Perform NAT last after parsing the packet. This makes no difference
currently, but is needed when dealing with registrations to make
sure we seen the unNATed addresses.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add support for per-method request/response handlers and perform SDP
parsing for INVITE/UPDATE requests and for all informational and
successful responses.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Use the URI parsing helper to get the numerical addresses and get rid of the
text based header translation.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Introduce a helper function to parse a SIP-URI in a header value, optionally
iterating through all headers of this kind.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Introduce new function for SIP header parsing that properly deals with
continuation lines and whitespace in headers and use it.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
The request URI is not a header and needs to be treated differently than
real SIP headers. Add a seperate function for parsing it and get rid of
the POS_REQ_URI/POS_REG_REQ_URI definitions.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
SDP and SIP headers are quite different, SIP can have continuation lines,
leading and trailing whitespace after the colon and is mostly case-insensitive
while SDP headers always begin on a new line and are followed by an equal
sign and the value, without any whitespace.
Introduce new SDP header parsing function and convert all users that used
the SIP header parsing function. This will allow to properly deal with the
special SIP cases in the SIP header parsing function later.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Replace sizeof/memcmp by strlen/strcmp. Use case-insensitive comparison
for SIP methods and the SIP/2.0 string, as specified in RFC 3261.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
The conntrack reference and ctinfo can be derived from the packet.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
After mangling the packet, the pointer to the data and the length of the data
portion may change and need to be adjusted.
Use double data pointers and a pointer to the length everywhere and add a
helper function to the NAT helper for performing the adjustments.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
"limit" marks the first character outside the bounds.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Introduce expectation classes and policies. An expectation class
is used to distinguish different types of expectations by the
same helper (for example audio/video/t.120). The expectation
policy is used to hold the maximum number of expectations and
the initial timeout for each class.
The individual classes are isolated from each other, which means
that for example an audio expectation will only evict other audio
expectations.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
This is useful for the SIP helper and signalling expectations.
We don't want to create a full-blown expectation with a wildcard
as source based on a single UDP packet, but need to know the
final port anyways. With inactive expectations we can register
the expectation and reserve the tuple, but wait for confirmation
from the registrar before activating it.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Without CONFIG_NET_NS, no namespace other than &init_net exists,
no need to store net in seq_net_private.
Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Introduce per-sock inlines: sock_net(), sock_net_set()
and per-inet_timewait_sock inlines: twsk_net(), twsk_net_set().
Without CONFIG_NET_NS, no namespace other than &init_net exists.
Let's explicitly define them to help compiler optimizations.
Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Introduce per-net_device inlines: dev_net(), dev_net_set().
Without CONFIG_NET_NS, no namespace other than &init_net exists.
Let's explicitly define them to help compiler optimizations.
Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
There were no packets in the namespace other than initial
previously. This will be changed in the neareast future. Netfilters
are not namespace aware and should be processed in the initial
namespace only for now.
Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
logical-bitwise & confusion
Signed-off-by: Roel Kluin <12o3l@tiscali.nl>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Commit ce7663d84:
[NETFILTER]: nfnetlink_queue: don't unregister handler of other subsystem
changed nf_unregister_queue_handler to return an error when attempting to
unregister a queue handler that is not identical to the one passed in.
This is correct in case we really do have a different queue handler already
registered, but some existing userspace code always does an unbind before
bind and aborts if that fails, so try to be nice and return success in
that case.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Similar to the nfnetlink_log problem, nfnetlink_queue incorrectly
returns -EPERM when binding or unbinding to an address family and
queueing instance 0 exists and is owned by a different process. Unlike
nfnetlink_log it previously completes the operation, but it is still
incorrect.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
When binding or unbinding to an address family, the res_id is usually set
to zero. When logging instance 0 already exists and is owned by a different
process, this makes nfunl_recv_config return -EPERM without performing
the bind operation.
Since no operation on the foreign logging instance itself was requested,
this is incorrect. Move bind/unbind commands before the queue instance
permissions checks.
Also remove an incorrect comment.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
There's a horrible slab abuse in net/netfilter/nf_conntrack_extend.c
that can be replaced with a call to ksize().
