When PPP net_if is changed to dormant, the PPP connection
is terminated asynchronously. This is used so that the driver
can terminate the PPP connection gracefully.
Note: net_if_down() is not used as it would require synchronous
operation which would block the system workqueue thus causing
deadlock.
Signed-off-by: Markus Lassila <markus.lassila@nordicsemi.no>
Turn the MBEDTLS_RSA_FULL selects into depends on.
This is how the other MBEDTLS_KEY_EXCHANGE_* Kconfig options are defined.
This is done to avoid circular dependencies.
At the same time update uses of the affected MBEDTLS_KEY_EXCHANGE_*
Kconfig options to enable/disable the dependencies which used to be
automatically handled.
Signed-off-by: Tomi Fontanilles <tomi.fontanilles@nordicsemi.no>
Error checking of otIp6AddUnicastAddress() and
otIp6SubscribeMulticastAddress() was added recently, however it wasn't
taken into account that those APIs return an error on attempt to
register an IPv6 address that is already present on the OT interface.
Therefore, add more specific error checks, to return silently in case
address was already present.
As those two APIs are not very consistent, and otIp6AddUnicastAddress()
returns OT_ERROR_INVALID_ARGS in such cases, add an extra check if the
address is already present before attempting to register the address.
Signed-off-by: Robert Lubos <robert.lubos@nordicsemi.no>
After the PR #85000 the calculation of NUM field of
Block 1 option on CoAP Ack packet started to advance
to next packet block.
We should not update the ctx->current field because it is
used for calculating the NUM field in response packet.
It should point to beginning of the payload,
so the response is correct.
Leshan server don't seem to care about this, but Coiote does.
Signed-off-by: Seppo Takalo <seppo.takalo@nordicsemi.no>
There was wrong implementation in pkt frag removing and inserting.
This was causing infinite loop in either net_buf_frag_last or
net_pkt_get_len. This happened only when the pkt frag removing
and inserting was executed too fast after sending pkt before
ethernet_send calling net_pkt_get_len.
Signed-off-by: Yangbo Lu <yangbo.lu@nxp.com>
Set connection manager to unstable. It is now used
by multiple parts in zephyr and numerous samples.
Signed-off-by: Fin Maaß <f.maass@vogl-electronic.com>
Add option to use the connection manager to
schedule a (re-) sync on connection and diabeling the
work, when there is no connection.
Signed-off-by: Fin Maaß <f.maass@vogl-electronic.com>
Use connection manager to deactivate and activate
net log backend.
This removes the warnings about dropped packages
completly.
Signed-off-by: Fin Maaß <f.maass@vogl-electronic.com>
Avoid early enabling of the syslog backend in
the the dhcpv4 options parser.
When CONFIG_NET_IPV4_ACD is enabled, the assigned ip address has
not been checked, when the other dhcpv4 options are parsed, this would
lead to the syslog backend being enabled before the src ip address
is valid, so we get lots of warnings about dropd packets, this fixes
it at least on start. We will still get the warnings, when the iface
goes down and then up later.
Signed-off-by: Fin Maaß <f.maass@vogl-electronic.com>
MLD APIs are commonly used across the codebase to configure IPv6
multicast addresses on network interfaces. Sending MLD reports however
works only for native interfaces as it uses low-level APIs. Therefore,
in order to make the APIs at least semi-functional for offloaded
interfaces as well (i.e. allow to configure multicast address on
the interface), return early in case interface is offloaded.
Signed-off-by: Robert Lubos <robert.lubos@nordicsemi.no>
IGMP APIs are commonly used across the codebase to configure IPv4
multicast addresses on network interfaces. Sending IGMP reports however
works only for native interfaces as it uses low-level APIs. Therefore,
in order to make the APIs at least semi-functional for offloaded
interfaces as well (i.e. allow to configure multicast address on
the interface), return early in case interface is offloaded.
