When we fast path datagram sends to avoid locking by putting
the inet_cork on the stack we use up lots of space that isn't
necessary.
This is because inet_cork contains a "struct flowi" which isn't
used in these code paths.
Split inet_cork to two parts, "inet_cork" and "inet_cork_full".
Only the latter of which has the "struct flowi" and is what is
stored in inet_sock.
Signed-off-by: David S. Miller <davem@davemloft.net>
Acked-by: Eric Dumazet <eric.dumazet@gmail.com>
We lack proper synchronization to manipulate inet->opt ip_options
Problem is ip_make_skb() calls ip_setup_cork() and
ip_setup_cork() possibly makes a copy of ipc->opt (struct ip_options),
without any protection against another thread manipulating inet->opt.
Another thread can change inet->opt pointer and free old one under us.
Use RCU to protect inet->opt (changed to inet->inet_opt).
Instead of handling atomic refcounts, just copy ip_options when
necessary, to avoid cache line dirtying.
We cant insert an rcu_head in struct ip_options since its included in
skb->cb[], so this patch is large because I had to introduce a new
ip_options_rcu structure.
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Cc: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
In order to allow simultaneous calls to ip_append_data on the same
socket, it must not modify any shared state in sk or inet (other
than those that are designed to allow that such as atomic counters).
This patch abstracts out write references to sk and inet_sk in
ip_append_data and its friends so that we may use the underlying
code in parallel.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Acked-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
TCP is going to record metrics for the connection,
so pre-COW the route metrics at route cache entry
creation time.
This avoids several atomic operations that have to
occur if we COW the metrics after the entry reaches
global visibility.
Signed-off-by: David S. Miller <davem@davemloft.net>
Followup of commit b178bb3dfc (net: reorder struct sock fields)
Optimize INET input path a bit further, by :
1) moving sk_refcnt close to sk_lock.
This reduces number of dirtied cache lines by one on 64bit arches (and
64 bytes cache line size).
2) moving inet_daddr & inet_rcv_saddr at the beginning of sk
(same cache line than hash / family / bound_dev_if / nulls_node)
This reduces number of accessed cache lines in lookups by one, and dont
increase size of inet and timewait socks.
inet and tw sockets now share same place-holder for these fields.
Before patch :
offsetof(struct sock, sk_refcnt) = 0x10
offsetof(struct sock, sk_lock) = 0x40
offsetof(struct sock, sk_receive_queue) = 0x60
offsetof(struct inet_sock, inet_daddr) = 0x270
offsetof(struct inet_sock, inet_rcv_saddr) = 0x274
After patch :
offsetof(struct sock, sk_refcnt) = 0x44
offsetof(struct sock, sk_lock) = 0x48
offsetof(struct sock, sk_receive_queue) = 0x68
offsetof(struct inet_sock, inet_daddr) = 0x0
offsetof(struct inet_sock, inet_rcv_saddr) = 0x4
compute_score() (udp or tcp) now use a single cache line per ignored
item, instead of two.
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
in_dev->mc_list is protected by one rwlock (in_dev->mc_list_lock).
This can easily be converted to a RCU protection.
Writers hold RTNL, so mc_list_lock is removed, not replaced by a
spinlock.
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Cc: Cypher Wu <cypher.w@gmail.com>
Cc: Américo Wang <xiyou.wangcong@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
this patch is implementing IP_NODEFRAG option for IPv4 socket.
The reason is, there's no other way to send out the packet with user
customized header of the reassembly part.
Signed-off-by: Jiri Olsa <jolsa@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Idea from Eric Dumazet.
As for placement inside of struct sock, I tried to choose a place
that otherwise has a 32-bit hole on 64-bit systems.
Signed-off-by: David S. Miller <davem@davemloft.net>
Acked-by: Eric Dumazet <eric.dumazet@gmail.com>
const qualifier on sock argument is misleading, since we can modify rxhash.