Cc: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Pekka Enberg <penberg@cs.helsinki.fi>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
From: Andrew Schulman <andrex@alumni.utexas.net>
xt_time_match() in net/netfilter/xt_time.c in kernel 2.6.24 never
matches on Sundays. On my host I have a rule like
iptables -A OUTPUT -m time --weekdays Sun -j REJECT
and it never matches. The problem is in localtime_2(), which uses
r->weekday = (4 + r->dse) % 7;
to map the epoch day onto a weekday in {0,...,6}. In particular this
gives 0 for Sundays. But 0 has to be wrong; a weekday of 0 can never
match. xt_time_match() has
if (!(info->weekdays_match & (1 << current_time.weekday)))
return false;
and when current_time.weekday = 0, the result of the & is always
zero, even when info->weekdays_match = XT_TIME_ALL_WEEKDAYS = 0xFE.
Signed-off-by: Jan Engelhardt <jengelh@computergmbh.de>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch is similar to nfnetlink_queue fixes. It fixes the computation
of skb size by using NLMSG_SPACE instead of NLMSG_ALIGN.
Signed-off-by: Eric Leblond <eric@inl.fr>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Size of the netlink skb was wrongly computed because the formula was using
NLMSG_ALIGN instead of NLMSG_SPACE. NLMSG_ALIGN does not add the room for
netlink header as NLMSG_SPACE does. This was causing a failure of message
building in some cases.
On my test system, all messages for packets in range [8*k+41, 8*k+48] where k
is an integer were invalid and the corresponding packets were dropped.
Signed-off-by: Eric Leblond <eric@inl.fr>
Signed-off-by: Patrick McHardy <kaber@trash.net>
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>
Some netfilter code and rxrpc one use seq_open() to open
a proc file, but seq_release_private to release one.
This is harmless, but ambiguous.
Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Jan Engelhardt <jengelh@computergmbh.de>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Jan Engelhardt <jengelh@computergmbh.de>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Since we're using RCU for the conntrack hash now, we need to avoid
getting preempted or interrupted by BHs while changing the stats.
Fixes warning reported by Tilman Schmidt <tilman@imap.cc> when using
preemptible RCU:
[ 48.180297] BUG: using smp_processor_id() in preemptible [00000000] code: ntpdate/3562
[ 48.180297] caller is __nf_conntrack_find+0x9b/0xeb [nf_conntrack]
[ 48.180297] Pid: 3562, comm: ntpdate Not tainted 2.6.25-rc2-mm1-testing #1
[ 48.180297] [<c02015b9>] debug_smp_processor_id+0x99/0xb0
[ 48.180297] [<fac643a7>] __nf_conntrack_find+0x9b/0xeb [nf_conntrack]
Tested-by: Tilman Schmidt <tilman@imap.cc>
Tested-by: Christian Casteyde <casteyde.christian@free.fr> [Bugzilla #10097]
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
The host address parts need to be converted to host-endian first
before arithmetic makes any sense on them.
Signed-off-by: Jan Engelhardt <jengelh@computergmbh.de>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
By allocating ->hinfo, we already have the needed indirection to cope
with the per-cpu xtables struct match_entry.
[Patrick: do this now before the revision 1 struct is used by userspace]
Signed-off-by: Jan Engelhardt <jengelh@computergmbh.de>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
The int ret variable is used only to trigger the BUG_ON() after
the skb_copy_bits() call, so check the call failure directly
and drop the variable.
Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
As reported by Tomas Simonaitis <tomas.simonaitis@gmail.com>,
inserting new data in skbs queued over {ip,ip6,nfnetlink}_queue
triggers a SKB_LINEAR_ASSERT in skb_put().
Going back through the git history, it seems this bug is present since
at least 2.6.12-rc2, probably even since the removal of
skb_linearize() for netfilter.
Linearize non-linear skbs through skb_copy_expand() when enlarging
them. Tested by Thomas, fixes bugzilla #9933.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch makes the needlessly global secmark_tg_destroy() static.
Signed-off-by: Adrian Bunk <bunk@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Jan Engelhardt <jengelh@computergmbh.de>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
The family for iprange_mt4 should be AF_INET, not AF_INET6.
Noticed by Jiri Moravec <jim.lkml@gmail.com>.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
The ->move operation has two bugs:
- It is called with the same extension as source and destination,
so it doesn't update the new extension.
