At some point OMA TLV integer encoding was optimised to use the smallest
size possible. This broke the objlnk encoding, which internally used
`put_s32()`, but should always be 4 bytes long. In case the result
32-bit integer would fit into 16 or 8 bytes, it'd get optimized.
Fix this, by creating and writing TLV manually for objlnk instead of
relying on int32 encoder.
Signed-off-by: Robert Lubos <robert.lubos@nordicsemi.no>
In case a string with an integer was provided to the function (no
decimal point), the function did not update the output pointer value.
Signed-off-by: Robert Lubos <robert.lubos@nordicsemi.no>
Verify the restult of the content writer functions and return an error
if content writer fails to read/write field in the message.
This solves an issue when for example malformed packet was sent to the
server if the payload did not fit in the message.
Signed-off-by: Robert Lubos <robert.lubos@nordicsemi.no>
It's now possible to return an error code instead of 0 where
appropriate.
Make `buf_read()` return -ENODATA instead of -ENOMEM if there's not
enough data in the packet to serve the request, which is more meaningful
for this scenario.
Signed-off-by: Robert Lubos <robert.lubos@nordicsemi.no>
Rework the content encoder/decoder API to allow to return negative
values as well. This allows a proper error reporting and error handling,
as it's now possible to differentiate when there is no content to write
(retuned 0) or and error occured, and further processing should be
aborted.
Signed-off-by: Robert Lubos <robert.lubos@nordicsemi.no>
Update test to ensure that any file can be tranfered,
not only files that are aligned with the packet size.
Signed-off-by: Efrain Calderon <efrain.calderon@aquarobur.com>
Signed-off-by: Robert Lubos <robert.lubos@nordicsemi.no>
When a file is download, `ctc->current` is increaed by the block
size, which is correct for all the blocks except the last one.
Once the download is compelted, ctx->current must match
`ctx->total_size`.
Signed-off-by: Efrain Calderon <efrain.calderon@aquarobur.com>
Test that verifies correct handling of Version exchanged procedure
called twice on the same connection failed. The reason was wrong
number of expected free procedure context objects. It was expected
that there will be one context not released.
That was wrong because the context is released just after the procedure
completes in idle state. The second and following calls to the procedure
do not start any PDU exchange, so the procedure ends immediately.
Signed-off-by: Piotr Pryga <piotr.pryga@nordicsemi.no>
The tests for direction finding connected mode were not building
correclty. There were missing Kconfig options that enable missing
functionality to be tested.
Tests were not failing because there were no expected positive
behavior tests added.
Added changes allow to validate all implemented test cases.
Signed-off-by: Piotr Pryga <piotr.pryga@nordicsemi.no>
There were no wait for free notification buffer (pdu object
used to send notifications to Host).
That may cause assertion in lp_comm_ntf function.
Signed-off-by: Piotr Pryga <piotr.pryga@nordicsemi.no>
To avoid violation of BT 5.3 Core Vol 6, Part B section 5.1.10.1
there was added a command pause mechanism that allows to postpone
handling of CTE REQ if there is pending PHY change procedure or
PHY change if there is pending CTE REQ procedure.
The commit adds unit tests for the functionality.
Signed-off-by: Piotr Pryga <piotr.pryga@nordicsemi.no>
Accodring to BT 5.3 Core Vol 4, Part E section 7.8.85 the CTE request
procedure has to be disabled as if Host issued the HCI_LE_Connection-
CTE_Request_Enable with enable property set to 0. It means a Controller
should automaticall disable the procedure.
There were no handling of this part of specification.
Signed-off-by: Piotr Pryga <piotr.pryga@nordicsemi.no>
There were missing pause of CTE request in case remote PHY change
was handled. That may end in violation of BT 5.3 Core Vol 6, Part B
section 5.1.10.1 Packet transmission restrictions.
Without the change there was a possibility to transmitt CTE when
PHY is about to change to PHY CODED.
Signed-off-by: Piotr Pryga <piotr.pryga@nordicsemi.no>
Add handling of complete event in idle state of localy innitiated
control procedures.
The transition will be used by CTE request disable if PHY changes
to CODED while the procedure is waiting in the waiting queue.
Signed-off-by: Piotr Pryga <piotr.pryga@nordicsemi.no>
CTE request may not be enabled when actual PHY does not support
transmission of constant tone extension. In case a connection is
asymetric, receive PHY is required to support transmission of CTE.
Receive PHY will be used by a device to get PDU with CTE.
In ll_df_set_conn_cte_req_enable was checking TX PHY type.
It should check RX PHY type.
Signed-off-by: Piotr Pryga <piotr.pryga@nordicsemi.no>
Add a function that is responsible for set CTE request disabled.
It is a refactoring, to have a single place where CTE request
is_enabled and req_interval are zeroed.
Signed-off-by: Piotr Pryga <piotr.pryga@nordicsemi.no>
There were missing set of is_enabled member of CTE request.
That allowed host to run HCI_LE_Connection_CTE_Request_Enable
command with enable parameter set to true sequently.
That violates BT 5.3 Core specification Vol 4 Part B section
7.8.8.85.
Signed-off-by: Piotr Pryga <piotr.pryga@nordicsemi.no>
GCC-based XCC uses GNU Assembler (xt-as). However, CMake doesn't
recognize it when invoking through xt-xcc. This results in CMake
going through all possible combinations of command line arguments
while invoking xt-xcc to determine assembler vendor. This multiple
invocation of xt-xcc unnecessarily lengthens the CMake phase of
build, especially when XCC needs to obtain license information
from remote licensing servers. So here forces the assembler ID
to be GNU to speed things up a bit.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
Gives more explanation on uart_stm32_err_check function.
