When writing buffers larger then page-size, there is already a routine
that checks wrap around and adjusts offsets, but this routine was
missing incrementing the data pointer, which would results in
rewriting the same page-size bytes over and over. This adds the proper
increment code.
Signed-off-by: Fabio Utzig <fabio.utzig@nordicsemi.no>
Silence a gcc warning due to possible return of unitialized variable in
erase function. This could only happen if size == 0, which doesn't seem
likely, but initializing the variable to zero should fix the issue.
Signed-off-by: Fabio Utzig <fabio.utzig@nordicsemi.no>
There is a typo in the function pointer assigned to get_parameters. It
should be flash_sam_get_parameters.
Signed-off-by: Kumar Gala <kumar.gala@linaro.org>
Commit id a538dcd8f8 got rid of the usage of these pointers so they are
useless now and can be removed in this drivers.
Signed-off-by: Tomasz Bursztyka <tomasz.bursztyka@linux.intel.com>
For some reason, use of the DT_NUM_INST macro in this driver was not
replaced with DT_NUM_INST_STATUS_OKAY. In consequence, this driver
could not be compiled successfully. Furthermore, because this now
undefined macro had been used inside UTIL_LISTIFY, gcc was not even
able to report an error, instead the compilation just "froze".
This patch replaces UTIL_LISTIFY(DT_NUM_INST()) with more appropriate
in this context DT_INST_FOREACH_STATUS_OKAY. It also adds an apparently
missing inclusion of sys/byteorder.h.
Signed-off-by: Andrzej Głąbek <andrzej.glabek@nordicsemi.no>
If shell is enabled then enable all sub-shells if their dependencies are
satisfied. This was done for some modules and subsystems but was not
consistent.
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
Several shell modules use cloned code to iterate over all devices and
identify the nth instance that meets some criteria. The code was
repetitive and included various errors. Abstract to a helper function
that performs the check consistently.
Signed-off-by: Peter Bigot <peter.bigot@nordicsemi.no>
Some flash controllers support operations only when buffers are word
aligned. An example of such is nRF QSPI, which checks input buffer using
nrfx_is_word_aligned() function inside nrfx_qspi_write().
Align test array on 4 bytes, so we will gain compatibility with more
flash controllers, including nRF QSPI.
Signed-off-by: Marcin Niestroj <m.niestroj@grinn-global.com>
With addition of flash_parameters structure, and supporting API call
to retrieve it, it is no longer needed to store write_block_size as
a part of flash_driver_api and it should be part of flash_parameters.
Signed-off-by: Dominik Ermel <dominik.ermel@nordicsemi.no>
With addition of flash_get_parameters API call, it is needed to provide
support for the API to flash drivers.
Signed-off-by: Dominik Ermel <dominik.ermel@nordicsemi.no>
The Nordic QSPI peripheral uses DMA transfers so data to write must be
located in SRAM. Add a Kconfig that enables copying data from NVMC to
a stack SRAM buffer so it can be written to flash.
Signed-off-by: Peter Bigot <peter.bigot@nordicsemi.no>
mcumgr and possibly mcuboot write single byte values to update the
state of objects. Rather than fail to do the write of values too
short for this peripheral detect the situation and write from a stack
buffer that meets the length criteria.
Signed-off-by: Sigvart Hovland <sigvart.m@gmail.com>
Signed-off-by: Peter Bigot <peter.bigot@nordicsemi.no>
Currently user is forced to configure an array of 4 IO pins. This makes
no sense when there are only 2 IO pins connected on board.
Configure 3rd and 4th pin in internal structure as
NRF_QSPI_PIN_NOT_CONNECTED if only 2 were specified in device-tree.
Signed-off-by: Marcin Niestroj <m.niestroj@grinn-global.com>
Erase can only succeed if the address is sector-aligned, and the span
to erase is an integer multiple of the sector size. Validate this
before starting the process of erasing things.
Signed-off-by: Peter Bigot <peter.bigot@nordicsemi.no>
Erase can only succeed if the address is sector-aligned, and the span
to erase is an integer multiple of the sector size. Validate this
before starting the process of erasing things.
Also reduce the check that the affected region is within the device
span to its minimal equivalent conditions.