- The address of the old extension is calculated incorrectly,
instead of (void *)ct->ext + ct->ext->offset[i] it uses
ct->ext + ct->ext->offset[i].
Fixes a crash on x86_64 reported by Chuck Ebbert <cebbert@redhat.com>
and Thomas Woerner <twoerner@redhat.com>.
Tested-by: Thomas Woerner <twoerner@redhat.com>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
TCP connection tracking in netfilter did not handle TCP reopening
properly: active close was taken into account for one side only and
not for any side, which is fixed now. The patch includes more comments
to explain the logic how the different cases are handled.
The bug was discovered by Jeff Chua.
Signed-off-by: Jozsef Kadlecsik <kadlec@blackhole.kfki.hu>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
The hashlimit_ipv6_mask() is called from under IP6_NF_IPTABLES config
option, but is not under it by itself.
gcc warns us about it :) :
net/netfilter/xt_hashlimit.c:473: warning: "hashlimit_ipv6_mask" defined but not used
Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
CHECK net/netfilter/xt_iprange.c
net/netfilter/xt_iprange.c:104:19: warning: restricted degrades to integer
net/netfilter/xt_iprange.c:104:37: warning: restricted degrades to integer
net/netfilter/xt_iprange.c:104:19: warning: restricted degrades to integer
net/netfilter/xt_iprange.c:104:37: warning: restricted degrades to integer
net/netfilter/xt_iprange.c:104:19: warning: restricted degrades to integer
net/netfilter/xt_iprange.c:104:37: warning: restricted degrades to integer
net/netfilter/xt_iprange.c:104:19: warning: restricted degrades to integer
net/netfilter/xt_iprange.c:104:37: warning: restricted degrades to integer
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Jan Engelhardt <jengelh@computergmbh.de>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Jan Engelhardt <jengelh@computergmbh.de>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Jan Engelhardt <jengelh@computergmbh.de>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Constify a few data tables use const qualifiers on variables where
possible in the nf_*_proto_tcp sources.
Signed-off-by: Jan Engelhardt <jengelh@computergmbh.de>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Annotate nf_conntrack_sane variables with const qualifier and remove
a few casts.
Signed-off-by: Jan Engelhardt <jengelh@computergmbh.de>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Jan Engelhardt <jengelh@computergmbh.de>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Jan Engelhardt <jengelh@computergmbh.de>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Jan Engelhardt <jengelh@computergmbh.de>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Constify data tables (predominantly in nf_conntrack_h323_types.c, but
also a few in nf_conntrack_h323_asn1.c) and use const qualifiers on
variables where possible in the h323 sources.
Signed-off-by: Jan Engelhardt <jengelh@computergmbh.de>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Propagate netns together with AF down to ->start/->next/->stop
iterators. Choose table based on netns and AF for showing.
Signed-off-by: Alexey Dobriyan <adobriyan@sw.ru>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
There are many small but still wrong things with /proc/net/*_tables_*
so I decided to do overhaul simultaneously making it more suitable for
per-netns /proc/net/*_tables_* implementation.
Fix
a) xt_get_idx() duplicating now standard seq_list_start/seq_list_next
iterators
b) tables/matches/targets list was chosen again and again on every ->next
c) multiple useless "af >= NPROTO" checks -- we simple don't supply invalid
AFs there and registration function should BUG_ON instead.
Regardless, the one in ->next() is the most useless -- ->next doesn't
run at all if ->start fails.
d) Don't use mutex_lock_interruptible() -- it can fail and ->stop is
executed even if ->start failed, so unlock without lock is possible.
As side effect, streamline code by splitting xt_tgt_ops into xt_target_ops,
xt_matches_ops, xt_tables_ops.
xt_tables_ops hooks will be changed by per-netns code. Code of
xt_matches_ops, xt_target_ops is identical except the list chosen for
iterating, but I think consolidating code for two files not worth it
given "<< 16" hacks needed for it.
[Patrick: removed unused enum in x_tables.c]
Signed-off-by: Alexey Dobriyan <adobriyan@sw.ru>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Introduces the xt_hashlimit match revision 1. It adds support for
kernel-level inversion and grouping source and/or destination IP
addresses, allowing to limit on a per-subnet basis. While this would
technically obsolete xt_limit, xt_hashlimit is a more expensive due
to the hashbucketing.
