Age | Commit message (Collapse) | Author |
|
This further refines the changes made to conntrack gc_worker in
commit e0df8cae6c16 ("netfilter: conntrack: refine gc worker heuristics").
The main idea of that change was to reduce the scan interval when evictions
take place.
However, on the reporters' setup, there are 1-2 million conntrack entries
in total and roughly 8k new (and closing) connections per second.
In this case we'll always evict at least one entry per gc cycle and scan
interval is always at 1 jiffy because of this test:
} else if (expired_count) {
gc_work->next_gc_run /= 2U;
next_run = msecs_to_jiffies(1);
being true almost all the time.
Given we scan ~10k entries per run its clearly wrong to reduce interval
based on nonzero eviction count, it will only waste cpu cycles since a vast
majorities of conntracks are not timed out.
Thus only look at the ratio (scanned entries vs. evicted entries) to make
a decision on whether to reduce or not.
Because evictor is supposed to only kick in when system turns idle after
a busy period, pick a high ratio -- this makes it 50%. We thus keep
the idea of increasing scan rate when its likely that table contains many
expired entries.
In order to not let timed-out entries hang around for too long
(important when using event logging, in which case we want to timely
destroy events), we now scan the full table within at most
GC_MAX_SCAN_JIFFIES (16 seconds) even in worst-case scenario where all
timed-out entries sit in same slot.
I tested this with a vm under synflood (with
sysctl net.netfilter.nf_conntrack_tcp_timeout_syn_recv=3).
While flood is ongoing, interval now stays at its max rate
(GC_MAX_SCAN_JIFFIES / GC_MAX_BUCKETS_DIV -> 125ms).
With feedback from Nicolas Dichtel.
Reported-by: Denys Fedoryshchenko <nuclearcat@nuclearcat.com>
Cc: Nicolas Dichtel <nicolas.dichtel@6wind.com>
Fixes: b87a2f9199ea82eaadc ("netfilter: conntrack: add gc worker to remove timed-out entries")
Signed-off-by: Florian Westphal <fw@strlen.de>
Tested-by: Nicolas Dichtel <nicolas.dichtel@6wind.com>
Acked-by: Nicolas Dichtel <nicolas.dichtel@6wind.com>
Tested-by: Denys Fedoryshchenko <nuclearcat@nuclearcat.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
|
|
Instead of breaking loop and instant resched, don't bother checking
this in first place (the loop calls cond_resched for every bucket anyway).
Suggested-by: Nicolas Dichtel <nicolas.dichtel@6wind.com>
Signed-off-by: Florian Westphal <fw@strlen.de>
Acked-by: Nicolas Dichtel <nicolas.dichtel@6wind.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
|
|
ktime is a union because the initial implementation stored the time in
scalar nanoseconds on 64 bit machine and in a endianess optimized timespec
variant for 32bit machines. The Y2038 cleanup removed the timespec variant
and switched everything to scalar nanoseconds. The union remained, but
become completely pointless.
Get rid of the union and just keep ktime_t as simple typedef of type s64.
The conversion was done with coccinelle and some manual mopping up.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: Peter Zijlstra <peterz@infradead.org>
|
|
Several cases of bug fixes in 'net' overlapping other changes in
'net-next-.
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
gcc correctly identified a theoretical uninitialized variable use:
net/netfilter/nf_conntrack_core.c: In function 'nf_conntrack_in':
net/netfilter/nf_conntrack_core.c:1125:14: error: 'l4proto' may be used uninitialized in this function [-Werror=maybe-uninitialized]
This could only happen when we 'goto out' before looking up l4proto,
and then enter the retry, implying that l3proto->get_l4proto()
returned NF_REPEAT. This does not currently get returned in any
code path and probably won't ever happen, but is not good to
rely on.
Moving the repeat handling up a little should have the same
behavior as today but avoids the warning by making that case
impossible to enter.
[ I have mangled this original patch to remove the check for tmpl, we
should inconditionally jump back to the repeat label in case we hit
NF_REPEAT instead. I have also moved the comment that explains this
where it belongs. --pablo ]
Fixes: 08733a0cb7de ("netfilter: handle NF_REPEAT from nf_conntrack_in()")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
|
|
Nicolas Dichtel says:
After commit b87a2f9199ea ("netfilter: conntrack: add gc worker to
remove timed-out entries"), netlink conntrack deletion events may be
sent with a huge delay.
