This uses bitarrays for allocating and deallocating virtual
addresses with k_mem_map() and k_mem_unmap(). This will
allow us to reuse virtual addresses.
Fixes#28900
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
This implements arch_page_phys_get() to translate mapped
virtual addresses back to physical addresses.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
This adds a new function prototype for arch_page_phys_get()
which will be used to translate mapped virtual addresses back
to physical memory addresses. This is needed for the future
k_mem_unmap() function which requires this to find
the corresponding page frame. It is faster to look through
the page tables instead of doing linear search of the page
frame array.
A weak function is provided in case arch_page_phys_get()
is not implemented at the arch level. This simply goes
through all the page frame and find the one which has
mapped to the virtual address.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
When we start allowing unmapping of memory region, there is no
exact way to know if k_mem_map() is called with guard page option
specified or not. So just unconditionally enable guard pages on
both sides of the memory region to hopefully catch access
violations.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
This adds a test for z_phys_unmap() to make sure that memory
can be unmapped and is no longer accessible.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
This provides a counterpart to z_phys_map() which can be used
to temporary map memory region during boot process, and
subsequently discards the mapping.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
This waits a bit for NRU eviction algorithm (which is the default)
to work its magic to clear the access bit of physical frames.
This increases the number of clean pages which can be evicted,
to make sure the number of clean pages evicted is not zero, which
would cause an assertion.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
The test itself is highly sensitive to the size of the kernel
image. When the kernel gets larger, the number of pages used by
the backing store needs to shrink. So here this is.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
In z_mem_manage_init(), z_free_page_count is only manipulated
after all reserved pages are marked, and will reflect
the actual number of page frames being added to the free page
frame list. Manipulating z_free_page_count before this is
going to mess up the accounting, so remove the code to
decrement z_free_page_count in arch_reserved_pages_update()
under x86.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
This adds some tests to make sure sys_bitarray_*() are
working correctly.
Signed-off-by: Lauren Murphy <lauren.murphy@intel.com>
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
This introduces bit arrays as a new data type. This is different
than sys_bitfield as it is working on raw arrays of 32-bit
data. The bit arrays encode additional data inside the struct
to avoid going beyond the declared number of bits, and also
provides locking.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
On RISC-V 64-bit, GCC complains about undefined reference
to 'ffs' via __builtin_ffs(). So implement a brute force
way to do it. Once the toolchain has __builtin_ffs(),
this can be reverted.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
This commit enhances the CAN API test. In the send_receive tests,
we are using two filters and frames at the same time to check
if the frame gets dispatched to the correct filter.
Signed-off-by: Alexander Wachter <alexander@wachter.cloud>
Implementation of the Bosch M_CAN IP driver.
This driver is just the base for a specific SoC implementation.
Signed-off-by: Alexander Wachter <alexander@wachter.cloud>
We do not want those .bin files, also remove intermediate files leading
to the final ELF, those are huge.
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
Add the missing .cached section on cAVS 2.0 and 2.5 to fix
compiler warnings.
Signed-off-by: Guennadi Liakhovetski <guennadi.liakhovetski@linux.intel.com>
PLATFORM_RESET_MHE_AT_BOOT and
PLATFORM_DISABLE_L2CACHE_AT_BOOT have to be defined on cAVS 1.8
too, without them secondary cores cannot be used.
Signed-off-by: Guennadi Liakhovetski <guennadi.liakhovetski@linux.intel.com>
uncache_to_cache(), cache_to_uncache() and is_uncached() macros
are currently unused in Zephyr, besides the former two are
dangerous because they add and subtract an offset instead of
masking and unmasking a bit, which can lead to wrong addresses
if applied repeatedly. Remove these macros for now, we can
re-add them later if needed in their fixed versions.
Signed-off-by: Guennadi Liakhovetski <guennadi.liakhovetski@linux.intel.com>
work_timeout() is a function, a statement like "(void)work_timeout;"
has no effect.
Signed-off-by: Guennadi Liakhovetski <guennadi.liakhovetski@linux.intel.com>
Currently only cAVS 1.5 Kconfig selects ARCH_HAS_COHERENCE,
which is wrong. cAVS 1.8, 2.0 and 2.5 have to select it too.
Signed-off-by: Guennadi Liakhovetski <guennadi.liakhovetski@linux.intel.com>
- bootutil_public: allow to confirm padded image without
copy-done flag
- Configure mimxrt106x_evk boards
fixes#34683
Signed-off-by: Andrzej Puzdrowski <andrzej.puzdrowski@nordicsemi.no>
Restrict sample execution in CI to boards that have
"st,stm32-lptim" enabled so we're sure it is a STM32 target.
Signed-off-by: Erwan Gouriou <erwan.gouriou@linaro.org>
The energy consumption on nrf52840 is unnecessarily high when
CONFIG_NORDIC_QSPI_NOR is used due to Anomaly 122. The nrf_qspi_nor
driver is unitialize after QSPI usage and initialize before using
it again.
Semaphore objects are used to allow multiple threads exclusive access
and efficient usage.