Kernel-level inversion: Previously you had to do user-level inversion:
iptables -N foo
iptables -A foo -m hashlimit --hashlimit(-upto) 5/s -j RETURN
iptables -A foo -j DROP
iptables -A INPUT -j foo
now it is simpler:
iptables -A INPUT -m hashlimit --hashlimit-over 5/s -j DROP
Signed-off-by: Jan Engelhardt <jengelh@computergmbh.de>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
The following feature was submitted some months ago. It forces the dump
of mark during the connection destruction event. The induced load is
quiet small and the patch is usefull to provide an easy way to filter
event on user side without having to keep an hash in userspace.
Signed-off-by: Eric Leblond <eric@inl.fr>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
-total: 81 errors, 3 warnings, 876 lines checked
+total: 44 errors, 3 warnings, 876 lines checked
There is still work to be done, but that's for another patch.
Signed-off-by: Jan Engelhardt <jengelh@computergmbh.de>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Rename all "conntrack" variables to "ct" for more consistency and
avoiding some overly long lines.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
early_drop() is only called *very* rarely, unfortunately gcc inlines it
into the hotpath because there is only a single caller. Explicitly mark
it noinline.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Avoid calling jhash three times and hash the entire tuple in one go.
__hash_conntrack | -485 # 760 -> 275, # inlines: 3 -> 1, size inlines: 717 -> 252
1 function changed, 485 bytes removed
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Ignoring specific entries in __nf_conntrack_find() is only needed by NAT
for nf_conntrack_tuple_taken(). Remove it from __nf_conntrack_find()
and make nf_conntrack_tuple_taken() search the hash itself.
Saves 54 bytes of text in the hotpath on x86_64:
__nf_conntrack_find | -54 # 321 -> 267, # inlines: 3 -> 2, size inlines: 181 -> 127
nf_conntrack_tuple_taken | +305 # 15 -> 320, lexblocks: 0 -> 3, # inlines: 0 -> 3, size inlines: 0 -> 181
nf_conntrack_find_get | -2 # 90 -> 88
3 functions changed, 305 bytes added, 56 bytes removed, diff: +249
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
With the RCU conversion only write_lock usages of nf_conntrack_lock are
left (except one read_lock that should actually use write_lock in the
H.323 helper). Switch to a spinlock.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Use RCU for expectation hash. This doesn't buy much for conntrack
runtime performance, but allows to reduce the use of nf_conntrack_lock
for /proc and nf_netlink_conntrack.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
The conntrack is unconfirmed, so we have an exclusive reference, which
means that the write_lock is definitely unneeded. A read_lock used to
be needed for the helper lookup, but since we're using RCU for helpers
now rcu_read_lock is enough.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Don't skip accounting for conntracks with the FIXED_TIMEOUT bit.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Properly drop nf_conntrack_lock on tuple parsing error.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
CHECK net/ipv4/netfilter/ip_tables.c
net/ipv4/netfilter/ip_tables.c:1453:8: warning: incorrect type in argument 3 (different signedness)
net/ipv4/netfilter/ip_tables.c:1453:8: expected int *size
net/ipv4/netfilter/ip_tables.c:1453:8: got unsigned int [usertype] *size
net/ipv4/netfilter/ip_tables.c:1458:44: warning: incorrect type in argument 3 (different signedness)
net/ipv4/netfilter/ip_tables.c:1458:44: expected int *size
net/ipv4/netfilter/ip_tables.c:1458:44: got unsigned int [usertype] *size
net/ipv4/netfilter/ip_tables.c:1603:2: warning: incorrect type in argument 2 (different signedness)
net/ipv4/netfilter/ip_tables.c:1603:2: expected unsigned int *i
net/ipv4/netfilter/ip_tables.c:1603:2: got int *<noident>
net/ipv4/netfilter/ip_tables.c:1627:8: warning: incorrect type in argument 3 (different signedness)
net/ipv4/netfilter/ip_tables.c:1627:8: expected int *size
net/ipv4/netfilter/ip_tables.c:1627:8: got unsigned int *size
net/ipv4/netfilter/ip_tables.c:1634:40: warning: incorrect type in argument 3 (different signedness)