Signed-off-by: Robert Lubos <robert.lubos@nordicsemi.no>
Use zsock_inet_ntop() instead of inet_ntop() to avoid dependency to the
POSIX subsystem in the library.
Signed-off-by: Robert Lubos <robert.lubos@nordicsemi.no>
The result of the transport init() function should be propagated to the
application, otherwise the initialization could fail silently.
Signed-off-by: Robert Lubos <robert.lubos@nordicsemi.no>
UDP transport for the MQTT SN libraries depends on IGMP/MLD APIs
unconditionally (via respective setsockopt calls) and without them being
enabled transport initialization would fail. Therefore, ensure
respective multicast libraries are always enabled if MQTT SN UDP
transport is used.
Signed-off-by: Robert Lubos <robert.lubos@nordicsemi.no>
Each incoming TCP packet has been completely handled in current
state. No need to do further process by 'goto next_state'.
Signed-off-by: Shrek Wang <inet_eman@outlook.com>
Many cellular modems attempt to negotiate an ACCM value of 0x00000000.
While the PPP driver rejects this by default, it does not propose an
alternative. As a result, some modems default to using 0x00000000 after
LCP negotiation. Because the PPP driver expects all control characters
to be escaped, this causes issues during decoding. This change
negotiates an ACCM value of 0xffffffff to ensure compatibility with such
modems.
Signed-off-by: Vytautas Virvičius <vytautas@virvicius.dev>
According to RFC 2131, DHCP clients should use the same xid as
received in the Offer message when sending DHCP Requests. Therefore,
when generating DHCP Request message, the xid value should not be
incremented.
One vague topic is whether the xid value should be updated when
sending Requests from Renewing or Rebinding states, however RFC makes no
exception for those states, and other implementations (dhclient, lwip)
seem to reuse the same xid in such cases, so comply with this behavior.
Signed-off-by: Robert Lubos <robert.lubos@nordicsemi.no>
Pad option (option code 0) can be present in between other options for
alignment. The option has a fixed 1-byte length (i. e. no length field),
therefore it did not fall under the common processing code for
unrecognized options (which include the length field at the second
byte). Therefore, not processing this option explicitly could disturb
other options processing, as the parser would wrongly interpret the next
option code as the length field. This commit adds Pad option handling to
fix the issue.
Signed-off-by: Robert Lubos <robert.lubos@nordicsemi.no>
In case the received DHCP message is malformed and contains invalid
message type, the code responsible for matching message type with a
string would assert. This shouldn't be the case that external conditions
(like receiving malformed packet) trigger asserts in the system.
Therefore modify that code, to return "invalid" string in such case
instead.
Signed-off-by: Robert Lubos <robert.lubos@nordicsemi.no>
Response "data_len" field needs to be set with the size of the
received data before calling the parser as it's used inside the on_body
callback, this commit fixes it.
Signed-off-by: Robert Lubos <robert.lubos@nordicsemi.no>
Current clock synchronization was always stepping clock. This was
causing large offset, and discontiguous ptp hardware clock time.
For TSN hardware, discontiguous ptp hardware clock time was not
able to be used for other TSN protocols.
This patch is to convert to frequency adjustment with a basic
PI control algorithm.
Signed-off-by: Yangbo Lu <yangbo.lu@nxp.com>
At the starting of ptp syncrhonization, there may be not Sync frame
TX/RX timestamps in first time link delay calculation.
So, need a check for that in case of wrong link delay calculated.
Signed-off-by: Yangbo Lu <yangbo.lu@nxp.com>
According to TCP Spec. RFC793, ACK flag should be always set
after sequences of both sides are sync-ed except for RST seg-
ment. It is not necessary to send FIN only packet in the
test case, using FIN | ACK instead.
Similarly, change the tcp_out(conn, FIN | ACK) in CLOSE_WAIT.
