The patch adds ERSPAN type II tunnel support.  The implementation
is based on the draft at [1].  One of the purposes is for Linux
box to be able to receive ERSPAN monitoring traffic sent from
the Cisco switch, by creating a ERSPAN tunnel device.
In addition, the patch also adds ERSPAN TX, so Linux virtual
switch can redirect monitored traffic to the ERSPAN tunnel device.
The traffic will be encapsulated into ERSPAN and sent out.

The implementation reuses tunnel key as ERSPAN session ID, and
field 'erspan' as ERSPAN Index fields:
./ip link add dev ers11 type erspan seq key 100 erspan 123 \
			local 172.16.1.200 remote 172.16.1.100

To use the above device as ERSPAN receiver, configure
Nexus 5000 switch as below:

monitor session 100 type erspan-source
  erspan-id 123
  vrf default
  destination ip 172.16.1.200
  source interface Ethernet1/11 both
  source interface Ethernet1/12 both
  no shut
monitor erspan origin ip-address 172.16.1.100 global

[1] https://tools.ietf.org/html/draft-foschiano-erspan-01
[2] iproute2 patch: http://marc.info/?l=linux-netdev&m=150306086924951&w=2
[3] test script: http://marc.info/?l=linux-netdev&m=150231021807304&w=2Signed-off-by: NWilliam Tu <u9012063@gmail.com>
Signed-off-by: NMeenakshi Vohra <mvohra@vmware.com>
Cc: Alexey Kuznetsov <kuznet@ms2.inr.ac.ru>
Cc: Hideaki YOSHIFUJI <yoshfuji@linux-ipv6.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Remove two references to ufo in the udp send path that are no longer
reachable now that ufo has been removed.

Commit 85f1bd9a ("udp: consistently apply ufo or fragmentation")
is a fix to ufo. It is safe to revert what remains of it.

Also, no skb can enter ip_append_page with skb_is_gso true now that
skb_shinfo(skb)->gso_type is no longer set in ip_append_page/_data.
Signed-off-by: NWillem de Bruijn <willemb@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Signed-off-by: NTonghao Zhang <xiangxia.m.yue@gmail.com>
Acked-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

tcp_peer_is_proven needs a proper route to make the
determination, but dst always is NULL. This bug may
be there at the beginning of git tree. This does not
look serious enough to deserve backports to stable
versions.
Signed-off-by: NTonghao Zhang <xiangxia.m.yue@gmail.com>
Acked-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This is useful for directly looking up a task based on class id rather than
having to scan through all open file descriptors.
Signed-off-by: NSasha Levin <alexander.levin@verizon.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

In some situations tcp_send_loss_probe() can realize that it's unable
to send a loss probe (TLP), and falls back to calling tcp_rearm_rto()
to schedule an RTO timer. In such cases, sometimes tcp_rearm_rto()
realizes that the RTO was eligible to fire immediately or at some
point in the past (delta_us <= 0). Previously in such cases
tcp_rearm_rto() was scheduling such "overdue" RTOs to happen at now +
icsk_rto, which caused needless delays of hundreds of milliseconds
(and non-linear behavior that made reproducible testing
difficult). This commit changes the logic to schedule "overdue" RTOs
ASAP, rather than at now + icsk_rto.

Fixes: 6ba8a3b1 ("tcp: Tail loss probe (TLP)")
Suggested-by: NYuchung Cheng <ycheng@google.com>
Signed-off-by: NNeal Cardwell <ncardwell@google.com>
Signed-off-by: NYuchung Cheng <ycheng@google.com>
Signed-off-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Syzkaller hit 'general protection fault in fib_dump_info' bug on
commit 4.13-rc5..

