Commit deaa0a6a93 ("net: Lookup actual route when oif is VRF device")
exposed a bug in __ip_route_output_key_hash for VRF devices: on FIB lookup
failure if the oif is specified the current logic drops to make_route on
the assumption that the route tables are wrong. For VRF/L3 master devices
this leads to wrong dst entries and route lookups. For example:
$ ip route ls table vrf-red
unreachable default
broadcast 10.2.1.0 dev eth1 proto kernel scope link src 10.2.1.2
10.2.1.0/24 dev eth1 proto kernel scope link src 10.2.1.2
local 10.2.1.2 dev eth1 proto kernel scope host src 10.2.1.2
broadcast 10.2.1.255 dev eth1 proto kernel scope link src 10.2.1.2
$ ip route get oif vrf-red 1.1.1.1
1.1.1.1 dev vrf-red src 10.0.0.2
cache
With this patch:
$ ip route get oif vrf-red 1.1.1.1
RTNETLINK answers: No route to host
which is the correct response based on the default route
Signed-off-by: David Ahern <dsa@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The network namespace is already passed into dst_output pass it into
dst->output lwt->output and friends.
Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Compute net and store it in a variable in the functions
ip_build_and_send_pkt and ip_queue_xmit so that it does not need to be
recomputed next time it is needed.
Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Store net in a variable in ip_tunnel_xmit so it does not need
to be recomputed when it is used again.
Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
It is confusing and silly hiding a parameter so modify all of
the callers to pass in the appropriate socket or skb->sk if
no socket is known.
Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
For consistency with the other similar methods in the kernel pass a
struct sock into the dst_ops .local_out method.
Simplifying the socket passing case is needed a prequel to passing a
struct net reference into .local_out.
Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Replace dst_output_okfn with dst_output
Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
After a packet has been encapsulated by a tunnel we should use the
tunnel sockets local multicast loopback flag to control if the
encapsulated packet should be locally loopback back.
Pass sk into ip_local_out_sk so that in the rare case we are dealing
with a tunneled packet whose tunnel destination address is a multicast
address the kernel properly decides to loopback this packet.
In practice I don't think this matters as ip_queue_xmit is used by
tcp, l2tp and sctp none of which I am aware of uses ip level
multicasting as they are all point to point communications protocols.
Let's fix this before someone uses ip_queue_xmit for a tunnel protocol
that does use multicast.
Fixes: aad88724c9 ("ipv4: add a sock pointer to dst->output() path.")
Fixes: b0270e9101 ("ipv4: add a sock pointer to ip_queue_xmit()")
Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In the rare case where sk != skb->sk ip_local_out_sk arranges
to call dst->output differently if the skb is queued or not.
This is a bug.
Fix this bug by passing the sk parameter of ip_local_out_sk through
from ip_local_out_sk to __ip_local_out_sk (skipping __ip_local_out).
Fixes: 7026b1ddb6 ("netfilter: Pass socket pointer down through okfn().")
Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Some applications use a listen() backlog of 1.
Prior kernels were silently enforcing a qlen_log of 4, so that we were
sending up to /proc/sys/net/ipv4/tcp_synack_retries SYNACK messages.
Fixes: ef547f2ac1 ("tcp: remove max_qlen_log")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
If the user specifies a VRF device in a get route query the custom route
pointing to the VRF device is returned:
$ ip route ls table vrf-red
unreachable default
broadcast 10.2.1.0 dev eth1 proto kernel scope link src 10.2.1.2
10.2.1.0/24 dev eth1 proto kernel scope link src 10.2.1.2
local 10.2.1.2 dev eth1 proto kernel scope host src 10.2.1.2
broadcast 10.2.1.255 dev eth1 proto kernel scope link src 10.2.1.2
$ ip route get oif vrf-red 10.2.1.40
10.2.1.40 dev vrf-red
cache
Add the flags to skip the custom route and go directly to the FIB. With
this patch the actual route is returned:
$ ip route get oif vrf-red 10.2.1.40
10.2.1.40 dev eth1 src 10.2.1.2
cache
Signed-off-by: David Ahern <dsa@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
ping originated on box through a VRF device is showing up in tcpdump
without a source address:
$ tcpdump -n -i vrf-blue
08:58:33.311303 IP 0.0.0.0 > 10.2.2.254: ICMP echo request, id 2834, seq 1, length 64
08:58:33.311562 IP 10.2.2.254 > 10.2.2.2: ICMP echo reply, id 2834, seq 1, length 64
Add the call to l3mdev_get_saddr to raw_sendmsg.
Signed-off-by: David Ahern <dsa@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add operation to l3mdev to lookup source address for a given flow.
Add support for the operation to VRF driver and convert existing
IPv4 hooks to use the new lookup.
