When the function dev_get_phys_port_name was added it missed a description
for it's len argument. Adding it.

Fixes: db24a904 ("net: add support for phys_port_name")
Signed-off-by: NLuis de Bethencourt <luisbg@osg.samsung.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Commit 22e0f8b9 ("net: sched: make bstats per cpu and estimator RCU safe")
added the argument cpu_bstats to functions gen_new_estimator and
gen_replace_estimator and now the descriptions of these are missing for the
documentation. Adding them.
Signed-off-by: NLuis de Bethencourt <luisbg@osg.samsung.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Function gnet_stats_copy_basic is missing the description of the cpu
argument in the documentation. Adding it.
Signed-off-by: NLuis de Bethencourt <luisbg@osg.samsung.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

If a packet is either locally encapsulated or processed through GRO
it is marked with the offloads that it requires. However, when it is
decapsulated these tunnel offload indications are not removed. This
means that if we receive an encapsulated TCP packet, aggregate it with
GRO, decapsulate, and retransmit the resulting frame on a NIC that does
not support encapsulation, we won't be able to take advantage of hardware
offloads even though it is just a simple TCP packet at this point.

This fixes the problem by stripping off encapsulation offload indications
when packets are decapsulated.

The performance impacts of this bug are significant. In a test where a
Geneve encapsulated TCP stream is sent to a hypervisor, GRO'ed, decapsulated,
and bridged to a VM performance is improved by 60% (5Gbps->8Gbps) as a
result of avoiding unnecessary segmentation at the VM tap interface.
Reported-by: NRamu Ramamurthy <sramamur@linux.vnet.ibm.com>
Fixes: 68c33163 ("v4 GRE: Add TCP segmentation offload for GRE")
Signed-off-by: NJesse Gross <jesse@kernel.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

When drivers express support for TSO of encapsulated packets, they
only mean that they can do it for one layer of encapsulation.
Supporting additional levels would mean updating, at a minimum,
more IP length fields and they are unaware of this.

No encapsulation device expresses support for handling offloaded
encapsulated packets, so we won't generate these types of frames
in the transmit path. However, GRO doesn't have a check for
multiple levels of encapsulation and will attempt to build them.

UDP tunnel GRO actually does prevent this situation but it only
handles multiple UDP tunnels stacked on top of each other. This
generalizes that solution to prevent any kind of tunnel stacking
that would cause problems.

Fixes: bf5a755f ("net-gre-gro: Add GRE support to the GRO stack")
Signed-off-by: NJesse Gross <jesse@kernel.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

ipip encapsulated packets can be merged together by GRO but the result
does not have the proper GSO type set or even marked as being
encapsulated at all. Later retransmission of these packets will likely
fail if the device does not support ipip offloads. This is similar to
the issue resolved in IPv6 sit in feec0cb3
("ipv6: gro: support sit protocol").
Reported-by: NPatrick Boutilier <boutilpj@ednet.ns.ca>
Fixes: 9667e9bb ("ipip: Add gro callbacks to ipip offload")
Tested-by: NPatrick Boutilier <boutilpj@ednet.ns.ca>
Acked-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NJesse Gross <jesse@kernel.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

SCTP is a protocol that is aligned to a word (4 bytes). Thus using bare
MTU can sometimes return values that are not aligned, like for loopback,
which is 65536 but ipv4_mtu() limits that to 65535. This mis-alignment
will cause the last non-aligned bytes to never be used and can cause
issues with congestion control.

So it's better to just consider a lower MTU and keep congestion control
calcs saner as they are based on PMTU.

Same applies to icmp frag needed messages, which is also fixed by this
patch.

One other effect of this is the inability to send MTU-sized packet
without queueing or fragmentation and without hitting Nagle. As the
check performed at sctp_packet_can_append_data():

if (chunk->skb->len + q->out_qlen >= transport->pathmtu - packet->overhead)
	/* Enough data queued to fill a packet */
	return SCTP_XMIT_OK;

with the above example of MTU, if there are no other messages queued,
one cannot send a packet that just fits one packet (65532 bytes) and
without causing DATA chunk fragmentation or a delay.

v2:
 - Added WORD_TRUNC macro
Signed-off-by: NMarcelo Ricardo Leitner <marcelo.leitner@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Currently, if a chunk is scheduled to be sent through a transport that
is currently unconfirmed, it will be leaked as it is dequeued from outq
and is not re-queued nor freed.