Guilty file: net/ipv4/fib_semantics.c

kasan: GPF could be caused by NULL-ptr deref or user memory access
general protection fault: 0000 [#1] SMP KASAN
Modules linked in:
CPU: 0 PID: 2808 Comm: syz-executor0 Not tainted 4.13.0-rc5 #1
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS
Ubuntu-1.8.2-1ubuntu1 04/01/2014
task: ffff880078562700 task.stack: ffff880078110000
RIP: 0010:fib_dump_info+0x388/0x1170 net/ipv4/fib_semantics.c:1314
RSP: 0018:ffff880078117010 EFLAGS: 00010206
RAX: dffffc0000000000 RBX: 00000000000000fe RCX: 0000000000000002
RDX: 0000000000000006 RSI: ffff880078117084 RDI: 0000000000000030
RBP: ffff880078117268 R08: 000000000000000c R09: ffff8800780d80c8
R10: 0000000058d629b4 R11: 0000000067fce681 R12: 0000000000000000
R13: ffff8800784bd540 R14: ffff8800780d80b5 R15: ffff8800780d80a4
FS:  00000000022fa940(0000) GS:ffff88007fc00000(0000)
knlGS:0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 00000000004387d0 CR3: 0000000079135000 CR4: 00000000000006f0
Call Trace:
  inet_rtm_getroute+0xc89/0x1f50 net/ipv4/route.c:2766
  rtnetlink_rcv_msg+0x288/0x680 net/core/rtnetlink.c:4217
  netlink_rcv_skb+0x340/0x470 net/netlink/af_netlink.c:2397
  rtnetlink_rcv+0x28/0x30 net/core/rtnetlink.c:4223
  netlink_unicast_kernel net/netlink/af_netlink.c:1265 [inline]
  netlink_unicast+0x4c4/0x6e0 net/netlink/af_netlink.c:1291
  netlink_sendmsg+0x8c4/0xca0 net/netlink/af_netlink.c:1854
  sock_sendmsg_nosec net/socket.c:633 [inline]
  sock_sendmsg+0xca/0x110 net/socket.c:643
  ___sys_sendmsg+0x779/0x8d0 net/socket.c:2035
  __sys_sendmsg+0xd1/0x170 net/socket.c:2069
  SYSC_sendmsg net/socket.c:2080 [inline]
  SyS_sendmsg+0x2d/0x50 net/socket.c:2076
  entry_SYSCALL_64_fastpath+0x1a/0xa5
  RIP: 0033:0x4512e9
  RSP: 002b:00007ffc75584cc8 EFLAGS: 00000216 ORIG_RAX:
  000000000000002e
  RAX: ffffffffffffffda RBX: 0000000000000002 RCX: 00000000004512e9
  RDX: 0000000000000000 RSI: 0000000020f2cfc8 RDI: 0000000000000003
  RBP: 000000000000000e R08: 0000000000000000 R09: 0000000000000000
  R10: 0000000000000000 R11: 0000000000000216 R12: fffffffffffffffe
  R13: 0000000000718000 R14: 0000000020c44ff0 R15: 0000000000000000
  Code: 00 0f b6 8d ec fd ff ff 48 8b 85 f0 fd ff ff 88 48 17 48 8b 45
  28 48 8d 78 30 48 b8 00 00 00 00 00 fc ff df 48 89 fa 48 c1 ea 03
  <0f>
  b6 04 02 84 c0 74 08 3c 03 0f 8e cb 0c 00 00 48 8b 45 28 44
  RIP: fib_dump_info+0x388/0x1170 net/ipv4/fib_semantics.c:1314 RSP:
  ffff880078117010
---[ end trace 254a7af28348f88b ]---

This patch adds a res->fi NULL check.

example run:
$ip route get 0.0.0.0 iif virt1-0
broadcast 0.0.0.0 dev lo
    cache <local,brd> iif virt1-0

