Add support for the IPV6_MULTICAST_LOOP socket option.
This option allows enabling or disabling the consumption of multicast
packets by the sender. By default, the socket option will be enabled.
Co-authored-by: Pieter De Gendt <pieter.degendt@gmail.com>
Signed-off-by: Sayooj K Karun <sayooj@aerlync.com>
TLS_CREDENTIAL_SERVER_CERTIFICATE credential type is misleading, as in
fact it just represents a public certificate, it does not matter if the
certificate belongs to a server or a client. And actually, it was
already used in-tree for clients as well, for example in LwM2M.
Therefore rename the credential type to a more generic
TLS_CREDENTIAL_PUBLIC_CERTIFICATE and deprecate the old one.
Signed-off-by: Robert Lubos <robert.lubos@nordicsemi.no>
If there are no virtual interfaces attached to this virtual
interface, check if there is a RX handler for this virtual
interface and pass data to it.
Signed-off-by: Jukka Rissanen <jukka.rissanen@nordicsemi.no>
Clear all previous events when register is called a second time. This is
the same behavior as before, if the call happens with `len` equal or
greater then the previous call but in the case if `len` is less then the
first call, now, all events are cleared. This is more predictable behavior.
Signed-off-by: Cla Mattia Galliard <cla-mattia.galliard@zuehlke.com>
We drop the packet if TTL or hop limit is 0, but we should
also unref the packet in this case because we return 0 to
the caller which is not then able to free the packet because
it thinks that the packet was sent properly.
Signed-off-by: Jukka Rissanen <jukka.rissanen@nordicsemi.no>
The check_ip() in net_core.c did not check that the packet
Ethernet type is either IPv4 or IPv6. This meant that we for
example checked TTL also for ARP packets which is pointless as
those are not IPv4 packets.
Fix this by checking the link layer protocol type of the packet
to be either IPv4 or IPv6 before doing L3 checks.
Signed-off-by: Jukka Rissanen <jukka.rissanen@nordicsemi.no>
Adding a helper to parse one or more IPv4/6 strings with optional
netmask or prefix length.
Example of the string:
2001:db8::1/64,192.0.2.1,2001:db8::2,192.0.2.2/24
Signed-off-by: Jukka Rissanen <jukka.rissanen@nordicsemi.no>
All connection information was printed only if connection debugging
was enabled which then did not print information about Wireguard or
DHCP handlers. Limiting the printouts like this is not really needed
so print all connection information with "net conn" command.
Signed-off-by: Jukka Rissanen <jukka.rissanen@nordicsemi.no>
It is useful in testing to allow user to change the default
network interface by "net iface default <index>".
Also print out the default interface for "net iface" and
"net iface default" commands.
Signed-off-by: Jukka Rissanen <jukka.rissanen@nordicsemi.no>
Sometimes it is useful to view low level network interface information
in the network shell. Enable such prints if network interface debugging
is enabled.
Signed-off-by: Jukka Rissanen <jukka.rissanen@nordicsemi.no>
The original flow of socket_service thread handling the Zperf UDP RX
packets is: zsock_poll() polls all sockets for events, if ctx->recv_q
of Zperf is not empty, it will call trigger_work() -> udp_recv_data()
-> zsock_recvfrom() to read only one UDP packet from ctx->recv_q, then
go back to zsock_poll() and run the same process again, which is
inefficient.
The alternative solution is, in udp_recv_data(), it should exhaust all
the packets in the current ctx->recv_q, and then go back to
zsock_poll() to run the same process again.
In our Wi-Fi test case, for WPA3 security mode of 5GHz, the STA UDP RX
throughput can be improved from 91.48 Mbps to 99.87 Mbps, the SAP UDP
RX throughput can be improved from 85.97 Mbps to 96.00 Mbps.
Signed-off-by: Maochen Wang <maochen.wang@nxp.com>
Currently, the L2 PPP won't work with AF_PACKET socket family as it only
supports packets from AF_INET/AF_INET6 families. Because of this, it's
not possible to use AF_PACKET RAW or DGRAm sockets with PPP interfaces,
as the packets they generate have family field set to AF_PACKET.
Fix this, by verifying the LL protocol field in the PPP L2 before
passing the packet the respective PPP driver. If the AF_PACKET packet is
received, and the protocol field is set to IP/IPv6, update the packet
family to AF_INET/AF_INET6 accordingly.
Signed-off-by: Robert Lubos <robert.lubos@nordicsemi.no>
Set NET_ADDR_PREFERRED before network event is generated, so that conn_mgr
properly generates NET_EVENT_L4_CONNECTED and NET_EVENT_L4_IPV6_CONNECTED
events.
