linux-stable.git/net/ipv4, branch v5.4.124

net: Only allow init netns to set default tcp cong to a restricted algo

2021-05-14T07:44:33+00:00

commit 8d432592f30fcc34ef5a10aac4887b4897884493 upstream.

tcp_set_default_congestion_control() is netns-safe in that it writes
to &net->ipv4.tcp_congestion_control, but it also sets
ca->flags |= TCP_CONG_NON_RESTRICTED which is not namespaced.
This has the unintended side-effect of changing the global
net.ipv4.tcp_allowed_congestion_control sysctl, despite the fact that it
is read-only: 97684f0970f6 ("net: Make tcp_allowed_congestion_control
readonly in non-init netns")

Resolve this netns "leak" by only allowing the init netns to set the
default algorithm to one that is restricted. This restriction could be
removed if tcp_allowed_congestion_control were namespace-ified in the
future.

This bug was uncovered with
https://github.com/JonathonReinhart/linux-netns-sysctl-verify

Fixes: 6670e1524477 ("tcp: Namespace-ify sysctl_tcp_default_congestion_control")
Signed-off-by: Jonathon Reinhart 
Signed-off-by: David S. Miller 
Signed-off-by: Greg Kroah-Hartman

inet: use bigger hash table for IP ID generation

2021-05-14T07:44:26+00:00

[ Upstream commit aa6dd211e4b1dde9d5dc25d699d35f789ae7eeba ]

In commit 73f156a6e8c1 ("inetpeer: get rid of ip_id_count")
I used a very small hash table that could be abused
by patient attackers to reveal sensitive information.

Switch to a dynamic sizing, depending on RAM size.

Typical big hosts will now use 128x more storage (2 MB)
to get a similar increase in security and reduction
of hash collisions.

As a bonus, use of alloc_large_system_hash() spreads
allocated memory among all NUMA nodes.

Fixes: 73f156a6e8c1 ("inetpeer: get rid of ip_id_count")
Reported-by: Amit Klein 
Signed-off-by: Eric Dumazet 
Cc: Willy Tarreau 
Signed-off-by: David S. Miller 
Signed-off-by: Sasha Levin

netfilter: arp_tables: add pre_exit hook for table unregister

2021-04-21T10:56:16+00:00

commit d163a925ebbc6eb5b562b0f1d72c7e817aa75c40 upstream.

Same problem that also existed in iptables/ip(6)tables, when
arptable_filter is removed there is no longer a wait period before the
table/ruleset is free'd.

Unregister the hook in pre_exit, then remove the table in the exit
function.
This used to work correctly because the old nf_hook_unregister API
did unconditional synchronize_net.

The per-net hook unregister function uses call_rcu instead.

Fixes: b9e69e127397 ("netfilter: xtables: don't hook tables by default")
Signed-off-by: Florian Westphal 
Signed-off-by: Pablo Neira Ayuso 
Signed-off-by: Greg Kroah-Hartman

netfilter: x_tables: fix compat match/target pad out-of-bound write

2021-04-16T09:46:38+00:00

commit b29c457a6511435960115c0f548c4360d5f4801d upstream.

xt_compat_match/target_from_user doesn't check that zeroing the area
to start of next rule won't write past end of allocated ruleset blob.

Remove this code and zero the entire blob beforehand.

Reported-by: syzbot+cfc0247ac173f597aaaa@syzkaller.appspotmail.com
Reported-by: Andy Nguyen 
Fixes: 9fa492cdc160c ("[NETFILTER]: x_tables: simplify compat API")
Signed-off-by: Florian Westphal 
Signed-off-by: Pablo Neira Ayuso 
Signed-off-by: Greg Kroah-Hartman

net: udp: Add support for getsockopt(..., ..., UDP_GRO, ..., ...);

2021-04-14T06:24:16+00:00

[ Upstream commit 98184612aca0a9ee42b8eb0262a49900ee9eef0d ]

Support for UDP_GRO was added in the past but the implementation for
getsockopt was missed which did lead to an error when we tried to
retrieve the setting for UDP_GRO. This patch adds the missing switch
case for UDP_GRO

Fixes: e20cf8d3f1f7 ("udp: implement GRO for plain UDP sockets.")
Signed-off-by: Norman Maurer 
Reviewed-by: David Ahern 
Signed-off-by: David S. Miller 
Signed-off-by: Sasha Levin

esp: delete NETIF_F_SCTP_CRC bit from features for esp offload

2021-04-14T06:24:13+00:00

[ Upstream commit 154deab6a3ba47792936edf77f2f13a1cbc4351d ]

Now in esp4/6_gso_segment(), before calling inner proto .gso_segment,
NETIF_F_CSUM_MASK bits are deleted, as HW won't be able to do the
csum for inner proto due to the packet encrypted already.

