Although LwM2M engine uses cooperative threads, the internal `send()`
implementation might trigger context switch when it calls a kernel
function, therefore resulting in `send()` call being entered from both
the LwM2M thread and the retransmit work.
Signed-off-by: Robert Lubos <robert.lubos@nordicsemi.no>
The context should only clear messages it owns, not all of them. Since
both context (LwM2M and FOTA) share common message pool, they might
interrupt their operation otherwise (i. e. cancel retransmissions).
Signed-off-by: Robert Lubos <robert.lubos@nordicsemi.no>
So far socket errors reported by poll/recvfrom were ignored, which could
lead to an unexpected behavior when socket was left in an undefined
state.
Fix this, by requesting a re-registration in the LWM2M state machine,
which will close the faulty socket and open a new one. Note, that simply
closing and re-opening a socket in the lwm2m engine would not work,
since this would silently invalidate any open observations on the
lwm2m server side (due to port number change). Triggering a fresh
registration will notify the server to update its observations.
Signed-off-by: Robert Lubos <robert.lubos@nordicsemi.no>
On OT network with poor coverage, very often request/observe packets
doesn't get it's ACK and consumes from pendings/replies/message stacks.
In such cases when LWM2M engine tries to recover by resetting its state,
it fails because of lack of free messages.
Signed-off-by: Kiril Petrov <retfie@gmail.com>
In networks with high latencies (like NB-IoT), it's quite common to
recieve duplicated response. It's not an error condition, a correct way
to handle it is to simply ignore the duplicate. Lower the log level for
this event, not to disturb users.
Signed-off-by: Robert Lubos <robert.lubos@nordicsemi.no>
The retaransmission logic was not correct in the lwm2m_engine, and could
lead to faulty behavior in case multiple messages were pending for
retransmission in the queue.
1. Since there is a singe delayed work item for entire retransmission
queue, `coap_pending_next_to_expire` should be called before
scheduling next timeout, to identify which message is going to expire
next (and when). Currently, the engine always set next timeout, based
on timeout from the message being currently re-transmitted.
2. In case the message was re-transmitted several times, and is removed
from the retansmission queue due to a timeout, next retransmission
should be scheduled, in case there are other messages on the queue.
3. Verify the timeout of the earliest message to expire in the
retransmission handler. In case messages from the beginning of the
queue were removed, we might need to schedule the retransmission
again, instead of sending message rightaway.
4. `lwm2m_send_message` is not handling retransmissions anyway, so
there's no need to check send attempts. Instead, verify
retransmission work item is already pending, and update its timeout
if needed.
Signed-off-by: Robert Lubos <robert.lubos@nordicsemi.no>
The LwM2M stack would previously ignore all OPAQUE resources when
reading them. This meant that it was impossible to read them, even if
there was a custom read callback.
Signed-off-by: Göran Weinholt <goran.weinholt@endian.se>
According to LWM2M specification, when Queue Mode is used, the LWM2M
client should keep the reciever on for specified time after sending A
CoAP message. This commit adds a new LWM2M event,
`LWM2M_RD_CLIENT_EVENT_QUEUE_MODE_RX_OFF`, to facilitate the process by
notifying the application when it's safe to turn the receiver off.
Signed-off-by: Robert Lubos <robert.lubos@nordicsemi.no>
Add Kconfig option to enable Queue Mode binding. With this option
enabled, the LWM2M client will register with `UQ` binding, instead of
`U`.
Signed-off-by: Robert Lubos <robert.lubos@nordicsemi.no>
With `CONFIG_NET_NATIVE`, offloaded drivers can specify capabilites with
`NET_IPV4/6` configs, so there is no longer need to handle socket
offloading separately.
Also, initialize hints structure with zeros, as according to man pages
unused fields should be set to 0:
`All the other fields in the structure pointed to by hints must contain
either 0 or a NULL pointer, as appropriate.`
Signed-off-by: Robert Lubos <robert.lubos@nordicsemi.no>
Add new config option `LWM2M_DNS_SUPPORT` to the LWM2M library, instead
of relying on `DNS_RESOLVER` which is only compatible with native
network stack. This allows to use DNS with offloaded interfaces
seamlessly.
Signed-off-by: Robert Lubos <robert.lubos@nordicsemi.no>
The LwM2M implementation for DNS resolving has checks which
configure hints based on whether IPv4 or IPv6 are enabled.
Neither of them need enabled if using NET_SOCKETS_OFFLOAD,
which then causes an error to be returned to due to
"hints.ai_family" not being set.
Also the offload API need to know when to free the allocated
"struct addrinfo" instead of calling free() generically,
thus let's use the freeaddrinfo() API for sockets which will
call into the offload API if needed.
Fixes: https://github.com/zephyrproject-rtos/zephyr/issues/18765
Signed-off-by: Jun Qing Zou <jun.qing.zou@nordicsemi.no>
When the bootstrap support was added, it looks like I somehow missed
the handling block in the engine.
