Spotted by Alexander Beregalov
Signed-off-by: NEric Dumazet <dada1@cosmosbay.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Signed-off-by: NHarvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Corey Minyard found a race added in commit 271b72c7
(udp: RCU handling for Unicast packets.)

 "If the socket is moved from one list to another list in-between the
 time the hash is calculated and the next field is accessed, and the
 socket has moved to the end of the new list, the traversal will not
 complete properly on the list it should have, since the socket will
 be on the end of the new list and there's not a way to tell it's on a
 new list and restart the list traversal.  I think that this can be
 solved by pre-fetching the "next" field (with proper barriers) before
 checking the hash."

This patch corrects this problem, introducing a new
sk_for_each_rcu_safenext() macro.
Signed-off-by: NEric Dumazet <dada1@cosmosbay.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Signed-off-by: NEric Dumazet <dada1@cosmosbay.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This patch mimics commit 57413ebc
(tcp: calculate tcp_mem based on low memory instead of all memory)

The udp_mem array which contains limits on the total amount of memory
used by UDP sockets is calculated based on nr_all_pages.  On a 32 bits
x86 system, we should base this on the number of lowmem pages.
Signed-off-by: NEric Dumazet <dada1@cosmosbay.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Goals are :

1) Optimizing handling of incoming Unicast UDP frames, so that no memory
 writes should happen in the fast path.

 Note: Multicasts and broadcasts still will need to take a lock,
 because doing a full lockless lookup in this case is difficult.

2) No expensive operations in the socket bind/unhash phases :
  - No expensive synchronize_rcu() calls.

  - No added rcu_head in socket structure, increasing memory needs,
  but more important, forcing us to use call_rcu() calls,
  that have the bad property of making sockets structure cold.
  (rcu grace period between socket freeing and its potential reuse
   make this socket being cold in CPU cache).
  David did a previous patch using call_rcu() and noticed a 20%
  impact on TCP connection rates.
  Quoting Cristopher Lameter :
   "Right. That results in cacheline cooldown. You'd want to recycle
    the object as they are cache hot on a per cpu basis. That is screwed
    up by the delayed regular rcu processing. We have seen multiple
    regressions due to cacheline cooldown.
    The only choice in cacheline hot sensitive areas is to deal with the
    complexity that comes with SLAB_DESTROY_BY_RCU or give up on RCU."

  - Because udp sockets are allocated from dedicated kmem_cache,
  use of SLAB_DESTROY_BY_RCU can help here.

Theory of operation :
---------------------

As the lookup is lockfree (using rcu_read_lock()/rcu_read_unlock()),
special attention must be taken by readers and writers.

Use of SLAB_DESTROY_BY_RCU is tricky too, because a socket can be freed,
reused, inserted in a different chain or in worst case in the same chain
while readers could do lookups in the same time.

In order to avoid loops, a reader must check each socket found in a chain
really belongs to the chain the reader was traversing. If it finds a
mismatch, lookup must start again at the begining. This *restart* loop
is the reason we had to use rdlock for the multicast case, because
we dont want to send same message several times to the same socket.

We use RCU only for fast path.
Thus, /proc/net/udp still takes spinlocks.
Signed-off-by: NEric Dumazet <dada1@cosmosbay.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

UDP sockets are hashed in a 128 slots hash table.

This hash table is protected by *one* rwlock.

This rwlock is readlocked each time an incoming UDP message is handled.

This rwlock is writelocked each time a socket must be inserted in
hash table (bind time), or deleted from this table (close time)

This is not scalable on SMP machines :

1) Even in read mode, lock() and unlock() are atomic operations and
 must dirty a contended cache line, shared by all cpus.

2) A writer might be starved if many readers are 'in flight'. This can
 happen on a machine with some NIC receiving many UDP messages. User
 process can be delayed a long time at socket creation/dismantle time.

This patch prepares RCU migration, by introducing 'struct udp_table
and struct udp_hslot', and using one spinlock per chain, to reduce
contention on central rwlock.

Introducing one spinlock per chain reduces latencies, for port
randomization on heavily loaded UDP servers. This also speedup
bindings to specific ports.

udp_lib_unhash() was uninlined, becoming to big.