Signed-off-by: Shrek Wang <inet_eman@outlook.com>
Move OpenThread-related code from
zephyr/subsys/net/l2/openthread/openthread.c to
zephyr/modules/openthread/platform/openthread.c.
The primary goal of this refactor is to enable the use
of OpenThread as an independent module, without the necessity
of Zephyr's networking layer.
This change is particularly beneficial for simple applications
that have their own implementation of the IEEE802.15.4 driver
and do not require a networking layer. These applications can
now disable Zephyr's L2 and IEEE802.15.4 shim layers and
directly use the OpenThread module, saving valuable kilobytes
of memory.
In this approach if the CONFIG_NET_L2_OPENTHREAD
Kconfig option is set, Zephyr's L2 and IEEE802.15.4 layers
will be used, and everything will function as before.
The main difference is the Zephyr's L2 layer now uses
the OpenThread module, no longer implementing it.
While most of the functions in include/net/openthread.h
have been deprecated, they are still available for use to
maintain backwards compatibility.
Signed-off-by: Arkadiusz Balys <arkadiusz.balys@nordicsemi.no>
Doing another init of the log_backend_net in
the net config init could lead to the
server, set during runtime, being overwritten
by the Kconfig default.
Signed-off-by: Fin Maaß <f.maass@vogl-electronic.com>
In case HTTP client read out more data from a socket that it processed
it will indicate there is leftover data in the receive buffer available.
Make use of it at the websocket level, so that no data is lost. As we
reuse the same receive buffer in this case, it's only needed to update
the count variable to indicate how many bytes are available.
Signed-off-by: Robert Lubos <robert.lubos@nordicsemi.no>
The HTTP client would read data from a socket up to the size of the
receiving buffer, however it may not process them all. This is usually
only the case if protocol switching takes place, where the data read may
belong to another protocol. Therefore we need a way to notify the caller
about any potential data that is already present in the buffer and needs
to be processed.
Introduce an new data_len member in struct http_request to provide the
information about the data already available in the receive buffer. If,
after HTTP response processing, the value is non-zero, the data will be
available in the beginning of the user provided receive buffer.
To make this possible however, we need to track how many bytes were
actually processed by the http_parser, therefore the code will no longer
ignore the http_parser_execute() return value. To simplify processing,
it's also been changed how the receive buffer is used. Instead of using
it in a ring-buffer-like way, the offset variable will track how many
bytes are available in the buffer, and in the rare occasions when not
all data from the buffer was processed by the HTTP parser, we'll
memmomve the remaining data to the beginning of the buffer.
Signed-off-by: Robert Lubos <robert.lubos@nordicsemi.no>
Current clock synchronization was always stepping clock. This was
causing large offset, and discontiguous ptp hardware clock time.
For TSN hardware, discontiguous ptp hardware clock time was not
able to be used for other TSN protocols.
This patch is to convert to frequency adjustment with a basic
PI control algorithm.
Signed-off-by: Yangbo Lu <yangbo.lu@nxp.com>
warning: format '%d' expects argument of type 'int', but argument 3 has
type 'size_t' {aka 'long unsigned int'} [-Wformat=]
Change the printf modifier to %zd for size_t variable will resolve the
warning.
Signed-off-by: Vijayakannan Ayyathurai <vijayakannan.ayyathurai@intel.com>
Default binding should take place before we actually make use of the
local address when registering packet socket "connection".
Also, instead of hard coding the protocol for default binding to
ETH_P_ALL, use the protocol that the socket was created with.
Signed-off-by: Robert Lubos <robert.lubos@nordicsemi.no>
In order to be able to receive packets on unbound packet sockets (which
should collect packet from all interfaces in such case), it's needed to
register receive callback at the socket layer as soon as the socket is
created.
In additional to that, the default binding for packet sockets need to be
revisited. Packet socket should not be bound to the default interface,
as this way the socket would only be receiving packets from that
particular interface. Instead, leave the interface unspecified in such
case.