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch implements receive flow steering (RFS). RFS steers
received packets for layer 3 and 4 processing to the CPU where
the application for the corresponding flow is running. RFS is an
extension of Receive Packet Steering (RPS).
The basic idea of RFS is that when an application calls recvmsg
(or sendmsg) the application's running CPU is stored in a hash
table that is indexed by the connection's rxhash which is stored in
the socket structure. The rxhash is passed in skb's received on
the connection from netif_receive_skb. For each received packet,
the associated rxhash is used to look up the CPU in the hash table,
if a valid CPU is set then the packet is steered to that CPU using
the RPS mechanisms.
The convolution of the simple approach is that it would potentially
allow OOO packets. If threads are thrashing around CPUs or multiple
threads are trying to read from the same sockets, a quickly changing
CPU value in the hash table could cause rampant OOO packets--
we consider this a non-starter.
To avoid OOO packets, this solution implements two types of hash
tables: rps_sock_flow_table and rps_dev_flow_table.
rps_sock_table is a global hash table. Each entry is just a CPU
number and it is populated in recvmsg and sendmsg as described above.
This table contains the "desired" CPUs for flows.
rps_dev_flow_table is specific to each device queue. Each entry
contains a CPU and a tail queue counter. The CPU is the "current"
CPU for a matching flow. The tail queue counter holds the value
of a tail queue counter for the associated CPU's backlog queue at
the time of last enqueue for a flow matching the entry.
Each backlog queue has a queue head counter which is incremented
on dequeue, and so a queue tail counter is computed as queue head
count + queue length. When a packet is enqueued on a backlog queue,
the current value of the queue tail counter is saved in the hash
entry of the rps_dev_flow_table.
And now the trick: when selecting the CPU for RPS (get_rps_cpu)
the rps_sock_flow table and the rps_dev_flow table for the RX queue
are consulted. When the desired CPU for the flow (found in the
rps_sock_flow table) does not match the current CPU (found in the
rps_dev_flow table), the current CPU is changed to the desired CPU
if one of the following is true:
- The current CPU is unset (equal to RPS_NO_CPU)
- Current CPU is offline
- The current CPU's queue head counter >= queue tail counter in the
rps_dev_flow table. This checks if the queue tail has advanced
beyond the last packet that was enqueued using this table entry.
This guarantees that all packets queued using this entry have been
dequeued, thus preserving in order delivery.
Making each queue have its own rps_dev_flow table has two advantages:
1) the tail queue counters will be written on each receive, so
keeping the table local to interrupting CPU s good for locality. 2)
this allows lockless access to the table-- the CPU number and queue
tail counter need to be accessed together under mutual exclusion
from netif_receive_skb, we assume that this is only called from
device napi_poll which is non-reentrant.
This patch implements RFS for TCP and connected UDP sockets.
It should be usable for other flow oriented protocols.
There are two configuration parameters for RFS. The
"rps_flow_entries" kernel init parameter sets the number of
entries in the rps_sock_flow_table, the per rxqueue sysfs entry
"rps_flow_cnt" contains the number of entries in the rps_dev_flow
table for the rxqueue. Both are rounded to power of two.
The obvious benefit of RFS (over just RPS) is that it achieves
CPU locality between the receive processing for a flow and the
applications processing; this can result in increased performance
(higher pps, lower latency).
The benefits of RFS are dependent on cache hierarchy, application
load, and other factors. On simple benchmarks, we don't necessarily
see improvement and sometimes see degradation. However, for more
complex benchmarks and for applications where cache pressure is
much higher this technique seems to perform very well.