Some cleanups were done to ease review of following patch
(RCUification of UDP Unicast lookups)
Signed-off-by: NEric Dumazet <dada1@cosmosbay.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Signed-off-by: NHarvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

call_rcu() will unconditionally rewrite RCU head anyway.
Applies to 
	struct neigh_parms
	struct neigh_table
	struct net
	struct cipso_v4_doi
	struct in_ifaddr
	struct in_device
	rt->u.dst
Signed-off-by: NAlexey Dobriyan <adobriyan@gmail.com>
Acked-by: NPaul E. McKenney <paulmck@linux.vnet.ibm.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

ifdef out
* struct sk_buff::sp		(pointer)
* struct dst_entry::xfrm	(pointer)
* struct sock::sk_policy	(2 pointers)
Signed-off-by: NAlexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This is a patch to provide on demand route cache rebuilding.  Currently, our
route cache is rebulid periodically regardless of need.  This introduced
unneeded periodic latency.  This patch offers a better approach.  Using code
provided by Eric Dumazet, we compute the standard deviation of the average hash
bucket chain length while running rt_check_expire.  Should any given chain
length grow to larger that average plus 4 standard deviations, we trigger an
emergency hash table rebuild for that net namespace.  This allows for the common
case in which chains are well behaved and do not grow unevenly to not incur any
latency at all, while those systems (which may be being maliciously attacked),
only rebuild when the attack is detected.  This patch take 2 other factors into
account:
1) chains with multiple entries that differ by attributes that do not affect the
hash value are only counted once, so as not to unduly bias system to rebuilding
if features like QOS are heavily used
2) if rebuilding crosses a certain threshold (which is adjustable via the added
sysctl in this patch), route caching is disabled entirely for that net
namespace, since constant rebuilding is less efficient that no caching at all

Tested successfully by me.
Signed-off-by: NNeil Horman <nhorman@tuxdriver.com>
Signed-off-by: NEric Dumazet <dada1@cosmosbay.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

David Miller noticed that commit
33ad798c '(tcp: options clean up')
did not move the req->cookie_ts check.
This essentially disabled commit 4dfc2817
'[Syncookies]: Add support for TCP options via timestamps.'.

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

This is not our bug! Sadly some devices cannot cope with the change
of TCP option ordering which was a result of the recent rewrite of
the option code (not that there was some particular reason steming
from the rewrite for the reordering) though any ordering of TCP
options is perfectly legal. Thus we restore the original ordering
to allow interoperability with/through such broken devices and add
some warning about this trap. Since the reordering just happened
without any particular reason, this change shouldn't cost us
anything.

There are already couple of known failure reports (within close
proximity of the last release), so the problem might be more
wide-spread than a single device. And other reports which may
be due to the same problem though the symptoms were less obvious.
Analysis of one of the case revealed (with very high probability)
that sack capability cannot be negotiated as the first option
(SYN never got a response).
Signed-off-by: NIlpo Järvinen <ilpo.jarvinen@helsinki.fi>
Reported-by: NAldo Maggi <sentiniate@tiscali.it>
Tested-by: NAldo Maggi <sentiniate@tiscali.it>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

While looking for the recent "sack issue" I also read all eff_sacks
usage that was played around by some relevant commit. I found
out that there's another thing that is asking for a fix (unrelated
to the "sack issue" though).

This feature has probably very little significance in practice.
Opposite direction timeout with bidirectional tcp comes to me as
the most likely scenario though there might be other cases as
well related to non-data segments we send (e.g., response to the
opposite direction segment). Also some ACK losses or option space
wasted for other purposes is necessary to prevent the earlier
SACK feedback getting to the sender.
Signed-off-by: NIlpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

(Supplements: ee999d8b)

NFPROTO_ARP actually has a different value from NF_ARP, so ensure all
callers use the new value so that packets _do_ get delivered to the
registered hooks.
Signed-off-by: NJan Engelhardt <jengelh@medozas.de>
Signed-off-by: NPatrick McHardy <kaber@trash.net>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Signed-off-by: NIlpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: NPatrick McHardy <kaber@trash.net>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Some code here depends on CONFIG_KMOD to not try to load
protocol modules or similar, replace by CONFIG_MODULES
where more than just request_module depends on CONFIG_KMOD
and and also use try_then_request_module in ebtables.
Signed-off-by: NJohannes Berg <johannes@sipsolutions.net>
Signed-off-by: NRusty Russell <rusty@rustcorp.com.au>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

rt_intern_hash() is doing an update of a RCU guarded hash chain
without using rcu_assign_pointer() or equivalent barrier.
Signed-off-by: NEric Dumazet <dada1@cosmosbay.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

name and nlen parameters passed to ->strategy hook are unused, remove
them.  In general ->strategy hook should know what it's doing, and don't
do something tricky for which, say, pointer to original userspace array
may be needed (name).
Signed-off-by: NAlexey Dobriyan <adobriyan@gmail.com>
Acked-by: David S. Miller <davem@davemloft.net> [ networking bits ]
Cc: Ralf Baechle <ralf@linux-mips.org>
Cc: David Howells <dhowells@redhat.com>
Cc: Matt Mackall <mpm@selenic.com>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

This patch removes the module dependency between ctnetlink and
nf_nat by means of an indirect call that is initialized when
nf_nat is loaded. Now, nf_conntrack_netlink only requires
nf_conntrack and nfnetlink.

