If new events are signaled during processing then they should be processed when processEvent is called again. The goal is to let other processing happen and not process sollely ble events.
This change has been forced by a change in latest softdevice that requires all key pointers to not be NULL unlike what is indicated in the documentation.
Previously EBAD (invalid exchange), mapping the error CORRUPT to EILSEQ
(illegal byte sequence) makes more sense as a description of the type of
error.
Add the allocate_key API. This replaces the previously added set_alloc_key API
(which allocates a key and sets the value at the same time).
Reason for the change: Key allocation will typically be used by other storage
features (like StorageLite), keeping the allocated keys in another location.
Previous API created problems in the case key allocation and value setting
couldn't be done at the same time (for instance, if the set value was
derived from the allocated key, such as hash or CMAC).
Right now, many users are trying out many different filesystems.
Unfortunately, this can leave partially written filesystems on disk
in various states.
A very common pattern for using embedded filesystems is to attempt
a mount, and on failure, format the storage with the filesystem.
Unfortunately, this simply doesn't work if you try to change the
filesystem being used on a piece of storage. Filesystems don't always
use the same regions of storage, and can leave enough metadata lying
around from old filesystems to trick a different mount into thinking a
valid filesystem exists on disk. The filesystems we have were never
designed to check for malicious modification and can't protect against
arbitrary changes.
That being said, it's caused enough problems for users, so as a
workaround this patch adds a disk erase to the FAT filesystem format.
The most common error happens when you use LittleFS, followed by FAT,
followed again by LittleFS.
No other combination of filesystem usage has shown a similar failure,
but it is possible after extensive filesystem use, so it is still
suggested to force a format of the storage when changing filesystems.
In the reception data path, we needed to check the MCPS CONFIRMATION type
not the MCPS INDICATION type. Indication message type is for downlink message type
which can be UNCONFIRMED even if we have sent a CONFIRMED one, e.g., an ACK.
Add support for Alternative ECDSA and ECDH, on the higher level,
over CC310. Note that CC generates ECC keys according to FIPS 186,
while mbed TLS generates according to RFC 6979 and RFC 4754,
which causes test vectors for curve p521 to fail
Fixes error: IOTCLT-2506 [GitHub] Cannot set registration time if server does not use max age option
Improvements; Extend blockwise message transfer status to have states for sending as well.
NOTE! These are internal changes required for cloud client. This has no direct relevance to any mbed-os functionality.
Our config file for lwIP had TCP_QUEUE_OOSEQ disabled - this can
cause significant performance problems, as observed during testing.
One lost packet can lock an input stream into a mode where the
transmitter keeps thinking packets are being lost, so keeps slowing
down. This caused test failures - a transfer that would normally
take 10s hit a 60s timeout.
Turning this on increases code size, but doesn't significantly increase
static memory use. The memory used for out-of-order packets comes from
the same pbuf pool as for outgoing TCP segments, so there is contention
when running bidirectionally.
Out-of-order processing is on by default for lwIP - this seems to be
another example of us excessively paring it back.
Application can give battery_level callback method what Lora stack
uses to query battery level for DevStatusReq MAC command response.
The problem was that this callback was never called.
This commit fixes this problem and if application does not set
battery_level callback at all, value 255 (= The end-device was not able to
measure the battery level.) will be returned to lora gateway.
The unified NRF51 target and feature BLE directories have been
reorganized to follow the naming and directory structure of the
NRF52 implementation.
This reorganization does not include TARGET_MCU_NRF51822 and
derived targets.
If the frame length is not what we are expecting, it is
found to be a good practise to actually continue with what we
have received rather than aborting. As we have already demodulated
the packet and RX slots are used up, ther is not so much benefit in
dropping that packet.
While configuring RX parameters for the radio, we need to feed in
rx windows 1 and 2 parameters which are computed when we do the transmission.
We are actually setting the physical value of the data rate rather than
data rate table index and the expectation was to set the data rate index.
The asia pacific region supports custom channel planning and
downlink channel request. By virtue of a mistake, this information
was missing and hence a custom channel support was not working.
Fixes issue #6783.
There had been essentially two state machines running in our stack
which was too cumbersome and was not alligned in any symmetry.
In this work we make sure that:
* There are no callbacks from the MAC layer to Stack controller layer.