Signed-off-by: Robert Lubos <robert.lubos@nordicsemi.no>
Allow to update the local address on a registered connection when
rebinding.
This is needed for packet sockets, as by default packet socket
will be bound to "any" interface (interface index 0), and interface
index is part of the local address registered for packet socket.
In order to be able to explicitly bind to a specific interface later, it
needs to be possible to update the local address registered for the
connection, as we need to update the interface index, which is used
by net_conn_packet_input() for packet filtering.
Signed-off-by: Robert Lubos <robert.lubos@nordicsemi.no>
keep_alive_timer_restart() only works in ESTABLISHED state. In
tcp_in() SYN_SENT and SYN_RECEIVED state, it won't work by calling
this function. So remove the call in that 2 states while adding it
in the bottom after changing the conn->state to ESTABLISHED.
Signed-off-by: Shrek Wang <inet_eman@outlook.com>
Slightly increase the default maximum size for zperf uploads to support
1kB of data payload together with the 40 bytes UDP client header.
Signed-off-by: Jordan Yates <jordan@embeint.com>
Add support for uploading custom data, instead of a static packet of
'z'. This can be used for more accurate profiling of uplink throughput
for a given application. For example, reading data from flash, or
application level encryption.
Signed-off-by: Jordan Yates <jordan@embeint.com>
The timestamp in the UDP header is expected to be a Unix timestamp, not
an application uptime. Add the option to specify the Unix time at
function entry so that the reported timestamps are accurate.
Signed-off-by: Jordan Yates <jordan@embeint.com>
The minimum packet size is the combination of the datagram and client
headers, not just the datagram header.
Signed-off-by: Jordan Yates <jordan@embeint.com>
Prevent the modulo operation causing a division by zero error by falling
back to the minimal ACK timeout value if the random factor is 1.0. Also
fall back to the same value if the random factor is erroneously smaller
than 1.0. Additionally make the upper bound of the random range
reachable.
Signed-off-by: Adrian Friedli <adrian.friedli@husqvarnagroup.com>
The _ is a reserved character in front of the symbols so remove
it from network management event macros. The remaining string
without the _ will identify the network API anyway so having
underscore there is not needed.
Signed-off-by: Jukka Rissanen <jukka.rissanen@nordicsemi.no>
This adjust the IPv6 source address selection so that it is possible
to select deprecated IPv6 address if no better preferred address is found.
From RFC 6724 chapter 5:
Rule 3: Avoid deprecated addresses.
If one of the two source addresses is "preferred" and one of them is
"deprecated" (in the RFC 4862 sense), then prefer the one that is
"preferred".
Rule 8: Use longest matching prefix.
If CommonPrefixLen(SA, D) > CommonPrefixLen(SB, D), then prefer SA.
Similarly, if CommonPrefixLen(SB, D) > CommonPrefixLen(SA, D), then
prefer SB.
So the fix allows deprecated address to be selected if it is a better
match than the preferred one. The reasoning here is that an address with
a longer matching prefix is generally considered topologically closer to
the destination. Using such a source address can lead to more efficient
routing, as it's more likely that the source and destination are within
the same network segment or a closely related one.
Signed-off-by: Jukka Rissanen <jukka.rissanen@nordicsemi.no>
Increased the heap mempool size for the socketpairs. This way there will
be enough memory available to allocate 2 * struct spair.
Signed-off-by: Ibe Van de Veire <ibe.vandeveire@basalte.be>
Make sure RS process is stopped if network interface goes down
as there is no point doing it any more.
Signed-off-by: Jukka Rissanen <jukka.rissanen@nordicsemi.no>
It might happen that if the network interface is going up/down
fast enough, the RS timer could get inserted to the active RS
timer list twice. This would then cause a forever loop in
rs_timeout() when traversing the active list.
Signed-off-by: Jukka Rissanen <jukka.rissanen@nordicsemi.no>
