This patch adds basic support for the PWM devices available on the Atmel
SAM family. Beside enabling the driver, everything is selected through
the device tree, including enabling the PWM0 and PWM1 devices. Thus
CONFIG_PWM_0 and CONFIG_PWM_1 are ignored.
Signed-off-by: Aurelien Jarno <aurelien@aurel32.net>
The SAM GMAC Ethernet driver currently keeps a reference to the packet
being sent in addition than keeping of references of the fragments. In
practice this is only needed when PTP is enabled, otherwise the driver
only need to prevent the fragment (or even their content) to be changed.
Signed-off-by: Aurelien Jarno <aurelien@aurel32.net>
The SAM GMAC Ethernet driver uses scatter gather DMA to transmit data.
Each fragment of a network packet is mapped from a set of descriptors
that is used by the controller to do the DMA transfer. This means that
the packet is not necessary sent when the send() function returns. For
that reason the driver calls net_pkt_ref() on the packet to prevent it
from being freed. It is then unreferenced with net_pkt_unref() in the
TX ISR when the packet has effectively been sent.
However this doesn't work if the packet is modified in the meantime,
like it will be done in PR #12563 to remove the Ethernet header
contained in the first fragment. To avoid that, call net_pkt_frag_ref()
on each fragment of the packet, and unreferenced them with
net_pkt_frag_unref() in the TX ISR when the packet has effectively been
sent.
Signed-off-by: Aurelien Jarno <aurelien@aurel32.net>
The function to set channel alarm is called counter_set_channel_alarm.
To match the name of the function performing the reverse operation this
commit renames counter_disable_channel_alarm() function to
counter_cancel_channel_alarm().
Signed-off-by: Piotr Mienkowski <piotr.mienkowski@gmail.com>
The controller does not support trigger. However, the check for
this condition was incorrectly (as GPIO_INT_LEVEL is 0). So fix
it.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
Interrupts default to trigger on level for historical reasons, so use of
GPIO_INT_LEVEL` as a mask results in a zero value. Use a mask macro to
isolate the trigger configuration.
Signed-off-by: Peter A. Bigot <pab@pabigot.com>
Use the new i2c_write_read() wrapper to simplify the code.
Also add several overlays used to test the sensor on a variety of
boards, and conf file support for trigger testing.
Signed-off-by: Peter A. Bigot <pab@pabigot.com>
Replace the sole use of i2c_burst_read_addr with a more generic API
function to allow the former to be deprecated along with its unreliable
sibling write and update functions.
Signed-off-by: Peter A. Bigot <pab@pabigot.com>
Provides a STM32 RTC driver using new counter API.
Driver does not support wrap related functions (set, get, ..)
Fixes#11373
Signed-off-by: Erwan Gouriou <erwan.gouriou@linaro.org>
The CMSDK Dual Timer can be used as a timer or as a counter.
The unified interface proposed in #8340 unifies counter.h and rtc.h to
provide a common interface.
This patch modifies the timer implementation of the dual timer to
make it compliant with the new proposed interface.
Signed-off-by: Vincenzo Frascino <vincenzo.frascino@linaro.org>
Signed-off-by: Kumar Gala <kumar.gala@linaro.org>
The CMSDK Timer can be used as a timer or as a counter.
The unified interface proposed in #8340 unifies counter.h and rtc.h to
provide a common interface.
This patch modifies the timer implementation of the single timer to
make it compliant with the new proposed interface.
Signed-off-by: Vincenzo Frascino <vincenzo.frascino@linaro.org>
Signed-off-by: Kumar Gala <kumar.gala@linaro.org>
Enables debug logging in the counter_basic_api test. Fixes a build error
in the nrf counter drivers when logging is enabled.
Signed-off-by: Maureen Helm <maureen.helm@nxp.com>
Adds a shim layer around the mcux rtc driver to adapt it to the zephyr
counter interface. Portions of this driver are reused from the existing
rtc driver in drivers/rtc/rtc_mcux.c.
The hardware supports a single alarm only and a fixed wrap value.
Signed-off-by: Maureen Helm <maureen.helm@nxp.com>
In case of default top value, driver was calling top value
callback only on the first period. This was not expected behavior.
Signed-off-by: Krzysztof Chruscinski <krzysztof.chruscinski@nordicsemi.no>
Add shim driver for i.MX EPIT (Enhanced Periodic Interrupt Timer)
peripheral which can be used for i.MX6SoloX, i.MX7D and other i.MX socs.
Origin: Original
Signed-off-by: Stanislav Poboril <stanislav.poboril@nxp.com>
Renamed:
- counter_set_wrap to counter_set_top_value
- counter_get_wrap to counter_get_top_value
- counter_get_max_wrap to counter_get_max_top_value
Updated nRF implementations and counter test.
Signed-off-by: Krzysztof Chruscinski <krzysztof.chruscinski@nordicsemi.no>
Modify alarm callback to return user_data and channel_id.
Set_alarm and disable_alarm updated accordingly. Renamed
counter_*_ch_alarm to counter_*_channel_alarm. Updated test
and nrf implementations.
Updated doxygen comments.
Signed-off-by: Krzysztof Chruscinski <krzysztof.chruscinski@nordicsemi.no>
nrf_rtc_timer was selecting counter RTC1 instance even though it
is not using counter API at all.