Signed-off-by: Peter Bigot <peter.bigot@nordicsemi.no>
Erase can only succeed if the address is sector-aligned, and the span
to erase is an integer multiple of the sector size. Validate this
before starting the process of erasing things.
Signed-off-by: Peter Bigot <peter.bigot@nordicsemi.no>
Having a completion wait function release a lock internally only when
the operation fails is confusing. Remove that feature, and make the
lock and unlock operations explicit and paired.
This makes it much more clear how to properly handle transactions that
require multiple calls to the HAL.
Signed-off-by: Peter Bigot <peter.bigot@nordicsemi.no>
The stack buffer used to pad small reads must still be aligned to a
word address as the underlying driver uses DMA transfers with that
requirement. Don't assume the compiler will naturally align it.
Signed-off-by: Peter Bigot <peter.bigot@nordicsemi.no>
The check for small transfers inadvertently allowed a transfer of zero
bytes, which should be an error (invalid parameter).
Signed-off-by: Peter Bigot <peter.bigot@nordicsemi.no>
mcuboot and possibly other tools read single byte values to determine
the state of objects. Rather than fail to do the read of values too
short for this peripheral detect the situation and read into a stack
buffer that meets the length criteria, and on success copy the data
into the provided buffer.
Signed-off-by: Peter Bigot <peter.bigot@nordicsemi.no>
Several reviewers agreed that DT_HAS_NODE_STATUS_OKAY(...) was an
undesirable API for the following reasons:
- it's inconsistent with the rest of the DT_NODE_HAS_FOO names
- DT_NODE_HAS_FOO_BAR_BAZ(node) was agreed upon as a shorthand
for macros which are equivalent to
DT_NODE_HAS_FOO(node) && DT_NODE_HAS_BAR(node) &&
- DT_NODE_HAS_BAZ(node), and DT_HAS_NODE_STATUS_OKAY is an odd duck
- DT_NODE_HAS_STATUS(..., okay) was viewed as more readable anyway
- it is seen as a somewhat aesthetically challenged name
Replace all users with DT_NODE_HAS_STATUS(..., okay), which is
semantically equivalent.
This is mostly done with sed, but a few remaining cases were done by
hand, along with whitespace, docs, and comment changes. These special
cases include the Nordic SOC static assert files.
Signed-off-by: Martí Bolívar <marti.bolivar@nordicsemi.no>
Use DT_INST_SPI_DEV_HAS_CS_GPIOS() in drivers to determine if we should
utilize CS_GPIO base SPI chipselect handling. This allows us to remove
Kconfig option for this feature.
Signed-off-by: Kumar Gala <kumar.gala@linaro.org>
When the device driver model got introduced, there were no concept of
SYS_INIT() which can be seen as software service. These were introduced
afterwards and reusing the device infrastructure for simplicity.
However, it meant to allocate a bit too much for something that only
required an initialization function to be called at right time.
Thus refactoring the devices structures relevantly:
- introducing struct init_entry which is a generic init end-point
- struct deviceconfig is removed and struct device owns everything now.
- SYS_INIT() generates only a struct init_entry via calling
INIT_ENTRY_DEFINE()
- DEVICE_AND_API_INIT() generates a struct device and calls
INIT_ENTRY_DEFINE()
- init objects sections are in ROM
- device objects sections are in RAM (but will end up in ROM once they
will be 'constified')
It also generate a tiny memory gain on both ROM and RAM, which is nice.
Perhaps kernel/device.c could be renamed to something more relevant.
Signed-off-by: Tomasz Bursztyka <tomasz.bursztyka@linux.intel.com>
Add a driver that can handle several instances of AT45 family chips,
which are enabled by specifying DT nodes for them with the "compatible"
property set to "atmel,at45" and other required properties like JEDEC
ID, chip capacity, block and page size etc. configured accordingly.
The driver is only capable of using "power of 2" binary page sizes in
those chips and at initialization configures them to work in that mode
(unless it is already done).
Signed-off-by: Andrzej Głąbek <andrzej.glabek@nordicsemi.no>
The sam0 has a page size of 64 bytes, making it incompatible with
the nvs driver that specifies a maximum write block size of
32 bytes. When CONFIG_SOC_FLASH_SAM0_EMULATE_BYTE_PAGES is set,
it should report a write block size of 1 byte, which is compatible
with nvs.