As I'm not aware of any situation that may lead to this situation, I'm
fixing this by freeing the chunk and also logging a trace so that we can
fix the other bug if it ever happens.
Signed-off-by: NMarcelo Ricardo Leitner <marcelo.leitner@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The SACK can be lost pretty much elsewhere, but if its allocation fail,
we know we are not sending it, so it is better to revert a_rwnd to its
previous value as this may give it a chance to issue a window update
later.
Signed-off-by: NMarcelo Ricardo Leitner <marcelo.leitner@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

TCP protocol is still used these days, and TCP uses
clones in its transmit path. We can not optimize linux
stack assuming it is mostly used in routers, or that TCP
is dead.

Fixes: 795bb1c0 ("net: bulk free infrastructure for NAPI context, use napi_consume_skb")
Signed-off-by: NEric Dumazet <edumazet@google.com>
Cc: Jesper Dangaard Brouer <brouer@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

RDS_TCP_DEFAULT_BUFSIZE has been unused since commit 1edd6a14
("RDS-TCP: Do not bloat sndbuf/rcvbuf in rds_tcp_tune").
Signed-off-by: NSowmini Varadhan <sowmini.varadhan@oracle.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Add per-net sysctl tunables to set the size of sndbuf and
rcvbuf on the kernel tcp socket.

The tunables are added at /proc/sys/net/rds/tcp/rds_tcp_sndbuf
and /proc/sys/net/rds/tcp/rds_tcp_rcvbuf.

These values must be set before accept() or connect(),
and there may be an arbitrary number of existing rds-tcp
sockets when the tunable is modified. To make sure that all
connections in the netns pick up the same value for the tunable,
we reset existing rds-tcp connections in the netns, so that
they can reconnect with the new parameters.
Signed-off-by: NSowmini Varadhan <sowmini.varadhan@oracle.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

eBPF defines this as BPF_TUNLEN_MAX and OVS just uses the hard-coded
value inside struct sw_flow_key. Thus, add and use IP_TUNNEL_OPTS_MAX
for this, which makes the code a bit more generic and allows to remove
BPF_TUNLEN_MAX from eBPF code.
Signed-off-by: NDaniel Borkmann <daniel@iogearbox.net>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

We can just add a small helper dst_tclassid() for retrieving the
dst->tclassid value. It makes the code a bit better in that we can
get rid of the ifdef from filter.c by moving this into the header.
Signed-off-by: NDaniel Borkmann <daniel@iogearbox.net>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Currently, the tc_classid from eBPF skb context is write-only, but there's
no good reason for tc programs to limit it to write-only. For example,
it can be used to transfer its state via tail calls where the resulting
tc_classid gets filled gradually.
Signed-off-by: NDaniel Borkmann <daniel@iogearbox.net>
Acked-by: NAlexei Starovoitov <ast@kernel.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

There are two issues with the current code. First one is that we need
to set res->class to 0 in case we use non-default classid matching.

This is important for the case where cls_bpf was initially set up with
an optional binding to a default class with tcf_bind_filter(), where
the underlying qdisc implements bind_tcf() that fills res->class and
tests for it later on when doing the classification. Convention for
these cases is that after tc_classify() was called, such qdiscs (atm,
drr, qfq, cbq, hfsc, htb) first test class, and if 0, then they lookup
based on classid.

Second, there's a bug with da mode, where res->classid is only assigned
a 16 bit minor, but it needs to expand to the full 32 bit major/minor
combination instead, therefore we need to expand with the bound major.
This is fine as classes belonging to a classful qdisc must share the
same major.

Fixes: 045efa82 ("cls_bpf: introduce integrated actions")
Signed-off-by: NDaniel Borkmann <daniel@iogearbox.net>
Acked-by: NAlexei Starovoitov <ast@kernel.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Currently output of MPLS packets on tunnel vports is not allowed by Open
vSwitch. This is because historically encapsulation was done in such a way
that the inner_protocol field of the skb needed to hold the inner protocol
for both MPLS and tunnel encapsulation in order for GSO segmentation to be
performed correctly.