net/ipv4/netfilter/ip_tables.c:1634:40: expected int *size
net/ipv4/netfilter/ip_tables.c:1634:40: got unsigned int *size
net/ipv4/netfilter/ip_tables.c:1653:8: warning: incorrect type in argument 5 (different signedness)
net/ipv4/netfilter/ip_tables.c:1653:8: expected unsigned int *i
net/ipv4/netfilter/ip_tables.c:1653:8: got int *<noident>
net/ipv4/netfilter/ip_tables.c:1666:2: warning: incorrect type in argument 2 (different signedness)
net/ipv4/netfilter/ip_tables.c:1666:2: expected unsigned int *i
net/ipv4/netfilter/ip_tables.c:1666:2: got int *<noident>
CHECK net/ipv4/netfilter/arp_tables.c
net/ipv4/netfilter/arp_tables.c:1285:40: warning: incorrect type in argument 3 (different signedness)
net/ipv4/netfilter/arp_tables.c:1285:40: expected int *size
net/ipv4/netfilter/arp_tables.c:1285:40: got unsigned int *size
net/ipv4/netfilter/arp_tables.c:1543:44: warning: incorrect type in argument 3 (different signedness)
net/ipv4/netfilter/arp_tables.c:1543:44: expected int *size
net/ipv4/netfilter/arp_tables.c:1543:44: got unsigned int [usertype] *size
CHECK net/ipv6/netfilter/ip6_tables.c
net/ipv6/netfilter/ip6_tables.c:1481:8: warning: incorrect type in argument 3 (different signedness)
net/ipv6/netfilter/ip6_tables.c:1481:8: expected int *size
net/ipv6/netfilter/ip6_tables.c:1481:8: got unsigned int [usertype] *size
net/ipv6/netfilter/ip6_tables.c:1486:44: warning: incorrect type in argument 3 (different signedness)
net/ipv6/netfilter/ip6_tables.c:1486:44: expected int *size
net/ipv6/netfilter/ip6_tables.c:1486:44: got unsigned int [usertype] *size
net/ipv6/netfilter/ip6_tables.c:1631:2: warning: incorrect type in argument 2 (different signedness)
net/ipv6/netfilter/ip6_tables.c:1631:2: expected unsigned int *i
net/ipv6/netfilter/ip6_tables.c:1631:2: got int *<noident>
net/ipv6/netfilter/ip6_tables.c:1655:8: warning: incorrect type in argument 3 (different signedness)
net/ipv6/netfilter/ip6_tables.c:1655:8: expected int *size
net/ipv6/netfilter/ip6_tables.c:1655:8: got unsigned int *size
net/ipv6/netfilter/ip6_tables.c:1662:40: warning: incorrect type in argument 3 (different signedness)
net/ipv6/netfilter/ip6_tables.c:1662:40: expected int *size
net/ipv6/netfilter/ip6_tables.c:1662:40: got unsigned int *size
net/ipv6/netfilter/ip6_tables.c:1680:8: warning: incorrect type in argument 5 (different signedness)
net/ipv6/netfilter/ip6_tables.c:1680:8: expected unsigned int *i
net/ipv6/netfilter/ip6_tables.c:1680:8: got int *<noident>
net/ipv6/netfilter/ip6_tables.c:1693:2: warning: incorrect type in argument 2 (different signedness)
net/ipv6/netfilter/ip6_tables.c:1693:2: expected unsigned int *i
net/ipv6/netfilter/ip6_tables.c:1693:2: got int *<noident>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Sparse complains when a function is not really static. Putting static
on the function prototype is not enough.
Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Some lock annotations, and make initializers static.
Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Teach sparse about locking here, and fix signed/unsigned warnings.
Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
The hashtable size is really unsigned so sparse complains when you pass
a signed integer. Change all uses to make it consistent.
Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add support for ranges to the new revision. This doesn't affect
compatibility since the new revision was not released yet.
Signed-off-by: Jan Engelhardt <jengelh@computergmbh.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
The TCPMSS target in Xtables should consider the MTU of the reverse
route on forwarded packets as part of the path MTU.
Point in case: IN=ppp0, OUT=eth0. MSS set to 1460 in spite of MTU of
ppp0 being 1392.
Signed-off-by: Jan Engelhardt <jengelh@computergmbh.de>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Typical table module registers xt_table structure (i.e. packet_filter)
and link it to list during it. We can't use one template for it because
corresponding list_head will become corrupted. We also can't unregister
with template because it wasn't changed at all and thus doesn't know in
which list it is.
