linux-stable.git/net/netfilter, branch v4.12.2

netfilter: ctnetlink: fix incorrect nf_ct_put during hash resize

2017-05-24T09:26:01+00:00

If nf_conntrack_htable_size was adjusted by the user during the ct
dump operation, we may invoke nf_ct_put twice for the same ct, i.e.
the "last" ct. This will cause the ct will be freed but still linked
in hash buckets.

It's very easy to reproduce the problem by the following commands:
  # while : ; do
  echo $RANDOM > /proc/sys/net/netfilter/nf_conntrack_buckets
  done
  # while : ; do
  conntrack -L
  done
  # iperf -s 127.0.0.1 &
  # iperf -c 127.0.0.1 -P 60 -t 36000

After a while, the system will hang like this:
  NMI watchdog: BUG: soft lockup - CPU#1 stuck for 22s! [bash:20184]
  NMI watchdog: BUG: soft lockup - CPU#0 stuck for 22s! [iperf:20382]
  ...

So at last if we find cb->args[1] is equal to "last", this means hash
resize happened, then we can set cb->args[1] to 0 to fix the above
issue.

Fixes: d205dc40798d ("[NETFILTER]: ctnetlink: fix deadlock in table dumping")
Signed-off-by: Liping Zhang 
Signed-off-by: Pablo Neira Ayuso

netfilter: nat: use atomic bit op to clear the _SRC_NAT_DONE_BIT

2017-05-23T20:54:51+00:00

We need to clear the IPS_SRC_NAT_DONE_BIT to indicate that the ct has
been removed from nat_bysource table. But unfortunately, we use the
non-atomic bit operation: "ct->status &= ~IPS_NAT_DONE_MASK". So
there's a race condition that we may clear the _DYING_BIT set by
another CPU unexpectedly.

Since we don't care about the IPS_DST_NAT_DONE_BIT, so just using
clear_bit to clear the IPS_SRC_NAT_DONE_BIT is enough.

Also note, this is the last user which use the non-atomic bit operation
to update the confirmed ct->status.

Reported-by: Florian Westphal 
Signed-off-by: Liping Zhang 
Signed-off-by: Pablo Neira Ayuso

netfilter: nft_set_rbtree: handle element re-addition after deletion

2017-05-23T20:54:14+00:00

The existing code selects no next branch to be inspected when
re-inserting an inactive element into the rb-tree, looping endlessly.
This patch restricts the check for active elements to the EEXIST case
only.

Fixes: e701001e7cbe ("netfilter: nft_rbtree: allow adjacent intervals with dynamic updates")
Reported-by: Wolfgang Bumiller 
Tested-by: Wolfgang Bumiller 
Signed-off-by: Pablo Neira Ayuso

netfilter: conntrack: fix false CRC32c mismatch using paged skb

2017-05-23T20:54:14+00:00

sctp_compute_cksum() implementation assumes that at least the SCTP header
is in the linear part of skb: modify conntrack error callback to avoid
false CRC32c mismatch, if the transport header is partially/entirely paged.

Fixes: cf6e007eef83 ("netfilter: conntrack: validate SCTP crc32c in PREROUTING")
Signed-off-by: Davide Caratti 
Signed-off-by: Pablo Neira Ayuso

Merge git://git.kernel.org/pub/scm/linux/kernel/git/pablo/nf

2017-05-21T17:00:02+00:00

Pablo Neira Ayuso says:

====================
Netfilter/IPVS fixes for net

The following patchset contains Netfilter/IPVS fixes for your net tree,
they are:

1) When using IPVS in direct-routing mode, normal traffic from the LVS
   host to a back-end server is sometimes incorrectly NATed on the way
   back into the LVS host. Patch to fix this from Julian Anastasov.

2) Calm down clang compilation warning in ctnetlink due to type
   mismatch, from Matthias Kaehlcke.

3) Do not re-setup NAT for conntracks that are already confirmed, this
   is fixing a problem that was introduced in the previous nf-next batch.
   Patch from Liping Zhang.

4) Do not allow conntrack helper removal from userspace cthelper
   infrastructure if already in used. This comes with an initial patch
   to introduce nf_conntrack_helper_put() that is required by this fix.
   From Liping Zhang.