Nicolas further points at this line:
goal = min(nf_conntrack_htable_size / GC_MAX_BUCKETS_DIV, GC_MAX_BUCKETS);
and indeed, this isn't optimal at all. Rationale here was to ensure that
we don't block other work items for too long, even if
nf_conntrack_htable_size is huge. But in order to have some guarantee
about maximum time period where a scan of the full conntrack table
completes we should always use a fixed slice size, so that once every
N scans the full table has been examined at least once.
We also need to balance this vs. the case where the system is either idle
(i.e., conntrack table (almost) empty) or very busy (i.e. eviction happens
from packet path).
So, after some discussion with Nicolas:
1. want hard guarantee that we scan entire table at least once every X s
-> need to scan fraction of table (get rid of upper bound)
2. don't want to eat cycles on idle or very busy system
-> increase interval if we did not evict any entries
3. don't want to block other worker items for too long
-> make fraction really small, and prefer small scan interval instead
4. Want reasonable short time where we detect timed-out entry when
system went idle after a burst of traffic, while not doing scans
all the time.
-> Store next gc scan in worker, increasing delays when no eviction
happened and shrinking delay when we see timed out entries.
The old gc interval is turned into a max number, scans can now happen
every jiffy if stale entries are present.
Longest possible time period until an entry is evicted is now 2 minutes
in worst case (entry expires right after it was deemed 'not expired').
Reported-by: Nicolas Dichtel <nicolas.dichtel@6wind.com>
Signed-off-by: Florian Westphal <fw@strlen.de>
Acked-by: Nicolas Dichtel <nicolas.dichtel@6wind.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
|
|
NF_REPEAT is only needed from nf_conntrack_in() under a very specific
case required by the TCP protocol tracker, we can handle this case
without returning to the core hook path. Handling of NF_REPEAT from the
nf_reinject() is left untouched.
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
|
|
When the maximum evictions number is reached, do not wait 5 seconds before
the next run.
CC: Florian Westphal <fw@strlen.de>
Signed-off-by: Nicolas Dichtel <nicolas.dichtel@6wind.com>
Acked-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next
Conflicts:
net/netfilter/core.c
net/netfilter/nf_tables_netdev.c
Resolve two conflicts before pull request for David's net-next tree:
1) Between c73c24849011 ("netfilter: nf_tables_netdev: remove redundant
ip_hdr assignment") from the net tree and commit ddc8b6027ad0
("netfilter: introduce nft_set_pktinfo_{ipv4, ipv6}_validate()").
2) Between e8bffe0cf964 ("net: Add _nf_(un)register_hooks symbols") and
Aaron Conole's patches to replace list_head with single linked list.
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
|
|
All of the callers of nf_hook_slow already hold the rcu_read_lock, so this
cleanup removes the recursive call. This is just a cleanup, as the locking
code gracefully handles this situation.
Signed-off-by: Aaron Conole <aconole@bytheb.org>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
|
|
|
|
When memory is exhausted, nfct_seqadj_ext_add may fail to add the
synproxy and seqadj extensions. The function nf_ct_seqadj_init doesn't
check if get valid seqadj pointer by the nfct_seqadj.
Now drop the packet directly when fail to add seqadj extension to
avoid dereference NULL pointer in nf_ct_seqadj_init from
init_conntrack().
Signed-off-by: Gao Feng <fgao@ikuai8.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
|
|
These counters sit in hot path and do show up in perf, this is especially
true for 'found' and 'searched' which get incremented for every packet
processed.
Information like
searched=212030105
new=623431
found=333613
delete=623327
does not seem too helpful nowadays:
- on busy systems found and searched will overflow every few hours
(these are 32bit integers), other more busy ones every few days.
- for debugging there are better methods, such as iptables' trace target,
the conntrack log sysctls. Nowadays we also have perf tool.
This removes packet path stat counters except those that
are expected to be 0 (or close to 0) on a normal system, e.g.
'insert_failed' (race happened) or 'invalid' (proto tracker rejects).
The insert stat is retained for the ctnetlink case.
The found stat is retained for the tuple-is-taken check when NAT has to
determine if it needs to pick a different source address.
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
|
|
After timer removal this just calls nf_ct_delete so remove the __ prefix
version and make nf_ct_kill a shorthand for nf_ct_delete.
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
|
|
If we evicted a large fraction of the scanned conntrack entries re-schedule
the next gc cycle for immediate execution.
