Add few extra definitions used by board DTSs and some that seems to make
sense for potential applications. These match the code and name of the
Linux event codes.
Signed-off-by: Fabio Baltieri <fabiobaltieri@google.com>
Since the sensor shell command was converted to use qt31_t, all the
integer values started to show up as rounded up by a fractional unit
when displayed, due to the conversion always rounding down.
Fix that by using the recently introduced DIV_ROUND_CLOSEST and handling
rounding up to next integer explicitly.
Before:
channel idx=44 gauge_state_of_charge value=83.999999
after:
channel idx=44 gauge_state_of_charge value=84.000000
Signed-off-by: Fabio Baltieri <fabiobaltieri@google.com>
At the moment, NET_EVENT_IPV6_PREFIX_ADD and NET_EVENT_IPV6_PREFIX_DEL
events provide provide information about an address without its length
and lifetime.
Signed-off-by: Konrad Derda <konrad.derda@nordicsemi.no>
Add overlays for esp32s2, esp32s3, esp32c3 based boards, which
has RGB led assembled. Those boards are capable of driving the
RGB led using SPI. The default MOSI idle level is different
to how DI driver expect it to be.
Signed-off-by: Marek Matej <marek.matej@espressif.com>
Fixes several minor display errors in the Segger SystemView output:
* Off-by-one error in some trace ids.
* Add missing syscall trace id.
* Displays easier to read syscall function names.
* Fixes syntax error in k_timer_start output.
Signed-off-by: Florian Grandel <fgrandel@code-for-humans.de>
Fixes a minor documentation issue and suggests/documents additional RTT
tracing and logging options.
Showcases logging via RTT UART which is now available w/o restrictions
due to the preceding change. This has been tested on CC13/26xx hardware.
Signed-off-by: Florian Grandel <fgrandel@code-for-humans.de>
The Zephyr-specific Segger SystemView configuration was not used as
the same file was available inside the external Segger module with
higher import priority.
Fixes the regression by moving the SystemView configuration to the same
place in the external module where RTT configuration already resides and
thereby creates a canonical include path to avoid further regressions of
the same kind.
Fixes: #61133
Signed-off-by: Florian Grandel <fgrandel@code-for-humans.de>
Let's not provide default toolchains for 'arc' intentionally: some
targets can't be built with GNU toolchain ("zephyr", "cross-compile",
"xtools" options) and for some targets we haven't provided MWDT
compiler / linker options in corresponding SoC file in Zephyr, so
these targets can't be built with ARC MWDT toolchain
("arcmwdt" option) by Zephyr build system.
Instead for 'arc' we rely on 'toolchain' option in board yaml
configuration.
Signed-off-by: Eugeniy Paltsev <Eugeniy.Paltsev@synopsys.com>
Signed-off-by: Evgeniy Paltsev <PaltsevEvgeniy@gmail.com>
atomic_t is a machine word signed integer but cq_count is meant to be an
unsigned positive only count of total completions. Cast when needed to
ensure the correct math and comparisons are being done.
Signed-off-by: Tom Burdick <thomas.burdick@intel.com>
Log transfer status and release transfer semaphore regardless if user
provided transfer completion callback or not. This fixes transfer slot
leak when transfer without callback completes.
Signed-off-by: Tomasz Moń <tomasz.mon@nordicsemi.no>
Add documentation to NXP i.MX RT series SOCs about external memories
present on the board. This is intended to help highlight how the SOC
enables external memory using configuration blocks exposed to the
bootrom.
Addtionally, add a check to these boards to warn a user if they are
attempting to utilize SDRAM as their RAM bank without a DCD present.
Signed-off-by: Daniel DeGrasse <daniel.degrasse@nxp.com>
Remove CONFIG_OCRAM_NOCACHE setting, as this is now possible to achieve
using devicetree linker regions, and there is no point in having a
specific Kconfig for one memory region on the RT series like this.
Signed-off-by: Daniel DeGrasse <daniel.degrasse@nxp.com>
Enable SOC fixed MPU regions by default for the RT10xx/RT11xx SOC lines.
Additionally, add code to handle defining the SDRAM0 region as
device type (non cacheable, non shareable). This behavior can
be disabled with CONFIG_NXP_IMX_EXTERNAL_SDRAM=y. Set this Kconfig
for all boards in tree using SDRAM.
This will resolve an issue present on the RT11xx series where
the core may execute speculative prefetches to the SDRAM region when
no SDRAM is present on the board, resulting in the system faulting.
