If !NULL, @skip_css makes cgroup_taskset_for_each() skip the matching
css.  The intention of the interface is to make it easy to skip css's
(cgroup_subsys_states) which already match the migration target;
however, this is entirely unnecessary as migration taskset doesn't
include tasks which are already in the target cgroup.  Drop @skip_css
from cgroup_taskset_for_each().
Signed-off-by: NTejun Heo <tj@kernel.org>
Acked-by: NLi Zefan <lizefan@huawei.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Arnaldo Carvalho de Melo <acme@ghostprotocols.net>
Cc: Daniel Borkmann <dborkman@redhat.com>

cgroup_subsys is a bit messier than it needs to be.

* The name of a subsys can be different from its internal identifier
  defined in cgroup_subsys.h.  Most subsystems use the matching name
  but three - cpu, memory and perf_event - use different ones.

* cgroup_subsys_id enums are postfixed with _subsys_id and each
  cgroup_subsys is postfixed with _subsys.  cgroup.h is widely
  included throughout various subsystems, it doesn't and shouldn't
  have claim on such generic names which don't have any qualifier
  indicating that they belong to cgroup.

* cgroup_subsys->subsys_id should always equal the matching
  cgroup_subsys_id enum; however, we require each controller to
  initialize it and then BUG if they don't match, which is a bit
  silly.

This patch cleans up cgroup_subsys names and initialization by doing
the followings.

* cgroup_subsys_id enums are now postfixed with _cgrp_id, and each
  cgroup_subsys with _cgrp_subsys.

* With the above, renaming subsys identifiers to match the userland
  visible names doesn't cause any naming conflicts.  All non-matching
  identifiers are renamed to match the official names.

  cpu_cgroup -> cpu
  mem_cgroup -> memory
  perf -> perf_event

* controllers no longer need to initialize ->subsys_id and ->name.
  They're generated in cgroup core and set automatically during boot.

* Redundant cgroup_subsys declarations removed.

* While updating BUG_ON()s in cgroup_init_early(), convert them to
  WARN()s.  BUGging that early during boot is stupid - the kernel
  can't print anything, even through serial console and the trap
  handler doesn't even link stack frame properly for back-tracing.

This patch doesn't introduce any behavior changes.

v2: Rebased on top of fe1217c4 ("net: net_cls: move cgroupfs
    classid handling into core").
Signed-off-by: NTejun Heo <tj@kernel.org>
Acked-by: NNeil Horman <nhorman@tuxdriver.com>
Acked-by: N"David S. Miller" <davem@davemloft.net>
Acked-by: N"Rafael J. Wysocki" <rjw@rjwysocki.net>
Acked-by: NMichal Hocko <mhocko@suse.cz>
Acked-by: NPeter Zijlstra <peterz@infradead.org>
Acked-by: NAristeu Rozanski <aris@redhat.com>
Acked-by: NIngo Molnar <mingo@redhat.com>
Acked-by: NLi Zefan <lizefan@huawei.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Balbir Singh <bsingharora@gmail.com>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Serge E. Hallyn <serue@us.ibm.com>
Cc: Vivek Goyal <vgoyal@redhat.com>
Cc: Thomas Graf <tgraf@suug.ch>

With module supported dropped from net_prio, no controller is using
cgroup module support.  None of actual resource controllers can be
built as a module and we aren't gonna add new controllers which don't
control resources.  This patch drops module support from cgroup.

* cgroup_[un]load_subsys() and cgroup_subsys->module removed.

* As there's no point in distinguishing IS_BUILTIN() and IS_MODULE(),
  cgroup_subsys.h now uses IS_ENABLED() directly.

* enum cgroup_subsys_id now exactly matches the list of enabled
  controllers as ordered in cgroup_subsys.h.

* cgroup_subsys[] is now a contiguously occupied array.  Size
  specification is no longer necessary and dropped.

* for_each_builtin_subsys() is removed and for_each_subsys() is
  updated to not require any locking.

* module ref handling is removed from rebind_subsystems().

* Module related comments dropped.

v2: Rebased on top of fe1217c4 ("net: net_cls: move cgroupfs
    classid handling into core").

v3: Added {} around the if (need_forkexit_callback) block in
    cgroup_post_fork() for readability as suggested by Li.
Signed-off-by: NTejun Heo <tj@kernel.org>
Acked-by: NLi Zefan <lizefan@huawei.com>

net_prio is the only cgroup which is allowed to be built as a module.
The savings from allowing one controller to be built as a module are
tiny especially given that cgroup module support itself adds quite a
bit of complexity.

Given that none of other controllers has much chance of being made a
module and that we're unlikely to add new modular controllers, the
added complexity is simply not justifiable.

As a first step to drop cgroup module support, this patch changes the
config option to bool from tristate and drops module related code from
it.

Also, while an earlier commit fe1217c4 ("net: net_cls: move
cgroupfs classid handling into core") dropped module support from
net_cls cgroup, it retained a call to cgroup_load_subsys(), which is
noop for built-in controllers.  Drop it along with
init_netclassid_cgroup().

v2: Removed modular version of task_netprioidx() in
    include/net/netprio_cgroup.h as suggested by Li Zefan.

v3: Rebased on top of fe1217c4 ("net: net_cls: move cgroupfs
    classid handling into core").  net_cls cgroup part is mostly
    dropped except for removal of init_netclassid_cgroup().
Signed-off-by: NTejun Heo <tj@kernel.org>
Acked-by: NNeil Horman <nhorman@tuxdriver.com>
Acked-by: N"David S. Miller" <davem@davemloft.net>
Acked-by: NLi Zefan <lizefan@huawei.com>
Cc: Thomas Graf <tgraf@suug.ch>

This patch fixed following Warning while executing "make htmldocs".

Warning(/net/core/skbuff.c:2164): No description found for parameter 'from'
Warning(/net/core/skbuff.c:2164): Excess function parameter 'source'
description in 'skb_zerocopy'
Replace "@source" with "@from" fixed the warning.
Signed-off-by: NMasanari Iida <standby24x7@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This moves part of Eric Dumazets skb_gso_seglen helper from tbf sched to
skbuff core so it may be reused by upcoming ip forwarding path patch.
Signed-off-by: NFlorian Westphal <fw@strlen.de>
Acked-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This check is not needed because the same check is done before
fill_slave_info is used in rtnl_link_slave_info_fill.
Also, by removing this check, kernel will fillup IFLA_INFO_SLAVE_KIND
even for slaves of masters which does not implement fill_slave_info.
Signed-off-by: NJiri Pirko <jiri@resnulli.us>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

When we have multiple devices attempting to sync the same address
to a single destination, each device should be permitted to sync
it once.  To accomplish this, pass the 'sync_cnt' of the source
address when adding the addresss to the lower device.  'sync_cnt'
tracks how many time a given address has been succefully synced.
This way, we know that if the 'sync_cnt' passed in is 0, we should
sync this address.

Also, turn 'synced' member back into the counter as was originally
done in
   commit 4543fbef.
   net: count hw_addr syncs so that unsync works properly.
It tracks how many time a given address has been added via a
'sync' operation.  For every successfull 'sync' the counter is
incremented, and for ever 'unsync', the counter is decremented.
This makes sure that the address will be properly removed from
the the lower device when all the upper devices have removed it.
Reported-by: NAndrey Dmitrov <andrey.dmitrov@oktetlabs.ru>
CC: Andrey Dmitrov <andrey.dmitrov@oktetlabs.ru>
CC: Alexandra N. Kossovsky <Alexandra.Kossovsky@oktetlabs.ru>
CC: Konstantin Ushakov <Konstantin.Ushakov@oktetlabs.ru>
Signed-off-by: NVlad Yasevich <vyasevic@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Signed-off-by: NJiri Pirko <jiri@resnulli.us>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Recent patch
bonding: add netlink attributes to slave link dev (1d3ee88a)

Introduced yet another device specific way to access slave information
over rtnetlink. There is one already there for bridge.

This patch introduces generic way to do this, for getting and setting
info as well by extending link_ops. Later on, this new interface will
be used for bridge ports as well.
Signed-off-by: NJiri Pirko <jiri@resnulli.us>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Signed-off-by: NJiri Pirko <jiri@resnulli.us>
Acked-by: NNicolas Dichtel <nicolas.dichtel@6wind.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Workqueue used in neighbour layer have no real dependency of scheduling these on
the cpu which scheduled them.

On a idle system, it is observed that an idle cpu wakes up many times just to
service this work. It would be better if we can schedule it on a cpu which the
scheduler believes to be the most appropriate one.

This patch replaces normal workqueues with power efficient versions. This
doesn't change existing behavior of code unless CONFIG_WQ_POWER_EFFICIENT is
enabled.
Signed-off-by: NViresh Kumar <viresh.kumar@linaro.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Having this struct in module memory could Oops when if the module is
unloaded while the buffer still persists in a pipe.

Since sock_pipe_buf_ops is essentially the same as fuse_dev_pipe_buf_steal
merge them into nosteal_pipe_buf_ops (this is the same as
default_pipe_buf_ops except stealing the page from the buffer is not
allowed).
Reported-by: NAl Viro <viro@zeniv.linux.org.uk>
Signed-off-by: NMiklos Szeredi <mszeredi@suse.cz>
Cc: stable@vger.kernel.org

Use ether_addr_copy instead of memcpy(a, b, ETH_ALEN) to
save some cycles on arm and powerpc.
Signed-off-by: NJoe Perches <joe@perches.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Use ether_addr_copy instead of memcpy(a, b, ETH_ALEN) to
save some cycles on arm and powerpc.
Signed-off-by: NJoe Perches <joe@perches.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Export the gro_find_receive/complete_by_type helpers to they can be invoked
by the gro callbacks of encapsulation protocols such as vxlan.
Signed-off-by: NOr Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Add GRO handlers for protocols that do UDP encapsulation, with the intent of
being able to coalesce packets which encapsulate packets belonging to
the same TCP session.

For GRO purposes, the destination UDP port takes the role of the ether type
field in the ethernet header or the next protocol in the IP header.

The UDP GRO handler will only attempt to coalesce packets whose destination
port is registered to have gro handler.

Use a mark on the skb GRO CB data to disallow (flush) running the udp gro receive
code twice on a packet. This solves the problem of udp encapsulated packets whose
inner VM packet is udp and happen to carry a port which has registered offloads.
Signed-off-by: NShlomo Pongratz <shlomop@mellanox.com>
Signed-off-by: NOr Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

softnet_data is already set to 0, no need to use memset or initialize
specific fields to 0 or NULL afterwards.
Signed-off-by: NSabrina Dubroca <sd@queasysnail.net>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

When I create a new namespace with 'ip netns add net0', or add/remove
new links in a namespace with 'ip link add/delete type veth', rx/tx
queues events can be got in all namespaces. That is because rx/tx queue
ktypes do not have namespace support, and their kobj parents are setted to
NULL. This patch is to fix it.
Reported-by: NLibo Chen <chenlibo@huawei.com>
Signed-off-by: NLibo Chen <chenlibo@huawei.com>
Signed-off-by: NWeilong Chen <chenweilong@huawei.com>
Acked-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

To silent "make htmldocs" warning.
Reported-by: Nkbuild test robot <fengguang.wu@intel.com>
Signed-off-by: NJason Wang <jasowang@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

For user space packet capturing libraries such as libpcap, there's
currently only one way to check which BPF extensions are supported
by the kernel, that is, commit aa1113d9 ("net: filter: return
-EINVAL if BPF_S_ANC* operation is not supported"). For querying all
extensions at once this might be rather inconvenient.

Therefore, this patch introduces a new option which can be used as
an argument for getsockopt(), and allows one to obtain information
about which BPF extensions are supported by the current kernel.

As David Miller suggests, we do not need to define any bits right
now and status quo can just return 0 in order to state that this
versions supports SKF_AD_PROTOCOL up to SKF_AD_PAY_OFFSET. Later
additions to BPF extensions need to add their bits to the
bpf_tell_extensions() function, as documented in the comment.
Signed-off-by: NMichal Sekletar <msekleta@redhat.com>
Cc: David Miller <davem@davemloft.net>
Reviewed-by: NDaniel Borkmann <dborkman@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

If link is IFF_SLAVE, extend link dev netlink attributes to include
slave attributes with new IFLA_SLAVE nest.  Add netlink notification
(RTM_NEWLINK) when slave status changes from backup to active, or
visa-versa.

Adds new ndo_get_slave op to net_device_ops to fill skb with IFLA_SLAVE
attributes.  Currently only used by bonding driver, but could be
used by other aggregating devices with slaves.
Signed-off-by: NScott Feldman <sfeldma@cumulusnetworks.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Extend existing support for netdevice receive queue sysfs attributes to
permit a device-specific attribute group. Initial use case for this
support will be to allow the virtio-net device to export per-receive
queue mergeable receive buffer size.
Signed-off-by: NMichael Dalton <mwdalton@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

skb_page_frag_refill currently permits only order-0 page allocs
unless GFP_WAIT is used. Change skb_page_frag_refill to attempt
higher-order page allocations whether or not GFP_WAIT is used. If
memory cannot be allocated, the allocator will fall back to
successively smaller page allocs (down to order-0 page allocs).

This change brings skb_page_frag_refill in line with the existing
page allocation strategy employed by netdev_alloc_frag, which attempts
higher-order page allocations whether or not GFP_WAIT is set, falling
back to successively lower-order page allocations on failure. Part
of migration of virtio-net to per-receive queue page frag allocators.
Acked-by: NMichael S. Tsirkin <mst@redhat.com>
Acked-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NMichael Dalton <mwdalton@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Currently, if a device changes its mtu, first the change happens (invloving
all the side effects), and after that the NETDEV_CHANGEMTU is sent so that
other devices can catch up with the new mtu. However, if they return
NOTIFY_BAD, then the change is reverted and error returned.

This is a really long and costy operation (sometimes). To fix this, add
NETDEV_PRECHANGEMTU notification which is called prior to any change
actually happening, and if any callee returns NOTIFY_BAD - the change is
aborted. This way we're skipping all the playing with apply/revert the mtu.

CC: "David S. Miller" <davem@davemloft.net>
CC: Jiri Pirko <jiri@resnulli.us>
CC: Eric Dumazet <edumazet@google.com>
CC: Nicolas Dichtel <nicolas.dichtel@6wind.com>
CC: Cong Wang <amwang@redhat.com>
Signed-off-by: NVeaceslav Falico <vfalico@redhat.com>
Acked-by: NJiri Pirko <jiri@resnulli.us>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

When initializing a gro_list for a packet, first check the rxhash of
the incoming skb against that of the skb's in the list. This should be
a very strong inidicator of whether the flow is going to be matched,
and potentially allows a lot of other checks to be short circuited.
Use skb_hash_raw so that we don't force the hash to be calculated.

Tested by running netperf 200 TCP_STREAMs between two machines with
GRO, HW rxhash, and 1G. Saw no performance degration, slight reduction
of time in dev_gro_receive.
Signed-off-by: NTom Herbert <therbert@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

At first Jakub Zawadzki noticed that some divisions by reciprocal_divide
were not correct. (off by one in some cases)
http://www.wireshark.org/~darkjames/reciprocal-buggy.c

He could also show this with BPF:
http://www.wireshark.org/~darkjames/set-and-dump-filter-k-bug.c

The reciprocal divide in linux kernel is not generic enough,
lets remove its use in BPF, as it is not worth the pain with
current cpus.
Signed-off-by: NEric Dumazet <edumazet@google.com>
Reported-by: NJakub Zawadzki <darkjames-ws@darkjames.pl>
Cc: Mircea Gherzan <mgherzan@gmail.com>
Cc: Daniel Borkmann <dxchgb@gmail.com>
Cc: Hannes Frederic Sowa <hannes@stressinduktion.org>
Cc: Matt Evans <matt@ozlabs.org>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: David S. Miller <davem@davemloft.net>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Currently, we don't rename the upper/lower_ifc symlinks in
/sys/class/net/*/ , which might result stale/duplicate links/names.

