This commit reworks the ARM AArch32 non-Cortex-M (i.e. Cortex-A and
Cortex-R) exception handling to establish the base exception handling
framework and support detailed exception information reporting.
Signed-off-by: Stephanos Ioannidis <root@stephanos.io>
This commit adds an "extension" header for the CMSIS to provide the
core register access functions and macros that are not currently
available in the CMSIS.
The functions and macros defined in this header should be eventually
added to the upstream CMSIS and, once added, removed in this header.
Signed-off-by: Stephanos Ioannidis <root@stephanos.io>
DAC (digital to analog converter) peripheral driver with a generic API
suitable for most MCUs (only basic DAC features considered).
Signed-off-by: Martin Jäger <martin@libre.solar>
Zephyr being an OS is typically expected to run at EL1. Arm core
can reset to EL3 which typically requires a firmware to run at EL3
and drop control to lower EL. In that case EL3 init is done by the
firmware allowing the lower EL software to have necessary control.
If Zephyr is entered at EL3 and it is desired to run at EL1, which
is indicated by 'CONFIG_SWITCH_TO_EL1', then Zephyr is responsible
for doing required EL3 initializations to allow lower EL necessary
control.
The entry sequence is modified to have control flow under single
'switch_el'.
Provisions added by giving weak funcions to do platform specific
init from EL3.
Signed-off-by: Sandeep Tripathy <sandeep.tripathy@broadcom.com>
extern "C" is missing.
This commit adds conditional extern "C" for cpp to
the beginning of the file and brackets the cpp template with
extern "C++"
Signed-off-by: Steven Slupsky <sslupsky@gmail.com>
The return value of the interrupt is never used, so remove the
functionality from the API. We ripple this change into the
ARCH_IRQ_CONNECT and ARCH_IRQ_DIRECT_CONNECT implementations. With
this change we can also remove the compound expression as that is
not needed anymore.
Signed-off-by: Kumar Gala <kumar.gala@linaro.org>
Several functions that are syscalls were passing structs by value
instead of by reference. Just changed that and implemented missing
verfication handlers for them.
Signed-off-by: Flavio Ceolin <flavio.ceolin@intel.com>
espi_read_flash should have implementation called
z_impl_espi_read_flash and not z_impl_espi_flash_read. Same thing for
espi_write_flash.
Signed-off-by: Flavio Ceolin <flavio.ceolin@intel.com>
We can't pass __COUNTER__ and then use it with '##' directly, because
the result is that the actual token __COUNTER__ is what gets pasted.
Fix that with another expansion, so __COUNTER__ is replaced with a
number before '## can' see it.
Signed-off-by: Martí Bolívar <marti.bolivar@nordicsemi.no>
To remove the need to have DT_NUM_IRQ_PRIO_BITS defined in every
dts_fixup.h we can just handle the few variant cases in irq.h. This
allows us to remove DT_NUM_MPU_REGIONS from all the dts_fixup.h files.
Signed-off-by: Kumar Gala <kumar.gala@linaro.org>
Fix thread fault, on user mode, when reading variable rt_clock_base.
For the moment, clock_settime is left without system call:
we don't want to expose clock_settime without figuring out access
control
Signed-off-by: Julien D'Ascenzio <julien.dascenzio@paratronic.fr>
This commit adds more pin function definitions (PINMUX_FUNC_I through
PINMUX_FUNC_P) to the pinmux interface header.
The SAM D5x and E5x series devices, for instances, define up to the
"function N" and this change is necessary to support such devices.
Signed-off-by: Stephanos Ioannidis <root@stephanos.io>
The TCP flag was conflicting with PPP flag in the same union.
This prevented TCP from working properly with PPP.
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
New function allows to set from the code the root command. It is
an equivalent of calling 'select <rootcmd>' except it sets command
for all shell instances.
Signed-off-by: Krzysztof Chruscinski <krzysztof.chruscinski@nordicsemi.no>
This defines the configuration for the DMA client
as specifed in the ./dts/bindings/dma/st,stm32-dma.yaml
Signed-off-by: Francois Ramu <francois.ramu@st.com>
Create support for muxed UARTs which are attached to a real
UART and which use GSM 07.10 muxing protocol to create virtual
channels that can be run on top of the real UART.
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
Some drivers support NOCACHE_MEMORY sections. To have a default
fallback for systems where this isn't available or disabled a
fallback define is added.