Let's add it now to fix boostrap support.
Fixes: https://github.com/zephyrproject-rtos/zephyr/issues/18080
Signed-off-by: Michael Scott <mike@foundries.io>
When presenting errors in lwm2m_engine_set() let's include the related
LwM2M path for easier debugging.
Signed-off-by: Michael Scott <mike@foundries.io>
LwM2M allows for multiple instance resources such the power source
resources in the device object. These types of resources have
always been very hard to work with, and frankly were poorly
implemented.
This led to other issues where it was very hard to have
non-sequential resource instances, and each resource of this type
needed special getter / setter methods such as:
lwm2m_device_add_pwrsrc()
lwm2m_device_set_pwrsrc_voltage_mv()
Going forward, as more LwM2M objects are implemented this just
doesn't scale well.
To fix this:
- split the resource instance data out from the resource data.
This includes the data pointer information and resource
instance id.
- add resource id and resource instance id to the event callback
functions so user's can see in more detail what resources and
resource instances are being handled.
- allow generic functions like lwm2m_engine_get_*() and
lwm2m_engine_set_*() to access resource instance data.
- adjust object resource initialization macros to map resource
instances to resources at the time of object instance
creation.
- fix up the lwm2m_client as a reflection of all of these changes.
Signed-off-by: Michael Scott <mike@foundries.io>
Due to work combining data into the lwm2m_message structure, we no
longer need to pass the lwm2m_engine_obj parameter between
formatters and most of the operation handlers.
So, let's remove it.
Signed-off-by: Michael Scott <mike@foundries.io>
Server records contain the default PMIN and PMAX settings for how
often we can send observe notifications. We are currently using
arbitrary defaults which cannot be changed when compiled or
during runtime.
Let's add Kconfig settings for the default settings to use and
also lookup the current values in the active server record when
an observe is added.
The actual PMIN/PMAX values can still be set via WRITE_ATTRIBUTE
operation.
Signed-off-by: Michael Scott <mike@foundries.io>
Currently the retransmit_work is not cancelled when closing a context,
making it operate on an invalid context.
LwM2M RD client also closes the context and initializes it again when
registration with the server fails, overwriting the active timeout and
breaking the timeout dlist.
Signed-off-by: Pieterjan Camerlynck <pieterjan.camerlynck@gmail.com>
Current implementation of LwM2M engine doesn't allow users a way
of overriding TLS credential load with custom function. This
would be needed by an offloaded TLS stack where we don't want
to use standard Zephyr functions.
Let's add a load_credential function pointer to the LwM2M client
context which will be called when it's available.
Fixes: https://github.com/zephyrproject-rtos/zephyr/issues/17408
Signed-off-by: Michael Scott <mike@foundries.io>
CONFIG_NET_IPV* checks are not needed in lwm2m_parse_peerinfo().
The functions used are always available. Worse, having these checks
forces the need to enable CONFIG_NET_IPV4 or IPV6 when it's not really
needed (LwM2M could be using an offloaded IP stack).
Fixes: https://github.com/zephyrproject-rtos/zephyr/issues/17401
Signed-off-by: Michael Scott <mike@foundries.io>
CONFIG_NET_IPV* checks are not needed in lwm2m_sprint_ip_addr(). The
functions used are always available. Worse, having these checks
forces the need to enable CONFIG_NET_IPV4 or IPV6 when it's not really
needed (LwM2M could be using an offloaded IP stack).
NOTE: Also fixes an issue where a NULL is returned when the IP address
is unknown. This usually ends up with a crash/abort in the logging
code.
Fixes: https://github.com/zephyrproject-rtos/zephyr/issues/17401
Signed-off-by: Michael Scott <mike@foundries.io>
This fixes the issue where observations are automatically reported using
the minimum period instead of the maximum. This causes notifications to
be sent more frequently than configured when the resource does not
change.
Signed-off-by: Pieterjan Camerlynck <pieterjan.camerlynck@gmail.com>
move misc/printk.h to sys/printk.h and
create a shim for backward-compatibility.
No functional changes to the headers.
A warning in the shim can be controlled with CONFIG_COMPAT_INCLUDES.
Related to #16539
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
Found a few annoying typos and figured I better run script and
fix anything it can find, here are the results...
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
Due to commit a211afb0 ("logging: Add option to detect missed
transient string duplication"), the logs for LwM2M subsystem
is now spamming missing log_strdup() calls.
Let's add log_strdup() where needed.
Signed-off-by: Michael Scott <mike@foundries.io>
This fixes an issue where if timestamp == service_due_timestamp,
we don't call the periodic service. Then the following call to
engine_next_service_timeout_ms() returns 0 because the service
is still due and lwm2m_engine_service() is called again.
This process repeats several times until the value of
k_uptime_get() changes and then the work is finally handled.
