linux-stable.git/net/ipv4/route.c, branch linux-3.4.y

ipv4: disable bh while doing route gc

2014-12-01T10:02:44+00:00

Further tests revealed that after moving the garbage collector to a work
queue and protecting it with a spinlock may leave the system prone to
soft lockups if bottom half gets very busy.

It was reproced with a set of firewall rules that REJECTed packets. If
the NIC bottom half handler ends up running on the same CPU that is
running the garbage collector on a very large cache, the garbage
collector will not be able to do its job due to the amount of work
needed for handling the REJECTs and also won't reschedule.

The fix is to disable bottom half during the garbage collecting, as it
already was in the first place (most calls to it came from softirqs).

Signed-off-by: Marcelo Ricardo Leitner 
Acked-by: Hannes Frederic Sowa 
Acked-by: David S. Miller 
Signed-off-by: Ben Hutchings 
Signed-off-by: Zefan Li

ipv4: avoid parallel route cache gc executions

2014-12-01T10:02:44+00:00

When rt_intern_hash() has to deal with neighbour cache overflowing,
it triggers the route cache garbage collector in an attempt to free
some references on neighbour entries.

Such call cannot be done async but should also not run in parallel with
an already-running one, so that they don't collapse fighting over the
hash lock entries.

This patch thus blocks parallel executions with spinlocks:
- A call from worker and from rt_intern_hash() are not the same, and
cannot be merged, thus they will wait each other on rt_gc_lock.
- Calls to gc from rt_intern_hash() may happen in parallel but we must
wait for it to finish in order to try again. This dedup and
synchrinozation is then performed by the locking just before calling
__do_rt_garbage_collect().

Signed-off-by: Marcelo Ricardo Leitner 
Acked-by: Hannes Frederic Sowa 
Signed-off-by: Ben Hutchings 
Signed-off-by: Zefan Li

ipv4: move route garbage collector to work queue

2014-12-01T10:02:43+00:00

Currently the route garbage collector gets called by dst_alloc() if it
have more entries than the threshold. But it's an expensive call, that
don't really need to be done by then.

Another issue with current way is that it allows running the garbage
collector with the same start parameters on multiple CPUs at once, which
is not optimal. A system may even soft lockup if the cache is big enough
as the garbage collectors will be fighting over the hash lock entries.

This patch thus moves the garbage collector to run asynchronously on a
work queue, much similar to how rt_expire_check runs.

There is one condition left that allows multiple executions, which is
handled by the next patch.

Signed-off-by: Marcelo Ricardo Leitner 
Acked-by: Hannes Frederic Sowa 
Signed-off-by: Ben Hutchings 
Signed-off-by: Zefan Li

ip: make IP identifiers less predictable

2014-08-14T00:42:35+00:00

[ Upstream commit 04ca6973f7c1a0d8537f2d9906a0cf8e69886d75 ]

In "Counting Packets Sent Between Arbitrary Internet Hosts", Jeffrey and
Jedidiah describe ways exploiting linux IP identifier generation to
infer whether two machines are exchanging packets.

With commit 73f156a6e8c1 ("inetpeer: get rid of ip_id_count"), we
changed IP id generation, but this does not really prevent this
side-channel technique.

This patch adds a random amount of perturbation so that IP identifiers
for a given destination [1] are no longer monotonically increasing after
an idle period.

Note that prandom_u32_max(1) returns 0, so if generator is used at most
once per jiffy, this patch inserts no hole in the ID suite and do not
increase collision probability.

This is jiffies based, so in the worst case (HZ=1000), the id can
rollover after ~65 seconds of idle time, which should be fine.

We also change the hash used in __ip_select_ident() to not only hash
on daddr, but also saddr and protocol, so that ICMP probes can not be
used to infer information for other protocols.

For IPv6, adds saddr into the hash as well, but not nexthdr.

If I ping the patched target, we can see ID are now hard to predict.