Signed-off-by: Daniel DeGrasse <daniel.degrasse@nxp.com>
Don't default CONFIG_DEVICE_CONFIGURATION_DATA to enabled for iMX RT
SOCs, as this configuration block is only used when the board needs
peripherals like the external memory controller setup from reset by the
bootrom.
Enable this feature on all in tree boards that will require it,
and document the change to the default value in release notes.
Signed-off-by: Daniel DeGrasse <daniel.degrasse@nxp.com>
extends the char buffer in the strsignal function to cover the
entire range of `int`
Had to use `-INT_MAX` as the compiler resolves
```
STRINGIFY(INT_MIN)
```
to:
```
(-2147483647 - 1)
```
instead of:
```
-2147483648
```
Signed-off-by: Yong Cong Sin <ycsin@meta.com>
After boot the channel used for loading the basefw might be left enabled
by ROM.
Make sure that all channels are in stopped state to have consistency.
On TGL during Zephyr boot one channel is left running:
0:0x0x72800: Channel 0 of host out DMA (used for bassefw loading)
dgcs: 0x4800100,
dgbba 0x6000,
dgbs 32768,
dgbrp 8192,
dgbwp 8192,
dgbsp 0,
dgmbs 0,
dgbllpi 0x0,
dglpibi 0x0
Signed-off-by: Peter Ujfalusi <peter.ujfalusi@linux.intel.com>
The RTIO API docs weren't being shown. I made a poor assumption thinking
I could simply include the top level doxygen group in index.rst which isn't
the case. Fix the groups, and remove a note about a poll function
that doesn't exist yet.
Signed-off-by: Tom Burdick <thomas.burdick@intel.com>
The controller and the driver support two hardware configurations:
- one three-channel (RGB) LED
- or three single-channel LEDs
Signed-off-by: Peter van der Perk <peter.vanderperk@nxp.com>
Instead of implementing a custom power off API (pm_system_off),
implement the sys_poweroff hook, and indicate power off is supported by
selecting HAS_POWEROFF. Note that according to the PSCI specification
(DEN0022E), the SYSTEM_OFF operation does not return, however, an error
is printed and system is halted in case this occurs.
Note that the pm_system_off has also been deleted, from now on, systems
supporting PSCI should enable CONFIG_POWEROFF and call the standard
sys_poweroff() API.
Signed-off-by: Gerard Marull-Paretas <gerard@teslabs.com>
Nordic SoCs do all power management automagically when going to idle (ie
k_cpu_idle()). The only extra state, system off, is now handled via
sys_shutdown(), so there's no need to support the PM subsystem.
Signed-off-by: Gerard Marull-Paretas <gerard@teslabs.com>
Make use of the new sys_poweroff API to actually power off the system.
This lets us remove many hacks present in the sample.
Signed-off-by: Gerard Marull-Paretas <gerard@teslabs.com>
Add a new API to perform an immediate system power off:
`sys_poweroff()`.
Until now, this functionality has been implemented via the system power
management module, but in a clunky fashion. The way system PM works is
by defining some idle states in devicetree, that, given some properties
(e.g. minimal residency, exit latency, etc.) are automatically selected
when system goes to idle based on the expected next wake-up. However,
system off is a power state that one typically wants to control manually
from the application because it implies state loss, and in most cases,
configuring some sort of wake-up source. So in general, it is not
desired to let the system enter this state automatically. This led to
the following stuff in-tree:
from `boards/arm/mimxrt595_evk/mimxrt595_evk_cm33.dts`:
```c
/*
* Deep power-down mode is supported in this SoC through
* 'PM_STATE_SOFT_OFF' state. There is no entry for this in device tree,
* user can call pm_state_force to enter this state.
*/
```
That is, state not being defined in devicetree so that PM subsystem
doesn't pick it automatically, but still implemented in in the PM hooks:
from `soc/arm/nxp_imx/rt5xx/power.c`, `pm_state_set()`:
```c
case PM_STATE_SOFT_OFF:
set_deepsleep_pin_config();
POWER_EnterDeepPowerDown(EXCLUDE_FROM_DEEP_POWERDOWN);
break;
```
And to actually make use of this state, users had to do this kind of
abominations:
```c
pm_state_force(0u, &(struct pm_state_info){ PM_STATE_SOFT_OFF, 0, 0 });
/* Now we need to go sleep. This will let the idle thread runs and
* the pm subsystem will use the forced state. To confirm that the
* forced state is used, lets set the same timeout used previously.