* Primitives are made local to the mac layer and are presented as
read-only to the stack controller layer.
* Interrupt handling and processing is moved to the stack controller layer.
* Reception is divided into smaller units, seperating handling of Join Accept
and normal data frames. MIC gets its own unit.
* Extraction of data and MAC commands from the payload is also being done now in
its own method.
* To ensure integrity of the stack, and sanctity of the radio payload, we copy the
radio payload buffer immediately in the rx interrupt and hoist a flag that prevents
another interrupt from happening for a short while when we are processing the previous
packet.
* If an automatic uplink is on going, we do not send a TX_DONE event to application
anymore as that is logically incorrect.
* state_controller() is the central engine for the state machine. To save code space and
memory, we are not handling each and every state in the state_controller(). Some of the states
which have no processing to be done, are explicitely set.
* For all the states who need special processing, seperate methods are added.
* Class A always run to completion to IDLE and CLass C always runs to completion as RECEIVING.
Making our LoRaWAN stack thread safe. If RTOS is not present, locks
don't do anything. ScopedLock is used to automate the lock release on
context expiry.
We went through an exercise of adding band information to
any new channel being added. Default channels were looked over.
This commits duly adds missing band information to default channels.
The procedure should be terminated whenever the server returns an error not equal
to ATTRIBUTE_NOT_FOUND. The block was effectivelly terminated but the
procedure was not. As a result the discovery was operating on already
freed memory.
If Lora stack is built with incorrect mbedtls settings, crypto methods
should assert and return error value. This fixes MBED_ASSERTs to
work correctly.
Now application can call connect and disconnect multiple times and resources are freed and constructed properly.
Also whole easycellular can be deleted and constructed again.
This patch refines permission applied to characteristic and descriptors; instead of a single level of permission , each characteristic receives a permission for the read operation, one for the write operation and another one for the update operation.
As a consequence, updates are not sent if the link does not cover the update permission requirement.
Descriptors also benefits individually from read and write permission.
The registration process has been breaked down into several functions that register the service attribute, characteristic declaration attributes, characteristic value attributes and characteristic descriptors.
Service registration now consider all characteristics permissions: read, write and update. Permissions are also considered when updates needs to be propagated to peers.
Handling of user authorization is also a change introduced by this refactoring.
A peripheral can act as a GATT client whether it is in the peripheral role or the central role therefore it doesn't make sense to enable signing only if roles will be reversed latter.
If the the upper layer has registered a peer csrk on the pal security manager then it must remove it once the connection is closed.
This API allows the upper layer to remove the peer csrk registered earlier.
A peer may not share a valid IRK and identity address during pairing (in that case the identity address received is all zeros). When this happens, the entry must be retrieved by looking at the address used by the peer during the connection.
Paths such as the following were causing issues:
/tea/hottea/.
/tea/hottea/..
Unfortunately the existing structure for path lookup didn't make it very
easy to introduce proper handling in this case without duplicating the
entire skip logic for paths. So the lfs_dir_find function had to be
restructured a bit.
One odd side-effect of this is that now lfs_dir_find includes the
initial fetch operation. This kinda breaks the fetch -> op pattern of
the dir functions, but does come with a nice code size reduction.
Nanostack related files moved under 'feature/nanostack'
Common libraries moved to 'features/frameworks'
Allow FEATURE_COMMON_PAL still to be defined in the build so
that we don't break any builds.
This patch adds independent management of security requirement for read, write and update operatiosn of a GattCharacteristic.
Requirements are defined after ble::att_security_requirement_t that maps LE security mode 1 and LE security mode 2.
The functions requireSecurity and getRequiredSecurity are deprecated as SecurityManager::ScurityMode_t does not map well with LE security modes.
band_t structure in phy parameters was not being filled in properly
and we were spilling over the array boundary for bands.
In addition to that two utility functions are added to perform safety
checking taking in the frequency and filling out band information plus
boundry value checks.
In LoRa TX power value 0 means the maximum allowed TX power and values >0
are limiting the allowed TX power to lower.
tx_config was incorrectly checking the power level and causing the maximum
TX power to be always used. Lora gateway can request node to use lower TX
power with LinkAdrReq MAC command.