Signed-off-by: Vincent van der Locht <vincent@vlotech.nl>
Signed-off-by: Gerson Fernando Budke <nandojve@gmail.com>
Thread and Zigbee differ in how do they handle Frame Pending Bit field
in their frames. Introduce a new enum, `ieee802154_fpb_mode`, which
allows to configure the radio driver in an appropriate mode.
Signed-off-by: Robert Lubos <robert.lubos@nordicsemi.no>
OpenThread expects information, whether Frame Pending Bit was set in the
ACK response sent by the radio driver. Carry this information in the
net_pkt structure.
Signed-off-by: Robert Lubos <robert.lubos@nordicsemi.no>
This patch introduces two major changes to the directed advertising
feature of the bluetooth host.
Deprecating the bt_conn_create_slave_le, and removing
bt_conn_le_create_slave which has never been released. This behaviour
has now been moved by to providing the peer direct address into the
advertising parameters.
Introducing directed advertising support for nonconnectable
directed extended advertising, both scannable and non-scannable.
A bug was also fixed in the the directed-adv command in the shell
when the argument "low" was given. The advertiseng parameter pointer
declared with BT_LE_ADV_CONN_DIR_LOW_DUTY was declared in a scope that
was no longer valid when it was used to start the advertiser.
Signed-off-by: Joakim Andersson <joakim.andersson@nordicsemi.no>
The implicit conversion of pointer-to-function to pointer-to-void is
not acceptable in C++. Provide a C-style explicit cast to force the
reinterpretation.
Signed-off-by: Peter Bigot <peter.bigot@nordicsemi.no>
With the generation of DT_N_<node-id>_{REG,IRQ}_IDX_<idx>_EXISTS
defines, we can now use the IS_ENABLED macro to implement DT_REG_HAS_IDX
and DT_IRQ_HAS_IDX. This matches how other DT_*_HAS_* macros are
implemented as well as lets us utilize these macros with COND_CODE_1.
Signed-off-by: Kumar Gala <kumar.gala@linaro.org>
Add a shim layer implementing the legacy k_mem_pool APIs backed by a
k_heap instead of the original implementation.
Signed-off-by: Andy Ross <andrew.j.ross@intel.com>
This adds a k_heap data structure, a synchronized wrapper around a
sys_heap memory allocator. As of this patch, it is an alternative
implementation to k_mem_pool() with somewhat better efficiency and
performance and more conventional (and convenient) behavior.
Note that commit involves some header motion to break dependencies.
The declaration for struct k_spinlock moves to kernel_structs.h, and a
bunch of includes were trimmed.
Signed-off-by: Andy Ross <andrew.j.ross@intel.com>
Almost all of the k_mem_pool API is implemented in terms of three
lower level primitives: K_MEM_POOL_DEFINE(), k_mem_pool_alloc() and
k_mem_pool_free_id(). These are themselves implemented on top of the
lower level sys_mem_pool abstraction.
Make this layering explicit by splitting the low level out into its
own files: mempool_sys.c/h.
Signed-off-by: Andy Ross <andrew.j.ross@intel.com>
Struct definitions contain no inlines that depend on other code so
should live early in the include tree. Upcoming refactoring needs
this to break header dependency cycles. The kernel_structs.h header
was designed for exactly this purpose.
Signed-off-by: Andy Ross <andrew.j.ross@intel.com>
The existing mem_pool implementation has been an endless source of
frustration. It's had alignment bugs, it's had racy behavior. It's
never been particularly fast. It's outrageously complicated to
configure statically. And while its fragmentation resistance and
overhead on small blocks is good, it's space efficiencey has always
been very poor due to the four-way buddy scheme.
This patch introduces sys_heap. It's a more or less conventional
segregated fit allocator with power-of-two buckets. It doesn't expose
its level structure to the user at all, simply taking an arbitrarily
aligned pointer to memory. It stores all metadata inside the heap
region. It allocates and frees by simple pointer and not block ID.
Static initialization is trivial, and runtime initialization is only a
few cycles to format and add one block to a list header.
It has excellent space efficiency. Chunks can be split arbitrarily in
8 byte units. Overhead is only four bytes per allocated chunk (eight
bytes for heaps >256kb or on 64 bit systems), plus a log2-sized array
of 2-word bucket headers. No coarse alignment restrictions on blocks,
they can be split and merged (in units of 8 bytes) arbitrarily.