*/
k_sleep(K_SECONDS(SLEEP_S));
printk("ERROR: System off failed\n");
while (true) {
/* spin to avoid fall-off behavior */
}
```
Signed-off-by: Gerard Marull-Paretas <gerard@teslabs.com>
These two boards have few oddly defined gpio-keys nodes that do not seem
to have anything to do with keys, and are just used as a placeholder to
define gpios.
Switch the gpio entries to regulators where appropriate, move the two
mag sensors to their own node, drop the rest.
Signed-off-by: Fabio Baltieri <fabiobaltieri@google.com>
So far the init priories were:
enetc_psi0=60 < enetc_vsin=61 < emdio=70 < ethernet-phy=80
because the Ethernet PSI driver was doing global initialization for the
whole NETC complex, including enabling MDIO function (due to the way
the HAL works).
Change to use the default init priorities:
mdio=60 < phy=70 < eth=enetc_psi0=80 < enetc_vsin=81
by executing at an early stage the NETC global initialization. This also
allows to match the DT hierarchy representation of NETC with the
effective priorities assigned.
Signed-off-by: Manuel Argüelles <manuel.arguelles@nxp.com>
Using a different channel for responding to a request is forbidden by spec.
The allocator was especially flawed as it iterated over all the EATT
channels to find one w/ a big enough MTU, but the sending was still done
over the same channel as the REQ.
Signed-off-by: Jonathan Rico <jonathan.rico@nordicsemi.no>
This to ensure we don't fail to send a response and never get an ATT
TIMEOUT due to ACL TX buffer starvation caused by other users of the stack.
Signed-off-by: Jonathan Rico <jonathan.rico@nordicsemi.no>
The ATT module has provisions to queue a packet/buffer for sending later if
it can't send it right away. For example if the conn.c tx context
allocation fails.
This unfortunately doesn't work if the buffer can't get allocated in the
first place, or if the ATT metadata can't also be allocated.
The metadata is allocated from a global pool set to the same number as
conn.c TX contexts. That can lead to a situation where other users of ATT
manage to queue a bunch of buffers (e.g. the app spamming GATT
notifications), depleting the number of ATT metadata slots so that none are
available.
When none are available, and we receive an ATT REQ, we try to allocate one,
fail, and drop the buffer (!). That pretty much guarantees an ATT timeout.
As a workaround for this, use a per-channel metadata slot, that is only
used for completing transactions.
Signed-off-by: Jonathan Rico <jonathan.rico@nordicsemi.no>
This is useful in multilink scenarios, especially since there is not user
callback when the ATT channel times out.
Adding a user-facing callback should ideally also be done, but just logging
the address already provides useful insight.
Signed-off-by: Jonathan Rico <jonathan.rico@nordicsemi.no>
Easier to debug that way.
Ideally we'd have more error codes/logging instead of just returning
UNLIKELY for most errors.
Signed-off-by: Jonathan Rico <jonathan.rico@nordicsemi.no>
To allow insighed into the correct functioning of the collision avoidance,
log the internal values and function calls.
Signed-off-by: Sjors Hettinga <s.a.hettinga@gmail.com>
To avoid a TCP connection from collapsing a link, implement a collision
avoidance algorithm. Initially TCP new Reno is implemented for its
simplicity.
Signed-off-by: Sjors Hettinga <s.a.hettinga@gmail.com>
Some boards may have less LED child nodes in the DT than the maximum
number of LEDs supported by the chipset. For these boards, the channel
test must skip the missing LEDs when preparing the color buffer and not
exit in error due to missing info.
Also, some boards may have a color mapping that is different from RGB
and this mapping should be used when providing the buffer to
led_set_color().
Signed-off-by: Mathieu Anquetin <mathieu.anquetin@groupe-cahors.com>
The led identifer should refer to devicetree ordering, not to the index
used by the controller. This way, it will be possible to deactivate
some leds or to reorganize the indexing if necessary.
Signed-off-by: Mathieu Anquetin <mathieu.anquetin@groupe-cahors.com>
Some boards may have connected the enable pin of the chipset to a GPIO.
On these boards, it is necessary to configure and set this GPIO before
using the chipset, otherwise the I2C circuitry is disabled.
Based on initial work from:
- Marek Janus <marek.janus@grinn-global.com>
- Rico Ganahl <rico.ganahl@bytesatwork.ch>
Signed-off-by: Mathieu Anquetin <mathieu.anquetin@groupe-cahors.com>