In order to have smaller and better testable methods, instead of having a big
switch-case -style add_mac_command(), each MAC command is now separated to an
own method.
* Updates driver library to v2.3.1 (2018q1) for bugfixes and convenience functions
* Provides library in correct format (2-byte wchar_t flag) for compiling with ARMCC (#6695 uncovered by #6577)
* Reverts to using a statically-allocated packet buffer since malloc is not thread-safe (and the asserts have been turned on)
* Add NRF_SD_BLE_API_VERSION=5 macro to MCU_NRF52832 target
* Added initial Softdevice v5 API support for nRF5XPalGattClient.cpp
* Added initial Softdevice v5 API support for nRF5xGap.h
* Added initial Softdevice v5 API support for nRF5xGattServer.cpp
* Make sure the ah() hash function used by the security manager is declared with C linkage
* Removed 'ARM porting pending' checks in Nordic SDK
* Added BLE configuration to mbed_lib.json for NRF52x targets
* Migrated APIs and events model for Nordic SDK14.2 & SD API v5 in btle.cpp; adopted polling dispatch model
* Migrated APIs and events model for Nordic SDK14.2 & SD API v5 in nRF5xn.cpp
* Migrated APIs and events model for Nordic SDK14.2 & SD API v5 in nRF5xGap.cpp
* Added const qualifier to hwCallbacks's p_ble_evt parameter in nRF5xGattServer
* Add workaround to enable Radio Notification module from Nordic SDK
* Disabled SWI1 interrupt manually from SWI driver module
* Added missing include directive for "nrf_nvic.h" in ble_radio_notification.c
* Set correct number of custom UUIDs
* Adjust Softdevice RAM requirements for NRF52832/S132
* RAM settings for the application were adjusted as follow:
* Start: 0x20003800
* Size: 0xC800
* Adjust Softdevice RAM requirements for NRF52840/S140
* RAM settings for the application were adjusted as follow:
* Start: 0x20003700 (was 0x20003000)
* Size: 0xC900
* Added 3dBm as a permitted TX Power value for NRF52832
* Fix address type recovery in GAP
* Return min non-connectable adv interval compatible with infrastrcture's expectations
* Moved BLE driver from TARGET_SDK_14_2/FEATURE_BLE to TARGET_SDK_14_2/TARGET_SOFTDEVICE_COMMON/ble
Currently lora stack will automatically send an empty uplink message to lora gateway in case of:
- Node received message with pending bit set.
- Node received MAC command which requires instant response (sticky MAC command)
- Node received confirmed message in class C mode
This commit makes this configurable via config item
"automatic-uplink-message": {
"help": "In case of pending bit, class c confirmed message or sticky MAC command, stack will automatically send empty uplink message",
"value": true
}
Default value is true. If sending an empty message fails, stack will send event AUTOMATIC_UPLINK_ERROR application.
If automatic uplink sending is disabled, stack will send application UPLINK_REQUIRED -event to indicate
application should issue a new uplink to gateway as soon as possible.
The DB entry lookup now looks at the identity address and/or connection address to find a DB entry associated to an address.
If the entry has not been found in the DB and a new entry is returned then the connection address is not stored for private addresses.
lwIP was using us_ticker unnecessarily, complicating the code and
potentially causing grief with power saving, and possible glitches
on timer wrap. Switch it to use the RTOS tick count.
Compiler reminded that a variable declaration was left behind when
the code using it was put behind #ifdef. Add the missing #ifdef.
Warning being fixed:
---8<---8<----
[Warning] ns_event_loop.c@44,0: #177-D: variable "event_thread_id"
was declared but never referenced
nanostack-hal.critical-section-usable-from-interrupt -tunable was
previously added to optionally make critical section code interrupt safe.
The IRQ safe critical section is a prequisite for interrupt safe timer
callbacks.
The same flag can be used to enable calling of the timer callbacks
directly from the timer interrupt context, without bouncing them via
event thread. This removes the code and RAM consumed by EventQueue
and the thread serving the high priority events.
If the system does not have any dependencies on mbed_shared_queues,
by setting this flag the static RAM usage is now further reduced
by ~1600 bytes and code size by 4KB.
Note: the default behavior is not changed, one needs to override the
"nanostack-hal.critical-section-usable-from-interrupt" to have "true".