Fix this by adding netdev_adjacent_rename_links(dev, oldname) which renames
all the upper/lower interface's links to dev from the upper/lower_oldname
to the new name.

We don't need a rollback because only we control these symlinks and if we
fail to rename them - sysfs will anyway complain.
Reported-by: NDing Tianhong <dingtianhong@huawei.com>
CC: Ding Tianhong <dingtianhong@huawei.com>
CC: "David S. Miller" <davem@davemloft.net>
CC: Eric Dumazet <edumazet@google.com>
CC: Nicolas Dichtel <nicolas.dichtel@6wind.com>
CC: Cong Wang <amwang@redhat.com>
Signed-off-by: NVeaceslav Falico <vfalico@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

They clean up the code a bit and can be used further.

CC: Ding Tianhong <dingtianhong@huawei.com>
CC: "David S. Miller" <davem@davemloft.net>
CC: Eric Dumazet <edumazet@google.com>
CC: Nicolas Dichtel <nicolas.dichtel@6wind.com>
CC: Cong Wang <amwang@redhat.com>
Signed-off-by: NVeaceslav Falico <vfalico@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

introduced by:
commit 1f9248e5
"neigh: convert parms to an array"
Signed-off-by: NJiri Pirko <jiri@resnulli.us>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This patch removes the net_random and net_srandom macros and replaces
them with direct calls to the prandom ones. As new commits only seem to
use prandom_u32 there is no use to keep them around.
This change makes it easier to grep for users of prandom_u32.
Signed-off-by: NAruna-Hewapathirane <aruna.hewapathirane@gmail.com>
Suggested-by: NHannes Frederic Sowa <hannes@stressinduktion.org>
Acked-by: NHannes Frederic Sowa <hannes@stressinduktion.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The existing net/netif_rx and net/netif_receive_skb trace events
provide little information about the skb, nor do they indicate how it
entered the stack.

Add trace events at entry of each of the exported functions, including
most fields that are likely to be interesting for debugging driver
datapath behaviour.  Split netif_rx() and netif_receive_skb() so that
internal calls are not traced.
Signed-off-by: NBen Hutchings <bhutchings@solarflare.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The existing net/net_dev_xmit trace event provides little information
about the skb that has been passed to the driver, and it is not
simple to add more since the skb may already have been freed at
the point the event is emitted.

Add a separate trace event before the skb is passed to the driver,
including most fields that are likely to be interesting for debugging
driver datapath behaviour.
Signed-off-by: NBen Hutchings <bhutchings@solarflare.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Signed-off-by: NBen Hutchings <bhutchings@solarflare.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This patch adds a function to set up the partial checksum offset for IP
packets (and optionally re-calculate the pseudo-header checksum) into the
core network code.
The implementation was previously private and duplicated between xen-netback
and xen-netfront, however it is not xen-specific and is potentially useful
to any network driver.
Signed-off-by: NPaul Durrant <paul.durrant@citrix.com>
Cc: David Miller <davem@davemloft.net>
Cc: Eric Dumazet <edumazet@google.com>
Cc: Veaceslav Falico <vfalico@redhat.com>
Cc: Alexander Duyck <alexander.h.duyck@intel.com>
Cc: Nicolas Dichtel <nicolas.dichtel@6wind.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Currently, after changing the MTU for a device, dev_set_mtu() calls
NETDEV_CHANGEMTU notification, however doesn't verify it's return code -
which can be NOTIFY_BAD - i.e. some of the net notifier blocks refused this
change, and continues nevertheless.

To fix this, verify the return code, and if it's an error - then revert the
MTU to the original one, notify again and pass the error code.