Since b2acd1dc ("openvswitch: Use regular GRE net_device instead of
vport") Open vSwitch makes use of lwt to output to tunnel netdevs which
perform encapsulation. As no drivers expose support for MPLS offloads this
means that GSO packets are segmented in software by validate_xmit_skb(),
which is called from __dev_queue_xmit(), before tunnel encapsulation occurs.
This means that the inner protocol of MPLS is no longer needed by the time
encapsulation occurs and the contention on the inner_protocol field of the
skb no longer occurs.

Thus it is now safe to output MPLS to tunnel vports.
Signed-off-by: NSimon Horman <simon.horman@netronome.com>
Reviewed-by: NJesse Gross <jesse@kernel.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Signed-off-by: NFengguang Wu <fengguang.wu@intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Signed-off-by: NFengguang Wu <fengguang.wu@intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Now SYN_RECV request sockets are installed in ehash table, an ICMP
handler can find a request socket while another cpu handles an incoming
packet transforming this SYN_RECV request socket into an ESTABLISHED
socket.

We need to remove the now obsolete WARN_ON(req->sk), since req->sk
is set when a new child is created and added into listener accept queue.

If this race happens, the ICMP will do nothing special.

Fixes: 079096f1 ("tcp/dccp: install syn_recv requests into ehash table")
Signed-off-by: NEric Dumazet <edumazet@google.com>
Reported-by: NBen Lazarus <blazarus@google.com>
Reported-by: NNeal Cardwell <ncardwell@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

vlan drivers lack proper propagation of gso_max_segs from
lower device.
Signed-off-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The success of CMA allocation largely depends on the success of
migration and key factor of it is page reference count.  Until now, page
reference is manipulated by direct calling atomic functions so we cannot
follow up who and where manipulate it.  Then, it is hard to find actual
reason of CMA allocation failure.  CMA allocation should be guaranteed
to succeed so finding offending place is really important.

In this patch, call sites where page reference is manipulated are
converted to introduced wrapper function.  This is preparation step to
add tracepoint to each page reference manipulation function.  With this
facility, we can easily find reason of CMA allocation failure.  There is
no functional change in this patch.

In addition, this patch also converts reference read sites.  It will
help a second step that renames page._count to something else and
prevents later attempt to direct access to it (Suggested by Andrew).
Signed-off-by: NJoonsoo Kim <iamjoonsoo.kim@lge.com>
Acked-by: NMichal Nazarewicz <mina86@mina86.com>
Acked-by: NVlastimil Babka <vbabka@suse.cz>
Cc: Minchan Kim <minchan@kernel.org>
Cc: Mel Gorman <mgorman@techsingularity.net>
Cc: "Kirill A. Shutemov" <kirill.shutemov@linux.intel.com>
Cc: Sergey Senozhatsky <sergey.senozhatsky.work@gmail.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

When the ETHTOOL_GLINKSETTINGS implementation finds that userland is
using the wrong number of words of link mode bitmaps (or is trying to
find out the right numbers) it sets the cmd field to 0 in the response
structure.

This is inconsistent with the implementation of every other ethtool
command, so let's remove that inconsistency before it gets into a
stable release.

Fixes: 3f1ac7a7 ("net: ethtool: add new ETHTOOL_xLINKSETTINGS API")
Signed-off-by: NBen Hutchings <ben@decadent.org.uk>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

local_bh_disable() + spin_lock() is equivalent to spin_lock_bh(), same for
the unlock/enable case, so replace the calls by the appropriate wrappers.
Signed-off-by: NNicholas Mc Guire <hofrat@osadl.org>
Acked-by: NMarcelo Ricardo Leitner <marcelo.leitner@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

$ make tags
  GEN     tags
ctags: Warning: drivers/acpi/processor_idle.c:64: null expansion of name pattern "\1"
ctags: Warning: drivers/xen/events/events_2l.c:41: null expansion of name pattern "\1"
ctags: Warning: kernel/locking/lockdep.c:151: null expansion of name pattern "\1"
ctags: Warning: kernel/rcu/rcutorture.c:133: null expansion of name pattern "\1"
ctags: Warning: kernel/rcu/rcutorture.c:135: null expansion of name pattern "\1"
ctags: Warning: kernel/workqueue.c:323: null expansion of name pattern "\1"
ctags: Warning: net/ipv4/syncookies.c:53: null expansion of name pattern "\1"
ctags: Warning: net/ipv6/syncookies.c:44: null expansion of name pattern "\1"
ctags: Warning: net/rds/page.c:45: null expansion of name pattern "\1"