On stm32 F4X, F1X, and F2X, when clearing the usart Error Flag
(PE, ORE, FE, NE), the LL Clear function applies a software sequence
which reads the usart SR then the usart DR.
Consequently the RXNE flag is affected (cleared) by the
uart_stm32_err_check function call.
Signed-off-by: Francois Ramu <francois.ramu@st.com>
The commit fixes smp_process_request_packet not checking header
stated message length against actual message length.
This could lead to an erratic behavior of an application when it tries
to parse CBOR data beyond actual size of message or potential
Zephyr crash.
When application has asserts enabled the error leads to inevitable
Zephyr crash, caused by net_buf checks.
Signed-off-by: Dominik Ermel <dominik.ermel@nordicsemi.no>
This patch adds fixes to the api so that it behaves as expected.
1 - The irq_tx_ready now only returns true if the tx interrupt is
enabled.
2 - The irq_tx_complete now functions as expected and returns true
only once all characters have been transmitted
Signed-off-by: Marius Scholtz <mariuss@ricelectronics.com>
The workaround doesn't seem to be necessary when using latest XDS110
firmware (3.0.0.13).
Signed-off-by: Gerard Marull-Paretas <gerard.marull@nordicsemi.no>
The workaround doesn't seem to be necessary when using latest XDS110
firmware (3.0.0.13).
Signed-off-by: Gerard Marull-Paretas <gerard.marull@nordicsemi.no>
This reverts commit deeb98da53.
A less invasive change has been implemented which does not
require changes to the GCC compiler flag file. So revert
the commit.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
Some older versions of XCC Clang would result in the following
error during compilation:
/tmp/file.s: Assembler messages:
/tmp/file.s:20: Error: file number 1 already allocated
clang-3.9: error: Xtensa-as command failed with exit code 1
due to a bug in LLVM: https://bugs.llvm.org/show_bug.cgi?id=11740.
This is fixed in upstream, https://reviews.llvm.org/D20002, in 2016.
However, it seems that it is only fixed after XCC RI-2018.0.
Instead of blanket disabling usage of '-g', use an environment
variable "XCC_NO_G_FLAG" to disable usage of flag '-g' to workaround
this issue. This needs to be manually set because there is no way to
know which XCC version is being used, and compiler flag checking for
'-g' would not result in error (and thus '-g' is not ignored).
This is only needed for older XCC Clang. For sufficiently new XCC
verisons, there is no need for this.
Note that this is an alternative implementation to commit
deeb98da53. This one does not
alter the GCC flag file, and instead, simply clear the debug
compiler property so "-g" is not used.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
Replace '%s' with '%ld', so that 'long' value is properly printed with
shell_error(). This suppresses following warning:
zephyr/drivers/lora/shell.c:181:23: warning: format '%s' expects \
argument of type 'char *', but argument 4 has type 'long int' \
[-Wformat=]
181 | shell_error(shell, "Invalid bandwidth: %s", lval);
| ^~~~~~~~~~~~~~~~~~~~~~~ ~~~~
| |
| long int
Signed-off-by: Marcin Niestroj <m.niestroj@grinn-global.com>
Fix net_pkt leak by increasing net_context the reference count earlier
in the zsock_accepted_cb() with instalment of the
zsock_received_cb() callback.
And consequently flushing recv_q and decrement net_context
reference count if zsock_accept_ctx() fails.
Signed-off-by: Daniel Nejezchleb <dnejezchleb@hwg.cz>
In case system workqueue processing is delayed for any reason, and
resolver callback is executed after getaddrinfo() call already timed
out, the system would crash as the callback makes use of the user data
allocated on the stack within getaddrinfo() function.
Prevent that, by cancelling the DNS request explicitly from the
getaddrinfo() context, therefore preventing the resolver callback
from being executed after the getaddrinfo() call ends.
Signed-off-by: Robert Lubos <robert.lubos@nordicsemi.no>
On nrf52832 disabling low frequency clock results in RTC COUNTER
reset. It is unexpected and system clock can be disrupted and
test may hang. Disable test which restarts LF clock for nrf52832.
Signed-off-by: Krzysztof Chruscinski <krzysztof.chruscinski@nordicsemi.no>
On nrf52832 LF clock cannot be stopped during runtime because
it resets RTC COUNTER. Testsuite run on nrf clock control driver
assumes that it will not happen. Disabling testing of LF clock
for nrf52832.
Signed-off-by: Krzysztof Chruscinski <krzysztof.chruscinski@nordicsemi.no>
Now that the forced flag is cleared with the kernel locked in the
suspend path, we need to clear it out when the suspend process fails.
This is being done before jump to the end label because in the
successful path the flag is already cleared.
Signed-off-by: Flavio Ceolin <flavio.ceolin@intel.com>
The "state forced" flag has to be cleared with the scheduler locked,
otherwise the idle thread can be scheduled out without clear it and
the next call to force a state will fail.
Fixes#41911
Signed-off-by: Flavio Ceolin <flavio.ceolin@intel.com>
Changes made to i2c, dma, and tracing for improved debugability
and hardware support. Adding driver level stats, dma api improvements,
and syscall tracing.
Signed-off-by: Tom Burdick <thomas.burdick@intel.com>
Deprecate the can_write() API wrapper function in favor of using
can_send() directly.
The can_write() API wrapper function:
- Does not support CAN-FD data lengths.
- Auto-selects between standard/extended CAN identifiers based on the
CAN ID (something which should be selected by the caller as the
selection affects filters on the receiving end).
- Does not support callbacks for frame transmission results.
- Has no in-tree users.
Signed-off-by: Henrik Brix Andersen <hebad@vestas.com>
Initial support for GigaDevice is now available upstream. This patch
lists relevant additions such as supported boards, drivers, etc.
Signed-off-by: Gerard Marull-Paretas <gerard@teslabs.com>