$ip route get 0.0.0.0 iif virt1-0 fibmatch
RTNETLINK answers: No route to host
Reported-by: Nidaifish <idaifish@gmail.com>
Reported-by: NDmitry Vyukov <dvyukov@google.com>
Fixes: b6179813 ("net: ipv4: RTM_GETROUTE: return matched fib result when requested")
Signed-off-by: NRoopa Prabhu <roopa@cumulusnetworks.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

refcount_t type and corresponding API should be
used instead of atomic_t when the variable is used as
a reference counter. This allows to avoid accidental
refcounter overflows that might lead to use-after-free
situations.
Signed-off-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Due to commit e6afc8ac ("udp: remove
headers from UDP packets before queueing"), when udp packets are being
peeked the requested extra offset is always 0 as there is no need to skip
the udp header.  However, when the offset is 0 and the next skb is
of length 0, it is only returned once.  The behaviour can be seen with
the following python script:

from socket import *;
f=socket(AF_INET6, SOCK_DGRAM | SOCK_NONBLOCK, 0);
g=socket(AF_INET6, SOCK_DGRAM | SOCK_NONBLOCK, 0);
f.bind(('::', 0));
addr=('::1', f.getsockname()[1]);
g.sendto(b'', addr)
g.sendto(b'b', addr)
print(f.recvfrom(10, MSG_PEEK));
print(f.recvfrom(10, MSG_PEEK));

Where the expected output should be the empty string twice.

Instead, make sk_peek_offset return negative values, and pass those values
to __skb_try_recv_datagram/__skb_try_recv_from_queue.  If the passed offset
to __skb_try_recv_from_queue is negative, the checked skb is never skipped.
__skb_try_recv_from_queue will then ensure the offset is reset back to 0
if a peek is requested without an offset, unless no packets are found.

Also simplify the if condition in __skb_try_recv_from_queue.  If _off is
greater then 0, and off is greater then or equal to skb->len, then
(_off || skb->len) must always be true assuming skb->len >= 0 is always
true.

Also remove a redundant check around a call to sk_peek_offset in af_unix.c,
as it double checked if MSG_PEEK was set in the flags.

V2:
 - Moved the negative fixup into __skb_try_recv_from_queue, and remove now
redundant checks
 - Fix peeking in udp{,v6}_recvmsg to report the right value when the
offset is 0

V3:
 - Marked new branch in __skb_try_recv_from_queue as unlikely.
Signed-off-by: NMatthew Dawson <matthew@mjdsystems.ca>
Acked-by: NWillem de Bruijn <willemb@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

While working on yet another syzkaller report, I found
that our IP_MAX_MTU enforcements were not properly done.

gcc seems to reload dev->mtu for min(dev->mtu, IP_MAX_MTU), and
final result can be bigger than IP_MAX_MTU :/

This is a problem because device mtu can be changed on other cpus or
threads.

While this patch does not fix the issue I am working on, it is
probably worth addressing it.
Signed-off-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Anuradha reported that statically added groups for interfaces enslaved
to a VRF device were not persisting. The problem is that igmp queries
and reports need to use the data in the in_dev for the real ingress
device rather than the VRF device. Update igmp_rcv accordingly.

Fixes: e58e4159 ("net: Enable support for VRF with ipv4 multicast")
Reported-by: NAnuradha Karuppiah <anuradhak@cumulusnetworks.com>
Signed-off-by: NDavid Ahern <dsahern@gmail.com>
Reviewed-by: NNikolay Aleksandrov <nikolay@cumulusnetworks.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Signed-off-by: NFlorian Westphal <fw@strlen.de>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

If fi->fib_metrics could not be allocated in fib_create_info()
we attempt to dereference a NULL pointer in free_fib_info_rcu() :

    m = fi->fib_metrics;
    if (m != &dst_default_metrics && atomic_dec_and_test(&m->refcnt))
            kfree(m);

Before my recent patch, we used to call kfree(NULL) and nothing wrong
happened.

Instead of using RCU to defer freeing while we are under memory stress,
it seems better to take immediate action.

This was reported by syzkaller team.