21:57:11.008086 IP (...)
    A > target: ICMP echo request, seq 1, length 64
21:57:11.010752 IP (... id 2081 ...)
    target > A: ICMP echo reply, seq 1, length 64

21:57:12.013133 IP (...)
    A > target: ICMP echo request, seq 2, length 64
21:57:12.015737 IP (... id 3039 ...)
    target > A: ICMP echo reply, seq 2, length 64

21:57:13.016580 IP (...)
    A > target: ICMP echo request, seq 3, length 64
21:57:13.019251 IP (... id 3437 ...)
    target > A: ICMP echo reply, seq 3, length 64

[1] TCP sessions uses a per flow ID generator not changed by this patch.

Signed-off-by: Eric Dumazet 
Reported-by: Jeffrey Knockel 
Reported-by: Jedidiah R. Crandall 
Cc: Willy Tarreau 
Cc: Hannes Frederic Sowa 
Signed-off-by: David S. Miller 
Signed-off-by: Greg Kroah-Hartman

inetpeer: get rid of ip_id_count

2014-08-14T00:42:35+00:00

[ Upstream commit 73f156a6e8c1074ac6327e0abd1169e95eb66463 ]

Ideally, we would need to generate IP ID using a per destination IP
generator.

linux kernels used inet_peer cache for this purpose, but this had a huge
cost on servers disabling MTU discovery.

1) each inet_peer struct consumes 192 bytes

2) inetpeer cache uses a binary tree of inet_peer structs,
   with a nominal size of ~66000 elements under load.

3) lookups in this tree are hitting a lot of cache lines, as tree depth
   is about 20.

4) If server deals with many tcp flows, we have a high probability of
   not finding the inet_peer, allocating a fresh one, inserting it in
   the tree with same initial ip_id_count, (cf secure_ip_id())

5) We garbage collect inet_peer aggressively.

IP ID generation do not have to be 'perfect'

Goal is trying to avoid duplicates in a short period of time,
so that reassembly units have a chance to complete reassembly of
fragments belonging to one message before receiving other fragments
with a recycled ID.

We simply use an array of generators, and a Jenkin hash using the dst IP
as a key.

ipv6_select_ident() is put back into net/ipv6/ip6_output.c where it
belongs (it is only used from this file)

secure_ip_id() and secure_ipv6_id() no longer are needed.

Rename ip_select_ident_more() to ip_select_ident_segs() to avoid
unnecessary decrement/increment of the number of segments.

Signed-off-by: Eric Dumazet 
Signed-off-by: David S. Miller 
Signed-off-by: Greg Kroah-Hartman

ipv4: initialise the itag variable in __mkroute_input

2014-06-07T23:02:00+00:00

[ Upstream commit fbdc0ad095c0a299e9abf5d8ac8f58374951149a ]

the value of itag is a random value from stack, and may not be initiated by
fib_validate_source, which called fib_combine_itag if CONFIG_IP_ROUTE_CLASSID
is not set

This will make the cached dst uncertainty

Signed-off-by: Li RongQing 
Acked-by: Alexei Starovoitov 
Signed-off-by: David S. Miller 
Signed-off-by: Greg Kroah-Hartman

ipv4: fix ineffective source address selection

2013-11-04T12:23:40+00:00

[ Upstream commit 0a7e22609067ff524fc7bbd45c6951dd08561667 ]

When sending out multicast messages, the source address in inet->mc_addr is
ignored and rewritten by an autoselected one. This is caused by a typo in
commit 813b3b5db831 ("ipv4: Use caller's on-stack flowi as-is in output
route lookups").

Signed-off-by: Jiri Benc 
Acked-by: Eric Dumazet 
Signed-off-by: David S. Miller 
Signed-off-by: Greg Kroah-Hartman

Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net

2012-04-03T00:53:39+00:00

Pull networking fixes from David Miller:

 1) Provide device string properly for USB i2400m wimax devices, also
    don't OOPS when providing firmware string.  From Phil Sutter.

 2) Add support for sh_eth SH7734 chips, from Nobuhiro Iwamatsu.

 3) Add another device ID to USB zaurus driver, from Guan Xin.

 4) Loop index start in pool vector iterator is wrong causing MAC to not
    get configured in bnx2x driver, fix from Dmitry Kravkov.

 5) EQL driver assumes HZ=100, fix from Eric Dumazet.

 6) Now that skb_add_rx_frag() can specify the truesize increment
    separately, do so in f_phonet and cdc_phonet, also from Eric
    Dumazet.