So, we duplicate template at the very first step of table registration.
Table modules will save it for use during unregistration time and actual
filtering.
Do it at once to not screw bisection.
P.S.: renaming i.e. packet_filter => __packet_filter is temporary until
full netnsization of table modules is done.
Signed-off-by: Alexey Dobriyan <adobriyan@sw.ru>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
In fact all we want is per-netns set of rules, however doing that will
unnecessary complicate routines such as ipt_hook()/ipt_do_table, so
make full xt_table array per-netns.
Every user stubbed with init_net for a while.
Signed-off-by: Alexey Dobriyan <adobriyan@sw.ru>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Switch from 0/-E to ptr/PTR_ERR convention.
Signed-off-by: Alexey Dobriyan <adobriyan@sw.ru>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Extend the xt_conntrack match revision 1 by port matching (all four
{orig,repl}{src,dst}) and by packet direction matching.
Signed-off-by: Jan Engelhardt <jengelh@computergmbh.de>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
CHECK net/netfilter/nf_conntrack_expect.c
net/netfilter/nf_conntrack_expect.c:429:13: warning: context imbalance in 'exp_seq_start' - wrong count at exit
net/netfilter/nf_conntrack_expect.c:441:13: warning: context imbalance in 'exp_seq_stop' - unexpected unlock
CHECK net/netfilter/nf_log.c
net/netfilter/nf_log.c:105:13: warning: context imbalance in 'seq_start' - wrong count at exit
net/netfilter/nf_log.c:125:13: warning: context imbalance in 'seq_stop' - unexpected unlock
CHECK net/netfilter/nfnetlink_queue.c
net/netfilter/nfnetlink_queue.c:363:7: warning: symbol 'size' shadows an earlier one
net/netfilter/nfnetlink_queue.c:217:9: originally declared here
net/netfilter/nfnetlink_queue.c:847:13: warning: context imbalance in 'seq_start' - wrong count at exit
net/netfilter/nfnetlink_queue.c:859:13: warning: context imbalance in 'seq_stop' - unexpected unlock
Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch introduces a mechanism for checking when labeled IPsec or SECMARK
are in use by keeping introducing a configuration reference counter for each
subsystem. In the case of labeled IPsec, whenever a labeled SA or SPD entry
is created the labeled IPsec/XFRM reference count is increased and when the
entry is removed it is decreased. In the case of SECMARK, when a SECMARK
target is created the reference count is increased and later decreased when the
target is removed. These reference counters allow SELinux to quickly determine
if either of these subsystems are enabled.
NetLabel already has a similar mechanism which provides the netlbl_enabled()
function.
This patch also renames the selinux_relabel_packet_permission() function to
selinux_secmark_relabel_packet_permission() as the original name and
description were misleading in that they referenced a single packet label which
is not the case.
Signed-off-by: Paul Moore <paul.moore@hp.com>
Signed-off-by: James Morris <jmorris@namei.org>
Convert packet schedulers to use the netlink API. Unfortunately a gradual
conversion is not possible without breaking compilation in the middle or
adding lots of casts, so this patch converts them all in one step. The
patch has been mostly generated automatically with some minor edits to
at least allow seperate conversion of classifiers and actions.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Create a specific helper for netlink kernel socket disposal. This just
let the code look better and provides a ground for proper disposal
inside a namespace.
Signed-off-by: Denis V. Lunev <den@openvz.org>
Tested-by: Alexey Dobriyan <adobriyan@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
- Remove declarations of non-existing variables and functions
- Move helper init/cleanup function declarations to nf_conntrack_helper.h
- Remove unneeded __nf_conntrack_attach declaration and make it static
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Since there now is generic support for shared sysctl paths, the only
remains are the net/netfilter and net/ipv4/netfilter paths. Move them
to net/netfilter/core.c and net/ipv4/netfilter.c and kill nf_sysctl.c.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Instead of keeping pointers to the timeout values in a table, simply
put the timeout values in the table directly.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Don't take and release the lock once per SCTP chunk, simply hold it
the entire time while iterating through the chunks.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
The name is misleading, it holds the new connection state, so rename it
to "newstate". Also rename "oldsctpstate" to "oldstate" for consistency.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Consolidate error paths and use proper symbolic return value instead
of magic values.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Eliminate a few lines over 80 characters by using a local variable to
hold the conntrack direction instead of using CTINFO2DIR everywhere.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Reduce the length of some overly long lines by renaming all
"conntrack" variables to "ct".