Fixes: 3fb07daf ("ipv4: add reference counting to metrics")
Signed-off-by: NEric Dumazet <edumazet@google.com>
Reported-by: NDmitry Vyukov <dvyukov@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Filtering the ACK packet was not put at the right place.

At this place, we already allocated a child and put it
into accept queue.

We absolutely need to call tcp_child_process() to release
its spinlock, or we will deadlock at accept() or close() time.

Found by syzkaller team (Thanks a lot !)

Fixes: 8fac365f ("tcp: Add a tcp_filter hook before handle ack packet")
Signed-off-by: NEric Dumazet <edumazet@google.com>
Reported-by: NDmitry Vyukov <dvyukov@google.com>
Cc: Chenbo Feng <fengc@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

__tcp_ulp_find_autoload returns tcp_ulp_ops after taking a reference on
the module. Then, if ->init fails, tcp_set_ulp propagates the error but
nothing releases that reference.

Fixes: 734942cc ("tcp: ULP infrastructure")
Signed-off-by: NSabrina Dubroca <sd@queasysnail.net>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

"ip route get $daddr iif eth0 from $saddr" causes:
 BUG: KASAN: use-after-free in ip_route_input_rcu+0x1535/0x1b50
 Call Trace:
  ip_route_input_rcu+0x1535/0x1b50
  ip_route_input_noref+0xf9/0x190
  tcp_v4_early_demux+0x1a4/0x2b0
  ip_rcv+0xbcb/0xc05
  __netif_receive_skb+0x9c/0xd0
  netif_receive_skb_internal+0x5a8/0x890

Problem is that inet_rtm_getroute calls either ip_route_input_rcu (if an
iif was provided) or ip_route_output_key_hash_rcu.

But ip_route_input_rcu, unlike ip_route_output_key_hash_rcu, already
associates the dst_entry with the skb.  This clears the SKB_DST_NOREF
bit (i.e. skb_dst_drop will release/free the entry while it should not).

Thus only set the dst if we called ip_route_output_key_hash_rcu().

I tested this patch by running:
 while true;do ip r get 10.0.1.2;done > /dev/null &
 while true;do ip r get 10.0.1.2 iif eth0  from 10.0.1.1;done > /dev/null &
... and saw no crash or memory leak.

Cc: Roopa Prabhu <roopa@cumulusnetworks.com>
Cc: David Ahern <dsahern@gmail.com>
Fixes: ba52d61e ("ipv4: route: restore skb_dst_set in inet_rtm_getroute")
Signed-off-by: NFlorian Westphal <fw@strlen.de>
Acked-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Similar to the loopback device, for packets sent through a VRF device
the index returned in ipi_ifindex needs to be the saved index in
rt_iif.
Signed-off-by: NDavid Ahern <dsahern@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

rt_iif is going to be set to either 0 or orig_oif. If orig_oif
is 0 it amounts to the same end result so remove the check.
Signed-off-by: NDavid Ahern <dsahern@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Attempts to connect to a local address with a socket bound
to a device with the local address hangs if there is no listener:

  $ ip addr sh dev eth1
  3: eth1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq state UP group default qlen 1000
    link/ether 02:e0:f9:1c:00:37 brd ff:ff:ff:ff:ff:ff
    inet 10.100.1.4/24 scope global eth1
       valid_lft forever preferred_lft forever
    inet6 2001:db8:1::4/120 scope global
       valid_lft forever preferred_lft forever
    inet6 fe80::e0:f9ff:fe1c:37/64 scope link
       valid_lft forever preferred_lft forever

  $ vrf-test -I eth1 -r 10.100.1.4
  <hangs when there is no server>

(don't let the command name fool you; vrf-test works without vrfs.)

The problem is that the original intended device, eth1 in this case, is
lost when the tcp reset is sent, so the socket lookup does not find a
match for the reset and the connect attempt hangs. Fix by adjusting
orig_oif for local traffic to the device from the fib lookup result.