So the UDP/TCP packet has to do the checksum on its own .gso_segment.
But SCTP is using CRC checksum, and for that NETIF_F_SCTP_CRC should
be deleted to make SCTP do the csum in own .gso_segment as well.

In Xiumei's testing with SCTP over IPsec/veth, the packets are kept
dropping due to the wrong CRC checksum.

Reported-by: Xiumei Mu 
Fixes: 7862b4058b9f ("esp: Add gso handlers for esp4 and esp6")
Signed-off-by: Xin Long 
Signed-off-by: Steffen Klassert 
Signed-off-by: Sasha Levin

Revert "netfilter: x_tables: Update remaining dereference to RCU"

2021-03-30T12:35:28+00:00

[ Upstream commit abe7034b9a8d57737e80cc16d60ed3666990bdbf ]

This reverts commit 443d6e86f821a165fae3fc3fc13086d27ac140b1.

This (and the following) patch basically re-implemented the RCU
mechanisms of patch 784544739a25. That patch was replaced because of the
performance problems that it created when replacing tables. Now, we have
the same issue: the call to synchronize_rcu() makes replacing tables
slower by as much as an order of magnitude.

Revert these patches and fix the issue in a different way.

Signed-off-by: Mark Tomlinson 
Signed-off-by: Pablo Neira Ayuso 
Signed-off-by: Sasha Levin

Revert "netfilter: x_tables: Switch synchronization to RCU"

2021-03-30T12:35:28+00:00

[ Upstream commit d3d40f237480abf3268956daf18cdc56edd32834 ]

This reverts commit cc00bcaa589914096edef7fb87ca5cee4a166b5c.

This (and the preceding) patch basically re-implemented the RCU
mechanisms of patch 784544739a25. That patch was replaced because of the
performance problems that it created when replacing tables. Now, we have
the same issue: the call to synchronize_rcu() makes replacing tables
slower by as much as an order of magnitude.

Prior to using RCU a script calling "iptables" approx. 200 times was
taking 1.16s. With RCU this increased to 11.59s.

Revert these patches and fix the issue in a different way.

Signed-off-by: Mark Tomlinson 
Signed-off-by: Pablo Neira Ayuso 
Signed-off-by: Sasha Levin

tcp: relookup sock for RST+ACK packets handled by obsolete req sock

2021-03-30T12:35:26+00:00

[ Upstream commit 7233da86697efef41288f8b713c10c2499cffe85 ]

Currently tcp_check_req can be called with obsolete req socket for which big
socket have been already created (because of CPU race or early demux
assigning req socket to multiple packets in gro batch).

Commit e0f9759f530bf789e984 ("tcp: try to keep packet if SYN_RCV race
is lost") added retry in case when tcp_check_req is called for PSH|ACK packet.
But if client sends RST+ACK immediatly after connection being
established (it is performing healthcheck, for example) retry does not
occur. In that case tcp_check_req tries to close req socket,
leaving big socket active.

Fixes: e0f9759f530 ("tcp: try to keep packet if SYN_RCV race is lost")
Signed-off-by: Alexander Ovechkin 
Reported-by: Oleg Senin 
Reviewed-by: Eric Dumazet 
Signed-off-by: David S. Miller 
Signed-off-by: Sasha Levin

cipso,calipso: resolve a number of problems with the DOI refcounts

2021-03-17T16:03:35+00:00

commit ad5d07f4a9cd671233ae20983848874731102c08 upstream.

The current CIPSO and CALIPSO refcounting scheme for the DOI
definitions is a bit flawed in that we:

1. Don't correctly match gets/puts in netlbl_cipsov4_list().
2. Decrement the refcount on each attempt to remove the DOI from the
   DOI list, only removing it from the list once the refcount drops
   to zero.

This patch fixes these problems by adding the missing "puts" to
netlbl_cipsov4_list() and introduces a more conventional, i.e.
not-buggy, refcounting mechanism to the DOI definitions.  Upon the
addition of a DOI to the DOI list, it is initialized with a refcount
of one, removing a DOI from the list removes it from the list and
drops the refcount by one; "gets" and "puts" behave as expected with
respect to refcounts, increasing and decreasing the DOI's refcount by
one.

Fixes: b1edeb102397 ("netlabel: Replace protocol/NetLabel linking with refrerence counts")
Fixes: d7cce01504a0 ("netlabel: Add support for removing a CALIPSO DOI.")
Reported-by: syzbot+9ec037722d2603a9f52e@syzkaller.appspotmail.com
Signed-off-by: Paul Moore 
Signed-off-by: David S. Miller 
Signed-off-by: Greg Kroah-Hartman