Netlink messages are appended, one object at a time, to the end of
the SKB. Therefore we need to test skb_tail_pointer() not skb->data
for alignment.
Fixes: 35c5845957 ("net: Add helpers for 64-bit aligning netlink attributes.")
Signed-off-by: David S. Miller <davem@davemloft.net>
HAVE_EFFICIENT_UNALIGNED_ACCESS needs CONFIG_ prefix.
Also add a comment in nla_align_64bit() explaining we have
to add a padding if current skb->data is aligned, as it
certainly can be confusing.
Fixes: 35c5845957 ("net: Add helpers for 64-bit aligning netlink attributes.")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When using LVS-NAT and SIP persistence-egine over UDP, the following
limitations are present with current implementation:
1) To actually have load-balancing based on Call-ID header, you need to
use one-packet-scheduling mode. But with one-packet-scheduling the
connection is deleted just after packet is forwarded, so SIP responses
coming from real-servers do not match any connection and SNAT is
not applied.
2) If you do not use "-o" option, IPVS behaves as normal UDP load
balancer, so different SIP calls (each one identified by a different
Call-ID) coming from the same ip-address/port go to the same
real-server. So basically you don’t have load-balancing based on
Call-ID as intended.
3) Call-ID is not learned when a new SIP call is started by a real-server
(inside-to-outside direction), but only in the outside-to-inside
direction. This would be a general problem for all SIP servers acting
as Back2BackUserAgent.
This patch aims to solve problems 1) and 3) while keeping OPS mode
mandatory for SIP-UDP, so that 2) is not a problem anymore.
The basic mechanism implemented is to make packets, that do not match any
existent connection but come from real-servers, create new connections
instead of let them pass without any effect.
When such packets pass through ip_vs_out(), if their source ip address and
source port match a configured real-server, a new connection is
automatically created in the same way as it would have happened if the
packet had come from outside-to-inside direction. A new connection template
is created too if the virtual-service is persistent and there is no
matching connection template found. The new connection automatically
created, if the service had "-o" option, is an OPS connection that lasts
only the time to forward the packet, just like it happens on the
ingress side.
The main part of this mechanism is implemented inside a persistent-engine
specific callback (at the moment only SIP persistent engine exists) and
is triggered only for UDP packets, since connection oriented protocols, by
using different set of ports (typically ephemeral ports) to open new
outgoing connections, should not need this feature.
The following requisites are needed for automatic connection creation; if
any is missing the packet simply goes the same way as before.
a) virtual-service is not fwmark based (this is because fwmark services
do not store address and port of the virtual-service, required to
build the connection data).
b) virtual-service and real-servers must not have been configured with
omitted port (this is again to have all data to create the connection).
Signed-off-by: Marco Angaroni <marcoangaroni@gmail.com>
Acked-by: Julian Anastasov <ja@ssi.bg>
Signed-off-by: Simon Horman <horms@verge.net.au>
skb->sk could point to timewait or request socket which has no sk_classid.
Detected as "BUG: KASAN: slab-out-of-bounds in cls_cgroup_classify".
Signed-off-by: Konstantin Khlebnikov <khlebnikov@yandex-team.ru>
Acked-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
nf_connlabel_set() takes the bit number that we would like to set.
nf_connlabels_get() however took the number of bits that we want to
support.
So e.g. nf_connlabels_get(32) support bits 0 to 31, but not 32.
This changes nf_connlabels_get() to take the highest bit that we want
to set.
Callers then don't have to cope with a potential integer wrap
when using nf_connlabels_get(bit + 1) anymore.
Current callers are fine, this change is only to make folloup
nft ct label set support simpler.
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Currently labels can only be set either by iptables connlabel
match or via ctnetlink.
Before adding nftables set support, clean up the clabel core and move
helpers that nft will not need after all to the xtables module.
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
This patch updates the IP tunnel core function iptunnel_handle_offloads so
that we return an int and do not free the skb inside the function. This
actually allows us to clean up several paths in several tunnels so that we
can free the skb at one point in the path without having to have a
secondary path if we are supporting tunnel offloads.