Which are all the result of the DEFINE_PER_CPU pattern:

  scripts/tags.sh:200:	'/\<DEFINE_PER_CPU([^,]*, *\([[:alnum:]_]*\)/\1/v/'
  scripts/tags.sh:201:	'/\<DEFINE_PER_CPU_SHARED_ALIGNED([^,]*, *\([[:alnum:]_]*\)/\1/v/'

The below cures them. All except the workqueue one are within reasonable
distance of the 80 char limit. TJ do you have any preference on how to
fix the wq one, or shall we just not care its too long?
Signed-off-by: NPeter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: NDavid S. Miller <davem@davemloft.net>
Acked-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>
Cc: Tejun Heo <tj@kernel.org>
Cc: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

The spin_lock()/spin_unlock() is synchronizing on the
nf_conntrack_locks_all_lock which is equivalent to
spin_unlock_wait() but the later should be more efficient.
Signed-off-by: NNicholas Mc Guire <hofrat@osadl.org>
Signed-off-by: NPablo Neira Ayuso <pablo@netfilter.org>

On loaded TCP servers, looking at millions of sockets can hold
cpu for many seconds, if the lookup condition is very narrow.

(eg : ss dst 1.2.3.4 )

Better add a cond_resched() to allow other processes to access
the cpu.
Signed-off-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Extend OVS conntrack interface to cover NAT.  New nested
OVS_CT_ATTR_NAT attribute may be used to include NAT with a CT action.
A bare OVS_CT_ATTR_NAT only mangles existing and expected connections.
If OVS_NAT_ATTR_SRC or OVS_NAT_ATTR_DST is included within the nested
attributes, new (non-committed/non-confirmed) connections are mangled
according to the rest of the nested attributes.

The corresponding OVS userspace patch series includes test cases (in
tests/system-traffic.at) that also serve as example uses.

This work extends on a branch by Thomas Graf at
https://github.com/tgraf/ovs/tree/nat.
Signed-off-by: NJarno Rajahalme <jarno@ovn.org>
Acked-by: NThomas Graf <tgraf@suug.ch>
Acked-by: NJoe Stringer <joe@ovn.org>
Signed-off-by: NPablo Neira Ayuso <pablo@netfilter.org>

There is no need to help connections that are not confirmed, so we can
delay helping new connections to the time when they are confirmed.
This change is needed for NAT support, and having this as a separate
patch will make the following NAT patch a bit easier to review.
Signed-off-by: NJarno Rajahalme <jarno@ovn.org>
Acked-by: NJoe Stringer <joe@ovn.org>
Signed-off-by: NPablo Neira Ayuso <pablo@netfilter.org>

Repeat the nf_conntrack_in() call when it returns NF_REPEAT.  This
avoids dropping a SYN packet re-opening an existing TCP connection.
Signed-off-by: NJarno Rajahalme <jarno@ovn.org>
Acked-by: NJoe Stringer <joe@ovn.org>
Signed-off-by: NPablo Neira Ayuso <pablo@netfilter.org>

Add a new function ovs_ct_find_existing() to find an existing
conntrack entry for which this packet was already applied to.  This is
only to be called when there is evidence that the packet was already
tracked and committed, but we lost the ct reference due to an
userspace upcall.

ovs_ct_find_existing() is called from skb_nfct_cached(), which can now
hide the fact that the ct reference may have been lost due to an
upcall.  This allows ovs_ct_commit() to be simplified.

This patch is needed by later "openvswitch: Interface with NAT" patch,
as we need to be able to pass the packet through NAT using the
original ct reference also after the reference is lost after an
upcall.
Signed-off-by: NJarno Rajahalme <jarno@ovn.org>
Acked-by: NJoe Stringer <joe@ovn.org>
Signed-off-by: NPablo Neira Ayuso <pablo@netfilter.org>

Only a successful nf_conntrack_in() call can effect a connection state
change, so it suffices to update the key only after the
nf_conntrack_in() returns.