Signed-off-by: Adam Serbinski <aserbinski@gmail.com>
The dpd-wake-sequence has a value that is an array of three integers,
which was formerly indexed by a suffix on the property name. This was
updated to new accessors but failed to separate the index from the
property name. Update the access idiom.
Signed-off-by: Peter Bigot <peter.bigot@nordicsemi.no>
Rename DT_HAS_NODE to DT_HAS_NODE_STATUS_OKAY so the semantics are
clear. As going forward DT_HAS_NODE will report if a NODE exists
regardless of its status.
Signed-off-by: Kumar Gala <kumar.gala@linaro.org>
Replace DT_FLASH_DEV_NAME with DT_CHOSEN_ZEPHYR_FLASH_CONTROLLER_LABEL.
We now set zephyr,flash-controller in the chosen node of the device
tree to the flash controller device.
NOTE: For a SoCs with on die flash, this points to the controller and
not the 'soc-nv-flash' node. Typically the controller is the
parent of the 'soc-nv-flash' node).
Signed-off-by: Kumar Gala <kumar.gala@linaro.org>
The code generally invoked qspi_wait_for_completion() within a locked
region without verifying that an operation was successfully initiated.
This caused a deadlock whenever the operation failed, e.g. because the
data buffer was not 4-byte aligned. Update that function to take the
result of the operation and either wait for completion or release the
lock, depending on its value.
Also uniformly use the correct type for Nordic HAL error values, and
refactor the erase module so that the correct values are displayed in
the diagnostic when something goes wrong.
Also check the alignment requirements for the flash address and
transfer size, which are highly constrained on this device. This
driver also requires 4-byte aligned data buffer; this is checked in
the Nordic HAL.
Signed-off-by: Peter A. Bigot <pab@pabigot.com>
Instead of various series compatibles, use single stm32 generic
compatible as reference for stm32 flash driver.
Signed-off-by: Erwan Gouriou <erwan.gouriou@linaro.org>
Convert driver from DT_FLASH_BASE_ADDRESS and DT_FLASH_DEV_NAME to use
DT_REG_ADDR and DT_INST_LABEL.
Signed-off-by: Kumar Gala <kumar.gala@linaro.org>
For a flash driver DT_DRV_COMPAT should be the compatible for the flash
controller and not the soc-nv-flash. Change to driver to use the flash
controllers compatible and get the soc-nv-flash properties as a child of
that controller.
Signed-off-by: Kumar Gala <kumar.gala@linaro.org>
For a flash driver DT_DRV_COMPAT should be the compatible for the flash
controller and not the soc-nv-flash. Change to driver to use the flash
controllers compatible and get the soc-nv-flash properties as a child of
that controller.
Signed-off-by: Kumar Gala <kumar.gala@linaro.org>
For a flash driver DT_DRV_COMPAT should be the compatible for the flash
controller and not the soc-nv-flash. Change to driver to use the flash
controllers compatible and get the soc-nv-flash properties as a child of
that controller.
The soc_flash_mcux supports several possible compatible so handle that
as part of how we set DT_DRV_COMPAT.
Signed-off-by: Kumar Gala <kumar.gala@linaro.org>
For a flash driver DT_DRV_COMPAT should be the compatible for the flash
controller and not the soc-nv-flash. Change to driver to use the flash
controllers compatible and get the soc-nv-flash properties as a child of
that controller.
Signed-off-by: Kumar Gala <kumar.gala@linaro.org>
The 32 KiBy bulk erase command was being invoked without respecting
the flag that indicates it's supported. Make the invocation
conditional.
Signed-off-by: Peter Bigot <peter.bigot@nordicsemi.no>
Use the UNALIGNED_GET() macro instead of the
flash_sam0_read_unaigned_u32() function to ensure
word alignment of the source address.
Signed-off-by: Steven Slupsky <sslupsky@gmail.com>
drivers: flash: sam0: fix whitespace
Fix checkpass whitespace error.
Signed-off-by: Steven Slupsky <sslupsky@gmail.com>
Update `flash` shell commands to take an optional device argument,
instead of using the chosen flash device.
Signed-off-by: Ivo Clarysse <ivo@bcdevices.com>
Replace DT_FLASH_WRITE_BLOCK_SIZE with new DT_INST_NODE_HAS_PROP and
DT_INST_PROP macro as we phase out DT_FLASH define usage.