It has good fragmentation resistance. Freed blocks are always
immediately merged with adjacent free blocks. Allocations are
attempted from a sample of the smallest bucket that might fit, falling
back rapidly to the smallest block guaranteed to fit. Split memory
remaining in the chunk is always returned immediately to the heap for
other allocation.
It has excellent performance with firmly bounded runtime. All
operations are constant time (though there is a search of the smallest
bucket that has a compile-time-configurable upper bound, setting this
to extreme values results in an effectively linear search of the
list), objectively fast (about a hundred instructions) and amenable to
locked operation. No more need for fragile lock relaxation trickery.
It also contains an extensive validation and stress test framework,
something that was sorely lacking in the previous implementation.
Note that sys_heap is not a compatible API with sys_mem_pool and
k_mem_pool. Partial wrappers for those (now-) legacy APIs will appear
later and a deprecation strategy needs to be chosen.
Signed-off-by: Andy Ross <andrew.j.ross@intel.com>
Add a devicetree/zephyr.h header, which is meant to contain
definitions for /chosen properties specific to Zephyr.
Currently, this just deals with zephyr,entropy. We add a
DT_CHOSEN_ZEPHYR_ENTROPY_LABEL macro which expands to the label for the
node pointed to by zephyr,entropy.
Signed-off-by: Martí Bolívar <marti.bolivar@nordicsemi.no>
Signed-off-by: Kumar Gala <kumar.gala@linaro.org>
Move the include of files from include/devicetree/*.h to the end so that
when those files are processes they have access to all previous defined
macros.
Signed-off-by: Kumar Gala <kumar.gala@linaro.org>
At least one flash driver requires that the source and destination
buffers be word-aligned. Annotate the synthesized definitions to make
sure this happens.
Signed-off-by: Peter Bigot <peter.bigot@nordicsemi.no>
Commit c408fa88a3 introduced changes
to the BUILD_ASSERT() macro so the MSG argument is optional.
However, in include/toolchain/common.h, the BUILD_ASSERT()
definition has not made the MSG argument as optional which results
in build errors complaing about too few arguments. Fix by adding
some dots there.
This is observed when using XCC (based on Clang 3.9).
Fixes#24008
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
Static analyzer (clang-tidy) complains about too few arguments for
the BUILD_ASSERT.
There is missing second argument to BUILD_ASSERT used inside of
the K_MEM_POOL_DEFINE.
Signed-off-by: Artur Lipowski <artur.lipowski@hidglobal.com>
z_isr_install is not suited to handle multi-level interrupt formats.
This update allows z_isr_install to accept irq numbers in zephyr format
and place them in the isr table appropriately.
Fixes issue #22145
Signed-off-by: Jaron Kelleher <jkelleher@fb.com>
Allow C++ code to evaluate time base conversion routines at compile time
by marking them as constexpr where possible.
Signed-off-by: Josh Gao <josh@jmgao.dev>
When configuring the I2S peripheral as a master, the DMA channel
direction must be configured to transfer data from memory to the
peripheral.
Currently the configuration of channel direction is always set for
peripheral to memory regardless of whether it is the TX or the RX
channel.
Signed-off-by: Abe Kohandel <abe@electronshepherds.com>
HWINFO drivers should be responsible for ensuring that
the data structure is a sequence of bytes. That is not
what the current sam0 and nordic drivers do. The drivers
read the data as u32_t and then memcpy the data to a
buffer. This ensures the data has the endianness of the
underlying MCU, which in this case is Cortex M0 which
is little endian.
This commit fixes the endianness so the data can be
interpreted as a "left to right sequence of bytes".
This commit updates the API doc to provide clarification
of the data structure.
Add to 2.3 release notes.
Fixes#23444, #24103
Signed-off-by: Steven Slupsky <sslupsky@gmail.com>
This change is to allow access to the firmware block context in order to
give the firmware update callback implementation an indication of when
to reset the flash context. Additionally, it allows for a validity check
between the total expected size downloaded and the actual size
downloaded. A simple implementation can check if the block context's
current downloaded blocks is equal to the expected block size in order
to determine if the flash context needs to be reset. This approach
seemed the simplest, and knowing the firmware block context can have
other purposes. This has been tested by accessing the block context
during the update in the block received callback and confirming that the
callback had information regarding the current downloaded bytes.
Fixes#16122
Signed-off-by: Kyle Sun <yaomon18@yahoo.com>