This patch puts nfnetlink_parse_nat_setup_hook into the
nf_conntrack_core to avoid dependencies between ctnetlink,
nf_conntrack_ipv4 and nf_conntrack_ipv6.

This patch also introduces the function ctnetlink_change_nat
that is only invoked from the creation path. Actually, the
nat handling cannot be invoked from the update path since
this is not allowed. By introducing this function, we remove
the useless nat handling in the update path and we avoid
deadlock-prone code.

This patch also adds the required EAGAIN logic for nfnetlink.
Signed-off-by: NPablo Neira Ayuso <pablo@netfilter.org>
Signed-off-by: NPatrick McHardy <kaber@trash.net>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Nir Tzachar <nir.tzachar@gmail.com> reported a warning when sending
fragments over loopback with NAT:

[ 6658.338121] WARNING: at net/ipv4/netfilter/nf_nat_standalone.c:89 nf_nat_fn+0x33/0x155()

The reason is that defragmentation is skipped for already tracked connections.
This is wrong in combination with NAT and ip_conntrack actually had some ifdefs
to avoid this behaviour when NAT is compiled in.

The entire "optimization" may seem a bit silly, for now simply restoring the
lost #ifdef is the easiest solution until we can come up with something better.
Signed-off-by: NPatrick McHardy <kaber@trash.net>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Clean up the various different email addresses of mine listed in the code
to a single current and valid address. As Dave says his network merges
for 2.6.28 are now done this seems a good point to send them in where
they won't risk disrupting real changes.
Signed-off-by: NAlan Cox <alan@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Brown paper bag error of calling memset with sizeof(p) instead
of sizeof(*p).
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

- use typeful helpers for IFLA_GRE_LOCAL/IFLA_GRE_REMOTE
- replace magic value by FIELD_SIZEOF
- use MODULE_ALIAS_RTNL_LINK macro
Signed-off-by: NPatrick McHardy <kaber@trash.net>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The flags are dumped twice, the keys not at all.
Signed-off-by: NPatrick McHardy <kaber@trash.net>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This patch accomplishes three minor tasks: add a new tag type for local
labeling, rename the CIPSO_V4_MAP_STD define to CIPSO_V4_MAP_TRANS and
replace some of the CIPSO "magic numbers" with constants from the header
file.  The first change allows CIPSO to support full LSM labels/contexts,
not just MLS attributes.  The second change brings the mapping names inline
with what userspace is using, compatibility is preserved since we don't
actually change the value.  The last change is to aid readability and help
prevent mistakes.
Signed-off-by: NPaul Moore <paul.moore@hp.com>

Previous work enabled the use of address based NetLabel selectors, which while
highly useful, brought the potential for additional per-packet overhead when
used.  This patch attempts to solve that by applying NetLabel socket labels
when sockets are connect()'d.  This should alleviate the per-packet NetLabel
labeling for all connected sockets (yes, it even works for connected DGRAM
sockets).
Signed-off-by: NPaul Moore <paul.moore@hp.com>
Reviewed-by: NJames Morris <jmorris@namei.org>

This patch builds upon the new NetLabel address selector functionality by
providing the NetLabel KAPI and CIPSO engine support needed to enable the
new packet-based labeling.  The only new addition to the NetLabel KAPI at
this point is shown below:

 * int netlbl_skbuff_setattr(skb, family, secattr)

... and is designed to be called from a Netfilter hook after the packet's
IP header has been populated such as in the FORWARD or LOCAL_OUT hooks.

This patch also provides the necessary SELinux hooks to support this new
functionality.  Smack support is not currently included due to uncertainty
regarding the permissions needed to expand the Smack network access controls.
Signed-off-by: NPaul Moore <paul.moore@hp.com>
Reviewed-by: NJames Morris <jmorris@namei.org>

NetLabel has always had a list of backpointers in the CIPSO DOI definition
structure which pointed to the NetLabel LSM domain mapping structures which
referenced the CIPSO DOI struct.  The rationale for this was that when an
administrator removed a CIPSO DOI from the system all of the associated
NetLabel LSM domain mappings should be removed as well; a list of
backpointers made this a simple operation.

Unfortunately, while the backpointers did make the removal easier they were
a bit of a mess from an implementation point of view which was making
further development difficult.  Since the removal of a CIPSO DOI is a
realtively rare event it seems to make sense to remove this backpointer
list as the optimization was hurting us more then it was helping.  However,
we still need to be able to track when a CIPSO DOI definition is being used
so replace the backpointer list with a reference count.  In order to
preserve the current functionality of removing the associated LSM domain
mappings when a CIPSO DOI is removed we walk the LSM domain mapping table,
removing the relevant entries.
Signed-off-by: NPaul Moore <paul.moore@hp.com>
Reviewed-by: NJames Morris <jmorris@namei.org>

While looking at UDP port randomization, I noticed it
was litle bit pessimistic, not looking at type of sockets
(IPV6/IPV4) and not looking at bound addresses if any.

