linux-toradex.git/include/linux/rtnetlink.h, branch v6.0-rc7

net: sched: use queue_mapping to pick tx queue

2022-04-19T10:20:45+00:00

This patch fixes issue:
* If we install tc filters with act_skbedit in clsact hook.
  It doesn't work, because netdev_core_pick_tx() overwrites
  queue_mapping.

  $ tc filter ... action skbedit queue_mapping 1

And this patch is useful:
* We can use FQ + EDT to implement efficient policies. Tx queues
  are picked by xps, ndo_select_queue of netdev driver, or skb hash
  in netdev_core_pick_tx(). In fact, the netdev driver, and skb
  hash are _not_ under control. xps uses the CPUs map to select Tx
  queues, but we can't figure out which task_struct of pod/containter
  running on this cpu in most case. We can use clsact filters to classify
  one pod/container traffic to one Tx queue. Why ?

  In containter networking environment, there are two kinds of pod/
  containter/net-namespace. One kind (e.g. P1, P2), the high throughput
  is key in these applications. But avoid running out of network resource,
  the outbound traffic of these pods is limited, using or sharing one
  dedicated Tx queues assigned HTB/TBF/FQ Qdisc. Other kind of pods
  (e.g. Pn), the low latency of data access is key. And the traffic is not
  limited. Pods use or share other dedicated Tx queues assigned FIFO Qdisc.
  This choice provides two benefits. First, contention on the HTB/FQ Qdisc
  lock is significantly reduced since fewer CPUs contend for the same queue.
  More importantly, Qdisc contention can be eliminated completely if each
  CPU has its own FIFO Qdisc for the second kind of pods.

  There must be a mechanism in place to support classifying traffic based on
  pods/container to different Tx queues. Note that clsact is outside of Qdisc
  while Qdisc can run a classifier to select a sub-queue under the lock.

  In general recording the decision in the skb seems a little heavy handed.
  This patch introduces a per-CPU variable, suggested by Eric.

  The xmit.skip_txqueue flag is firstly cleared in __dev_queue_xmit().
  - Tx Qdisc may install that skbedit actions, then xmit.skip_txqueue flag
    is set in qdisc->enqueue() though tx queue has been selected in
    netdev_tx_queue_mapping() or netdev_core_pick_tx(). That flag is cleared
    firstly in __dev_queue_xmit(), is useful:
  - Avoid picking Tx queue with netdev_tx_queue_mapping() in next netdev
    in such case: eth0 macvlan - eth0.3 vlan - eth0 ixgbe-phy:
    For example, eth0, macvlan in pod, which root Qdisc install skbedit
    queue_mapping, send packets to eth0.3, vlan in host. In __dev_queue_xmit() of
    eth0.3, clear the flag, does not select tx queue according to skb->queue_mapping
    because there is no filters in clsact or tx Qdisc of this netdev.
    Same action taked in eth0, ixgbe in Host.
  - Avoid picking Tx queue for next packet. If we set xmit.skip_txqueue
    in tx Qdisc (qdisc->enqueue()), the proper way to clear it is clearing it
    in __dev_queue_xmit when processing next packets.

  For performance reasons, use the static key. If user does not config the NET_EGRESS,
  the patch will not be compiled.

  +----+      +----+      +----+
  | P1 |      | P2 |      | Pn |
  +----+      +----+      +----+
    |           |           |
    +-----------+-----------+
                |
                | clsact/skbedit
                |      MQ
                v
    +-----------+-----------+
    | q0        | q1        | qn
    v           v           v
  HTB/FQ      HTB/FQ  ...  FIFO

Cc: Jamal Hadi Salim 
Cc: Cong Wang 
Cc: Jiri Pirko 
Cc: "David S. Miller" 
Cc: Jakub Kicinski 
Cc: Jonathan Lemon 
Cc: Eric Dumazet 
Cc: Alexander Lobakin 
Cc: Paolo Abeni 
Cc: Talal Ahmad 
Cc: Kevin Hao 
Cc: Ilias Apalodimas 
Cc: Kees Cook 
Cc: Kumar Kartikeya Dwivedi 
Cc: Antoine Tenart 
Cc: Wei Wang 
Cc: Arnd Bergmann 
Suggested-by: Eric Dumazet 
Signed-off-by: Tonghao Zhang 
Acked-by: Jamal Hadi Salim 
Signed-off-by: Paolo Abeni

net: rtnetlink: Add UAPI toggle for IFLA_OFFLOAD_XSTATS_L3_STATS

2022-03-03T10:37:23+00:00

The offloaded HW stats are designed to allow per-netdevice enablement and
disablement. Add an attribute, IFLA_STATS_SET_OFFLOAD_XSTATS_L3_STATS,
which should be carried by the RTM_SETSTATS message, and expresses a desire
to toggle L3 offload xstats on or off.