This triggers during tests where load is high, then drops to zero and
many connections will be in TW/CLOSE state with < 30 second timeouts.
Without this change it will take several minutes until conntrack count
comes back to normal.
Signed-off-by: Florian Westphal <fw@strlen.de>
Acked-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
|
|
Conntrack gc worker to evict stale entries.
GC happens once every 5 seconds, but we only scan at most 1/64th of the
table (and not more than 8k) buckets to avoid hogging cpu.
This means that a complete scan of the table will take several minutes
of wall-clock time.
Considering that the gc run will never have to evict any entries
during normal operation because those will happen from packet path
this should be fine.
We only need gc to make sure userspace (conntrack event listeners)
eventually learn of the timeout, and for resource reclaim in case the
system becomes idle.
We do not disable BH and cond_resched for every bucket so this should
not introduce noticeable latencies either.
A followup patch will add a small change to speed up GC for the extreme
case where most entries are timed out on an otherwise idle system.
v2: Use cond_resched_rcu_qs & add comment wrt. missing restart on
nulls value change in gc worker, suggested by Eric Dumazet.
v3: don't call cancel_delayed_work_sync twice (again, Eric).
Signed-off-by: Florian Westphal <fw@strlen.de>
Acked-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
|
|
With stats enabled this eats 80 bytes on x86_64 per nf_conn entry, as
Eric Dumazet pointed out during netfilter workshop 2016.
Eric also says: "Another reason was the fact that Thomas was about to
change max timer range [..]" (500462a9de657f8, 'timers: Switch to
a non-cascading wheel').
Remove the timer and use a 32bit jiffies value containing timestamp until
entry is valid.
During conntrack lookup, even before doing tuple comparision, check
the timeout value and evict the entry in case it is too old.
The dying bit is used as a synchronization point to avoid races where
multiple cpus try to evict the same entry.
Because lookup is always lockless, we need to bump the refcnt once
when we evict, else we could try to evict already-dead entry that
is being recycled.
This is the standard/expected way when conntrack entries are destroyed.
Followup patches will introduce garbage colliction via work queue
and further places where we can reap obsoleted entries (e.g. during
netlink dumps), this is needed to avoid expired conntracks from hanging
around for too long when lookup rate is low after a busy period.
Signed-off-by: Florian Westphal <fw@strlen.de>
Acked-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
|
|
In case nf_conntrack_tuple_taken did not find a conflicting entry
check that all entries in this hash slot were tested and restart
in case an entry was moved to another chain.
Reported-by: Eric Dumazet <edumazet@google.com>
Fixes: ea781f197d6a ("netfilter: nf_conntrack: use SLAB_DESTROY_BY_RCU and get rid of call_rcu()")
Signed-off-by: Florian Westphal <fw@strlen.de>
Acked-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
|
|
This is required to iterate over the hash table in cttimeout, ctnetlink
and nf_conntrack_ipv4.
>> ERROR: "nf_conntrack_htable_size" [net/netfilter/nfnetlink_cttimeout.ko] undefined!
ERROR: "nf_conntrack_htable_size" [net/netfilter/nf_conntrack_netlink.ko] undefined!
ERROR: "nf_conntrack_htable_size" [net/ipv4/netfilter/nf_conntrack_ipv4.ko] undefined!
Fixes: adf0516845bcd0 ("netfilter: remove ip_conntrack* sysctl compat code")
Reported-by: kbuild test robot <fengguang.wu@intel.com>
Reported-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
|
|
Since commit 64b87639c9cb ("netfilter: conntrack: fix race between
nf_conntrack proc read and hash resize") introduce the
nf_conntrack_get_ht, so there's no need to check nf_conntrack_generation
again and again to get the hash table and hash size. And convert
nf_conntrack_get_ht to inline function here.
Suggested-by: Pablo Neira Ayuso <pablo@netfilter.org>
Signed-off-by: Liping Zhang <liping.zhang@spreadtrum.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
|
|
This backward compatibility has been around for more than ten years,
since Yasuyuki Kozakai introduced IPv6 in conntrack. These days, we have
alternate /proc/net/nf_conntrack* entries, the ctnetlink interface and
the conntrack utility got adopted by many people in the user community
according to what I observed on the netfilter user mailing list.
So let's get rid of this.
Note that nf_conntrack_htable_size and unsigned int nf_conntrack_max do
not need to be exported as symbol anymore.