Below are some benchmark results which show the potential benfit of
this patch. The netperf test has 500 instances of netperf TCP_RR
test with 1 byte req. and resp. The RPC test is an request/response
test similar in structure to netperf RR test ith 100 threads on
each host, but does more work in userspace that netperf.
e1000e on 8 core Intel
No RFS or RPS 104K tps at 30% CPU
No RFS (best RPS config): 290K tps at 63% CPU
RFS 303K tps at 61% CPU
RPC test tps CPU% 50/90/99% usec latency Latency StdDev
No RFS/RPS 103K 48% 757/900/3185 4472.35
RPS only: 174K 73% 415/993/2468 491.66
RFS 223K 73% 379/651/1382 315.61
Signed-off-by: Tom Herbert <therbert@google.com>
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch adds the kernel portions needed to implement
RFC 5082 Generalized TTL Security Mechanism (GTSM).
It is a lightweight security measure against forged
packets causing DoS attacks (for BGP).
This is already implemented the same way in BSD kernels.
For the necessary Quagga patch
http://www.gossamer-threads.com/lists/quagga/dev/17389
Description from Cisco
http://www.cisco.com/en/US/docs/ios/12_3t/12_3t7/feature/guide/gt_btsh.html
It does add one byte to each socket structure, but I did
a little rearrangement to reuse a hole (on 64 bit), but it
does grow the structure on 32 bit
This should be documented on ip(4) man page and the Glibc in.h
file also needs update. IPV6_MINHOPLIMIT should also be added
(although BSD doesn't support that).
Only TCP is supported, but could also be added to UDP, DCCP, SCTP
if desired.
Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In order to have better cache layouts of struct sock (separate zones
for rx/tx paths), we need this preliminary patch.
Goal is to transfert fields used at lookup time in the first
read-mostly cache line (inside struct sock_common) and move sk_refcnt
to a separate cache line (only written by rx path)
This patch adds inet_ prefix to daddr, rcv_saddr, dport, num, saddr,
sport and id fields. This allows a future patch to define these
fields as macros, like sk_refcnt, without name clashes.
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
After some discussion offline with Christoph Lameter and David Stevens
regarding multicast behaviour in Linux, I'm submitting a slightly
modified patch from the one Christoph submitted earlier.
This patch provides a new socket option IP_MULTICAST_ALL.
In this case, default behaviour is _unchanged_ from the current
Linux standard. The socket option is set by default to provide
original behaviour. Sockets wishing to receive data only from
multicast groups they join explicitly will need to clear this
socket option.
Signed-off-by: Nivedita Singhvi <niv@us.ibm.com>
Signed-off-by: Christoph Lameter<cl@linux.com>
Acked-by: David Stevens <dlstevens@us.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Current TCP code relies on the local port of the listening socket
being the same as the destination address of the incoming
connection. Port redirection used by many transparent proxying
techniques obviously breaks this, so we have to store the original
destination port address.
This patch extends struct inet_request_sock and stores the incoming
destination port value there. It also modifies the handshake code to
use that value as the source port when sending reply packets.
Signed-off-by: KOVACS Krisztian <hidden@sch.bme.hu>
Signed-off-by: David S. Miller <davem@davemloft.net>
The TCP stack sends out SYN+ACK/ACK/RST reply packets in response to
incoming packets. The non-local source address check on output bites
us again, as replies for transparently redirected traffic won't have a
chance to leave the node.
This patch selectively sets the FLOWI_FLAG_ANYSRC flag when doing the
route lookup for those replies. Transparent replies are enabled if the
listening socket has the transparent socket flag set.
Signed-off-by: KOVACS Krisztian <hidden@sch.bme.hu>
Signed-off-by: David S. Miller <davem@davemloft.net>
inet_iif() in inet_sock.h requires route.h. Since users of inet_iif()
usually require other route.h functionality anyway this patch moves
inet_iif() to route.h.
Signed-off-by: KOVACS Krisztian <hidden@sch.bme.hu>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch introduces the IP_TRANSPARENT socket option: enabling that
will make the IPv4 routing omit the non-local source address check on
output. Setting IP_TRANSPARENT requires NET_ADMIN capability.