CC: Jiri Pirko <jiri@resnulli.us>
CC: "David S. Miller" <davem@davemloft.net>
CC: Eric Dumazet <edumazet@google.com>
CC: Alexander Duyck <alexander.h.duyck@intel.com>
CC: Nicolas Dichtel <nicolas.dichtel@6wind.com>
Signed-off-by: NVeaceslav Falico <vfalico@redhat.com>
Reviewed-by: NJiri Pirko <jiri@resnulli.us>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

GRO layer has a limit of 8 flows being held in GRO list,
for performance reason.

When a packet comes for a flow not yet in the list,
and list is full, we immediately give it to upper
stacks, lowering aggregation performance.

With TSO auto sizing and FQ packet scheduler, this situation
happens more often.

This patch changes strategy to simply evict the oldest flow of
the list. This works better because of the nature of packet
trains for which GRO is efficient. This also has the effect
of lowering the GRO latency if many flows are competing.

Tested :

Used a 40Gbps NIC, with 4 RX queues, and 200 concurrent TCP_STREAM
netperf.

Before patch, aggregate rate is 11Gbps (while a single flow can reach
30Gbps)

After patch, line rate is reached.
Signed-off-by: NEric Dumazet <edumazet@google.com>
Cc: Jerry Chu <hkchu@google.com>
Cc: Neal Cardwell <ncardwell@google.com>
Acked-by: NNeal Cardwell <ncardwell@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Currently, the tx queue were selected implicitly in ndo_dfwd_start_xmit(). The
will cause several issues:

- NETIF_F_LLTX were removed for macvlan, so txq lock were done for macvlan
  instead of lower device which misses the necessary txq synchronization for
  lower device such as txq stopping or frozen required by dev watchdog or
  control path.
- dev_hard_start_xmit() was called with NULL txq which bypasses the net device
  watchdog.
- dev_hard_start_xmit() does not check txq everywhere which will lead a crash
  when tso is disabled for lower device.

Fix this by explicitly introducing a new param for .ndo_select_queue() for just
selecting queues in the case of l2 forwarding offload. netdev_pick_tx() was also
extended to accept this parameter and dev_queue_xmit_accel() was used to do l2
forwarding transmission.

With this fixes, NETIF_F_LLTX could be preserved for macvlan and there's no need
to check txq against NULL in dev_hard_start_xmit(). Also there's no need to keep
a dedicated ndo_dfwd_start_xmit() and we can just reuse the code of
dev_queue_xmit() to do the transmission.

In the future, it was also required for macvtap l2 forwarding support since it
provides a necessary synchronization method.

Cc: John Fastabend <john.r.fastabend@intel.com>
Cc: Neil Horman <nhorman@tuxdriver.com>
Cc: e1000-devel@lists.sourceforge.net
Signed-off-by: NJason Wang <jasowang@redhat.com>
Acked-by: NNeil Horman <nhorman@tuxdriver.com>
Acked-by: NJohn Fastabend <john.r.fastabend@intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

0-DAY kernel build testing backend reported below error:
All error/warnings:

   net/core/pktgen.c: In function 'pktgen_if_write':
>> >> net/core/pktgen.c:1487:10: error: 'struct pktgen_dev' has no member named 'spi'
>> >> net/core/pktgen.c:1488:43: error: 'struct pktgen_dev' has no member named 'spi'

Fix this by encapuslating the code with CONFIG_XFRM.

Cc: Fengguang Wu <fengguang.wu@intel.com>
Signed-off-by: NFan Du <fan.du@windriver.com>
Signed-off-by: NSteffen Klassert <steffen.klassert@secunet.com>

This patch built on top of Commit 299603e8
("net-gro: Prepare GRO stack for the upcoming tunneling support") to add
the support of the standard GRE (RFC1701/RFC2784/RFC2890) to the GRO
stack. It also serves as an example for supporting other encapsulation
protocols in the GRO stack in the future.

The patch supports version 0 and all the flags (key, csum, seq#) but
will flush any pkt with the S (seq#) flag. This is because the S flag
is not support by GSO, and a GRO pkt may end up in the forwarding path,
thus requiring GSO support to break it up correctly.