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
|
|
Pull networking updates from David Miller:
1) Unified UDP encapsulation offload methods for drivers, from
Alexander Duyck.
2) Make DSA binding more sane, from Andrew Lunn.
3) Support QCA9888 chips in ath10k, from Anilkumar Kolli.
4) Several workqueue usage cleanups, from Bhaktipriya Shridhar.
5) Add XDP (eXpress Data Path), essentially running BPF programs on RX
packets as soon as the device sees them, with the option to mirror
the packet on TX via the same interface. From Brenden Blanco and
others.
6) Allow qdisc/class stats dumps to run lockless, from Eric Dumazet.
7) Add VLAN support to b53 and bcm_sf2, from Florian Fainelli.
8) Simplify netlink conntrack entry layout, from Florian Westphal.
9) Add ipv4 forwarding support to mlxsw spectrum driver, from Ido
Schimmel, Yotam Gigi, and Jiri Pirko.
10) Add SKB array infrastructure and convert tun and macvtap over to it.
From Michael S Tsirkin and Jason Wang.
11) Support qdisc packet injection in pktgen, from John Fastabend.
12) Add neighbour monitoring framework to TIPC, from Jon Paul Maloy.
13) Add NV congestion control support to TCP, from Lawrence Brakmo.
14) Add GSO support to SCTP, from Marcelo Ricardo Leitner.
15) Allow GRO and RPS to function on macsec devices, from Paolo Abeni.
16) Support MPLS over IPV4, from Simon Horman.
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next: (1622 commits)
xgene: Fix build warning with ACPI disabled.
be2net: perform temperature query in adapter regardless of its interface state
l2tp: Correctly return -EBADF from pppol2tp_getname.
net/mlx5_core/health: Remove deprecated create_singlethread_workqueue
net: ipmr/ip6mr: update lastuse on entry change
macsec: ensure rx_sa is set when validation is disabled
tipc: dump monitor attributes
tipc: add a function to get the bearer name
tipc: get monitor threshold for the cluster
tipc: make cluster size threshold for monitoring configurable
tipc: introduce constants for tipc address validation
net: neigh: disallow transition to NUD_STALE if lladdr is unchanged in neigh_update()
MAINTAINERS: xgene: Add driver and documentation path
Documentation: dtb: xgene: Add MDIO node
dtb: xgene: Add MDIO node
drivers: net: xgene: ethtool: Use phy_ethtool_gset and sset
drivers: net: xgene: Use exported functions
drivers: net: xgene: Enable MDIO driver
drivers: net: xgene: Add backward compatibility
drivers: net: phy: xgene: Add MDIO driver
...
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull locking updates from Ingo Molnar:
"The locking tree was busier in this cycle than the usual pattern - a
couple of major projects happened to coincide.
The main changes are:
- implement the atomic_fetch_{add,sub,and,or,xor}() API natively
across all SMP architectures (Peter Zijlstra)
- add atomic_fetch_{inc/dec}() as well, using the generic primitives
(Davidlohr Bueso)
- optimize various aspects of rwsems (Jason Low, Davidlohr Bueso,
Waiman Long)
- optimize smp_cond_load_acquire() on arm64 and implement LSE based
atomic{,64}_fetch_{add,sub,and,andnot,or,xor}{,_relaxed,_acquire,_release}()
on arm64 (Will Deacon)
- introduce smp_acquire__after_ctrl_dep() and fix various barrier
mis-uses and bugs (Peter Zijlstra)
- after discovering ancient spin_unlock_wait() barrier bugs in its
implementation and usage, strengthen its semantics and update/fix
usage sites (Peter Zijlstra)
- optimize mutex_trylock() fastpath (Peter Zijlstra)
- ... misc fixes and cleanups"
* 'locking-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (67 commits)
locking/atomic: Introduce inc/dec variants for the atomic_fetch_$op() API
locking/barriers, arch/arm64: Implement LDXR+WFE based smp_cond_load_acquire()
locking/static_keys: Fix non static symbol Sparse warning
locking/qspinlock: Use __this_cpu_dec() instead of full-blown this_cpu_dec()
locking/atomic, arch/tile: Fix tilepro build
locking/atomic, arch/m68k: Remove comment
locking/atomic, arch/arc: Fix build
locking/Documentation: Clarify limited control-dependency scope
locking/atomic, arch/rwsem: Employ atomic_long_fetch_add()
locking/atomic, arch/qrwlock: Employ atomic_fetch_add_acquire()
locking/atomic, arch/mips: Convert to _relaxed atomics
locking/atomic, arch/alpha: Convert to _relaxed atomics
locking/atomic: Remove the deprecated atomic_{set,clear}_mask() functions
locking/atomic: Remove linux/atomic.h:atomic_fetch_or()
locking/atomic: Implement atomic{,64,_long}_fetch_{add,sub,and,andnot,or,xor}{,_relaxed,_acquire,_release}()
locking/atomic: Fix atomic64_relaxed() bits
locking/atomic, arch/xtensa: Implement atomic_fetch_{add,sub,and,or,xor}()
locking/atomic, arch/x86: Implement atomic{,64}_fetch_{add,sub,and,or,xor}()
locking/atomic, arch/tile: Implement atomic{,64}_fetch_{add,sub,and,or,xor}()
locking/atomic, arch/sparc: Implement atomic{,64}_fetch_{add,sub,and,or,xor}()
...