Signed-off-by: Krzysztof Chruscinski <krzysztof.chruscinski@nordicsemi.no>
Upstream trunk moved to a newer Nordic HAL that changed file and
constant names.
Signed-off-by: Peter A. Bigot <pab@pabigot.com>
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
The system timer uses RTC1, but does not implement the counter API with
it. Instead of auto-enabling the counter API on the system timer make
the two conflict until/unless both APIs are supported by the peripheral.
Signed-off-by: Peter A. Bigot <pab@pabigot.com>
Using `COUNTER_2_NAME` when all other properties are
`COUNTER_TIMER2_foo` is confusing. Make the names consistent.
Signed-off-by: Peter A. Bigot <pab@pabigot.com>
Add driver support for kinetis ID device.
The length depends on the SoC.
`SIM_GetUniqueId` was not used because the struct would reorder
the ID and makes the driver more complicated because the length
of the struct depends on the SoC.
Signed-off-by: Alexander Wachter <alexander.wachter@student.tugraz.at>
Adds a shim layer around the mcux elcdif driver to adapt it to the
zephyr display interface. Although the hardware and underlying mcux sdk
driver can support additional configurations, some shortcuts are
currently made in the shim that force a given pixel format, lcd data
bus width, and signal polarity. This works with the rocktech lcd module
used on imx rt boards, but will need to be updated for other display
panels.
Signed-off-by: Maureen Helm <maureen.helm@nxp.com>
The GPIO controller only supports edge triggering according to
the descriptions of the associated registers. So errors out
when level trigger is requested. Also adds the option to do
double edges triggering as the controller supports this.
Fixes#12763
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
Driver for the I2C peripheral in the SiFive Blocks RTL
Repository (https://github.com/sifive/sifive-blocks).
Signed-off-by: Nathaniel Graff <nathaniel.graff@sifive.com>
Fix issue #9028: last bit of SPI/I2S transaction may be corrupted.
Impacted STM32 SOC series: F0/F1/F2/F3/F4/L0.
Notes:
- F2/F4/L0: set gpio to very_high speed ('11')
- F0/F3: set gpio to high speed ('11').
- F1: set gpio to 50MHz.
Signed-off-by: Armando Visconti <armando.visconti@st.com>
Add SPI_6 Kconfig symbol as this is the higher supported instance on
STM32.
This makes symbol CONFIG_SPI_6, used in stm32 driver a valid symbol.
Signed-off-by: Erwan Gouriou <erwan.gouriou@linaro.org>
Enable HW PWM driver instead of the SW one on nRF SoCs where the PWM
peripheral is present.
Default PWM instances are also enabled on Nordic DK boards so that it
is possible to build the basic fade_led sample for them without extra
adjustments.
After the above changes are applied, some configuration alterations
in basic samples blink_led and fade_led become no longer needed.
These are removed. And the blink_led sample is corrected so that it
works with the nRF HW PWM driver as well.
Signed-off-by: Andrzej Głąbek <andrzej.glabek@nordicsemi.no>
To avoid confusion, callbacks using ordinal pin numbers
is going to be reverted. So the driver has to be re-worked
to expose multiple devices so each device has 32 pins.
Also fixes#12765
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
This allows the shared_irq driver to be configured by device tree.
With previous implementation, only the board configuration can
override the IRQ trigger, as the trigger config is a "choice" rather
than "config". With this patch, the driver can be fully configued at
the SoC level.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
The selection of the Cortex M systick driver to be used
as a system clock driver is controlled by
CONFIG_CORTEX_M_SYSTICK.
To replace it by another driver CONFIG_CORTEX_M_SYSTICK
must be set to 'n'. Unfortunately this also controls
the interrupt vector for the systick interrupt. It is
now routed to __reserved. More bad the interrupt vector
can not be set by IRQ_CONNECT as it is one of the hard
coded interrupts in the interrupt table.
Route the hard coded systick interrupt to z_clock_isr
and make z_clock_isr a weak symbol that can be overwritten
by an alternative systick system clock driver.
Signed-off-by: Bobby Noelte <b0661n0e17e@gmail.com>
The iterator over registered callbacks failed to account for the
possibility that the callback would remove itself from the list. If
this occurred any remaining callbacks would no longer be reachable from
the node. Switch to the slist iterator that is safe for self-removal.
Note that the slist API remains unsafe for removal of subsequent nodes.
Even with the corrected code removal of the next callback registration
(cached in tmp) will result in it being called anyway, with the
remaining unremoved registrations not being called. If the next
callback were removed and re-registered on a different device, the
callbacks would be invoked for the wrong device.
Resolve this by a documentation change describing the conditions under
which a change to callback registration from within a callback are
permitted. Add a similar note regarding the effect of adding a
callback. The current event invocation behavior for callbacks added
within an event is explicitly left unspecified, though in the current
slist implementation newly added callbacks will not be invoked until the
next event.
Closes#10186
Signed-off-by: Peter A. Bigot <pab@pabigot.com>
This commit fixes the issue with excess current being drawn
during sleep due to active HFCLK with external crystal.
Clock-related operations were moved to cable attachmend and
detachment handlers to ensure that HFCLK is not requested
when it's not needed.
Signed-off-by: Paweł Zadrożniak <pawel.zadrozniak@nordicsemi.no>
