Caching saddr and daddr before pskb_pull() is problematic
since skb->head can change.
Remove these temporary variables:
- We only access &ipv6_hdr(skb)->saddr and &ipv6_hdr(skb)->daddr
when net_dbg_ratelimited() is called in the slow path.
- Avoid potential future misuse after pskb_pull() call.
Fixes: 4b3418fba0 ("ipv6: icmp: include addresses in debug messages")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Fernando Fernandez Mancera <fmancera@suse.de>
Reviewed-by: Joe Damato <joe@dama.to>
Reviewed-by: Ido Schimmel <idosch@nvidia.com>
Link: https://patch.msgid.link/20260416103505.2380753-1-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
As IPv6 is built-in only, there is no need to maintain the sender
registration infrastructure used to allow built-in subsystems to send
ICMPv6 messages when IPv6 was compiled as a module.
Drop the registration mechanism and the __icmpv6_send() sender
implementation. While icmpv6_send() users could be converted to
icmp6_send() that doesn't seems necessary as none of them are using the
force_saddr parameter.
Signed-off-by: Fernando Fernandez Mancera <fmancera@suse.de>
Tested-by: Ricardo B. Marlière <rbm@suse.com>
Link: https://patch.msgid.link/20260325120928.15848-5-fmancera@suse.de
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Sashiko AI-review observed:
In ip6_err_gen_icmpv6_unreach(), the skb is an outer IPv4 ICMP error packet
where its cb contains an IPv4 inet_skb_parm. When skb is cloned into skb2
and passed to icmp6_send(), it uses IP6CB(skb2).
IP6CB interprets the IPv4 inet_skb_parm as an inet6_skb_parm. The cipso
offset in inet_skb_parm.opt directly overlaps with dsthao in inet6_skb_parm
at offset 18.
If an attacker sends a forged ICMPv4 error with a CIPSO IP option, dsthao
would be a non-zero offset. Inside icmp6_send(), mip6_addr_swap() is called
and uses ipv6_find_tlv(skb, opt->dsthao, IPV6_TLV_HAO).
This would scan the inner, attacker-controlled IPv6 packet starting at that
offset, potentially returning a fake TLV without checking if the remaining
packet length can hold the full 18-byte struct ipv6_destopt_hao.
Could mip6_addr_swap() then perform a 16-byte swap that extends past the end
of the packet data into skb_shared_info?
Should the cb array also be cleared in ip6_err_gen_icmpv6_unreach() and
ip6ip6_err() to prevent this?
This patch implements the first suggestion.
I am not sure if ip6ip6_err() needs to be changed.
A separate patch would be better anyway.
Fixes: ca15a078bd ("sit: generate icmpv6 error when receiving icmpv4 error")
Reported-by: Ido Schimmel <idosch@nvidia.com>
Closes: https://sashiko.dev/#/patchset/20260326155138.2429480-1-edumazet%40google.com
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Oskar Kjos <oskar.kjos@hotmail.com>
Reviewed-by: Ido Schimmel <idosch@nvidia.com>
Link: https://patch.msgid.link/20260326202608.2976021-1-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Following part was needed before the blamed commit, because
inet_getpeer_v6() second argument was the prefix.
/* Give more bandwidth to wider prefixes. */
if (rt->rt6i_dst.plen < 128)
tmo >>= ((128 - rt->rt6i_dst.plen)>>5);
Now inet_getpeer_v6() retrieves hosts, we need to remove
@tmo adjustement or wider prefixes likes /24 allow 8x
more ICMP to be sent for a given ratelimit.
As we had this issue for a while, this patch changes net.ipv6.icmp.ratelimit
default value from 1000ms to 100ms to avoid potential regressions.
Also add a READ_ONCE() when reading net->ipv6.sysctl.icmpv6_time.
Fixes: fd0273d793 ("ipv6: Remove external dependency on rt6i_dst and rt6i_src")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Kuniyuki Iwashima <kuniyu@google.com>
Cc: Martin KaFai Lau <martin.lau@kernel.org>
Link: https://patch.msgid.link/20260216142832.3834174-4-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Yizhou Zhao reported that simply having one RAW socket on protocol
IPPROTO_RAW (255) was dangerous.
socket(AF_INET, SOCK_RAW, 255);
A malicious incoming ICMP packet can set the protocol field to 255
and match this socket, leading to FNHE cache changes.
inner = IP(src="192.168.2.1", dst="8.8.8.8", proto=255)/Raw("TEST")
pkt = IP(src="192.168.1.1", dst="192.168.2.1")/ICMP(type=3, code=4, nexthopmtu=576)/inner
"man 7 raw" states:
A protocol of IPPROTO_RAW implies enabled IP_HDRINCL and is able
to send any IP protocol that is specified in the passed header.