Signed-off-by: Kumar Gala <kumar.gala@linaro.org>
Replace DT_FLASH_DEV_BASE_ADDRESS with new DT_REG_ADDR/DT_INST macro as
we phase out DT_FLASH define usage.
Signed-off-by: Kumar Gala <kumar.gala@linaro.org>
Replace use of DT_FLASH_DEV_NAME with DT_INST_LABEL in drivers as we
want to phase out DT_FLASH_DEV_NAME.
Signed-off-by: Kumar Gala <kumar.gala@linaro.org>
The duration is used in math with the uptime clock, not as a timeout.
Correct value expression and rename to clarify units.
Signed-off-by: Peter Bigot <peter.bigot@nordicsemi.no>
Use valid stm32 flash driver compatibles instead of flash
area compatible.
For 'write_block_size' property, use reference to soc-nv-flash.
fixes#23997
Signed-off-by: Erwan Gouriou <erwan.gouriou@linaro.org>
When enabled, instead of erasing entire flash page at once, page will
be erased in defined time slices. Erasing single page stalls CPU
for significant time share (~80ms) and partial erase divides the
operation in to the shorter time periods, resuming CPU operation in
meantime and enabling better scheduling of time sensitive operations.
Signed-off-by: Dominik Ermel <dominik.ermel@nordicsemi.no>
Delay radio abort called in the flash driver to emulate the
behavior of pre-empt timeout in Bluetooth LL split
controller. Without this, the driver aborted radio events
in its reserved time space.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
Kernel timeouts have always been a 32 bit integer despite the
existence of generation macros, and existing code has been
inconsistent about using them. Upcoming commits are going to make the
timeout arguments opaque, so fix things up to be rigorously correct.
Changes include:
+ Adding a K_TIMEOUT_EQ() macro for code that needs to compare timeout
values for equality (e.g. with K_FOREVER or K_NO_WAIT).
+ Adding a k_msleep() synonym for k_sleep() which can continue to take
integral arguments as k_sleep() moves away to timeout arguments.
+ Pervasively using the K_MSEC(), K_SECONDS(), et. al. macros to
generate timeout arguments.
+ Removing the usage of K_NO_WAIT as the final argument to
K_THREAD_DEFINE(). This is just a count of milliseconds and we need
to use a zero.
This patch include no logic changes and should not affect generated
code at all.
Signed-off-by: Andy Ross <andrew.j.ross@intel.com>
Replace all occurences of BUILD_ASSERT_MSG() with BUILD_ASSERT()
as a result of merging BUILD_ASSERT() and BUILD_ASSERT_MSG().
Signed-off-by: Oleg Zhurakivskyy <oleg.zhurakivskyy@intel.com>
Convert older DT_INST_ macro use in atmel sam0 drivers to the new
include/devicetree.h DT_INST macro APIs.
Signed-off-by: Kumar Gala <kumar.gala@linaro.org>
Convert older DT_INST_ macro use in openisa drivers to the new
include/devicetree.h DT_INST macro APIs.
Signed-off-by: Kumar Gala <kumar.gala@linaro.org>
Convert older DT_INST_ macro use in atmel sam drivers to the new
include/devicetree.h DT_INST macro APIs.
Signed-off-by: Kumar Gala <kumar.gala@linaro.org>
Convert older DT_INST_ macro use in STM32 drivers to the new
include/devicetree.h DT_INST macro APIs.
Signed-off-by: Kumar Gala <kumar.gala@linaro.org>
Convert older DT_INST_ macro use in silab drivers to the new
include/devicetree.h DT_INST macro APIs.
Signed-off-by: Kumar Gala <kumar.gala@linaro.org>
In the default configuration of the stm32g4 is the dual bank mode.
In dual bank mode, the flash is split into two banks with 256k each.
The erase pages have a range form 0 to 127, where each page has 2k.
If one wants to erase an area above 256k, the driver has to switch
to bank 2 befor erasing. Otherwise it will erase the page moulo 127.
Signed-off-by: Alexander Wachter <alexander.wacher@leica-geosystems.com>
This commit introduces some logs in the stm32 flash implementation.
Thith the logs it is easier to locate problems when they arrise.