As part of the above, add an exported function rtnl_offload_xstats_notify()
that drivers can use when they have installed or deinstalled the counters
backing the HW stats.

At this point, it is possible to enable, disable and query L3 offload
xstats on netdevices. (However there is no driver actually implementing
these.)

Signed-off-by: Petr Machata 
Signed-off-by: Ido Schimmel 
Signed-off-by: David S. Miller

net: sched: extend Qdisc with rcu

2018-09-26T03:17:35+00:00

Currently, Qdisc API functions assume that users have rtnl lock taken. To
implement rtnl unlocked classifiers update interface, Qdisc API must be
extended with functions that do not require rtnl lock.

Extend Qdisc structure with rcu. Implement special version of put function
qdisc_put_unlocked() that is called without rtnl lock taken. This function
only takes rtnl lock if Qdisc reference counter reached zero and is
intended to be used as optimization.

Signed-off-by: Vlad Buslov 
Acked-by: Jiri Pirko 
Signed-off-by: David S. Miller

net: core: netlink: add helper refcount dec and lock function

2018-09-26T03:17:35+00:00

Rtnl lock is encapsulated in netlink and cannot be accessed by other
modules directly. This means that reference counted objects that rely on
rtnl lock cannot use it with refcounter helper function that atomically
releases decrements reference and obtains mutex.

This patch implements simple wrapper function around refcount_dec_and_lock
that obtains rtnl lock if reference counter value reached 0.

Signed-off-by: Vlad Buslov 
Acked-by: Jiri Pirko 
Signed-off-by: David S. Miller

net: Introduce net_rwsem to protect net_namespace_list

2018-03-29T17:47:53+00:00

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: Kirill Tkhai 
Signed-off-by: David S. Miller

net: Rename net_sem to pernet_ops_rwsem

2018-03-27T17:18:09+00:00

net_sem is some undefined area name, so it will be better
to make the area more defined.

Rename it to pernet_ops_rwsem for better readability and
better intelligibility.

Signed-off-by: Kirill Tkhai 
Signed-off-by: David S. Miller

net: Add rtnl_lock_killable()

2018-03-16T16:31:19+00:00

rtnl_lock() is widely used mutex in kernel. Some of kernel code
does memory allocations under it. In case of memory deficit this
may invoke OOM killer, but the problem is a killed task can't
exit if it's waiting for the mutex. This may be a reason of deadlock
and panic.

This patch adds a new primitive, which responds on SIGKILL, and
it allows to use it in the places, where we don't want to sleep
forever.

Signed-off-by: Kirill Tkhai 
Signed-off-by: David S. Miller

net: Kill net_mutex

2018-02-20T18:23:13+00:00

We take net_mutex, when there are !async pernet_operations
registered, and read locking of net_sem is not enough. But
we may get rid of taking the mutex, and just change the logic
to write lock net_sem in such cases. This obviously reduces
the number of lock operations, we do.

Signed-off-by: Kirill Tkhai 
Signed-off-by: David S. Miller

net: Introduce net_sem for protection of pernet_list

2018-02-13T15:36:04+00:00

Currently, the mutex is mostly used to protect pernet operations
list. It orders setup_net() and cleanup_net() with parallel
{un,}register_pernet_operations() calls, so ->exit{,batch} methods
of the same pernet operations are executed for a dying net, as
were used to call ->init methods, even after the net namespace
is unlinked from net_namespace_list in cleanup_net().

But there are several problems with scalability. The first one
is that more than one net can't be created or destroyed
at the same moment on the node. For big machines with many cpus
running many containers it's very sensitive.

The second one is that it's need to synchronize_rcu() after net
is removed from net_namespace_list():

Destroy net_ns:
cleanup_net()
  mutex_lock(&net_mutex)
  list_del_rcu(&net->list)
  synchronize_rcu()                                  <--- Sleep there for ages
  list_for_each_entry_reverse(ops, &pernet_list, list)
    ops_exit_list(ops, &net_exit_list)
  list_for_each_entry_reverse(ops, &pernet_list, list)
    ops_free_list(ops, &net_exit_list)
  mutex_unlock(&net_mutex)

This primitive is not fast, especially on the systems with many processors
and/or when preemptible RCU is enabled in config. So, all the time, while
cleanup_net() is waiting for RCU grace period, creation of new net namespaces
is not possible, the tasks, who makes it, are sleeping on the same mutex:

Create net_ns:
copy_net_ns()
  mutex_lock_killable(&net_mutex)                    <--- Sleep there for ages

I observed 20-30 seconds hangs of "unshare -n" on ordinary 8-cpu laptop
with preemptible RCU enabled after CRIU tests round is finished.