We should perform same tests than when binding to a
specific port.

This permits a cleanup of udp_lib_get_port()
Signed-off-by: NEric Dumazet <dada1@cosmosbay.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Maybe it's just me but I guess those md5 people made a mess
out of it by having *_md5_hash_* to use daddr, saddr order
instead of the one that is natural (and equal to what csum
functions use). For the segment were sending, the original
addresses are reversed so buff's saddr == skb's daddr and
vice-versa.

Maybe I can finally proceed with unification of some code
after fixing it first... :-)
Signed-off-by: NIlpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This patch makes the RX/TX byte counters for IPIP, GRE and SIT more
consistent.  Previously we included the external IP headers on the
way out but not when the packet is inbound.

The new scheme is to count payload only in both directions.  For
IPIP and SIT this simply means the exclusion of the external IP
header.  For GRE this means that we exclude the GRE header as
well.
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This patch adds support for Ethernet over GRE encapsulation.
This is exposed to user-space with a new link type of "gretap"
instead of "gre".  It will create an ARPHRD_ETHER device in
lieu of the usual ARPHRD_IPGRE.

Note that to preserver backwards compatibility all Transparent
Ethernet Bridging packets are passed to an ARPHRD_IPGRE tunnel
if its key matches and there is no ARPHRD_ETHER device whose
key matches more closely.
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This patch adds a netlink interface that will eventually displace
the existing ioctl interface.  It utilises the elegant rtnl_link_ops
mechanism.

This also means that user-space no longer needs to rely on the
tunnel interface being of type GRE to identify GRE tunnels.  The
identification can now occur using rtnl_link_ops.
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This patch moves the dev->mtu setting out of ipgre_tunnel_bind_dev.
This is in prepartion of using rtnl_link where we'll need to make
the MTU setting conditional on whether the user has supplied an
MTU.  This also requires the move of the ipgre_tunnel_bind_dev
call out of the dev->init function so that we can access the user
parameters later.

This patch also adds a check to prevent setting the MTU below
the minimum of 68.
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Now that we have dev->needed_headroom, we can use it instead of
having a bogus dev->hard_header_len.  This also allows us to
include dev->hard_header_len in the MTU computation so that when
we do have a meaningful hard_harder_len in future it is included
automatically in figuring out the MTU.

Incidentally, this fixes a bug where we ignored the needed_headroom
field of the underlying device in calculating our own hard_header_len.
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Commit cb7f6a7b ("IPVS: Move IPVS to
net/netfilter/ipvs") has left a stray file in the old location of ipvs.
Signed-off-by: NSven Wegener <sven.wegener@stealer.net>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

I noticed sysctl_local_port_range[] and its associated seqlock
sysctl_local_port_range_lock were on separate cache lines.
Moreover, sysctl_local_port_range[] was close to unrelated
variables, highly modified, leading to cache misses.

Moving these two variables in a structure can help data
locality and moving this structure to read_mostly section
helps sharing of this data among cpus.

Cleanup of extern declarations (moved in include file where
they belong), and use of inet_get_local_port_range()
accessor instead of direct access to ports values.
Signed-off-by: NEric Dumazet <dada1@cosmosbay.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Current UDP port allocation is suboptimal.
We select the shortest chain to chose a port (out of 512)
that will hash in this shortest chain.

First, it can lead to give not so ramdom ports and ease
give attackers more opportunities to break the system.

Second, it can consume a lot of CPU to scan all table
in order to find the shortest chain.

Third, in some pathological cases we can fail to find
a free port even if they are plenty of them.

This patch zap the search for a short chain and only
use one random seed. Problem of getting long chains
should be addressed in another way, since we can
obtain long chains with non random ports.

Based on a report and patch from Vitaly Mayatskikh
Signed-off-by: NEric Dumazet <dada1@cosmosbay.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

While looking for some common code I came across difference
in checksum calculation between tcp_v6_send_(reset|ack) I
couldn't explain. I checked both v4 and v6 and found out that
both seem to have the same "feature". I couldn't find anything
in rfc nor anywhere else which would state that md5 option
should be ignored like it was in case of reset so I came to
a conclusion that this is probably a genuine bug. I suspect
that addition of md5 just was fooled by the excessive
copy-paste code in those functions and the reset part was
never tested well enough to find out the problem.
Signed-off-by: NIlpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>