32ab5999.1664350652.git.pabeni@redhat.comSigned-off-by: NJakub Kicinski <kuba@kernel.org>

We tell driver developers to always pass NAPI_POLL_WEIGHT
as the weight to netif_napi_add(). This may be confusing
to newcomers, drop the weight argument, those who really
need to tweak the weight can use netif_napi_add_weight().

Acked-by: Marc Kleine-Budde <mkl@pengutronix.de> # for CAN
Link: https://lore.kernel.org/r/20220927132753.750069-1-kuba@kernel.orgSigned-off-by: NJakub Kicinski <kuba@kernel.org>

Some network drivers use __dev_mc_sync()/__dev_uc_sync() and therefore
program the hardware only with addresses with a non-zero sync_cnt.

Some of the above drivers also need to save/restore the address
filtering lists when certain events happen, and they need to walk
through the struct net_device :: uc and struct net_device :: mc lists.
But these lists contain unsynced addresses too.

To keep the appearance of an elementary form of data encapsulation,
provide iterators through these lists that only look at entries with a
non-zero sync_cnt, instead of filtering entries out from device drivers.
Signed-off-by: NVladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: NFlorian Fainelli <f.fainelli@gmail.com>
Signed-off-by: NPaolo Abeni <pabeni@redhat.com>

All callers are now gone.
Signed-off-by: NJakub Kicinski <kuba@kernel.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Add some documentation for netdev_tx_sent_queue() and
netdev_tx_completed_queue()

Stating that netdev_tx_completed_queue() must be called once
per TX completion round is apparently not obvious for everybody.
Signed-off-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Use try_cmpxchg instead of cmpxchg (*ptr, old, new) == old in
napi_if_scheduled_mark_missed. x86 CMPXCHG instruction returns
success in ZF flag, so this change saves a compare after cmpxchg
(and related move instruction in front of cmpxchg).

Also, try_cmpxchg implicitly assigns old *ptr value to "old" when cmpxchg
fails, enabling further code simplifications.

Cc: Eric Dumazet <edumazet@google.com>
Cc: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: NUros Bizjak <ubizjak@gmail.com>
Link: https://lore.kernel.org/r/20220822143243.2798-1-ubizjak@gmail.comSigned-off-by: NJakub Kicinski <kuba@kernel.org>

While reading sysctl_devconf_inherit_init_net, it can be changed
concurrently.  Thus, we need to add READ_ONCE() to its readers.

Fixes: 856c395c ("net: introduce a knob to control whether to inherit devconf config")
Signed-off-by: NKuniyuki Iwashima <kuniyu@amazon.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

While reading sysctl_fb_tunnels_only_for_init_net, it can be changed
concurrently.  Thus, we need to add READ_ONCE() to its readers.

Fixes: 79134e6c ("net: do not create fallback tunnels for non-default namespaces")
Signed-off-by: NKuniyuki Iwashima <kuniyu@amazon.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

DECnet is an obsolete network protocol that receives more attention
from kernel janitors than users. It belongs in computer protocol
history museum not in Linux kernel.

It has been "Orphaned" in kernel since 2010. The iproute2 support
for DECnet was dropped in 5.0 release. The documentation link on
Sourceforge says it is abandoned there as well.

Leave the UAPI alone to keep userspace programs compiling.
This means that there is still an empty neighbour table
for AF_DECNET.

The table of /proc/sys/net entries was updated to match
current directories and reformatted to be alphabetical.
Signed-off-by: NStephen Hemminger <stephen@networkplumber.org>
Acked-by: NDavid Ahern <dsahern@kernel.org>
Acked-by: NNikolay Aleksandrov <razor@blackwall.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The "1<<31" shift has a sign extension bug so IFF_TX_SKB_NO_LINEAR is
0xffffffff80000000 instead of 0x0000000080000000.

Fixes: c2ff53d8 ("net: Add priv_flags for allow tx skb without linear")
Signed-off-by: NDan Carpenter <dan.carpenter@oracle.com>
Reviewed-by: NXuan Zhuo <xuanzhuo@linux.alibaba.com>
Link: https://lore.kernel.org/r/YrRrcGttfEVnf85Q@kiliSigned-off-by: NJakub Kicinski <kuba@kernel.org>

As explained in commit 316580b6 ("u64_stats: provide u64_stats_t type")
we should use u64_stats_t and related accessors to avoid load/store tearing.
Signed-off-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NJakub Kicinski <kuba@kernel.org>

Netdev reference helpers have a dev_ prefix for historic
reasons. Renaming the old helpers would be too much churn
but we can rename the tracking ones which are relatively
recent and should be the default for new code.

Rename:
 dev_hold_track()    -> netdev_hold()
 dev_put_track()     -> netdev_put()
 dev_replace_track() -> netdev_ref_replace()

Link: https://lore.kernel.org/r/20220608043955.919359-1-kuba@kernel.orgSigned-off-by: NJakub Kicinski <kuba@kernel.org>

Most protocol-specific pointers in struct net_device are under
a respective ifdef. Wireless is the notable exception. Since
there's a sizable number of custom-built kernels for datacenter
workloads which don't build wireless it seems reasonable to
ifdefy those pointers as well.

While at it move IPv4 and IPv6 pointers up, those are special
for obvious reasons.
Acked-by: NJohannes Berg <johannes@sipsolutions.net>
Acked-by: Stefan Schmidt <stefan@datenfreihafen.org> # ieee802154
Acked-by: NSven Eckelmann <sven@narfation.org>
Signed-off-by: NJakub Kicinski <kuba@kernel.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

When calling dev_fill_forward_path on a pppoe device, the provided destination
address is invalid. In order for the bridge fdb lookup to succeed, the pppoe
code needs to update ctx->daddr to the correct value.
Fix this by storing the address inside struct net_device_path_ctx

Fixes: f6efc675 ("net: ppp: resolve forwarding path for bridge pppoe devices")
Signed-off-by: NFelix Fietkau <nbd@nbd.name>
Signed-off-by: NPablo Neira Ayuso <pablo@netfilter.org>

A cpu can observe sd->defer_count reaching 128,
and call smp_call_function_single_async()

Problem is that the remote CPU can clear sd->defer_count
before the IPI is run/acknowledged.

Other cpus can queue more packets and also decide
to call smp_call_function_single_async() while the pending
IPI was not yet delivered.

This is a common issue with smp_call_function_single_async().
Callers must ensure correct synchronization and serialization.

I triggered this issue while experimenting smaller threshold.
Performing the call to smp_call_function_single_async()
under sd->defer_lock protection did not solve the problem.

Commit 5a18ceca ("smp: Allow smp_call_function_single_async()
to insert locked csd") replaced an informative WARN_ON_ONCE()
with a return of -EBUSY, which is often ignored.
Test of CSD_FLAG_LOCK presence is racy anyway.

Fixes: 68822bdf ("net: generalize skb freeing deferral to per-cpu lists")
Signed-off-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>