Currently the "packet_offload" structure only contains L3 (ETH_P_IP/
ETH_P_IPV6) GRO offload support so the encapped pkts are limited to
IP pkts (i.e., w/o L2 hdr). But support for other protocol type can
be easily added, so is the support for GRE variations like NVGRE.

The patch also support csum offload. Specifically if the csum flag is on
and the h/w is capable of checksumming the payload (CHECKSUM_COMPLETE),
the code will take advantage of the csum computed by the h/w when
validating the GRE csum.

Note that commit 60769a5d "ipv4: gre:
add GRO capability" already introduces GRO capability to IPv4 GRE
tunnels, using the gro_cells infrastructure. But GRO is done after
GRE hdr has been removed (i.e., decapped). The following patch applies
GRO when pkts first come in (before hitting the GRE tunnel code). There
is some performance advantage for applying GRO as early as possible.
Also this approach is transparent to other subsystem like Open vSwitch
where GRE decap is handled outside of the IP stack hence making it
harder for the gro_cells stuff to apply. On the other hand, some NICs
are still not capable of hashing on the inner hdr of a GRE pkt (RSS).
In that case the GRO processing of pkts from the same remote host will
all happen on the same CPU and the performance may be suboptimal.

I'm including some rough preliminary performance numbers below. Note
that the performance will be highly dependent on traffic load, mix as
usual. Moreover it also depends on NIC offload features hence the
following is by no means a comprehesive study. Local testing and tuning
will be needed to decide the best setting.

All tests spawned 50 copies of netperf TCP_STREAM and ran for 30 secs.
(super_netperf 50 -H 192.168.1.18 -l 30)

An IP GRE tunnel with only the key flag on (e.g., ip tunnel add gre1
mode gre local 10.246.17.18 remote 10.246.17.17 ttl 255 key 123)
is configured.

The GRO support for pkts AFTER decap are controlled through the device
feature of the GRE device (e.g., ethtool -K gre1 gro on/off).

1.1 ethtool -K gre1 gro off; ethtool -K eth0 gro off
thruput: 9.16Gbps
CPU utilization: 19%

1.2 ethtool -K gre1 gro on; ethtool -K eth0 gro off
thruput: 5.9Gbps
CPU utilization: 15%

1.3 ethtool -K gre1 gro off; ethtool -K eth0 gro on
thruput: 9.26Gbps
CPU utilization: 12-13%

1.4 ethtool -K gre1 gro on; ethtool -K eth0 gro on
thruput: 9.26Gbps
CPU utilization: 10%

The following tests were performed on a different NIC that is capable of
csum offload. I.e., the h/w is capable of computing IP payload csum
(CHECKSUM_COMPLETE).

2.1 ethtool -K gre1 gro on (hence will use gro_cells)

2.1.1 ethtool -K eth0 gro off; csum offload disabled
thruput: 8.53Gbps
CPU utilization: 9%

2.1.2 ethtool -K eth0 gro off; csum offload enabled
thruput: 8.97Gbps
CPU utilization: 7-8%

2.1.3 ethtool -K eth0 gro on; csum offload disabled
thruput: 8.83Gbps
CPU utilization: 5-6%

2.1.4 ethtool -K eth0 gro on; csum offload enabled
thruput: 8.98Gbps
CPU utilization: 5%

2.2 ethtool -K gre1 gro off

2.2.1 ethtool -K eth0 gro off; csum offload disabled
thruput: 5.93Gbps
CPU utilization: 9%

2.2.2 ethtool -K eth0 gro off; csum offload enabled
thruput: 5.62Gbps
CPU utilization: 8%

2.2.3 ethtool -K eth0 gro on; csum offload disabled
thruput: 7.69Gbps
CPU utilization: 8%

2.2.4 ethtool -K eth0 gro on; csum offload enabled
thruput: 8.96Gbps
CPU utilization: 5-6%
Signed-off-by: NH.K. Jerry Chu <hkchu@google.com>
Reviewed-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>