In addition it should resolve some double-free issues I have found in the
tunnels paths as I believe it is possible for us to end up triggering such
an event in the case of fou or gue.
Signed-off-by: Alexander Duyck <aduyck@mirantis.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Due to the fact that the udp socket is destructed asynchronously in a
work queue, we have some nondeterministic behavior during shutdown of
vxlan tunnels and creating new ones. Fix this by keeping the destruction
process synchronous in regards to the user space process so IFF_UP can
be reliably set.
udp_tunnel_sock_release destroys vs->sock->sk if reference counter
indicates so. We expect to have the same lifetime of vxlan_sock and
vxlan_sock->sock->sk even in fast paths with only rcu locks held. So
only destruct the whole socket after we can be sure it cannot be found
by searching vxlan_net->sock_list.
Cc: Eric Dumazet <eric.dumazet@gmail.com>
Cc: Jiri Benc <jbenc@redhat.com>
Cc: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com>
Signed-off-by: Hannes Frederic Sowa <hannes@stressinduktion.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
For some main variables in sctp.ko, we couldn't export it to other modules,
so we have to define some api to access them.
It will include sctp transport and endpoint's traversal.
There are some transport traversal functions for sctp_diag, we can also
use it for sctp_proc. cause they have the similar situation to traversal
transport.
v2->v3:
- rhashtable_walk_init need the parameter gfp, because of recent upstrem
update
Signed-off-by: Xin Long <lucien.xin@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
sctp_diag will dump some important details of sctp's assoc or ep, we use
sctp_info to describe them, sctp_get_sctp_info to get them, and export
it to sctp_diag.ko.
v2->v3:
- we will not use list_for_each_safe in sctp_get_sctp_info, cause
all the callers of it will use lock_sock.
- fix the holes in struct sctp_info with __reserved* field.
because sctp_diag is a new feature, and sctp_info is just for now,
it may be changed in the future.
Signed-off-by: Xin Long <lucien.xin@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
SCTP already serializes access to rcvbuf through its sock lock:
sctp_recvmsg takes it right in the start and release at the end, while
rx path will also take the lock before doing any socket processing. On
sctp_rcv() it will check if there is an user using the socket and, if
there is, it will queue incoming packets to the backlog. The backlog
processing will do the same. Even timers will do such check and
re-schedule if an user is using the socket.
Simplifying this will allow us to remove sctp_skb_list_tail and get ride
of some expensive lockings. The lists that it is used on are also
mangled with functions like __skb_queue_tail and __skb_unlink in the
same context, like on sctp_ulpq_tail_event() and sctp_clear_pd().
sctp_close() will also purge those while using only the sock lock.
Therefore the lockings performed by sctp_skb_list_tail() are not
necessary. This patch removes this function and replaces its calls with
just skb_queue_splice_tail_init() instead.
The biggest gain is at sctp_ulpq_tail_event(), because the events always
contain a list, even if it's queueing a single skb and this was
triggering expensive calls to spin_lock_irqsave/_irqrestore for every
data chunk received.
As SCTP will deliver each data chunk on a corresponding recvmsg, the
more effective the change will be.
Before this patch, with chunks with 30 bytes:
netperf -t SCTP_STREAM -H 192.168.1.2 -cC -l 60 -- -m 30 -S 400000
400000 -s 400000 400000
on a 10Gbit link with 1500 MTU:
SCTP STREAM TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 192.168.1.1 () port 0 AF_INET
Recv Send Send Utilization Service Demand
Socket Socket Message Elapsed Send Recv Send Recv
Size Size Size Time Throughput local remote local remote
bytes bytes bytes secs. 10^6bits/s % S % S us/KB us/KB
425984 425984 30 60.00 137.45 7.34 7.36 52.504 52.608
With it:
SCTP STREAM TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 192.168.1.1 () port 0 AF_INET
Recv Send Send Utilization Service Demand
Socket Socket Message Elapsed Send Recv Send Recv
Size Size Size Time Throughput local remote local remote
bytes bytes bytes secs. 10^6bits/s % S % S us/KB us/KB
425984 425984 30 60.00 179.10 7.97 6.70 43.740 36.788
Signed-off-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When removing sk_refcnt manipulation on synflood, I missed that
using skb_set_owner_w() was racy, if sk->sk_wmem_alloc had already
transitioned to 0.