With this patch you get the more user friendly:
  $ vrf-test -I eth1 -r 10.100.1.4
  connect failed: 111: Connection refused

orig_oif is saved to the newly created rtable as rt_iif and when set
it is used as the dif for socket lookups. It is set based on flowi4_oif
passed in to ip_route_output_key_hash_rcu and will be set to either
the loopback device, an l3mdev device, nothing (flowi4_oif = 0 which
is the case in the example above) or a netdev index depending on the
lookup path. In each case, resetting orig_oif to the device in the fib
result for the RTN_LOCAL case allows the actual device to be preserved
as the skb tx and rx is done over the loopback or VRF device.
Signed-off-by: NDavid Ahern <dsahern@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

On systems that use mark-based routing it may be necessary for
routing lookups to use marks in order for packets to be routed
correctly. An example of such a system is Android, which uses
socket marks to route packets via different networks.

Currently, routing lookups in tunnel mode always use a mark of
zero, making routing incorrect on such systems.

This patch adds a new output_mark element to the xfrm state and
a corresponding XFRMA_OUTPUT_MARK netlink attribute. The output
mark differs from the existing xfrm mark in two ways:

1. The xfrm mark is used to match xfrm policies and states, while
   the xfrm output mark is used to set the mark (and influence
   the routing) of the packets emitted by those states.
2. The existing mark is constrained to be a subset of the bits of
   the originating socket or transformed packet, but the output
   mark is arbitrary and depends only on the state.

The use of a separate mark provides additional flexibility. For
example:

- A packet subject to two transforms (e.g., transport mode inside
  tunnel mode) can have two different output marks applied to it,
  one for the transport mode SA and one for the tunnel mode SA.
- On a system where socket marks determine routing, the packets
  emitted by an IPsec tunnel can be routed based on a mark that
  is determined by the tunnel, not by the marks of the
  unencrypted packets.
- Support for setting the output marks can be introduced without
  breaking any existing setups that employ both mark-based
  routing and xfrm tunnel mode. Simply changing the code to use
  the xfrm mark for routing output packets could xfrm mark could
  change behaviour in a way that breaks these setups.

If the output mark is unspecified or set to zero, the mark is not
set or changed.

Tested: make allyesconfig; make -j64
Tested: https://android-review.googlesource.com/452776Signed-off-by: NLorenzo Colitti <lorenzo@google.com>
Signed-off-by: NSteffen Klassert <steffen.klassert@secunet.com>

When iteratively building a UDP datagram with MSG_MORE and that
datagram exceeds MTU, consistently choose UFO or fragmentation.

Once skb_is_gso, always apply ufo. Conversely, once a datagram is
split across multiple skbs, do not consider ufo.

Sendpage already maintains the first invariant, only add the second.
IPv6 does not have a sendpage implementation to modify.

A gso skb must have a partial checksum, do not follow sk_no_check_tx
in udp_send_skb.

Found by syzkaller.

Fixes: e89e9cf5 ("[IPv4/IPv6]: UFO Scatter-gather approach")
Reported-by: NAndrey Konovalov <andreyknvl@google.com>
Signed-off-by: NWillem de Bruijn <willemb@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Commit dcd87999 ("igmp: net: Move igmp namespace init to correct file")
moved the igmp sysctls initialization from tcp_sk_init to igmp_net_init. This
function is only called as part of per-namespace initialization, only if
CONFIG_IP_MULTICAST is defined, otherwise igmp_mc_init() call in ip_init is
compiled out, casuing the igmp pernet ops to not be registerd and those sysctl
being left initialized with 0. However, there are certain functions, such as
ip_mc_join_group which are always compiled and make use of some of those
sysctls. Let's do a partial revert of the aforementioned commit and move the
sysctl initialization into inet_init_net, that way they will always have
sane values.

Fixes: dcd87999 ("igmp: net: Move igmp namespace init to correct file")
Link: https://bugzilla.kernel.org/show_bug.cgi?id=196595Reported-by: NGerardo Exequiel Pozzi <vmlinuz386@gmail.com>
Signed-off-by: NNikolay Borisov <nborisov@suse.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This change allows us to later indicate to rtnetlink core that certain
doit functions should be called without acquiring rtnl_mutex.