This change is needed for the later NAT patches.
Signed-off-by: NJarno Rajahalme <jarno@ovn.org>
Acked-by: NJoe Stringer <joe@ovn.org>
Signed-off-by: NPablo Neira Ayuso <pablo@netfilter.org>

This makes the code easier to understand and the following patches
more focused.
Signed-off-by: NJarno Rajahalme <jarno@ovn.org>
Acked-by: NJoe Stringer <joe@ovn.org>
Signed-off-by: NPablo Neira Ayuso <pablo@netfilter.org>

NAT checksum recalculation code assumes existence of skb_dst, which
becomes a problem for a later patch in the series ("openvswitch:
Interface with NAT.").  Simplify this by removing the check on
skb_dst, as the checksum will be dealt with later in the stack.
Suggested-by: NPravin Shelar <pshelar@nicira.com>
Signed-off-by: NJarno Rajahalme <jarno@ovn.org>
Signed-off-by: NPablo Neira Ayuso <pablo@netfilter.org>

Remove the definition of IP_CT_NEW_REPLY from the kernel as it does
not make sense.  This allows the definition of IP_CT_NUMBER to be
simplified as well.
Signed-off-by: NJarno Rajahalme <jarno@ovn.org>
Signed-off-by: NPablo Neira Ayuso <pablo@netfilter.org>

Rework the netdev event handler, similar to what the Mellanox Spectrum
driver does, to easily welcome more events later (for example
NETDEV_PRECHANGEUPPER) and use netdev helpers (such as
netif_is_bridge_master).
Signed-off-by: NVivien Didelot <vivien.didelot@savoirfairelinux.com>
Acked-by: NJiri Pirko <jiri@mellanox.com>
Acked-by: NIdo Schimmel <idosch@mellanox.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

netdev_upper_dev_unlink() which notifies NETDEV_CHANGEUPPER, returns
void, as well as del_nbp(). So there's no advantage to catch an eventual
error from the port_bridge_leave routine at the DSA level.

Make this routine void for the DSA layer and its existing drivers.
Signed-off-by: NVivien Didelot <vivien.didelot@savoirfairelinux.com>
Acked-by: NJiri Pirko <jiri@mellanox.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Rename DSA port_join_bridge and port_leave_bridge routines to
respectively port_bridge_join and port_bridge_leave in order to respect
an implicit Port::Bridge namespace.
Signed-off-by: NVivien Didelot <vivien.didelot@savoirfairelinux.com>
Acked-by: NJiri Pirko <jiri@mellanox.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Zefir Kurtisi reported kernel panic with an openwrt specific patch.
However, it turns out that mainline has a similar bug waiting to happen.

Once NF_HOOK() returns the skb is in undefined state and must not be
used.   Moreover, the okfn must consume the skb to support async
processing (NF_QUEUE).

Current okfn in this spot doesn't consume it and caller assumes that
NF_HOOK return value tells us if skb was freed or not, but thats wrong.

It "works" because no in-tree user registers a NFPROTO_BRIDGE hook at
LOCAL_IN that returns STOLEN or NF_QUEUE verdicts.

Once we add NF_QUEUE support for nftables bridge this will break --
NF_QUEUE holds the skb for async processing, caller will erronoulsy
return RX_HANDLER_PASS and on reinject netfilter will access free'd skb.

Fix this by pushing skb up the stack in the okfn instead.

NB: It also seems dubious to use LOCAL_IN while bypassing PRE_ROUTING
completely in this case but this is how its been forever so it seems
preferable to not change this.

Cc: Felix Fietkau <nbd@openwrt.org>
Cc: Zefir Kurtisi <zefir.kurtisi@neratec.com>
Signed-off-by: NFlorian Westphal <fw@strlen.de>
Tested-by: NZefir Kurtisi <zefir.kurtisi@neratec.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The fixed phys delete function simply removed the fixed phy from the
internal linked list and freed the memory. It however did not
unregister the associated phy device. This meant it was still possible
to find the phy device on the mdio bus.

Make fixed_phy_del() an internal function and add a
fixed_phy_unregister() to unregisters the phy device and then uses
fixed_phy_del() to free resources.

Modify DSA to use this new API function, so we don't leak phys.
Signed-off-by: NAndrew Lunn <andrew@lunn.ch>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>