We should hold sk_refcnt instead, but this is a big deal under attack.
(Doing so increase performance from 3.2 Mpps to 3.8 Mpps only)
In this patch, I chose to not attach a socket to syncookies skb.
Performance is now 5 Mpps instead of 3.2 Mpps.
Following patch will remove last known false sharing in
tcp_rcv_state_process()
Fixes: 3b24d854cb ("tcp/dccp: do not touch listener sk_refcnt under synflood")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
With the SO_REUSEPORT socket option, it is possible to create sockets
in the AF_INET and AF_INET6 domains which are bound to the same IPv4 address.
This is only possible with SO_REUSEPORT and when not using IPV6_V6ONLY on
the AF_INET6 sockets.
Prior to the commits referenced below, an incoming IPv4 packet would
always be routed to a socket of type AF_INET when this mixed-mode was used.
After those changes, the same packet would be routed to the most recently
bound socket (if this happened to be an AF_INET6 socket, it would
have an IPv4 mapped IPv6 address).
The change in behavior occurred because the recent SO_REUSEPORT optimizations
short-circuit the socket scoring logic as soon as they find a match. They
did not take into account the scoring logic that favors AF_INET sockets
over AF_INET6 sockets in the event of a tie.
To fix this problem, this patch changes the insertion order of AF_INET
and AF_INET6 addresses in the TCP and UDP socket lists when the sockets
have SO_REUSEPORT set. AF_INET sockets will be inserted at the head of the
list and AF_INET6 sockets with SO_REUSEPORT set will always be inserted at
the tail of the list. This will force AF_INET sockets to always be
considered first.
Fixes: e32ea7e747 ("soreuseport: fast reuseport UDP socket selection")
Fixes: 125e80b88687 ("soreuseport: fast reuseport TCP socket selection")
Reported-by: Maciej Żenczykowski <maze@google.com>
Signed-off-by: Craig Gallek <kraig@google.com>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch adds a release_cb for UDPv6. It does a route lookup
and updates sk->sk_dst_cache if it is needed. It picks up the
left-over job from ip6_sk_update_pmtu() if the sk was owned
by user during the pmtu update.
It takes a rcu_read_lock to protect the __sk_dst_get() operations
because another thread may do ip6_dst_store() without taking the
sk lock (e.g. sendmsg).
Fixes: 45e4fd2668 ("ipv6: Only create RTF_CACHE routes after encountering pmtu exception")
Signed-off-by: Martin KaFai Lau <kafai@fb.com>
Reported-by: Wei Wang <weiwan@google.com>
Cc: Cong Wang <xiyou.wangcong@gmail.com>
Cc: Eric Dumazet <edumazet@google.com>
Cc: Wei Wang <weiwan@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
There is a case in connected UDP socket such that
getsockopt(IPV6_MTU) will return a stale MTU value. The reproducible
sequence could be the following:
1. Create a connected UDP socket
2. Send some datagrams out
3. Receive a ICMPV6_PKT_TOOBIG
4. No new outgoing datagrams to trigger the sk_dst_check()
logic to update the sk->sk_dst_cache.
5. getsockopt(IPV6_MTU) returns the mtu from the invalid
sk->sk_dst_cache instead of the newly created RTF_CACHE clone.
This patch updates the sk->sk_dst_cache for a connected datagram sk
during pmtu-update code path.