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Use unsigned long instead of char for the bitmap and removed lots
of casts.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Reindent switch cases properly, get rid of weird constructs like "!(x == y)",
put logical operations on the end of the line instead of the next line, get
rid of superfluous braces.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Instead of keeping pointers to the timeout values in a table, simply
put the timeout values in the table directly.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
The TCP and SCTP conntrack state transition tables only holds
small numbers, but gcc uses 4 byte per entry for the enum. Switching
to an u8 reduces the size from 480 to 120 bytes for TCP and from
576 to 144 bytes for SCTP.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Adds IPv6 support to xt_iprange, making it possible to match on IPv6
address ranges with ip6tables.
Signed-off-by: Jan Engelhardt <jengelh@computergmbh.de>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch moves ipt_iprange to xt_iprange, in preparation for adding
IPv6 support to xt_iprange.
Signed-off-by: Jan Engelhardt <jengelh@computergmbh.de>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Updates the MODULE_DESCRIPTION() tags for all Netfilter modules,
actually describing what the module does and not just
"netfilter XYZ target".
Signed-off-by: Jan Engelhardt <jengelh@computergmbh.de>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Jan Engelhardt <jengelh@computergmbh.de>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Jan Engelhart <jengelh@computergmbh.de>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
In the PACKET_LOOPBACK case, the skb data was always interpreted as
IPv4, but that is not valid for IPv6, obviously. Fix this by adding an
extra condition to check for AF_INET.
Signed-off-by: Jan Engelhardt <jengelh@computergmbh.de>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Introduces the xt_mark match revision 1. It uses fixed types,
eventually obsoleting revision 0 some day (uses nonfixed types).
Signed-off-by: Jan Engelhardt <jengelh@computergmbh.de>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Introduces the xt_conntrack match revision 1. It uses fixed types, the
new nf_inet_addr and comes with IPv6 support, thereby completely
superseding xt_state.
Signed-off-by: Jan Engelhardt <jengelh@computergmbh.de>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Introduces the xt_connmark match revision 1. It uses fixed types,
eventually obsoleting revision 0 some day (uses nonfixed types).
(Unfixed types like "unsigned long" do not play well with mixed
user-/kernelspace "bitness", e.g. 32/64, as is common on SPARC64,
and need extra compat code.)
Signed-off-by: Jan Engelhardt <jengelh@computergmbh.de>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Introduces the xt_MARK target revision 2. It uses fixed types, and
also uses the more expressive XOR logic.
Signed-off-by: Jan Engelhardt <jengelh@computergmbh.de>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Introduces the xt_CONNMARK target revision 1. It uses fixed types, and
also uses the more expressive XOR logic. Futhermore, it allows to
selectively pick bits from both the ctmark and the nfmark in the SAVE
and RESTORE operations.
Signed-off-by: Jan Engelhardt <jengelh@computergmbh.de>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Fix incorrect mask value passed to ipv4_change_dsfield/ipv6_change_dsfield.
Signed-off-by: Jan Engelhardt <jengelh@computergmbh.de>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch changes the behavior of xt_TOS v1 so that the mask value
the user supplies means "zero out these bits" rather than "keep these
bits". This is more easy on the user, as (I would assume) people keep
more bits than zeroing, so, an example:
Action: Set bit 0x01.
before (&): iptables -j TOS --set-tos 0x01/0xFE
after (&~): iptables -j TOS --set-tos 0x01/0x01
This is not too "tragic" with xt_TOS, but where larger fields are used
(e.g. proposed xt_MARK v2), `--set-xmar 0x01/0x01` vs. `--set-xmark
0x01/0xFFFFFFFE` really makes a difference. Other target(!) modules,
such as xt_TPROXY also use &~ rather than &, so let's get to a common
ground.
(Since xt_TOS has not yet left the development tree en direction to
mainline, the semantic can be changed as proposed without breaking
iptables.)