 7) virtio_net accidently uses net_ratelimit() not only on the kernel
    warning but also the statistic bump, fix from Rick Jones.

 8) ip_route_input_mc() uses fixed init_net namespace, oops, use
    dev_net(dev) instead.  Fix from Benjamin LaHaise.

 9) dev_forward_skb() needs to clear the incoming interface index of the
    SKB so that it looks like a new incoming packet, also from Benjamin
    LaHaise.

10) iwlwifi mistakenly initializes a channel entry as 2GHZ instead of
    5GHZ, fix from Stanislav Yakovlev.

11) Missing kmalloc() return value checks in orinoco, from Santosh
    Nayak.

12) ath9k doesn't check for HT capabilities in the right way, it is
    checking ht_supported instead of the ATH9K_HW_CAP_HT flag.  Fix from
    Sujith Manoharan.

13) Fix x86 BPF JIT emission of 16-bit immediate field of AND
    instructions, from Feiran Zhuang.

14) Avoid infinite loop in GARP code when registering sysfs entries.
    From David Ward.

15) rose protocol uses memcpy instead of memcmp in a device address
    comparison, oops.  Fix from Daniel Borkmann.

16) Fix build of lpc_eth due to dev_hw_addr_rancom() interface being
    renamed to eth_hw_addr_random().  From Roland Stigge.

17) Make ipv6 RTM_GETROUTE interpret RTA_IIF attribute the same way
    that ipv4 does.  Fix from Shmulik Ladkani.

18) via-rhine has an inverted bit test, causing suspend/resume
    regressions.  Fix from Andreas Mohr.

19) RIONET assumes 4K page size, fix from Akinobu Mita.

20) Initialization of imask register in sky2 is buggy, because bits are
    "or'd" into an uninitialized local variable.  Fix from Lino
    Sanfilippo.

21) Fix FCOE checksum offload handling, from Yi Zou.

22) Fix VLAN processing regression in e1000, from Jiri Pirko.

* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net: (52 commits)
  sky2: dont overwrite settings for PHY Quick link
  tg3: Fix 5717 serdes powerdown problem
  net: usb: cdc_eem: fix mtu
  net: sh_eth: fix endian check for architecture independent
  usb/rtl8150 : Remove duplicated definitions
  rionet: fix page allocation order of rionet_active
  via-rhine: fix wait-bit inversion.
  ipv6: Fix RTM_GETROUTE's interpretation of RTA_IIF to be consistent with ipv4
  net: lpc_eth: Fix rename of dev_hw_addr_random
  net/netfilter/nfnetlink_acct.c: use linux/atomic.h
  rose_dev: fix memcpy-bug in rose_set_mac_address
  Fix non TBI PHY access; a bad merge undid bug fix in a previous commit.
  net/garp: avoid infinite loop if attribute already exists
  x86 bpf_jit: fix a bug in emitting the 16-bit immediate operand of AND
  bonding: emit event when bonding changes MAC
  mac80211: fix oper channel timestamp updation
  ath9k: Use HW HT capabilites properly
  MAINTAINERS: adding maintainer for ipw2x00
  net: orinoco: add error handling for failed kmalloc().
  net/wireless: ipw2x00: fix a typo in wiphy struct initilization
  ...

Remove all #inclusions of asm/system.h

2012-03-28T17:30:03+00:00

Remove all #inclusions of asm/system.h preparatory to splitting and killing
it.  Performed with the following command:

perl -p -i -e 's!^#\s*include\s*.*\n!!' `grep -Irl '^#\s*include\s*' *`

Signed-off-by: David Howells

net/ipv4: fix IPv4 multicast over network namespaces

2012-03-28T08:45:37+00:00

When using multicast over a local bridge feeding a number of LXC guests
using veth, the LXC guests are unable to get a response from other guests
when pinging 224.0.0.1.  Multicast packets did not appear to be getting
delivered to the network namespaces of the guest hosts, and further
inspection showed that the incoming route was pointing to the loopback
device of the host, not the guest.  This lead to the wrong network namespace
being picked up by sockets (like ICMP).  Fix this by using the correct
network namespace when creating the inbound route entry.

Signed-off-by: Benjamin LaHaise 
Signed-off-by: David S. Miller