Note that the sk->sk_v6_daddr is used to do the route lookup
instead of skb->data (i.e. iph). It is because a UDP socket can become
connected after sending out some datagrams in un-connected state. or
It can be connected multiple times to different destinations. Hence,
iph may not be related to where sk is currently connected to.
It is done under '!sock_owned_by_user(sk)' condition because
the user may make another ip6_datagram_connect() (i.e changing
the sk->sk_v6_daddr) while dst lookup is happening in the pmtu-update
code path.
For the sock_owned_by_user(sk) == true case, the next patch will
introduce a release_cb() which will update the sk->sk_dst_cache.
Test:
Server (Connected UDP Socket):
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Route Details:
[root@arch-fb-vm1 ~]# ip -6 r show | egrep '2fac'
2fac::/64 dev eth0 proto kernel metric 256 pref medium
2fac:face::/64 via 2fac::face dev eth0 metric 1024 pref medium
A simple python code to create a connected UDP socket:
import socket
import errno
HOST = '2fac::1'
PORT = 8080
s = socket.socket(socket.AF_INET6, socket.SOCK_DGRAM)
s.bind((HOST, PORT))
s.connect(('2fac:face::face', 53))
print("connected")
while True:
try:
data = s.recv(1024)
except socket.error as se:
if se.errno == errno.EMSGSIZE:
pmtu = s.getsockopt(41, 24)
print("PMTU:%d" % pmtu)
break
s.close()
Python program output after getting a ICMPV6_PKT_TOOBIG:
[root@arch-fb-vm1 ~]# python2 ~/devshare/kernel/tasks/fib6/udp-connect-53-8080.py
connected
PMTU:1300
Cache routes after recieving TOOBIG:
[root@arch-fb-vm1 ~]# ip -6 r show table cache
2fac:face::face via 2fac::face dev eth0 metric 0
cache expires 463sec mtu 1300 pref medium
Client (Send the ICMPV6_PKT_TOOBIG):
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
scapy is used to generate the TOOBIG message. Here is the scapy script I have
used:
>>> p=Ether(src='da:75:4d:36:ac:32', dst='52:54:00:12:34:66', type=0x86dd)/IPv6(src='2fac::face', dst='2fac::1')/ICMPv6PacketTooBig(mtu=1300)/IPv6(src='2fac::
1',dst='2fac:face::face', nh='UDP')/UDP(sport=8080,dport=53)
>>> sendp(p, iface='qemubr0')
Fixes: 45e4fd2668 ("ipv6: Only create RTF_CACHE routes after encountering pmtu exception")
Signed-off-by: Martin KaFai Lau <kafai@fb.com>
Reported-by: Wei Wang <weiwan@google.com>
Cc: Cong Wang <xiyou.wangcong@gmail.com>
Cc: Eric Dumazet <edumazet@google.com>
Cc: Wei Wang <weiwan@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
User needs to monitor shared buffer occupancy. For that, he issues a
snapshot command in order to instruct hardware to catch current and
maximal occupancy values, and clear command in order to clear the
historical maximal values.
Also port-pool and tc-pool-bind command response messages are extended to
carry occupancy values.
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Reviewed-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Define userspace API and drivers API for configuration of shared
buffers. Four basic objects are defined:
shared buffer - attributes are size, number of pools and TCs
pool - chunk of sharedbuffer definition, it has some size and either
static or dynamic threshold
port pool threshold - to set per-port threshold for each pool
port tc threshold bind - to bind port and TC to specified pool
with threshold.
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Reviewed-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The structure can be packed denser by doing minor rearrangement
of existing elements.
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Currently processing of multiple chunks in a single SCTP packet leads to
multiple calls to sk_data_ready, causing multiple wake up signals which
are costy and doesn't make it wake up any faster.
With this patch it will note that the wake up is pending and will do it
before leaving the state machine interpreter, latest place possible to
do it realiably and cleanly.
Note that sk_data_ready events are not dependent on asocs, unlike waking
up writers.
v2: series re-checked
v3: use local vars to cleanup the code, suggested by Jakub Sitnicki
Signed-off-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
It wastes space and gets worse as we add new flags, so convert bit-wide
flags to a bitfield.