|
|
Pablo Neira Ayuso says:
====================
Netfilter/IPVS updates for net-next
The following patchset contains Netfilter/IPVS updates for net-next,
they are:
1) Count pre-established connections as active in "least connection"
schedulers such that pre-established connections to avoid overloading
backend servers on peak demands, from Michal Kubecek via Simon Horman.
2) Address a race condition when resizing the conntrack table by caching
the bucket size when fulling iterating over the hashtable in these
three possible scenarios: 1) dump via /proc/net/nf_conntrack,
2) unlinking userspace helper and 3) unlinking custom conntrack timeout.
From Liping Zhang.
3) Revisit early_drop() path to perform lockless traversal on conntrack
eviction under stress, use del_timer() as synchronization point to
avoid two CPUs evicting the same entry, from Florian Westphal.
4) Move NAT hlist_head to nf_conn object, this simplifies the existing
NAT extension and it doesn't increase size since recent patches to
align nf_conn, from Florian.
5) Use rhashtable for the by-source NAT hashtable, also from Florian.
6) Don't allow --physdev-is-out from OUTPUT chain, just like
--physdev-out is not either, from Hangbin Liu.
7) Automagically set on nf_conntrack counters if the user tries to
match ct bytes/packets from nftables, from Liping Zhang.
8) Remove possible_net_t fields in nf_tables set objects since we just
simply pass the net pointer to the backend set type implementations.
9) Fix possible off-by-one in h323, from Toby DiPasquale.
10) early_drop() may be called from ctnetlink patch, so we must hold
rcu read size lock from them too, this amends Florian's patch #3
coming in this batch, from Liping Zhang.
11) Use binary search to validate jump offset in x_tables, this
addresses the O(n!) validation that was introduced recently
resolve security issues with unpriviledge namespaces, from Florian.
12) Fix reference leak to connlabel in error path of nft_ct, from Zhang.
13) Three updates for nft_log: Fix log prefix leak in error path. Bail
out on loglevel larger than debug in nft_log and set on the new
NF_LOG_F_COPY_LEN flag when snaplen is specified. Again from Zhang.
14) Allow to filter rule dumps in nf_tables based on table and chain
names.
15) Simplify connlabel to always use 128 bits to store labels and
get rid of unused function in xt_connlabel, from Florian.
16) Replace set_expect_timeout() by mod_timer() from the h323 conntrack
helper, by Gao Feng.
17) Put back x_tables module reference in nft_compat on error, from
Liping Zhang.
18) Add a reference count to the x_tables extensions cache in
nft_compat, so we can remove them when unused and avoid a crash
if the extensions are rmmod, again from Zhang.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Just several instances of overlapping changes.
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
The clash resolution is not easy to apply if the NAT table is
registered. Even if no NAT rules are installed, the nul-binding ensures
that a unique tuple is used, thus, the packet that loses race gets a
different source port number, as described by:
http://marc.info/?l=netfilter-devel&m=146818011604484&w=2
Clash resolution with NAT is also problematic if addresses/port range
ports are used since the conntrack that wins race may describe a
different mangling that we may have earlier applied to the packet via
nf_nat_setup_info().
Fixes: 71d8c47fc653 ("netfilter: conntrack: introduce clash resolution on insertion race")
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Tested-by: Marc Dionne <marc.c.dionne@gmail.com>
|
|
User can add ct entry via nfnetlink(IPCTNL_MSG_CT_NEW), and if the total
number reach the nf_conntrack_max, we will try to drop some ct entries.