Receiving of all IP protocols via IPPROTO_RAW is not possible
using raw sockets.
Make sure we drop these malicious packets.
Fixes: 1da177e4c3 ("Linux-2.6.12-rc2")
Reported-by: Yizhou Zhao <zhaoyz24@mails.tsinghua.edu.cn>
Link: https://lore.kernel.org/netdev/20251109134600.292125-1-zhaoyz24@mails.tsinghua.edu.cn/
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: David Ahern <dsahern@kernel.org>
Reviewed-by: Ido Schimmel <idosch@nvidia.com>
Link: https://patch.msgid.link/20260203192509.682208-1-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
When replying to a ICMPv6 echo request that comes from localhost address
the right output ifindex is 1 (lo) and not rt6i_idev dev index. Use the
skb device ifindex instead. This fixes pinging to a local address from
localhost source address.
$ ping6 -I ::1 2001:1:1::2 -c 3
PING 2001:1:1::2 (2001:1:1::2) from ::1 : 56 data bytes
64 bytes from 2001:1:1::2: icmp_seq=1 ttl=64 time=0.037 ms
64 bytes from 2001:1:1::2: icmp_seq=2 ttl=64 time=0.069 ms
64 bytes from 2001:1:1::2: icmp_seq=3 ttl=64 time=0.122 ms
2001:1:1::2 ping statistics
3 packets transmitted, 3 received, 0% packet loss, time 2032ms
rtt min/avg/max/mdev = 0.037/0.076/0.122/0.035 ms
Fixes: 1b70d792cf ("ipv6: Use rt6i_idev index for echo replies to a local address")
Signed-off-by: Fernando Fernandez Mancera <fmancera@suse.de>
Reviewed-by: David Ahern <dsahern@kernel.org>
Link: https://patch.msgid.link/20260121194409.6749-1-fmancera@suse.de
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Add the ability to append the incoming IP interface information to
ICMPv6 error messages in accordance with RFC 5837 and RFC 4884. This is
required for more meaningful traceroute results in unnumbered networks.
The feature is disabled by default and controlled via a new sysctl
("net.ipv6.icmp.errors_extension_mask") which accepts a bitmask of ICMP
extensions to append to ICMP error messages. Currently, only a single
value is supported, but the interface and the implementation should be
able to support more extensions, if needed.
Clone the skb and copy the relevant data portions before modifying the
skb as the caller of icmp6_send() still owns the skb after the function
returns. This should be fine since by default ICMP error messages are
rate limited to 1000 per second and no more than 1 per second per
specific host.
Trim or pad the packet to 128 bytes before appending the ICMP extension
structure in order to be compatible with legacy applications that assume
that the ICMP extension structure always starts at this offset (the
minimum length specified by RFC 4884).
Since commit 20e1954fe2 ("ipv6: RFC 4884 partial support for SIT/GRE
tunnels") it is possible for icmp6_send() to be called with an skb that
already contains ICMP extensions. This can happen when we receive an
ICMPv4 message with extensions from a tunnel and translate it to an
ICMPv6 message towards an IPv6 host in the overlay network. I could not
find an RFC that supports this behavior, but it makes sense to not
overwrite the original extensions that were appended to the packet.
Therefore, avoid appending extensions if the length field in the
provided ICMPv6 header is already filled.
Export netdev_copy_name() using EXPORT_IPV6_MOD_GPL() to make it
available to IPv6 when it is built as a module.
Reviewed-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: David Ahern <dsahern@kernel.org>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Link: https://patch.msgid.link/20251027082232.232571-3-idosch@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Refactor icmpv6_xrlim_allow() and ip6_dst_hoplimit()
so that we acquire rcu_read_lock() a bit longer
to be able to use dst_dev_rcu() instead of dst_dev().
__ip6_rt_update_pmtu() and rt6_do_redirect can directly
use dst_dev_rcu() in sections already holding rcu_read_lock().
Small changes to use dst_dev_net_rcu() in
ip6_default_advmss(), ipv6_sock_ac_join(),
ip6_mc_find_dev() and ndisc_send_skb().
Fixes: 4a6ce2b6f2 ("net: introduce a new function dst_dev_put()")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: David Ahern <dsahern@kernel.org>
Link: https://patch.msgid.link/20250828195823.3958522-3-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
icmp6_send() must acquire rcu_read_lock() sooner to ensure
the dev_net() call done from a safe context.