Currently it already saves 4 bytes in sctp_sock, which are left as holes
in it for now. The whole struct needs packing, which should be done in
another patch.
Note that do_auto_asconf cannot be merged, as explained in the comment
before it.
Signed-off-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com>
Acked-by: Neil Horman <nhorman@tuxdriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The phys in phys_port_mask suggests this mask is about PHYs. In fact,
it means physical ports. Rename to enabled_port_mask, indicating
external enabled ports of the switch, which is hopefully less
confusing.
Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Tested-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Acked-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Johannes Berg says:
====================
To synchronize with Kalle, here's just a big change that affects
all drivers - removing the duplicated enum ieee80211_band and
replacing it by enum nl80211_band. On top of that, just a small
documentation update.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
In case of link-layer specific handling for 802.15.4 we need to cast to
802.15.4 sepcific structures. Simple add this header when include the
6lowpan header.
Signed-off-by: Alexander Aring <aar@pengutronix.de>
Reviewed-by: Stefan Schmidt<stefan@osg.samsung.com>
Acked-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
This patch moves the 802.15.4 link layer specific structures to generic
6lowpan. This is necessary for special 802.15.4 6lowpan handling in
6lowpan generic layer.
Reviewed-by: Stefan Schmidt <stefan@osg.samsung.com>
Signed-off-by: Alexander Aring <aar@pengutronix.de>
Acked-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
This patch changes the naming for interface private data for lowpan
intefaces. The current private data scheme is:
-------------------------------------------------
| 6LoWPAN Generic | LinkLayer 6LoWPAN |
-------------------------------------------------
the current naming schemes are:
- 6LoWPAN Generic:
- lowpan_priv
- LinkLayer 6LoWPAN:
- BTLE
- lowpan_dev
- 802.15.4:
- lowpan_dev_info
the new naming scheme with this patch will be:
- 6LoWPAN Generic:
- lowpan_dev
- LinkLayer 6LoWPAN:
- BTLE
- lowpan_btle_dev
- 802.15.4:
- lowpan_802154_dev
Signed-off-by: Alexander Aring <aar@pengutronix.de>
Reviewed-by: Stefan Schmidt<stefan@osg.samsung.com>
Acked-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
Pablo Neira Ayuso says:
====================
Netfilter updates for net-next
The following patchset contains the first batch of Netfilter updates for
your net-next tree.
1) Define pr_fmt() in nf_conntrack, from Weongyo Jeong.
2) Define and register netfilter's afinfo for the bridge family,
this comes in preparation for native nfqueue's bridge for nft,
from Stephane Bryant.
3) Add new attributes to store layer 2 and VLAN headers to nfqueue,
also from Stephane Bryant.
4) Parse new NFQA_VLAN and NFQA_L2HDR nfqueue netlink attributes
coming from userspace, from Stephane Bryant.
5) Use net->ipv6.devconf_all->hop_limit instead of hardcoded hop_limit
in IPv6 SYNPROXY, from Liping Zhang.
6) Remove unnecessary check for dst == NULL in nf_reject_ipv6,
from Haishuang Yan.
7) Deinline ctnetlink event report functions, from Florian Westphal.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Not performance critical, it is only invoked when an expectation is
added/destroyed.
While at it, kill unused nf_ct_expect_event() wrapper.
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Way too large; move it to nf_conntrack_ecache.c.
Reduces total object size by 1216 byte on my machine.
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
This enum is already perfectly aliased to enum nl80211_band, and
the only reason for it is that we get IEEE80211_NUM_BANDS out of
it. There's no really good reason to not declare the number of
bands in nl80211 though, so do that and remove the cfg80211 one.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
The roaming cases for the Connect command were not fully covered and
neither Connect nor Associate command uses of the prev_bssid parameter
were very clear. Add details to describe how the prev_bssid argument is
supposed to be used and when the driver should use association or
reassociation.
Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Vivek reported a kernel exception deleting a VRF with an active
connection through it. The root cause is that the socket has a cached
reference to a dst that is destroyed. Converting the dst_destroy to
dst_release and letting proper reference counting kick in does not
work as the dst has a reference to the device which needs to be released
as well.
I talked to Hannes about this at netdev and he pointed out the ipv4 and
ipv6 dst handling has dst_ifdown for just this scenario. Rather than
continuing with the reinvented dst wheel in VRF just remove it and
leverage the ipv4 and ipv6 versions.
Fixes: 193125dbd8 ("net: Introduce VRF device driver")
Fixes: 35402e3136 ("net: Add IPv6 support to VRF device")
Signed-off-by: David Ahern <dsa@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In the rxrpc_connection and rxrpc_call structs, there's one field to hold
the abort code, no matter whether that value was generated locally to be
sent or was received from the peer via an abort packet.
Split the abort code fields in two for cleanliness sake and add an error
field to hold the Linux error number to the rxrpc_call struct too
(sometimes this is generated in a context where we can't return it to
userspace directly).
Furthermore, add a skb mark to indicate a packet that caused a local abort
to be generated so that recvmsg() can pick up the correct abort code. A
future addition will need to be to indicate to userspace the difference
between aborts via a control message.
Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Move some miscellaneous bits out into their own file to make it easier to
split the call handling.
Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Multipath route lookups should consider knowledge about next hops and not
select a hop that is known to be failed.
Example:
[h2] [h3] 15.0.0.5
| |
3| 3|
[SP1] [SP2]--+
1 2 1 2
| | /-------------+ |
| \ / |
| X |
| / \ |
| / \---------------\ |
1 2 1 2
12.0.0.2 [TOR1] 3-----------------3 [TOR2] 12.0.0.3
4 4
\ /
\ /
\ /
-------| |-----/
1 2
[TOR3]
3|
|
[h1] 12.0.0.1
host h1 with IP 12.0.0.1 has 2 paths to host h3 at 15.0.0.5:
root@h1:~# ip ro ls
...
12.0.0.0/24 dev swp1 proto kernel scope link src 12.0.0.1
15.0.0.0/16
nexthop via 12.0.0.2 dev swp1 weight 1
nexthop via 12.0.0.3 dev swp1 weight 1
...
If the link between tor3 and tor1 is down and the link between tor1
and tor2 then tor1 is effectively cut-off from h1. Yet the route lookups
in h1 are alternating between the 2 routes: ping 15.0.0.5 gets one and
ssh 15.0.0.5 gets the other. Connections that attempt to use the
12.0.0.2 nexthop fail since that neighbor is not reachable:
root@h1:~# ip neigh show
...
12.0.0.3 dev swp1 lladdr 00:02:00:00:00:1b REACHABLE
12.0.0.2 dev swp1 FAILED
...
The failed path can be avoided by considering known neighbor information
when selecting next hops. If the neighbor lookup fails we have no
knowledge about the nexthop, so give it a shot. If there is an entry
then only select the nexthop if the state is sane. This is similar to
what fib_detect_death does.
To maintain backward compatibility use of the neighbor information is
based on a new sysctl, fib_multipath_use_neigh.
Signed-off-by: David Ahern <dsa@cumulusnetworks.com>
Reviewed-by: Julian Anastasov <ja@ssi.bg>
Signed-off-by: David S. Miller <davem@davemloft.net>
Currently on high rate SCTP streams the heartbeat timer refresh can
consume quite a lot of resources as timer updates are costly and it
contains a random factor, which a) is also costly and b) invalidates
mod_timer() optimization for not editing a timer to the same value.
It may even cause the timer to be slightly advanced, for no good reason.
As suggested by David Laight this patch now removes this timer update
from hot path by leaving the timer on and re-evaluating upon its
expiration if the heartbeat is still needed or not, similarly to what is
done for TCP. If it's not needed anymore the timer is re-scheduled to
the new timeout, considering the time already elapsed.