But in this case(the main function call path is ctnetlink_create_conntrack
-> nf_conntrack_alloc -> early_drop), rcu_read_lock is not held, so race
with hash resize will happen.
Fixes: 242922a02717 ("netfilter: conntrack: simplify early_drop")
Cc: Florian Westphal <fw@strlen.de>
Signed-off-by: Liping Zhang <liping.zhang@spreadtrum.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
|
|
We don't need to acquire the bucket lock during early drop, we can
use lockless traveral just like ____nf_conntrack_find.
The timer deletion serves as synchronization point, if another cpu
attempts to evict same entry, only one will succeed with timer deletion.
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
|
|
When we do "cat /proc/net/nf_conntrack", and meanwhile resize the conntrack
hash table via /sys/module/nf_conntrack/parameters/hashsize, race will
happen, because reader can observe a newly allocated hash but the old size
(or vice versa). So oops will happen like follows:
BUG: unable to handle kernel NULL pointer dereference at 0000000000000017
IP: [<ffffffffa0418e21>] seq_print_acct+0x11/0x50 [nf_conntrack]
Call Trace:
[<ffffffffa0412f4e>] ? ct_seq_show+0x14e/0x340 [nf_conntrack]
[<ffffffff81261a1c>] seq_read+0x2cc/0x390
[<ffffffff812a8d62>] proc_reg_read+0x42/0x70
[<ffffffff8123bee7>] __vfs_read+0x37/0x130
[<ffffffff81347980>] ? security_file_permission+0xa0/0xc0
[<ffffffff8123cf75>] vfs_read+0x95/0x140
[<ffffffff8123e475>] SyS_read+0x55/0xc0
[<ffffffff817c2572>] entry_SYSCALL_64_fastpath+0x1a/0xa4
It is very easy to reproduce this kernel crash.
1. open one shell and input the following cmds:
while : ; do
echo $RANDOM > /sys/module/nf_conntrack/parameters/hashsize
done
2. open more shells and input the following cmds:
while : ; do
cat /proc/net/nf_conntrack
done
3. just wait a monent, oops will happen soon.
The solution in this patch is based on Florian's Commit 5e3c61f98175
("netfilter: conntrack: fix lookup race during hash resize"). And
add a wrapper function nf_conntrack_get_ht to get hash and hsize
suggested by Florian Westphal.
Signed-off-by: Liping Zhang <liping.zhang@spreadtrum.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
|
|
Pablo Neira Ayuso says:
====================
Netfilter updates for net-next
The following patchset contains Netfilter updates for net-next,
they are:
1) Don't use userspace datatypes in bridge netfilter code, from
Tobin Harding.
2) Iterate only once over the expectation table when removing the
helper module, instead of once per-netns, from Florian Westphal.
3) Extra sanitization in xt_hook_ops_alloc() to return error in case
we ever pass zero hooks, xt_hook_ops_alloc():
4) Handle NFPROTO_INET from the logging core infrastructure, from
Liping Zhang.
5) Autoload loggers when TRACE target is used from rules, this doesn't
change the behaviour in case the user already selected nfnetlink_log
as preferred way to print tracing logs, also from Liping Zhang.
6) Conntrack slabs with SLAB_HWCACHE_ALIGN to allow rearranging fields
by cache lines, increases the size of entries in 11% per entry.
From Florian Westphal.
7) Skip zone comparison if CONFIG_NF_CONNTRACK_ZONES=n, from Florian.
8) Remove useless defensive check in nf_logger_find_get() from Shivani
Bhardwaj.
9) Remove zone extension as place it in the conntrack object, this is
always include in the hashing and we expect more intensive use of
zones since containers are in place. Also from Florian Westphal.
10) Owner match now works from any namespace, from Eric Bierdeman.
11) Make sure we only reply with TCP reset to TCP traffic from
nf_reject_ipv4, patch from Liping Zhang.
12) Introduce --nflog-size to indicate amount of network packet bytes
that are copied to userspace via log message, from Vishwanath Pai.
This obsoletes --nflog-range that has never worked, it was designed
to achieve this but it has never worked.
13) Introduce generic macros for nf_tables object generation masks.
14) Use generation mask in table, chain and set objects in nf_tables.