The main assumption made was that all QSPI API is stateless, in the
sense that it is not required to store any peripheral state before
uninit. Also, the QSPI driver was supposed to be synchronous, except
for the erase operation, in which the nrf signals its start, instead of
its end. While the flash is performing the erase, an uninit followed
by an init doesn't work. For that reason, polling is done before
every uninit.
Tests were made with a simple LittleFS application in a custom board
using flash MX25R3235F and another with the LittleFS flash sample
using nrf52840 DK that has a MX25R6435F. Current consumption dropped
from 630 uA to ~10uA in both cases.
Signed-off-by: Rodrigo Brochado <git.rodrigobrochado@gmail.com>
Previously to this commit, nothing was done to restore the bus to an
healthy level.
However, I2C devices sometimes needs help to recover from an aborted
transaction.
Therefore, we now add a call to nrfx_{twi, twim}_bus_recover().
Signed-off-by: Xavier Chapron <xavier.chapron@stimio.fr>
The k_work handler cannot manipulate the used k_work. This means
that it is not easy to cleanup the net_pkt because it contains
k_work in it. Because of this, use k_fifo instead between
RX thread and network driver, and between application and TX
thread.
A echo-server/client run with IPv4 and UDP gave following
results:
Using k_work
------------
TX traffic class statistics:
TC Priority Sent pkts bytes time
[0] BK (1) 21922 5543071 103 us [0->41->26->34=101 us]
[1] BE (0) 0 0 -
RX traffic class statistics:
TC Priority Recv pkts bytes time
[0] BK (0) 0 0 -
[1] BE (0) 21925 6039151 97 us [0->21->16->37->20=94 us]
Using k_fifo
------------
TX traffic class statistics:
TC Priority Sent pkts bytes time
[0] BK (1) 15079 3811118 94 us [0->36->23->32=91 us]
[1] BE (0) 0 0 -
RX traffic class statistics:
TC Priority Recv pkts bytes time
[0] BK (1) 0 0 -
[1] BE (0) 15073 4150947 79 us [0->17->12->32->14=75 us]
So using k_fifo gives about 10% better performance with same workload.
Fixes#34690
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
Following commits will remove k_work from net_pkt, so convert
PPP L2 to use k_fifo when sending PPP data.
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
Following commits will remove k_work from net_pkt, so convert
6locan L2 to use k_fifo between application and TX thread, and
driver and RX error handler.
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
Puts all filter in common section and fix filters. Platforms in
integration_platforms must also be allowed first, otherwise the tests
will never run.
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
Adding static to local functions. Using void * for %p argument
which often triggered compilation warning.
Signed-off-by: Krzysztof Chruscinski <krzysztof.chruscinski@nordicsemi.no>
Test was failing because of uninitialized variable and wrong
pointer being used (pointer to char).
Signed-off-by: Krzysztof Chruscinski <krzysztof.chruscinski@nordicsemi.no>
Fixes for C++ includes:
- avoid calling static inline template function in static
assert on certain platforms which consider it non const expression
- Add 0 to variable before calling C++ argument storing function
to promote a variable.
Fix for coverity:
- avoid using sizeof(n + 0) by using local variable (gcc extension)
Signed-off-by: Krzysztof Chruscinski <krzysztof.chruscinski@nordicsemi.no>
Increase the number of works to help PR throughput as we get
close to code freeze. Will revert this back after code freeze.
Signed-off-by: Kumar Gala <kumar.gala@linaro.org>
Fixed check condition for GPIO pin number.
This check is done
before reading or writing value for the pin.
Verified on ehl_crb.
Signed-off-by: U Divya <u.divya@intel.com>
Added support for GPIO driver for Intel Elkhart Lake
board.
The GPIO driver will support pin value read/write operations,
pin direction and interrupt configuration. ACPI enumeration
support and support for different GPIO communities is also
present.
Verified on ehl_crb.
Signed-off-by: U Divya <u.divya@intel.com>
Contrary to the documentation giving a semaphore while an IRQ lock is
held does not release the lock and give control to another thread.
The release-lock behavior is observed only if the lock-holding thread
sleeps.
However the opportunity to reschedule will have been lost so it may be
necessary to explicitly yield to allow the higher-priority thread to
be serviced.
Signed-off-by: Peter Bigot <peter.bigot@nordicsemi.no>
ADC emulator is designed to be used in tests on native_posix board. It
supports 1-16 bit resolution range and all GAINs from enum adc_gain.
Reference voltages and number of emulated channels are set through dts.
Using special API from drivers/adc/adc_emul.h it is possible to set
constant voltage value returned by given ADC channel or set custom
function which allows to simulate complex output.
Also reference voltages can be changed in runtime using the API.
The CL also includes:
- Add adc definitions of ADC emulator in
tests/drivers/adc/adc_api/src/test_adc.c for supporting test suites.
- Add test for ADC emulator API in tests/drivers/adc/adc_emul/
Signed-off-by: Tomasz Michalec <tm@semihalf.com>