Signed-off-by: KOVACS Krisztian <hidden@sch.bme.hu>
Signed-off-by: David S. Miller <davem@davemloft.net>
There are many possible ways to add this "salt", thus I made this
patch to be the last in the series to change it if required.
Currently I propose to use the struct net pointer itself as this
salt, but since this pointer is most often cache-line aligned, shift
this right to eliminate the bits, that are most often zeroed.
After this, simply add this mix to prepared hashfn-s.
For CONFIG_NET_NS=n case this salt is 0 and no changes in hashfn
appear.
Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Although this hash takes addresses into account, the ehash chains
can also be too long when, for instance, communications via lo occur.
So, prepare the inet_hashfn to take struct net into account.
Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Wei Yongjun noticed that we may call reqsk_free on request sock objects where
the opt fields may not be initialized, fix it by introducing inet_reqsk_alloc
where we initialize ->opt to NULL and set ->pktopts to NULL in
inet6_reqsk_alloc.
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
ip_options->is_data is assigned only and never checked. The structure is
not a part of kernel interface to the userspace. So, it is safe to remove
this field.
Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
(Anonymous) unions can help us to avoid ugly casts.
A common cast it the (struct rtable *)skb->dst one.
Defining an union like :
union {
struct dst_entry *dst;
struct rtable *rtable;
};
permits to use skb->rtable in place.
Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
If all of the entropy is in the local and foreign addresses,
but xor'ing together would cancel out that entropy, the
current hash performs poorly.
Suggested by Cosmin Ratiu:
Basically, the situation is as follows: There is a client
machine and a server machine. Both create 15000 virtual
interfaces, open up a socket for each pair of interfaces and
do SIP traffic. By profiling I noticed that there is a lot of
time spent walking the established hash chains with this
particular setup.
The addresses were distributed like this: client interfaces
were 198.18.0.1/16 with increments of 1 and server interfaces
were 198.18.128.1/16 with increments of 1. As I said, there
were 15000 interfaces. Source and destination ports were 5060
for each connection. So in this case, ports don't matter for
hashing purposes, and the bits from the address pairs used
cancel each other, meaning there are no differences in the
whole lot of pairs, so they all end up in the same hash chain.
Signed-off-by: David S. Miller <davem@davemloft.net>
UDP currently uses skb->dev->ifindex which may provide the wrong
information when the socket bound to a specific interface.
This patch makes inet_iif() accessible to UDP and makes UDP use it.
The scenario we are trying to fix is when a client is running on
the same system and the server and both client and server bind to
a non-loopback device.
Signed-off-by: Vlad Yasevich <vladislav.yasevich@hp.com>
Acked-by: David L Stevens <dlstevens@us.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The days are gone when this was not an issue, there are folks out
there with huge bot networks that can be used to attack the
established hash tables on remote systems.
So just like the routing cache and connection tracking
hash, use Jenkins hash with random secret input.
Signed-off-by: David S. Miller <davem@davemloft.net>
inet_lookup() annotated along with helper functions (__inet_lookup(),
__inet_lookup_established(), inet_lookup_established(),
inet_lookup_listener(), __inet_lookup_listener() and inet_ehashfn())
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
->faddr is net-endian; annotated as such, variables inferred to be net-endian
annotated.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
The value is_setbyuser from struct ip_options is never used and set
only one time (http://linux-net.osdl.org/index.php/TODO#IPV4).
This little patch removes it from the kernel source.
Signed-off-by: Louis Nyffenegger <louis.nyffenegger@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Changes to the core network stack to support the NetLabel subsystem. This
includes changes to the IPv4 option handling to support CIPSO labels.
Signed-off-by: Paul Moore <paul.moore@hp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
To help in reducing the number of include dependencies, several files were
touched as they were getting needed headers indirectly for stuff they use.
Thanks also to Alan Menegotto for pointing out that net/dccp/proto.c had
linux/dccp.h include twice.
Signed-off-by: Arnaldo Carvalho de Melo <acme@mandriva.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller