linux-stable.git/include/net, branch v3.16.52

tcp: fix tcp_mtu_probe() vs highest_sack

2018-01-01T20:52:02+00:00

commit 2b7cda9c35d3b940eb9ce74b30bbd5eb30db493d upstream.

Based on SNMP values provided by Roman, Yuchung made the observation
that some crashes in tcp_sacktag_walk() might be caused by MTU probing.

Looking at tcp_mtu_probe(), I found that when a new skb was placed
in front of the write queue, we were not updating tcp highest sack.

If one skb is freed because all its content was copied to the new skb
(for MTU probing), then tp->highest_sack could point to a now freed skb.

Bad things would then happen, including infinite loops.

This patch renames tcp_highest_sack_combine() and uses it
from tcp_mtu_probe() to fix the bug.

Note that I also removed one test against tp->sacked_out,
since we want to replace tp->highest_sack regardless of whatever
condition, since keeping a stale pointer to freed skb is a recipe
for disaster.

Fixes: a47e5a988a57 ("[TCP]: Convert highest_sack to sk_buff to allow direct access")
Signed-off-by: Eric Dumazet 
Reported-by: Alexei Starovoitov 
Reported-by: Roman Gushchin 
Reported-by: Oleksandr Natalenko 
Acked-by: Alexei Starovoitov 
Acked-by: Neal Cardwell 
Acked-by: Yuchung Cheng 
Signed-off-by: David S. Miller 
[bwh: Backported to 3.16: adjust context]
Signed-off-by: Ben Hutchings

udp: perform source validation for mcast early demux

2018-01-01T20:51:45+00:00

commit bc044e8db7962e727a75b591b9851ff2ac5cf846 upstream.

The UDP early demux can leverate the rx dst cache even for
multicast unconnected sockets.

In such scenario the ipv4 source address is validated only on
the first packet in the given flow. After that, when we fetch
the dst entry  from the socket rx cache, we stop enforcing
the rp_filter and we even start accepting any kind of martian
addresses.

Disabling the dst cache for unconnected multicast socket will
cause large performace regression, nearly reducing by half the
max ingress tput.

Instead we factor out a route helper to completely validate an
skb source address for multicast packets and we call it from
the UDP early demux for mcast packets landing on unconnected
sockets, after successful fetching the related cached dst entry.

This still gives a measurable, but limited performance
regression:

		rp_filter = 0		rp_filter = 1
edmux disabled:	1182 Kpps		1127 Kpps
edmux before:	2238 Kpps		2238 Kpps
edmux after:	2037 Kpps		2019 Kpps

The above figures are on top of current net tree.
Applying the net-next commit 6e617de84e87 ("net: avoid a full
fib lookup when rp_filter is disabled.") the delta with
rp_filter == 0 will decrease even more.

Fixes: 421b3885bf6d ("udp: ipv4: Add udp early demux")
Signed-off-by: Paolo Abeni 
Signed-off-by: David S. Miller 
Signed-off-by: Ben Hutchings

IPv4: early demux can return an error code

2018-01-01T20:51:45+00:00

commit 7487449c86c65202b3b725c4524cb48dd65e4e6f upstream.

Currently no error is emitted, but this infrastructure will
used by the next patch to allow source address validation
for mcast sockets.
Since early demux can do a route lookup and an ipv4 route
lookup can return an error code this is consistent with the
current ipv4 route infrastructure.

Signed-off-by: Paolo Abeni 
Signed-off-by: David S. Miller 
[bwh: Backported to 3.16:
 - Drop change to net_protocol::early_demux_handler
 - Keep using NET_INC_STATS_BH() in ip_rcv_finish()
 - Fix up additional return statement in udp_v4_early_demux()
 - Adjust context]
Signed-off-by: Ben Hutchings

Revert "net: use lib/percpu_counter API for fragmentation mem accounting"

2017-11-26T13:50:30+00:00

commit fb452a1aa3fd4034d7999e309c5466ff2d7005aa upstream.

This reverts commit 6d7b857d541ecd1d9bd997c97242d4ef94b19de2.

There is a bug in fragmentation codes use of the percpu_counter API,
that can cause issues on systems with many CPUs.

The frag_mem_limit() just reads the global counter (fbc->count),
without considering other CPUs can have upto batch size (130K) that
haven't been subtracted yet.  Due to the 3MBytes lower thresh limit,
this become dangerous at >=24 CPUs (3*1024*1024/130000=24).

The correct API usage would be to use __percpu_counter_compare() which
does the right thing, and takes into account the number of (online)
CPUs and batch size, to account for this and call __percpu_counter_sum()
when needed.

We choose to revert the use of the lib/percpu_counter API for frag
memory accounting for several reasons:

1) On systems with CPUs > 24, the heavier fully locked
   __percpu_counter_sum() is always invoked, which will be more
   expensive than the atomic_t that is reverted to.

Given systems with more than 24 CPUs are becoming common this doesn't
seem like a good option.  To mitigate this, the batch size could be
decreased and thresh be increased.

2) The add_frag_mem_limit+sub_frag_mem_limit pairs happen on the RX
   CPU, before SKBs are pushed into sockets on remote CPUs.  Given
   NICs can only hash on L2 part of the IP-header, the NIC-RXq's will
   likely be limited.  Thus, a fair chance that atomic add+dec happen
   on the same CPU.

Revert note that commit 1d6119baf061 ("net: fix percpu memory leaks")
removed init_frag_mem_limit() and instead use inet_frags_init_net().
After this revert, inet_frags_uninit_net() becomes empty.

Fixes: 6d7b857d541e ("net: use lib/percpu_counter API for fragmentation mem accounting")
Fixes: 1d6119baf061 ("net: fix percpu memory leaks")
Signed-off-by: Jesper Dangaard Brouer 
Acked-by: Florian Westphal 
Signed-off-by: David S. Miller 
[bwh: Backported to 3.16: adjust context]
Signed-off-by: Ben Hutchings

ipv6: fix sparse warning on rt6i_node

2017-11-11T13:33:32+00:00

commit 4e587ea71bf924f7dac621f1351653bd41e446cb upstream.

Commit c5cff8561d2d adds rcu grace period before freeing fib6_node. This
generates a new sparse warning on rt->rt6i_node related code:
  net/ipv6/route.c:1394:30: error: incompatible types in comparison
  expression (different address spaces)
  ./include/net/ip6_fib.h:187:14: error: incompatible types in comparison
  expression (different address spaces)

This commit adds "__rcu" tag for rt6i_node and makes sure corresponding
rcu API is used for it.
After this fix, sparse no longer generates the above warning.

Fixes: c5cff8561d2d ("ipv6: add rcu grace period before freeing fib6_node")
Signed-off-by: Wei Wang 
Acked-by: Eric Dumazet 
Acked-by: Martin KaFai Lau 
Signed-off-by: David S. Miller 
[bwh: Backported to 3.16: drop changes in rt6_cache_allowed_for_pmtu()]
Signed-off-by: Ben Hutchings

ipv6: add rcu grace period before freeing fib6_node

2017-11-11T13:33:28+00:00

commit c5cff8561d2d0006e972bd114afd51f082fee77c upstream.

We currently keep rt->rt6i_node pointing to the fib6_node for the route.
And some functions make use of this pointer to dereference the fib6_node
from rt structure, e.g. rt6_check(). However, as there is neither
refcount nor rcu taken when dereferencing rt->rt6i_node, it could
potentially cause crashes as rt->rt6i_node could be set to NULL by other
CPUs when doing a route deletion.
This patch introduces an rcu grace period before freeing fib6_node and
makes sure the functions that dereference it takes rcu_read_lock().

Note: there is no "Fixes" tag because this bug was there in a very
early stage.

Signed-off-by: Wei Wang 
Acked-by: Eric Dumazet 
Acked-by: Martin KaFai Lau 
Signed-off-by: David S. Miller 
[bwh: Backported to 3.16: adjust context]
Signed-off-by: Ben Hutchings

ipv6: Add rt6_get_cookie() function

2017-11-11T13:33:28+00:00

commit b197df4f0f3782782e9ea8996e91b65ae33e8dd9 upstream.

Instead of doing the rt6->rt6i_node check whenever we need
to get the route's cookie.  Refactor it into rt6_get_cookie().
It is a prep work to handle FLOWI_FLAG_KNOWN_NH and also
percpu rt6_info later.

Signed-off-by: Martin KaFai Lau 
Cc: Hannes Frederic Sowa 
Cc: Steffen Klassert 
Cc: Julian Anastasov 
Signed-off-by: David S. Miller 
[bwh: Backported to 3.16: adjust context]
Signed-off-by: Ben Hutchings

net_sched: fix order of queue length updates in qdisc_replace()

2017-11-11T13:33:23+00:00

commit 68a66d149a8c78ec6720f268597302883e48e9fa upstream.

This important to call qdisc_tree_reduce_backlog() after changing queue
length. Parent qdisc should deactivate class in ->qlen_notify() called from
qdisc_tree_reduce_backlog() but this happens only if qdisc->q.qlen in zero.

Missed class deactivations leads to crashes/warnings at picking packets
from empty qdisc and corrupting state at reactivating this class in future.

Signed-off-by: Konstantin Khlebnikov 
Fixes: 86a7996cc8a0 ("net_sched: introduce qdisc_replace() helper")
Acked-by: Cong Wang 
Signed-off-by: David S. Miller 
Signed-off-by: Ben Hutchings

ipv4: add reference counting to metrics

2017-11-11T13:33:21+00:00

commit 3fb07daff8e99243366a081e5129560734de4ada upstream.

Andrey Konovalov reported crashes in ipv4_mtu()

I could reproduce the issue with KASAN kernels, between
10.246.7.151 and 10.246.7.152 :

1) 20 concurrent netperf -t TCP_RR -H 10.246.7.152 -l 1000 &

2) At the same time run following loop :
while :
do
 ip ro add 10.246.7.152 dev eth0 src 10.246.7.151 mtu 1500
 ip ro del 10.246.7.152 dev eth0 src 10.246.7.151 mtu 1500
done

Cong Wang attempted to add back rt->fi in commit
82486aa6f1b9 ("ipv4: restore rt->fi for reference counting")
but this proved to add some issues that were complex to solve.

Instead, I suggested to add a refcount to the metrics themselves,
being a standalone object (in particular, no reference to other objects)

I tried to make this patch as small as possible to ease its backport,
instead of being super clean. Note that we believe that only ipv4 dst
need to take care of the metric refcount. But if this is wrong,
this patch adds the basic infrastructure to extend this to other
families.

Many thanks to Julian Anastasov for reviewing this patch, and Cong Wang
for his efforts on this problem.

Fixes: 2860583fe840 ("ipv4: Kill rt->fi")
Signed-off-by: Eric Dumazet 
Reported-by: Andrey Konovalov 
Reviewed-by: Julian Anastasov 
Acked-by: Cong Wang 
Signed-off-by: David S. Miller 
[bwh: Backported to 3.16:
 - Give DST_METRICS_REFCOUNTED a value of 4 and struct dst_metrics an
   alignment of 8
 - Update dst_metrics cache size
 - Adjust context]
Signed-off-by: Ben Hutchings

dst: Increase alignment of metrics to allow extra flag on pointers

2017-11-11T13:33:20+00:00

For the backport of "ipv4: add reference counting to metrics", we will
need a third flag on metrics pointers.  This was not needed upstream
as the DST_METRICS_FORCE_OVERWRITE flag has been eliminated there.
In order to use three flag bits we need to increase the alignment of
metrics from 4 to 8 bytes.

Signed-off-by: Ben Hutchings