Signed-off-by: David Ahern <dsa@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
VRF device needs the same path selection following lookup to set source
address. Rather than duplicating code, move existing code into a
function that is exported to modules.
Code move only; no functional change.
Signed-off-by: David Ahern <dsa@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This fixes
net/built-in.o: In function `fib_rebalance':
fib_semantics.c:(.text+0x9df14): undefined reference to `__divdi3'
and
net/built-in.o: In function `fib_rebalance':
net/ipv4/fib_semantics.c:572: undefined reference to `__aeabi_ldivmod'
Fixes: 0e884c78ee ("ipv4: L3 hash-based multipath")
Signed-off-by: Peter Nørlund <pch@ordbogen.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Eric W. Biederman says:
====================
net: Pass net through ip fragmention
This is the next installment of my work to pass struct net through the
output path so the code does not need to guess how to figure out which
network namespace it is in, and ultimately routes can have output
devices in another network namespace.
This round focuses on passing net through ip fragmentation which we seem
to call from about everywhere. That is the main ip output paths, the
bridge netfilter code, and openvswitch. This has to happend at once
accross the tree as function pointers are involved.
First some prep work is done, then ipv4 and ipv6 are converted and then
temporary helper functions are removed.
====================
Acked-by: Nicolas Dichtel <nicolas.dichtel@6wind.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
ICMP packets are inspected to let them route together with the flow they
belong to, minimizing the chance that a problematic path will affect flows
on other paths, and so that anycast environments can work with ECMP.
Signed-off-by: Peter Nørlund <pch@ordbogen.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Replaces the per-packet multipath with a hash-based multipath using
source and destination address.
Signed-off-by: Peter Nørlund <pch@ordbogen.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
inet_reqsk_alloc() is used to allocate a temporary request
in order to generate a SYNACK with a cookie. Then later,
syncookie validation also uses a temporary request.
These paths already took a reference on listener refcount,
we can avoid a couple of atomic operations.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
There are multiple races that need fixes :
1) skb_get() + queue skb + kfree_skb() is racy
An accept() can be done on another cpu, data consumed immediately.
tcp_recvmsg() uses __kfree_skb() as it is assumed all skb found in
socket receive queue are private.
Then the kfree_skb() in tcp_rcv_state_process() uses an already freed skb
2) tcp_reqsk_record_syn() needs to be done before tcp_try_fastopen()
for the same reasons.
3) We want to send the SYNACK before queueing child into accept queue,
otherwise we might reintroduce the ooo issue fixed in
commit 7c85af8810 ("tcp: avoid reorders for TFO passive connections")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Everything should now be ready to finally allow SYN
packets processing without holding listener lock.
Tested:
3.5 Mpps SYNFLOOD. Plenty of cpu cycles available.
Next bottleneck is the refcount taken on listener,
that could be avoided if we remove SLAB_DESTROY_BY_RCU
strict semantic for listeners, and use regular RCU.
13.18% [kernel] [k] __inet_lookup_listener
9.61% [kernel] [k] tcp_conn_request
8.16% [kernel] [k] sha_transform
5.30% [kernel] [k] inet_reqsk_alloc
4.22% [kernel] [k] sock_put
3.74% [kernel] [k] tcp_make_synack
2.88% [kernel] [k] ipt_do_table
2.56% [kernel] [k] memcpy_erms
2.53% [kernel] [k] sock_wfree
2.40% [kernel] [k] tcp_v4_rcv
2.08% [kernel] [k] fib_table_lookup
1.84% [kernel] [k] tcp_openreq_init_rwin
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
If a listener with thousands of children in accept queue
is dismantled, it can take a while to close all of them.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This control variable was set at first listen(fd, backlog)
call, but not updated if application tried to increase or decrease
backlog. It made sense at the time listener had a non resizeable
hash table.
Also rounding to powers of two was not very friendly.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
It is enough to check listener sk_state, no need for an extra
condition.
max_qlen_log can be moved into struct request_sock_queue
We can remove syn_wait_lock and the alignment it enforced.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
If a listen backlog is very big (to avoid syncookies), then
the listener sk->sk_wmem_alloc is the main source of false
sharing, as we need to touch it twice per SYNACK re-transmit
and TX completion.
(One SYN packet takes listener lock once, but up to 6 SYNACK
are generated)
By attaching the skb to the request socket, we remove this
source of contention.
Tested:
listen(fd, 10485760); // single listener (no SO_REUSEPORT)
16 RX/TX queue NIC
Sustain a SYNFLOOD attack of ~320,000 SYN per second,
Sending ~1,400,000 SYNACK per second.
Perf profiles now show listener spinlock being next bottleneck.