This change should have no effect, we simply replace the last (now
unused) calcit argument with the new flag.
Signed-off-by: NFlorian Westphal <fw@strlen.de>
Reviewed-by: NHannes Frederic Sowa <hannes@stressinduktion.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

skb_warn_bad_offload triggers a warning when an skb enters the GSO
stack at __skb_gso_segment that does not have CHECKSUM_PARTIAL
checksum offload set.

Commit b2504a5d ("net: reduce skb_warn_bad_offload() noise")
observed that SKB_GSO_DODGY producers can trigger the check and
that passing those packets through the GSO handlers will fix it
up. But, the software UFO handler will set ip_summed to
CHECKSUM_NONE.

When __skb_gso_segment is called from the receive path, this
triggers the warning again.

Make UFO set CHECKSUM_UNNECESSARY instead of CHECKSUM_NONE. On
Tx these two are equivalent. On Rx, this better matches the
skb state (checksum computed), as CHECKSUM_NONE here means no
checksum computed.

See also this thread for context:
http://patchwork.ozlabs.org/patch/799015/

Fixes: b2504a5d ("net: reduce skb_warn_bad_offload() noise")
Signed-off-by: NWillem de Bruijn <willemb@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

With new TCP_FASTOPEN_CONNECT socket option, there is a possibility
to call tcp_connect() while socket sk_dst_cache is either NULL
or invalid.

 +0 socket(..., SOCK_STREAM, IPPROTO_TCP) = 4
 +0 fcntl(4, F_SETFL, O_RDWR|O_NONBLOCK) = 0
 +0 setsockopt(4, SOL_TCP, TCP_FASTOPEN_CONNECT, [1], 4) = 0
 +0 connect(4, ..., ...) = 0

<< sk->sk_dst_cache becomes obsolete, or even set to NULL >>

 +1 sendto(4, ..., 1000, MSG_FASTOPEN, ..., ...) = 1000

We need to refresh the route otherwise bad things can happen,
especially when syzkaller is running on the host :/

Fixes: 19f6d3f3 ("net/tcp-fastopen: Add new API support")
Reported-by: NDmitry Vyukov <dvyukov@google.com>
Signed-off-by: NEric Dumazet <edumazet@google.com>
Cc: Wei Wang <weiwan@google.com>
Cc: Yuchung Cheng <ycheng@google.com>
Acked-by: NWei Wang <weiwan@google.com>
Acked-by: NYuchung Cheng <ycheng@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Add a second device index, sdif, to raw socket lookups. sdif is the
index for ingress devices enslaved to an l3mdev. It allows the lookups
to consider the enslaved device as well as the L3 domain when searching
for a socket.
Signed-off-by: NDavid Ahern <dsahern@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Add a second device index, sdif, to udp socket lookups. sdif is the
index for ingress devices enslaved to an l3mdev. It allows the lookups
to consider the enslaved device as well as the L3 domain when searching
for a socket.

Early demux lookups are handled in the next patch as part of INET_MATCH
changes.
Signed-off-by: NDavid Ahern <dsahern@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Signed-off-by: NDavid Ahern <dsahern@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Add a second device index, sdif, to raw socket lookups. sdif is the
index for ingress devices enslaved to an l3mdev. It allows the lookups
to consider the enslaved device as well as the L3 domain when searching
for a socket.
Signed-off-by: NDavid Ahern <dsahern@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Add a second device index, sdif, to inet socket lookups. sdif is the
index for ingress devices enslaved to an l3mdev. It allows the lookups
to consider the enslaved device as well as the L3 domain when searching
for a socket.