Fixes: 1a5e13a79b ("net: if: Release the interface lock early when
starting IPv4 ACD")
Signed-off-by: Marcin Niestroj <m.niestroj@emb.dev>
Set NET_ADDR_PREFERRED before network event is generated, so that conn_mgr
properly generates NET_EVENT_L4_CONNECTED and NET_EVENT_L4_IPV4_CONNECTED
events.
Fixes: 1a5e13a79b ("net: if: Release the interface lock early when
starting IPv4 ACD")
Signed-off-by: Marcin Niestroj <m.niestroj@emb.dev>
Previously the net_linkaddr struct had pointers to the link address.
This is error prone and difficult to handle if cloning the packet as
those pointers can point to wrong place. Mitigate this issue by
allocating the space for link address in net_linkaddr struct. This will
increase the size of the net_pkt by 4 octets for IEEE 802.15.4 where the
link address length is 8, but there no increase in size if link address
is 6 bytes like in Ethernet/Wi-Fi.
Signed-off-by: Jukka Rissanen <jukka.rissanen@nordicsemi.no>
Fix that a lwm2m message is reset without executing a registered
callback on lwm2m_engine_context_close.
Signed-off-by: Sebastian Schoch <sebastian.schoch@husqvarnagroup.com>
Ignore bootstrap 'transition done' state transition
(ENGINE_BOOTSTRAP_TRANS_DONE) from an invalid state. The client must be
in state ENGINE_BOOTSTRAP_REG_DONE to process a bootstrap finish command
received from the server.
The LwM2M client does not implement CoAP deduplication (rfc 7252,
section 4.5) and retransmissions of the bootstrap finish command `/bs`
would restart the registration client with undesired side effects.
Signed-off-by: Marc Lasch <marc.lasch@husqvarnagroup.com>
Conform rfc6762 a mDNS responder should answer clients
which are not using the mDNS port in the source address
with unicast UDP to the same port as described in chapter 6.7
Fixes: #81657
Signed-off-by: Vincent van der Locht <vincent@synchronicit.nl>
Use Wi-Fi connecting to Qualcomm IPQ8074 AP, and run the UDP RX traffic
with Zperf, but zperf does not return the throughput number after
traffic completion nor any session started prints. After traffic
completion, ping from STA to AP and vice versa does not work. The
socket_service thread is found blocked forever at zsock_wait_data()
after dns_dispatcher_svc_handler() is called via trigger_work().
The root cause of this issue is:
STA received one DHCPv4 packet containing DHCPV4_OPTIONS_DNS_SERVER,
it will create DNS socket and registered to socket_service. Then STA
received another IPv6 router advertisement packet containing
NET_ICMPV6_ND_OPT_RDNSS, it will close socket and change the state of
poll_events to K_POLL_STATE_CANCELLED(8), then registered to
socket_service with same fd. In socket_service thread, zsock_poll()
called zsock_poll_update_ctx() when handled ZFD_IOCTL_POLL_UPDATE, and
it checked the state of poll_events was not K_POLL_STATE_NOT_READY(0),
then it will set pfd->revents to '|= ZSOCK_POLLIN'. Finally
trigger_work() can be called as 'ctx.events[i].revents > 0' is matched.
The fix of this issue is that, in zsock_poll_update_ctx(), it should
check the state of poll_events is neither K_POLL_STATE_NOT_READY nor
K_POLL_STATE_CANCELLED before setting revents as ZSOCK_POLLIN, to avoid
trigger_work be unexpectedly called.
Signed-off-by: Maochen Wang <maochen.wang@nxp.com>
The ethernet header size with VLAN tag is 18 bytes,
so the length check should accommodate this.
Signed-off-by: Christoph Seitz <christoph.seitz@infineon.com>
Do not update packet length for unsupported packet families.
AF_PACKET family used the IPv6 length update, which breaks the
packet length if there is a payload with bytes resembling a
IPv6 length field less then 6 or a IPv4 length field less
then 46.
Signed-off-by: Christoph Seitz <christoph.seitz@infineon.com>
Avoid a "defined but not used" warning on
`init_next_pending_timeseries_data` when
LWM2M_RESOURCE_DATA_CACHE_SUPPORT is enabled but not LWM2M_VERSION_1_1.
Signed-off-by: Etienne de Maricourt <edmecomemail@gmail.com>
If mDNS resolver is enabled but mDNS responder is not, then
mDNS multicast address group is not joined. This would prevent
the mDNS resolver to receive the responses. Fix this by
joining the mDNS multicast group if mDNS responder is not
enabled (because the responder will join the group itself).