20.29% [kernel] [k] queued_spin_lock_slowpath
10.06% [kernel] [k] __inet_lookup_established
5.12% [kernel] [k] reqsk_timer_handler
3.22% [kernel] [k] get_next_timer_interrupt
3.00% [kernel] [k] tcp_make_synack
2.77% [kernel] [k] ipt_do_table
2.70% [kernel] [k] run_timer_softirq
2.50% [kernel] [k] ip_finish_output
2.04% [kernel] [k] cascade
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In this patch, we insert request sockets into TCP/DCCP
regular ehash table (where ESTABLISHED and TIMEWAIT sockets
are) instead of using the per listener hash table.
ACK packets find SYN_RECV pseudo sockets without having
to find and lock the listener.
In nominal conditions, this halves pressure on listener lock.
Note that this will allow for SO_REUSEPORT refinements,
so that we can select a listener using cpu/numa affinities instead
of the prior 'consistent hash', since only SYN packets will
apply this selection logic.
We will shrink listen_sock in the following patch to ease
code review.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Ying Cai <ycai@google.com>
Cc: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When request sockets are no longer in a per listener hash table
but on regular TCP ehash, we need to access listener uid
through req->rsk_listener
get_openreq6() also gets a const for its request socket argument.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Once listener is lockless, its sk_state can change anytime.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
We'll soon have to call tcp_v[46]_inbound_md5_hash() twice.
Also add const attribute to the socket, as it might be the
unlocked listener for SYN packets.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This fixes a typo : We want to store the NAPI id on child socket.
Presumably nobody really uses busy polling, on short lived flows.
Fixes: 3d97379a67 ("tcp: move sk_mark_napi_id() at the right place")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
long term plan is to remove struct listen_sock when its hash
table is no longer there.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
qlen_inc & young_inc were protected by listener lock,
while qlen_dec & young_dec were atomic fields.
Everything needs to be atomic for upcoming lockless listener.
Also move qlen/young in request_sock_queue as we'll get rid
of struct listen_sock eventually.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
struct request_sock_queue fields are currently protected
by the listener 'lock' (not a real spinlock)
We need to add a private spinlock instead, so that softirq handlers
creating children do not have to worry with backlog notion
that the listener 'lock' carries.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The fib_table_lookup tracepoint found 2 places where the flowi4_flags is
not initialized.
Signed-off-by: David Ahern <dsa@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Pablo Neira Ayuso says:
====================
Netfilter/IPVS updates for net-next
The following pull request contains Netfilter/IPVS updates for net-next
containing 90 patches from Eric Biederman.
The main goal of this batch is to avoid recurrent lookups for the netns
pointer, that happens over and over again in our Netfilter/IPVS code. The idea
consists of passing netns pointer from the hook state to the relevant functions
and objects where this may be needed.
You can find more information on the IPVS updates from Simon Horman's commit
merge message:
c3456026ad ("Merge tag 'ipvs2-for-v4.4' of https://git.kernel.org/pub/scm/linux/kernel/git/horms/ipvs-next").
Exceptionally, this time, I'm not posting the patches again on netdev, Eric
already Cc'ed this mailing list in the original submission. If you need me to
make, just let me know.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Replace calls to vrf_dev_get_rth with l3mdev_get_rtable.
The check on the flow flags is handled in the l3mdev operation.
Signed-off-by: David Ahern <dsa@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Replace calls to vrf_dev_table and friends with l3mdev_fib_table
and kin.
Signed-off-by: David Ahern <dsa@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Replace calls to vrf_master_ifindex_rcu and vrf_master_ifindex with either
l3mdev_master_ifindex_rcu or l3mdev_master_ifindex.
The pattern:
oif = vrf_master_ifindex(dev) ? : dev->ifindex;
is replaced with
oif = l3mdev_fib_oif(dev);
And remove the now unused vrf macros.
Signed-off-by: David Ahern <dsa@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Rename IFF_VRF_MASTER to IFF_L3MDEV_MASTER and update the name of the
netif_is_vrf and netif_index_is_vrf macros.
Signed-off-by: David Ahern <dsa@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
While auditing TCP stack for upcoming 'lockless' listener changes,
I found I had to change fastopen_init_queue() to properly init the object
before publishing it.
Otherwise an other cpu could try to lock the spinlock before it gets
properly initialized.
Instead of adding appropriate barriers, just remove dynamic memory
allocations :
- Structure is 28 bytes on 64bit arches. Using additional 8 bytes
for holding a pointer seems overkill.
- Two listeners can share same cache line and performance would suffer.
If we really want to save few bytes, we would instead dynamically allocate
whole struct request_sock_queue in the future.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
tcp_syn_flood_action() will soon be called with unlocked socket.
In order to avoid SYN flood warning being emitted multiple times,
use xchg().
Extend max_qlen_log and synflood_warned fields in struct listen_sock
to u32
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
These functions do not change the listener socket.
Goal is to make sure tcp_conn_request() is not messing with
listener in a racy way.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