Other ICMPv6 uses of dev_net() seem safe, change them to
dev_net_rcu() to get LOCKDEP support to catch bugs.
Fixes: 9a43b709a2 ("[NETNS][IPV6] icmp6 - make icmpv6_socket per namespace")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Link: https://patch.msgid.link/20250205155120.1676781-12-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
All inet_getpeer() callers except ip4_frag_init() don't need
to acquire a permanent refcount on the inetpeer.
They can switch to full RCU protection.
Move the refcount_inc_not_zero() into ip4_frag_init(),
so that all the other callers no longer have to
perform a pair of expensive atomic operations on
a possibly contended cache line.
inet_putpeer() no longer needs to be exported.
After this patch, my DUT can receive 8,400,000 UDP packets
per second targeting closed ports, using 50% less cpu cycles
than before.
Also change two calls to l3mdev_master_ifindex() by
l3mdev_master_ifindex_rcu() (Ido ideas)
Fixes: 8c2bd38b95 ("icmp: change the order of rate limits")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Link: https://patch.msgid.link/20241215175629.1248773-5-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
ICMP messages are ratelimited :
After the blamed commits, the two rate limiters are applied in this order:
1) host wide ratelimit (icmp_global_allow())
2) Per destination ratelimit (inetpeer based)
In order to avoid side-channels attacks, we need to apply
the per destination check first.
This patch makes the following change :
1) icmp_global_allow() checks if the host wide limit is reached.
But credits are not yet consumed. This is deferred to 3)
2) The per destination limit is checked/updated.
This might add a new node in inetpeer tree.
3) icmp_global_consume() consumes tokens if prior operations succeeded.
This means that host wide ratelimit is still effective
in keeping inetpeer tree small even under DDOS.
As a bonus, I removed icmp_global.lock as the fast path
can use a lock-free operation.
Fixes: c0303efeab ("net: reduce cycles spend on ICMP replies that gets rate limited")
Fixes: 4cdf507d54 ("icmp: add a global rate limitation")
Reported-by: Keyu Man <keyu.man@email.ucr.edu>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: David Ahern <dsahern@kernel.org>
Cc: Jesper Dangaard Brouer <hawk@kernel.org>
Cc: stable@vger.kernel.org
Link: https://patch.msgid.link/20240829144641.3880376-2-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
This commit comes at the tail end of a greater effort to remove the
empty elements at the end of the ctl_table arrays (sentinels) which
will reduce the overall build time size of the kernel and run time
memory bloat by ~64 bytes per sentinel (further information Link :
https://lore.kernel.org/all/ZO5Yx5JFogGi%2FcBo@bombadil.infradead.org/)
* Remove sentinel element from ctl_table structs.
* Remove the zeroing out of an array element (to make it look like a
sentinel) in sysctl_route_net_init And ipv6_route_sysctl_init.
This is not longer needed and is safe after commit c899710fe7
("networking: Update to register_net_sysctl_sz") added the array size
to the ctl_table registration.
* Remove extra sentinel element in the declaration of devinet_vars.
* Removed the "-1" in __devinet_sysctl_register, sysctl_route_net_init,
ipv6_sysctl_net_init and ipv4_sysctl_init_net that adjusted for having
an extra empty element when looping over ctl_table arrays
* Replace the for loop stop condition in __addrconf_sysctl_register that
tests for procname == NULL with one that depends on array size
* Removing the unprivileged user check in ipv6_route_sysctl_init is
safe as it is replaced by calling ipv6_route_sysctl_table_size;
introduced in commit c899710fe7 ("networking: Update to
register_net_sysctl_sz")
* Use a table_size variable to keep the value of ARRAY_SIZE
Signed-off-by: Joel Granados <j.granados@samsung.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Instead of (struct rt6_info *)dst casts, we can use :
#define dst_rt6_info(_ptr) \
container_of_const(_ptr, struct rt6_info, dst)
Some places needed missing const qualifiers :
ip6_confirm_neigh(), ipv6_anycast_destination(),
ipv6_unicast_destination(), has_gateway()
v2: added missing parts (David Ahern)
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: David Ahern <dsahern@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
np->ucast_oif is read locklessly in some contexts.
Make all accesses to this field lockless, adding appropriate
annotations.
This also makes setsockopt( IPV6_UNICAST_IF ) lockless.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: David Ahern <dsahern@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
np->mcast_oif is read locklessly in some contexts.
Make all accesses to this field lockless, adding appropriate
annotations.
This also makes setsockopt( IPV6_MULTICAST_IF ) lockless.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: David Ahern <dsahern@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