This allows fixes interferences with ongoing preparation phase of
the commit protocol and object listings going on at the same time.
This update is introduced in three patches, one per object.
15) Check if the object is active in the next generation for element
deactivation in the rbtree implementation, given that deactivation
happens from the commit phase path we have to observe the future
status of the object.
16) Support for deletion of just added elements in the hash set type.
17) Allow to resize hashtable from /proc entry, not only from the
obscure /sys entry that maps to the module parameter, from Florian
Westphal.
18) Get rid of NFT_BASECHAIN_DISABLED, this code is not exercised
anymore since we tear down the ruleset whenever the netdevice
goes away.
19) Support for matching inverted set lookups, from Arturo Borrero.
20) Simplify the iptables_mangle_hook() by removing a superfluous
extra branch.
21) Introduce ether_addr_equal_masked() and use it from the netfilter
codebase, from Joe Perches.
22) Remove references to "Use netfilter MARK value as routing key"
from the Netfilter Kconfig description given that this toggle
doesn't exists already for 10 years, from Moritz Sichert.
23) Introduce generic NF_INVF() and use it from the xtables codebase,
from Joe Perches.
24) Setting logger to NONE via /proc was not working unless explicit
nul-termination was included in the string. This fixes seems to
leave the former behaviour there, so we don't break backward.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Can overflow so we might allocate very small table when bucket count is
high on a 32bit platform.
Note: resize is only possible from init_netns.
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
|
|
No need to restrict this to module parameter.
We export a copy of the real hash size -- when user alters the value we
allocate the new table, copy entries etc before we update the real size
to the requested one.
This is also needed because the real size is used by concurrent readers
and cannot be changed without synchronizing the conntrack generation
seqcnt.
We only allow changing this value from the initial net namespace.
Tested using http-client-benchmark vs. httpterm with concurrent
while true;do
echo $RANDOM > /proc/sys/net/netfilter/nf_conntrack_buckets
done
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
|
|
Curently we store zone information as a conntrack extension.
This has one drawback: for every lookup we need to fetch the zone data
from the extension area.
This change place the zone data directly into the main conntrack object
structure and then removes the zone conntrack extension.
The zone data is just 4 bytes, it fits into a padding hole before
the tuplehash info, so we do not even increase the nf_conn structure size.
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
|
|
increases struct size by 32 bytes (288 -> 320), but it is the right thing,
else any attempt to (re-)arrange nf_conn members by cacheline won't work.
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
|
|
I forgot to move the kmem_cache_destroy into the exit path.
Fixes: 0c5366b3a8c7 ("netfilter: conntrack: use single slab cache)
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
|
|
Even with spin_unlock_wait() fixed, nf_conntrack_lock{,_all}() is
borken as it misses a bunch of memory barriers to order the whole
global vs local locks scheme.
Even x86 (and other TSO archs) are affected.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
[ Updated the comments. ]
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: David S. Miller <davem@davemloft.net>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
|
|
The nf_conntrack_core.c fix in 'net' is not relevant in 'net-next'
because we no longer have a per-netns conntrack hash.
The ip_gre.c conflict as well as the iwlwifi ones were cases of
overlapping changes.
Conflicts:
drivers/net/wireless/intel/iwlwifi/mvm/tx.c
net/ipv4/ip_gre.c
net/netfilter/nf_conntrack_core.c
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
The slab name ends up being visible in the directory structure under
/sys, and even if you don't have access rights to the file you can see
the filenames.
Just use a 64-bit counter instead of the pointer to the 'net' structure
to generate a unique name.
This code will go away in 4.7 when the conntrack code moves to a single
kmemcache, but this is the backportable simple solution to avoiding
leaking kernel pointers to user space.
Fixes: 5b3501faa874 ("netfilter: nf_conntrack: per netns nf_conntrack_cachep")
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Acked-by: Eric Dumazet <eric.dumazet@gmail.com>
Cc: stable@vger.kernel.org
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
A recent commit introduced an unconditional use of an uninitialized
variable, as reported in this gcc warning:
net/netfilter/nf_conntrack_core.c: In function '__nf_conntrack_confirm':
net/netfilter/nf_conntrack_core.c:632:33: error: 'ctinfo' may be used uninitialized in this function [-Werror=maybe-uninitialized]
bytes = atomic64_read(&counter[CTINFO2DIR(ctinfo)].bytes);
^
net/netfilter/nf_conntrack_core.c:628:26: note: 'ctinfo' was declared here
enum ip_conntrack_info ctinfo;
The problem is that a local variable shadows the function parameter.
