The `channels` member should not be set to the number of hardware
channels *n*, but to *n* - 1 instead. The last channel is implicitly
used in `timer_read()`. Hence out of *n* hardware channels, only *n* - 1
are available to the application.
This fixes a bug introduced by 4d02e15247
which incorrectly set the channel number to *n* rather than to
*n* - 1.
- All nRF52 timers support 32 bit mode, so use that
- All nRF52 timers support at least 4 channels, the timers NRF_TIMER3
and NRF_TIMER4 even support 6 channels.
- Add a warning that `TIMER_DEV(1)` is used by the IEEE 802.15.4 driver
The peripheral configuration has been completely reworked to resolve
pin conflicts while provided as much of the peripherals as possible.
The changes include:
- Move `I2C_DEV(0)` from PB6/PB7 to PB8/PB9 to solve pin conflict with
`QDEC_DEV(2)`.
- Use pins PB0, PB1, PB4, and PB5 for PWM instead PA8, PA9, PA10, and
PA11
- PA9 and PA10 is in pin conflict with `UART_DEV(0)` which is used
for stdio with `stdio_uart`, PA8 was in conflict with
`QDEC_DEV(0)`, PA11 was in conflict with USB D-
- Use PB6, PB7 as `QDEC_DEV(0)` (previously `QDEC_DEV(2)`), as this is
the only completely conflict free setting
- Use PB4/PB5 instead of PA6/PA7 for QDEC_DEV(1)
- This fixes a pin conflict with `SPI_DEV(0)` MISO (and
`ADC_LINE(4)`)
- Only provide QDEC at PB4/PB5 when PWM is not used to avoid conflict
- Only provide QDEC at PA8/PA9 when UART is not used to avoid conflict
- Use SPI2 (PB15, PB14, PB13, PB12) as `SPI_DEV(0)` instead of SPI1,
use SPI1 (PA7, PA6, PA5, PA4) as `SPI_DEV(1)`
- Only provide `SPI_DEV(1)` if the ADC is not in used to resolve a
pin conflict
- Move PB0 and PB1 at the end of the ADC lines (previously
`ADC_LINE(6)` and `ADC_LINE(7)`, now `ADC_LINE(8)` and `ADC_LINE(9)`)
- Only provide them when PWM is not in use (to resolve pin conflict
with PWM)
- Also do not provide them for the Blackpill boards, which are
missing pins PB0 and PB1 on the headers
To make life of users easier, a Pinout diagram with the new
configuration was added.
Nightlies are failing due to kconfig mismatch.
It would seem this is a result of bringing in the USB stuff.
I assume that this uses ztimer periph_timer as a backend as periph_timer is already selected.
However, kconfig only resolves one and not recursively making it hard to match.
For not a hack is added to override for these boards.