TCP moves the data in the cb. Prior to tcp_v4_rcv (e.g., early demux) the
ingress index is obtained from IPCB using inet_sdif and after the cb move
in  tcp_v4_rcv the tcp_v4_sdif helper is used.
Signed-off-by: NDavid Ahern <dsahern@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Add a second device index, sdif, to udp socket lookups. sdif is the
index for ingress devices enslaved to an l3mdev. It allows the lookups
to consider the enslaved device as well as the L3 domain when searching
for a socket.

Early demux lookups are handled in the next patch as part of INET_MATCH
changes.
Signed-off-by: NDavid Ahern <dsahern@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Most TCP congestion controls are using identical logic to undo
cwnd except BBR. This patch consolidates these similar functions
to the one used currently by Reno and others.
Suggested-by: NNeal Cardwell <ncardwell@google.com>
Signed-off-by: NYuchung Cheng <ycheng@google.com>
Signed-off-by: NNeal Cardwell <ncardwell@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Using ssthresh to revert cwnd is less reliable when ssthresh is
bounded to 2 packets. This patch uses an existing variable in TCP
"prior_cwnd" that snapshots the cwnd right before entering fast
recovery and RTO recovery in Reno.  This fixes the issue discussed
in netdev thread: "A buggy behavior for Linux TCP Reno and HTCP"
https://www.spinics.net/lists/netdev/msg444955.htmlSuggested-by: NNeal Cardwell <ncardwell@google.com>
Reported-by: NWei Sun <unlcsewsun@gmail.com>
Signed-off-by: NYuchung Cheng <ncardwell@google.com>
Signed-off-by: NNeal Cardwell <ncardwell@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

__ip_options_echo() does not need anymore skb->dst, so we can
avoid explicitly preserving it for its own sake.

This is almost a revert of commit 0ddf3fb2 ("udp: preserve
skb->dst if required for IP options processing") plus some
lifting to fit later changes.
Signed-off-by: NPaolo Abeni <pabeni@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

ip_options_echo() does not use anymore the skb->dst and don't
need to keep the dst around for options's sake only.
This reverts commit 34b2cef2.
Signed-off-by: NPaolo Abeni <pabeni@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

__ip_options_echo() uses the current network namespace, and
currently retrives it via skb->dst->dev.

This commit adds an explicit 'net' argument to __ip_options_echo()
and update all the call sites to provide it, usually via a simpler
sock_net().

After this change, __ip_options_echo() no more needs to access
skb->dst and we can drop a couple of hack to preserve such
info in the rx path.
Signed-off-by: NPaolo Abeni <pabeni@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

While computing the response option set for LSRR, ip_options_echo()
also changes the ingress packet LSRR addresses list, setting
the last one to the dst specific address for the ingress packet
- via memset(start[ ...
The only visible effect of such change - beyond possibly damaging
shared/cloned skbs - is modifying the data carried by ICMP replies
changing the header information for reported the ingress packet,
which violates RFC1122 3.2.2.6.
All the others call sites just ignore the ingress packet IP options
after calling ip_options_echo()
Note that the last element in the LSRR option address list for the
reply packet will be properly set later in the ip output path
via ip_options_build().
This buggy memset() predates git history and apparently was present
into the initial ip_options_echo() implementation in linux 1.3.30 but
still looks wrong.

The removal of the fib_compute_spec_dst() call will help
completely dropping the skb->dst usage by __ip_options_echo() with a
later patch.
Signed-off-by: NPaolo Abeni <pabeni@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Enable support for MSG_ZEROCOPY to the TCP stack. TSO and GSO are
both supported. Only data sent to remote destinations is sent without
copying. Packets looped onto a local destination have their payload
copied to avoid unbounded latency.