Signed-off-by: Jan Engelhardt <jengelh@computergmbh.de>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Most of the netfilter modules are not considered experimental anymore,
the only ones I want to keep marked as EXPERIMENTAL are:
- TCPOPTSTRIP target, which is brand new.
- SANE helper, which is quite new.
- CLUSTERIP target, which I believe hasn't had much testing despite
being in the kernel for quite a long time.
- SCTP match and conntrack protocol, which are a mess and need to
be reviewed and cleaned up before I would trust them.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
net/netfilter/xt_policy.c:
policy_mt | -906
1 function changed, 906 bytes removed, diff: -906
net/netfilter/xt_policy.c:
match_xfrm_state | +427
1 function changed, 427 bytes added, diff: +427
net/netfilter/xt_policy.o:
2 functions changed, 427 bytes added, 906 bytes removed, diff: -479
Alternatively, this could be done by combining identical
parts of the match_policy_in/out()
Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>
The conntracks subsystem has a similar infrastructure
to maintain ctl_paths, but since we already have it
on the generic level, I think it's OK to switch to
using it.
So, basically, this patch just replaces the ctl_table-s
with ctl_path-s, nf_register_sysctl_table with
register_sysctl_paths() and removes no longer needed code.
After this the net/netfilter/nf_sysctl.c file contains
the paths only.
Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Acked-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
This includes the most simple cases for netfilter.
The first part is tne queue modules for ipv4 and ipv6,
on which the net/ipv4/ and net/ipv6/ paths are reused
from the appropriate ipv4 and ipv6 code.
The conntrack module is also patched, but this hunk is
very small and simple.
Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Acked-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
The NETFILTER_ADVANCED option hides lots of the rather obscure netfilter
options when disabled and provides defaults (M) that should allow to
run a distribution firewall without further thinking.
Defaults to 'y' to avoid breaking current configurations.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Apply Eric Dumazet's jhash optimizations where applicable. Quoting Eric:
Thanks to jhash, hash value uses full 32 bits. Instead of returning
hash % size (implying a divide) we return the high 32 bits of the
(hash * size) that will give results between [0 and size-1] and same
hash distribution.
On most cpus, a multiply is less expensive than a divide, by an order
of magnitude.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch generalizes the (CONFIG_IP6_NF_IPTABLES || CONFIG_IP6_NF_IPTABLES_MODULE)
test done in hashlimit_init_dst() to all the xt_hashlimit module.
This permits a size reduction of "struct dsthash_dst". This saves memory and
cpu for IPV4 only hosts.
Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
1) Using jhash2() instead of jhash() is a litle bit faster if applicable.
2) Thanks to jhash, hash value uses full 32 bits.
Instead of returning hash % size (implying a divide)
we return the high 32 bits of the (hash * size) that will
give results between [0 and size-1] and same hash distribution.
On most cpus, a multiply is less expensive than a divide, by an order
of magnitude.
Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Jan Engelhardt <jengelh@computergmbh.de>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Parenthesize macro parameters.
Signed-off-by: Jan Engelhardt <jengelh@computergmbh.de>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
A few netfilter modules provide their own union of IPv4 and IPv6
address storage. Will unify that in this patch series.
(1/4): Rename union nf_conntrack_address to union nf_inet_addr and
move it to x_tables.h.
Signed-off-by: Jan Engelhardt <jengelh@computergmbh.de>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Use %u format specifiers as ->family is unsigned.
Signed-off-by: Jan Engelhardt <jengelh@computergmbh.de>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Similar to Maciej Soltysiak's ipt_LOG patch, include GID in addition
to UID in netlink message.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Currently we return EINVAL for "instance exists", "allocation failed" and
"module unloaded below us", which is completely inapproriate.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Similar to the nfnetlink_queue fixes:
The peer_pid must be checked in all cases when a logging instance exists,
additionally we must check whether an instance exists before attempting
to configure it to avoid NULL ptr dereferences.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Whatever that comment tries to say, I don't get it and it looks like
a leftover from the time when RCU wasn't used properly.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
nf_nat_setup_info gets the hook number and translates that to the
manip type to perform. This is a relict from the time when one
manip per hook could exist, the exact hook number doesn't matter
anymore, its converted to the manip type. Most callers already
know what kind of NAT they want to perform, so pass the maniptype
in directly.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
When the timer is late its timeout might be before the current time,
in which case a very large value is dumped.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>