Any data structure used in LoRaWANBase class should be available
in a separate header in order to make the code easy to port and
easy to read as the developer doesn't need to know about all the
internal data structures being used in Mbed LoRaWAN stack.
Compile: lwip_stack.c
In file included from ../features/FEATURE_LWIP/lwip-interface/lwip_stack.c:41:0:
../features/FEATURE_LWIP/lwip-interface/lwip_stack.c: In function 'mbed_lwip_bringup_2':
../features/FEATURE_LWIP/lwip-interface/ppp_lwip.h:58:44: warning: statement with no effect [-Wunused-value]
#define ppp_lwip_disconnect() ERR_IF
^
../features/FEATURE_LWIP/lwip-interface/lwip_stack.c:858:21: note: in expansion of macro 'ppp_lwip_disconnect'
ppp_lwip_disconnect();
^~~~~~~~~~~~~~~~~~~
../features/FEATURE_LWIP/lwip-interface/ppp_lwip.h:58:44: warning: statement with no effect [-Wunused-value]
#define ppp_lwip_disconnect() ERR_IF
^
../features/FEATURE_LWIP/lwip-interface/lwip_stack.c:875:21: note: in expansion of macro 'ppp_lwip_disconnect'
ppp_lwip_disconnect();
^~~~~~~~~~~~~~~~~~~
Contains one error fix:
IOTCLT-2469 CoAP UDP retransmission does not work for blocks after first one for requests (Eg. registration POST)
NOTE! These are internal changes required for cloud client. This has no direct relevance to any mbed-os functionality.
As pointed out by davidefer, the lookahead pointer modular arithmetic
does not work around integer overflow when the pointer size is not a
multiple of the block count.
To avoid overflow problems, the easy solution is to stop trying to
work around integer overflows and keep the lookahead offset inside the
block device. To make this work, the ack was modified into a resetable
counter that is decremented every block allocation.
As a plus, quite a bit of the allocation logic ended up simplified.
One of the big simplifications in littlefs's implementation is the
complete lack of tracking free blocks, allowing operations to simply
drop blocks that are no longer in use.
However, this means the lookahead buffer can easily contain outdated
blocks that were previously deleted. This is usually fine, as littlefs
will rescan the storage if it can't find a free block in the lookahead
buffer, but after changes that caused littlefs to more conservatively
respect the alloc acks (e611cf5), any scanned blocks after an ack would
be incorrectly trusted.
The fix is to eagerly scan ahead in the lookahead when we allocate so
that alloc acks are better able to discredit old lookahead blocks. Since
usually alloc acks are tightly coupled to allocations of one or two blocks,
this allows littlefs to properly rescan every set of allocations.
This may still be a concern if there is a long series of worn out
blocks, but in the worst case littlefs will conservatively avoid using
blocks it's not sure about.
Found by davidefer
This commit adds API to enable and configure the device privacy.
It deprecates address random types present in Gap::AddressType as these types are not appropriate for scan reports, connection initiation and the connection event. Now user should use the function Gap::getRandomAddressType to find the type of a random address.
The function gap::setAddress is deprecated as it is not portable and can colide with privacy.
Bug was reported by @tpet93 and it was correctly diagnosed that channels
were not being set in the phy_params structure. This commit fixes the bug
and sets up channels correctly.
This is a fix for issue #6389.
Currently when application receives RX_DONE event from stack, it has to provide the correct port
value to receive method in order to read the received message. The problem is that current
API does not provide any way to know in to which port message was received.
This commit introduces a new receive() method, which instead of checking these values, will return
them to application.
Yossi Levy
Vincent Coubard
Donatien Garnier
paul-szczepanek-arm
Cruz Monrreal
Kimmo Vaisanen
Christopher Haster
David Saada
Juhani Puurula
Bartek Szatkowski
Qinghao Shi
Hasnain Virk
Ron Eldor
Amanda Butler
Ari Parkkila
Antti Yli-Tokola
Kevin Bracey
Krzysztof Stachowiak
Martin Kojtal
Steven Cartmell
Marcus Chang
Teppo Järvelin
Brendan McDonnell
Mirela Chirica
Seppo Takalo
Steven Cooreman
Teemu Kultala
Paul Thompson
Tero Jääskö
jeromecoutant
ccli8
Antti Kauppila