This removes the local variable, which looks like what Pablo originally
intended.
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Fixes: 71d8c47fc653 ("netfilter: conntrack: introduce clash resolution on insertion race")
Acked-by: Pablo Neira Ayuso <pablo@netfilter.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
An earlier patch changed lookup side to also net_eq() namespaces after
obtaining a reference on the conntrack, so a single kmemcache can be used.
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
|
|
This patch introduces nf_ct_resolve_clash() to resolve race condition on
conntrack insertions.
This is particularly a problem for connection-less protocols such as
UDP, with no initial handshake. Two or more packets may race to insert
the entry resulting in packet drops.
Another problematic scenario are packets enqueued to userspace via
NFQUEUE after the raw table, that make it easier to trigger this
race.
To resolve this, the idea is to reset the conntrack entry to the one
that won race. Packet and bytes counters are also merged.
The 'insert_failed' stats still accounts for this situation, after
this patch, the drop counter is bumped whenever we drop packets, so we
can watch for unresolved clashes.
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
|
|
Introduce a helper function to update conntrack counters.
__nf_ct_kill_acct() was unnecessarily subtracting skb_network_offset()
that is expected to be zero from the ipv4/ipv6 hooks.
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
|
|
Remove unnecessary check for non-nul pointer in destroy_conntrack()
given that __nf_ct_l4proto_find() returns the generic protocol tracker
if the protocol is not supported.
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
|
|
When iterating, skip conntrack entries living in a different netns.
We could ignore netns and kill some other non-assured one, but it
has two problems:
- a netns can kill non-assured conntracks in other namespace
- we would start to 'over-subscribe' the affected/overlimit netns.
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
|
|
We already include netns address in the hash and compare the netns pointers
during lookup, so even if namespaces have overlapping addresses entries
will be spread across the table.
Assuming 64k bucket size, this change saves 0.5 mbyte per namespace on a
64bit system.
NAT bysrc and expectation hash is still per namespace, those will
changed too soon.
Future patch will also make conntrack object slab cache global again.
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
|
|
Once we place all conntracks into a global hash table we want them to be
spread across entire hash table, even if namespaces have overlapping ip
addresses.
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
|
|
Once we place all conntracks in the same hash table we must also compare
the netns pointer to skip conntracks that belong to a different namespace.
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
|
|
This prepares for upcoming change that places all conntracks into a
single, global table. For this to work we will need to also compare
net pointer during lookup. To avoid open-coding such check use the
nf_ct_key_equal helper and then later extend it to also consider net_eq.
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
|
|
Once we place all conntracks into same table iteration becomes more
costly because the table contains conntracks that we are not interested
in (belonging to other netns).
So don't bother scanning if the current namespace has no entries.
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
|
|
When resizing the conntrack hash table at runtime via
echo 42 > /sys/module/nf_conntrack/parameters/hashsize, we are racing with
the conntrack lookup path -- reads can happen in parallel and nothing
prevents readers from observing a the newly allocated hash but the old
size (or vice versa).
So access to hash[bucket] can trigger OOB read access in case the table got
expanded and we saw the new size but the old hash pointer (or it got shrunk
and we got new hash ptr but the size of the old and larger table):
kasan: GPF could be caused by NULL-ptr deref or user memory access general protection fault: 0000 [#1] SMP KASAN
CPU: 0 PID: 3 Comm: ksoftirqd/0 Not tainted 4.6.0-rc2+ #107
[..]
Call Trace:
[<ffffffff822c3d6a>] ? nf_conntrack_tuple_taken+0x12a/0xe90
[<ffffffff822c3ac1>] ? nf_ct_invert_tuplepr+0x221/0x3a0
[<ffffffff8230e703>] get_unique_tuple+0xfb3/0x2760
Use generation counter to obtain the address/length of the same table.
Also add a synchronize_net before freeing the old hash.
AFAICS, without it we might access ct_hash[bucket] after ct_hash has been
freed, provided that lockless reader got delayed by another event:
CPU1 CPU2
seq_begin
seq_retry
<delay> resize occurs
free oldhash
for_each(oldhash[size])
Note that resize is only supported in init_netns, it took over 2 minutes
of constant resizing+flooding to produce the warning, so this isn't a
big problem in practice.
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
|