For this, we now record the last tx timestamp per transport, updated in
the same spots as hb timer was restarted on tx. Also split up
sctp_transport_reset_timers into sctp_transport_reset_t3_rtx and
sctp_transport_reset_hb_timer, so we can re-arm T3 without re-arming the
heartbeat one.
On loopback with MTU of 65535 and data chunks with 1636, so that we
have a considerable amount of chunks without stressing system calls,
netperf -t SCTP_STREAM -l 30, perf looked like this before:
Samples: 103K of event 'cpu-clock', Event count (approx.): 25833000000
Overhead Command Shared Object Symbol
+ 6,15% netperf [kernel.vmlinux] [k] copy_user_enhanced_fast_string
- 5,43% netperf [kernel.vmlinux] [k] _raw_write_unlock_irqrestore
- _raw_write_unlock_irqrestore
- 96,54% _raw_spin_unlock_irqrestore
- 36,14% mod_timer
+ 97,24% sctp_transport_reset_timers
+ 2,76% sctp_do_sm
+ 33,65% __wake_up_sync_key
+ 28,77% sctp_ulpq_tail_event
+ 1,40% del_timer
- 1,84% mod_timer
+ 99,03% sctp_transport_reset_timers
+ 0,97% sctp_do_sm
+ 1,50% sctp_ulpq_tail_event
And after this patch, now with netperf -l 60:
Samples: 230K of event 'cpu-clock', Event count (approx.): 57707250000
Overhead Command Shared Object Symbol
+ 5,65% netperf [kernel.vmlinux] [k] memcpy_erms
+ 5,59% netperf [kernel.vmlinux] [k] copy_user_enhanced_fast_string
- 5,05% netperf [kernel.vmlinux] [k] _raw_spin_unlock_irqrestore
- _raw_spin_unlock_irqrestore
+ 49,89% __wake_up_sync_key
+ 45,68% sctp_ulpq_tail_event
- 2,85% mod_timer
+ 76,51% sctp_transport_reset_t3_rtx
+ 23,49% sctp_do_sm
+ 1,55% del_timer
+ 2,50% netperf [sctp] [k] sctp_datamsg_from_user
+ 2,26% netperf [sctp] [k] sctp_sendmsg
Throughput-wise, from 6800mbps without the patch to 7050mbps with it,
~3.7%.
Signed-off-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The switchdev design implies that a software error should not happen in
the commit phase since it must have been previously reported in the
prepare phase. If an hardware error occurs during the commit phase,
there is nothing switchdev can do about it.
The DSA layer separates port_vlan_prepare and port_vlan_add for
simplicity and convenience. If an hardware error occurs during the
commit phase, there is no need to report it outside the driver itself.
Make the DSA port_vlan_add routine return void for explicitness.
Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The switchdev design implies that a software error should not happen in
the commit phase since it must have been previously reported in the
prepare phase. If an hardware error occurs during the commit phase,
there is nothing switchdev can do about it.
The DSA layer separates port_fdb_prepare and port_fdb_add for simplicity
and convenience. If an hardware error occurs during the commit phase,
there is no need to report it outside the DSA driver itself.
Make the DSA port_fdb_add routine return void for explicitness.
Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The DSA layer doesn't care about the return code of the port_stp_update
routine, so make it void in the layer and the DSA drivers.
Replace the useless dsa_slave_stp_update function with a
dsa_slave_stp_state function used to reply to the switchdev
SWITCHDEV_ATTR_ID_PORT_STP_STATE attribute.
In the meantime, rename port_stp_update to port_stp_state_set to
explicit the state change.
Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Johannes Berg says:
====================
For the 4.7 cycle, we have a number of changes:
* Bob's mesh mode rhashtable conversion, this includes
the rhashtable API change for allocation flags
* BSSID scan, connect() command reassoc support (Jouni)
* fast (optimised data only) and support for RSS in mac80211 (myself)
* various smaller changes
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