The solution is to convert net_mutex to the rw_semaphore and add fine grain
locks to really small number of pernet_operations, what really need them.

Then, pernet_operations::init/::exit methods, modifying the net-related data,
will require down_read() locking only, while down_write() will be used
for changing pernet_list (i.e., when modules are being loaded and unloaded).

This gives signify performance increase, after all patch set is applied,
like you may see here:

%for i in {1..10000}; do unshare -n bash -c exit; done

*before*
real 1m40,377s
user 0m9,672s
sys 0m19,928s

*after*
real 0m17,007s
user 0m5,311s
sys 0m11,779

(5.8 times faster)

This patch starts replacing net_mutex to net_sem. It adds rw_semaphore,
describes the variables it protects, and makes to use, where appropriate.
net_mutex is still present, and next patches will kick it out step-by-step.

Signed-off-by: Kirill Tkhai 
Acked-by: Andrei Vagin 
Signed-off-by: David S. Miller

Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next

2018-01-31T22:31:10+00:00

Pull networking updates from David Miller:

 1) Significantly shrink the core networking routing structures. Result
    of http://vger.kernel.org/~davem/seoul2017_netdev_keynote.pdf

 2) Add netdevsim driver for testing various offloads, from Jakub
    Kicinski.

 3) Support cross-chip FDB operations in DSA, from Vivien Didelot.

 4) Add a 2nd listener hash table for TCP, similar to what was done for
    UDP. From Martin KaFai Lau.

 5) Add eBPF based queue selection to tun, from Jason Wang.

 6) Lockless qdisc support, from John Fastabend.

 7) SCTP stream interleave support, from Xin Long.

 8) Smoother TCP receive autotuning, from Eric Dumazet.

 9) Lots of erspan tunneling enhancements, from William Tu.

10) Add true function call support to BPF, from Alexei Starovoitov.

11) Add explicit support for GRO HW offloading, from Michael Chan.

12) Support extack generation in more netlink subsystems. From Alexander
    Aring, Quentin Monnet, and Jakub Kicinski.

13) Add 1000BaseX, flow control, and EEE support to mvneta driver. From
    Russell King.

14) Add flow table abstraction to netfilter, from Pablo Neira Ayuso.

15) Many improvements and simplifications to the NFP driver bpf JIT,
    from Jakub Kicinski.

16) Support for ipv6 non-equal cost multipath routing, from Ido
    Schimmel.

17) Add resource abstration to devlink, from Arkadi Sharshevsky.

18) Packet scheduler classifier shared filter block support, from Jiri
    Pirko.

19) Avoid locking in act_csum, from Davide Caratti.

20) devinet_ioctl() simplifications from Al viro.

21) More TCP bpf improvements from Lawrence Brakmo.

22) Add support for onlink ipv6 route flag, similar to ipv4, from David
    Ahern.

* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next: (1925 commits)
  tls: Add support for encryption using async offload accelerator
  ip6mr: fix stale iterator
  net/sched: kconfig: Remove blank help texts
  openvswitch: meter: Use 64-bit arithmetic instead of 32-bit
  tcp_nv: fix potential integer overflow in tcpnv_acked
  r8169: fix RTL8168EP take too long to complete driver initialization.
  qmi_wwan: Add support for Quectel EP06
  rtnetlink: enable IFLA_IF_NETNSID for RTM_NEWLINK
  ipmr: Fix ptrdiff_t print formatting
  ibmvnic: Wait for device response when changing MAC
  qlcnic: fix deadlock bug
  tcp: release sk_frag.page in tcp_disconnect
  ipv4: Get the address of interface correctly.
  net_sched: gen_estimator: fix lockdep splat
  net: macb: Handle HRESP error
  net/mlx5e: IPoIB, Fix copy-paste bug in flow steering refactoring
  ipv6: addrconf: break critical section in addrconf_verify_rtnl()
  ipv6: change route cache aging logic
  i40e/i40evf: Update DESC_NEEDED value to reflect larger value
  bnxt_en: cleanup DIM work on device shutdown
  ...