Tested:
  A 10x TCP_STREAM between two hosts showed a reduction in netserver
  process cycles by up to 70%, depending on packet size. Systemwide,
  savings are of course much less pronounced, at up to 20% best case.

  msg_zerocopy.sh 4 tcp:

  without zerocopy
    tx=121792 (7600 MB) txc=0 zc=n
    rx=60458 (7600 MB)

  with zerocopy
    tx=286257 (17863 MB) txc=286257 zc=y
    rx=140022 (17863 MB)

  This test opens a pair of sockets over veth, one one calls send with
  64KB and optionally MSG_ZEROCOPY and on the other reads the initial
  bytes. The receiver truncates, so this is strictly an upper bound on
  what is achievable. It is more representative of sending data out of
  a physical NIC (when payload is not touched, either).
Signed-off-by: NWillem de Bruijn <willemb@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Fix a TCP loss recovery performance bug raised recently on the netdev
list, in two threads:

(i)  July 26, 2017: netdev thread "TCP fast retransmit issues"
(ii) July 26, 2017: netdev thread:
     "[PATCH V2 net-next] TLP: Don't reschedule PTO when there's one
     outstanding TLP retransmission"

The basic problem is that incoming TCP packets that did not indicate
forward progress could cause the xmit timer (TLP or RTO) to be rearmed
and pushed back in time. In certain corner cases this could result in
the following problems noted in these threads:

 - Repeated ACKs coming in with bogus SACKs corrupted by middleboxes
   could cause TCP to repeatedly schedule TLPs forever. We kept
   sending TLPs after every ~200ms, which elicited bogus SACKs, which
   caused more TLPs, ad infinitum; we never fired an RTO to fill in
   the holes.

 - Incoming data segments could, in some cases, cause us to reschedule
   our RTO or TLP timer further out in time, for no good reason. This
   could cause repeated inbound data to result in stalls in outbound
   data, in the presence of packet loss.

This commit fixes these bugs by changing the TLP and RTO ACK
processing to:

 (a) Only reschedule the xmit timer once per ACK.

 (b) Only reschedule the xmit timer if tcp_clean_rtx_queue() deems the
     ACK indicates sufficient forward progress (a packet was
     cumulatively ACKed, or we got a SACK for a packet that was sent
     before the most recent retransmit of the write queue head).

This brings us back into closer compliance with the RFCs, since, as
the comment for tcp_rearm_rto() notes, we should only restart the RTO
timer after forward progress on the connection. Previously we were
restarting the xmit timer even in these cases where there was no
forward progress.

As a side benefit, this commit simplifies and speeds up the TCP timer
arming logic. We had been calling inet_csk_reset_xmit_timer() three
times on normal ACKs that cumulatively acknowledged some data:

1) Once near the top of tcp_ack() to switch from TLP timer to RTO:
        if (icsk->icsk_pending == ICSK_TIME_LOSS_PROBE)
               tcp_rearm_rto(sk);

2) Once in tcp_clean_rtx_queue(), to update the RTO:
        if (flag & FLAG_ACKED) {
               tcp_rearm_rto(sk);

3) Once in tcp_ack() after tcp_fastretrans_alert() to switch from RTO
   to TLP:
        if (icsk->icsk_pending == ICSK_TIME_RETRANS)
               tcp_schedule_loss_probe(sk);

This commit, by only rescheduling the xmit timer once per ACK,
simplifies the code and reduces CPU overhead.

This commit was tested in an A/B test with Google web server
traffic. SNMP stats and request latency metrics were within noise
levels, substantiating that for normal web traffic patterns this is a
rare issue. This commit was also tested with packetdrill tests to
verify that it fixes the timer behavior in the corner cases discussed
in the netdev threads mentioned above.

This patch is a bug fix patch intended to be queued for -stable
relases.

Fixes: 6ba8a3b1 ("tcp: Tail loss probe (TLP)")
Reported-by: NKlavs Klavsen <kl@vsen.dk>
Reported-by: NMao Wenan <maowenan@huawei.com>
Signed-off-by: NNeal Cardwell <ncardwell@google.com>
Signed-off-by: NYuchung Cheng <ycheng@google.com>
Signed-off-by: NNandita Dukkipati <nanditad@google.com>
Acked-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>