Previously, the resolution of k_uptime_get() was in ms. A recent
change to this API defaults Zephyr so that the resolution is
set via CONFIG_SYS_CLOCK_TICKS_PER_SEC (default 100).
This means the value of k_uptime_get() only changes every 10ms.
Reported-by: Github User pieterjanc
Signed-off-by: Michael Scott <mike@foundries.io>
"It's a Trap!" -- Admiral Ackbar
When moving to the BSD-socket APIs, the original thread running LwM2M
periodic services such as observes and lifetime updates, was replaced
with a re-occuring workqueue job. To save the overhead of creating a
new thread, I used the system workqueue for these jobs.
This was a mistake. If these jobs hit a semaphore or wait for some
reason, it cannot be prempted due to the priority of the system work
queue.
Let's instead add this service handling to the thread that we already
use for polling sockets. This also removes a configuration issue where
the system workqueue stack size needed to be increased. This can now
be adjusted via the LWM2M_ENGINE_STACK_SIZE knob.
Directly fixes semaphore usage in the socket-based DNS code.
This was introduced as a bugfix for non-responsive DNS server hanging
the Zephyr device forever. However, this probably fixes randomly
seeming hangs on the device.
Signed-off-by: Michael Scott <mike@foundries.io>
Normally, this bug wasn't apparent as the value is type-casted
to a float32/64 type. However, once we start persisting these
values they need the correct length.
Signed-off-by: Michael Scott <mike@foundries.io>
When a context is closed to a server, we should clean up any
existing observes along with it. Otherwise these will try to fire
afterward.
Signed-off-by: Michael Scott <mike@foundries.io>
send() returns -1 upon error and sets errno appropriately. Let's
not bother saving the return code and instead share errno back
to the user.
Signed-off-by: Michael Scott <mike@foundries.io>
Errno value is only significant when `recvfrom` function indicated an
error (by returning -1). We should not depend on it's value if no error
is notified.
As the return value of `recvfrom` is already checked, misused errno
verification can simply be removed.
Signed-off-by: Robert Lubos <robert.lubos@nordicsemi.no>
There are issues using lowercase min and max macros when compiling a C++
application with a third-party toolchain such as GNU ARM Embedded when
using some STL headers i.e. <chrono>.
This is because there are actual C++ functions called min and max
defined in some of the STL headers and these macros interfere with them.
By changing the macros to UPPERCASE, which is consistent with almost all
other pre-processor macros this naming conflict is avoided.
All files that use these macros have been updated.
Signed-off-by: Carlos Stuart <carlosstuart1970@gmail.com>
Previously, the net_app layer handled DNS support as a part of
network initialization. With the move to BSD-socket APIs,
we need to add support for DNS to the LwM2M library.
Signed-off-by: Michael Scott <mike@foundries.io>
We can save some resources by removing the periodic service thread
and replacing it by queuing the services to the work queue.
Before (reel_board using BT + DTLS)
Memory region Used Size Region Size %age Used
FLASH: 289464 B 1 MB 27.61%
SRAM: 75620 B 256 KB 28.85%
IDT_LIST: 136 B 2 KB 6.64%
After
Memory region Used Size Region Size %age Used
FLASH: 289576 B 1 MB 27.62%
SRAM: 74596 B 256 KB 28.46%
IDT_LIST: 136 B 2 KB 6.64%
Signed-off-by: Michael Scott <mike@foundries.io>
Now that the security data can be loaded into and used from the
security / server objects, we can add support for LwM2M bootstrap.
This is a mode where initially a connection can be made to a server
which can update several LwM2M (including security and server
data) and then trigger a "bootstrap complete". Once this happens
the client will start it's connection process over but now with
the new information.
Signed-off-by: Michael Scott <mike@foundries.io>
In order to support bootstrap mode, we need to store server data
in the security / server objects. Once the connection to the
bootstrap server is made, it will clear these objects and add
new server connection data.
Signed-off-by: Michael Scott <mike@foundries.io>
Update the parsing functions for JSON used by the JSON data
formatter and enable it in the LwM2M engine.
Signed-off-by: Michael Scott <mike@foundries.io>
net_app contexts save the remote address and we use this during
observe notifications and pending handling. If we move to another
network layer such as sockets, then the remote address becomes
harder to reference. Let's save it as a part of the client
context.
Signed-off-by: Michael Scott <mike@foundries.io>
As part of the migration from net_app APIs to socket APIs, let's
stop referencing the net_pkt fragments throughout the LwM2M library.
Establish a msg_data flat buffer inside lwm2m_message and use that
instead.
NOTE: As a part of this change we remove the COAP_NET_PKT setting.
The COAP library reverts to COAP_SOCK behavior.
This doesn't mean we use sockets in LwM2M (yet), it only means we
use the socket-compatible COAP library which parses flat buffers
instead of net_pkt fragments.
Signed-off-by: Michael Scott <mike@foundries.io>