Signed-off-by: Alexander Wachter <alexander.wacher@leica-geosystems.com>
Don't use DT_INST_0_SOC_NV_FLASH_LABEL as device name.
Use DT_FLASH_DEV_NAME instead
Fixes#23678
Signed-off-by: Erwan Gouriou <erwan.gouriou@linaro.org>
Coverity doesn't invoke the build for this module with a configuration
that provides a non-zero delay for entering or releasing deep power
down, so it sees a check that a value known to be non-negative
magically became negative without being touched, and so screams about
dead code.
Refactor to make the adjustments unconditional which has the same
effect in any reasonable level of optimization while making it
slightly less obvious that the value may not change.
Signed-off-by: Peter Bigot <peter.bigot@nordicsemi.no>
This reverts commit 8739517107.
Pull Request #23437 was merged by mistake with an invalid manifest.
Signed-off-by: Carles Cufi <carles.cufi@nordicsemi.no>
Replace all occurences of BUILD_ASSERT_MSG() with BUILD_ASSERT()
as a result of merging BUILD_ASSERT() and BUILD_ASSERT_MSG().
Signed-off-by: Oleg Zhurakivskyy <oleg.zhurakivskyy@intel.com>
Convert driver to use DT_INST_ defines.
Removed DT_FLASH_DEV fixup macros, except DT_FLASH_DEV_NAME
used in applications.
Signed-off-by: Erwan Gouriou <erwan.gouriou@linaro.org>
There is timeout guard, in form of k_sem_get within work_in_time_slice,
used to detect if flash operation takes longer than expected maximum
and to return error code in such case. The processing of this timeout
is missing cancellation of scheduled ticker job that would attempt to
perform another operation. This causes problem as context for ticker
is created, on stack, within erase/write_in_timeslice which
would conclude when timeout occurs, folding the stack in process.
Signed-off-by: Dominik Ermel <dominik.ermel@nordicsemi.no>
I think people might be reading differences into 'if' and 'depends on'
that aren't there, like maybe 'if' being needed to "hide" a symbol,
while 'depends on' just adds a dependency.
There are no differences between 'if' and 'depends on'. 'if' is just a
shorthand for 'depends on'. They work the same when it comes to creating
implicit menus too.
The way symbols get "hidden" is through their dependencies not being
satisfied ('if'/'depends on' get copied up as a dependency on the
prompt).
Since 'if' and 'depends on' are the same, an 'if' with just a single
symbol in it can be replaced with a 'depends on'. IMO, it's best to
avoid 'if' there as a style choice too, because it confuses people into
thinking there's deep Kconfig magic going on that requires 'if'.
Going for 'depends on' can also remove some nested 'if's, which
generates nicer symbol information and docs, because nested 'if's really
are so simple/dumb that they just add the dependencies from both 'if's
to all symbols within.
Replace a bunch of single-symbol 'if's with 'depends on' to despam the
Kconfig files a bit and make it clearer how things work. Also do some
other minor related dependency refactoring.
The replacement isn't complete. Will fix up the rest later. Splitting it
a bit to make it more manageable.
(Everything above is true for choices, menus, and comments as well.)
Detected by tweaking the Kconfiglib parsing code. It's impossible to
detect after parsing, because 'if' turns into 'depends on'.
Signed-off-by: Ulf Magnusson <Ulf.Magnusson@nordicsemi.no>
The SAM flash driver was developed for the SAME70 SoCs and hasn't ever
been validated on the other SAM families (SAM3X, SAM4S). If we try and
build on one of the other SoC families we get build errors. So limit it
to just SAME70 for now.
Signed-off-by: Kumar Gala <kumar.gala@linaro.org>
Most JEDEC NOR flash devices uses not only typical SPI mode
(MISO,MOSI,SCK and CS), but also QSPI mode (IO0,IO1,IO2,IO3,SCK and CS).
QSPI mode uses more data lines and as a result provide higher
throughput. If this were not enough, Nordic chips provide
hardware acceleration for read/write/erase functions, what
gives significant performance boost.
It does a lot of things "behind the scene", i.e when user has written
some data to the flash and would like to read them back, it has to wait
until the flash is ready by reading WIP bit in Status Register.
This driver does it automatically.