5) Zero the pad when copying data to userspace, otherwise iptables fails
   to remove rules. This is a follow up on the patchset that sorts out
   the internal match/target structure pointer leak to userspace. Patch
   from the same author, Willem de Bruijn. This also comes with a build
   failure when CONFIG_COMPAT is not on, coming in the last patch of
   this series.

6) SYNPROXY crashes with conntrack entries that are created via
   ctnetlink, more specifically via conntrackd state sync. Patch from
   Eric Leblond.

7) RCU safe iteration on set element dumping in nf_tables, from
   Liping Zhang.

8) Missing sanitization of immediate date for the bitwise and cmp
   expressions in nf_tables.

9) Refcounting logic for chain and objects from set elements does not
   integrate into the nf_tables 2-phase commit protocol.

10) Missing sanitization of target verdict in ebtables arpreply target,
    from Gao Feng.
====================

Signed-off-by: David S. Miller

netfilter: xtables: fix build failure from COMPAT_XT_ALIGN outside CONFIG_COMPAT

2017-05-18T11:10:03+00:00

The patch in the Fixes references COMPAT_XT_ALIGN in the definition
of XT_DATA_TO_USER, outside an #ifdef CONFIG_COMPAT block.

Split XT_DATA_TO_USER into separate compat and non compat variants and
define the first inside an CONFIG_COMPAT block.

This simplifies both variants by removing branches inside the macro.

Fixes: 324318f0248c ("netfilter: xtables: zero padding in data_to_user")
Reported-by: Stephen Rothwell 
Signed-off-by: Willem de Bruijn 
Signed-off-by: Pablo Neira Ayuso

netfilter: nf_tables: revisit chain/object refcounting from elements

2017-05-15T10:51:41+00:00

Andreas reports that the following incremental update using our commit
protocol doesn't work.

 # nft -f incremental-update.nft
 delete element ip filter client_to_any { 10.180.86.22 : goto CIn_1 }
 delete chain ip filter CIn_1
 ... Error: Could not process rule: Device or resource busy

The existing code is not well-integrated into the commit phase protocol,
since element deletions do not result in refcount decrement from the
preparation phase. This results in bogus EBUSY errors like the one
above.

Two new functions come with this patch:

* nft_set_elem_activate() function is used from the abort path, to
  restore the set element refcounting on objects that occurred from
  the preparation phase.

* nft_set_elem_deactivate() that is called from nft_del_setelem() to
  decrement set element refcounting on objects from the preparation
  phase in the commit protocol.

The nft_data_uninit() has been renamed to nft_data_release() since this
function does not uninitialize any data store in the data register,
instead just releases the references to objects. Moreover, a new
function nft_data_hold() has been introduced to be used from
nft_set_elem_activate().

Reported-by: Andreas Schultz 
Signed-off-by: Pablo Neira Ayuso

netfilter: nf_tables: missing sanitization in data from userspace

2017-05-15T10:51:40+00:00

Do not assume userspace always sends us NFT_DATA_VALUE for bitwise and
cmp expressions. Although NFT_DATA_VERDICT does not make any sense, it
is still possible to handcraft a netlink message using this incorrect
data type.

Signed-off-by: Pablo Neira Ayuso

netfilter: nf_tables: can't assume lock is acquired when dumping set elems

2017-05-15T10:51:39+00:00

When dumping the elements related to a specified set, we may invoke the
nf_tables_dump_set with the NFNL_SUBSYS_NFTABLES lock not acquired. So
we should use the proper rcu operation to avoid race condition, just
like other nft dump operations.

Signed-off-by: Liping Zhang 
Signed-off-by: Pablo Neira Ayuso

netfilter: synproxy: fix conntrackd interaction

2017-05-15T10:51:39+00:00

This patch fixes the creation of connection tracking entry from
netlink when synproxy is used. It was missing the addition of
the synproxy extension.

This was causing kernel crashes when a conntrack entry created by
conntrackd was used after the switch of traffic from active node
to the passive node.

Signed-off-by: Eric Leblond 
Signed-off-by: Pablo Neira Ayuso