Fixes#86477
Signed-off-by: Jukka Rissanen <jukka.rissanen@nordicsemi.no>
There is a deadlock issue when calling stop using address conflict
detection. This is due to the fact that some net_mgmt events are fired
and trigger the dhcpv4_acd_event_handler() with lock held even if they
are of no interest for this callback.
Therefore, before acquiring the lock, make sure the event we received
is one we are expecting.
Also, do the same for dhcpv4_iface_event_handler().
Signed-off-by: Mathieu Anquetin <mathieu.anquetin@groupe-cahors.com>
Add compression support using the accept-encoding
header to the http server static filesystem resource.
Signed-off-by: Carlo Kirchmeier <carlo.kirchmeier@zuehlke.com>
http_server_http2.c does not guard check if file system is available
or not before using file system api. This PR will add guard for that
Signed-off-by: bac phan <phanhaibac98@gmail.com>
Rename this driver to eth_native_tap, including renaming all its
options.
The old options remain until v4.4, but as deprecated.
Signed-off-by: Alberto Escolar Piedras <alberto.escolar.piedras@nordicsemi.no>
The CONFIG_POSIX_C_LIB_EXT will get support for fnmatch() function.
The old CONFIG_FNMATCH is deprecated.
Signed-off-by: Jukka Rissanen <jukka.rissanen@nordicsemi.no>
When we receive piggybacked Ack, it means that we have
received all information for this request.
There is no need to track exchange lifetime anymore.
Once we reset the internal request, it is free to be used
for the application. So if we only have few requests allocated,
it would be slow to send those as get_free_request() only
gives request structures that don't have a lifetime left.
Signed-off-by: Seppo Takalo <seppo.takalo@nordicsemi.no>
- Add possibility to choose implementation of OpenThread L2 and
set it to Zephyr's by default
- Remove unused `OPENTHREAD_SECURITY_INTERNAL` Kconfig
- Add missing choice names for logging and coprocessor
Signed-off-by: Maciej Baczmanski <maciej.baczmanski@nordicsemi.no>
Use the newly added timeout in various send functions from within
net_context_sendto.
Signed-off-by: Cla Mattia Galliard <cla-mattia.galliard@zuehlke.com>
Allows to send with different timeouts to not block caller in some
situations. Stable API is kept and just calls `try`-variant with a timeout
of `K_FOREVER`.
Signed-off-by: Cla Mattia Galliard <cla-mattia.galliard@zuehlke.com>
The coap_client has a static/internal function, has_ongoing_exchange(),
which can be an useful addition to the public API. This would provide
a mechanism to determine when it is safe to close a socket gracefully.
Signed-off-by: Povilas Selevicius <povilas.selevicius@quadigi.com>
Add a variant of the SNTP API that does not block while waiting for a
response from the SNTP server. Instead it takes advantage of
`CONFIG_NET_SOCKETS_SERVICE` to read the response asynchronously when
the response is received.
Signed-off-by: Jordan Yates <jordan@embeint.com>
Extract the SNTP query send logic out of `sntp_query` so that it can be
used by functions that don't synchronously wait for the response.
Signed-off-by: Jordan Yates <jordan@embeint.com>
When a TCP connection is refused during zsock_connect, errno is set
to -ENOTCONN, but errno should be set to -ECONNREFUSED. This change
causes the ECONNREFUSED status to be propagated from tcp_in to
net_tcp_connect, which eventually causes errno to be set
to -ECONNREFUSED.
Signed-off-by: Noah Olson <noah@wavelynx.com>
The root cause of this issue is a modification of `struct ring_buf` in
3075a7d9. Even though all the fields of the struct are marked as
internal, the LwM2M code is using some of them to roll back the state of
the ring buffer on failure.
Signed-off-by: Etienne de Maricourt <edmecomemail@gmail.com>
In order to avoid any mutex deadlocks between iface->lock and
TX lock, release the interface lock before calling a function
that will acquire TX lock. See previous commit for similar issue
in RS timer handling. So here we create a separate list of ACD
addresses that are to be started when network interface comes up
without iface->lock held.
Signed-off-by: Jukka Rissanen <jukka.rissanen@nordicsemi.no>
In order to avoid any mutex deadlocks between iface->lock and
TX lock, release the interface lock before calling a function
that will acquire TX lock. See previous commit for similar issue
in RS timer handling. So here we create a separate list of multicast
addresses that are to be rejoined when network interface comes up
and then rejoin the groups without iface->lock held.
Signed-off-by: Jukka Rissanen <jukka.rissanen@nordicsemi.no>