Signed-off-by: Kamil Lazowski <Kamil.Lazowski@nordicsemi.no>
Remove soc/arm/st_stm32/stm32YY/flash_registers.h files.
Change register accesses in stm32 flash drivers to use FLASH_TypeDef
from modules/hal/stm32/stm32cube/stm32YYxx/soc/stm32xxxxxx.h.
Fixes#16235
Signed-off-by: Sarvesh Patkar <psarvesh314@gmail.com>
The build infrastructure should not be adding the drivers subdirectory
to the include path. Fix the legacy uses that depended on that
addition.
Signed-off-by: Peter Bigot <peter.bigot@nordicsemi.no>
The build infrastructure should not be adding the drivers subdirectory
to the include path. Fix the legacy uses that depended on that
addition.
Signed-off-by: Peter Bigot <peter.bigot@nordicsemi.no>
After startup the native_posix flash was filed by 0x00 instead
of 0xff, which is not coherent with erased byte value which
is 0xff for the native_posix target.
Such inconsistent behavior makes for instance running
a storage system designed(or configured) for a flash device
with 0xff as erased byte values problematic.
This patch makes flash erased after new flash device file creation.
Signed-off-by: Andrzej Puzdrowski <andrzej.puzdrowski@nordicsemi.no>
Patch introduce references to LittleFS instead of NFFS where it
was suitable. In other places NFFS mentions were removed
Signed-off-by: Andrzej Puzdrowski <andrzej.puzdrowski@nordicsemi.no>
According to NRF product specifications the smallest write unit is
32-bits. Added parameter 'CONFIG_EMULATE_ONE_BYTE_WRITE_ACCESS'
which allows one byte flash write using more complex operation.
Signed-off-by: Adam Kondraciuk <adam.kondraciuk@nordicsemi.no>
Grouped all nRF driver sub-option under SOC_FLASH_NRF
menuconfig.
This makes menuconfig cleaner.
Signed-off-by: Andrzej Puzdrowski <andrzej.puzdrowski@nordicsemi.no>
Following update of stm32wb package, FLASH_FLAG_SR_ERROR
has been renamed to FLASH_FLAG_SR_ERRORS.
Update driver to fix compilation issue.
Signed-off-by: Erwan Gouriou <erwan.gouriou@linaro.org>
The amount of lock regions differs between different sam0 MCUs.
saml10: 2
saml11: 4
samd2x: 16
saml2x: 16
samd5x: 32
ASF does not provide a definition for this, so create a new one
in dts.
Signed-off-by: Benjamin Valentin <benjamin.valentin@ml-pa.com>
This adds support for the flash peripheral found in the SAME5x/SAMD5x
line of MCUs.
The peripheral is very simmilar to the one found on SAMD2x with only
a few register names changed.
Signed-off-by: Benjamin Valentin <benjamin.valentin@ml-pa.com>
In STM32 flash driver, don't use semaphore if multithreading
isn't defined.
If multithreading isn't defined, a call to sem_give function
generates a hardware fault exception.
Signed-off-by: Benoit Leforestier <benoit.leforestier@gmail.com>
The current configuration causes the STM32 flash support always to be
built, even if an unrelated flash driver, by example the SPI_NOR driver
is selected.
This behaviour gets especially problematic (build failure) if the flash
hardware of the given MCU is not supported (e.g. STM32F2).
The suggested change should ensure that STM32 flash support only is
built when it actually is selected.
Signed-off-by: Marco Peter <marco@peter-net.ch>
NRF devices hardware has flash protection which doesn't
reflect flash API definition well. So fare protection
mechanism was emulated by the software and the driver deals
with hardware flash protection on its own.
Recent change to protection behavior requirement allows
to remove flash API behavior emulation at all.
Signed-off-by: Andrzej Puzdrowski <andrzej.puzdrowski@nordicsemi.no>
Remove leading/trailing blank lines in .c, .h, .py, .rst, .yml, and
.yaml files.
Will avoid failures with the new CI test in
https://github.com/zephyrproject-rtos/ci-tools/pull/112, though it only
checks changed files.
Move the 'target-notes' target in boards/xtensa/odroid_go/doc/index.rst
to get rid of the trailing blank line there. It was probably misplaced.
Signed-off-by: Ulf Magnusson <Ulf.Magnusson@nordicsemi.no>
Fabio Utzig