the only change this results in in the example binaries is in the
hackrf-jellybean/systick example, where the the check in
systick_set_clocksource for overflowing from the stm32 area gets used.
these register definitions are common to all cortex mcus. some of the
registers might not be implemented everywhere (especially the floating
point registers), but defining them does no harm.
this modification does not result in any changes in the example
binaries.
now tries to mkdir its way to the output files
this wouldn't be a problem currently if it wasn't for the efm32 data
lingering in the wrong branch, but otoh it's just on the safe side in
case we meet architectures that don't need other specializations at all.
this makes the previous hackish vector.c assemblies into dispatched files
(using the same mechanism as for nvic, just this time in lib).
the old irq.h files that were generated manually from the old vector.c files
were dropped in the process, as were the nvic.h files, and replaced with very
simple yaml lists that generate the headers.
file generation takes place both in the include/ and the lib/ part, as some of
it is definitely header stuff (the NVIC_name_IRQ defines), and some of it needs
to be included in specific compilation units (the weak pragmas).
* created include/libopencm3/cm3/nvic.h from the respective stm32 and
lpc43xx files. the chip specific definitions were left in place (they
were already split out in sthe stm32 case).
* created lib/cm3/nvic.c from the respective stm32 and lpc43xx files. a
hack from the lpc43xx was taken over (for manipulating the internal
interrupts); for now it'll work.
* created a include/libopencm3/dispatch/ directory where the dispatching
of files with common interfaces but different implenentations can
happen; for now, an nvic.h there includes the respective irq name
definitions. (future implementations might have some automation or
preprocessor magic there; so far, it's manual dispatching based on
defines.)
* for efm32, an nvic.h gets generated from an interrupt list, the
rationale for code generation is, in this case, that this can't be
done easily in c preprocessor, and it's really just a list of
definitions and not code proper.
* examples now include <libopencm3/cm3/nvic.h> instead of
<libopencm3/stm32/nvic.h>
Conflicts:
lib/lm3s/vector.c -- split out to lm3s/irq.h
lib/stm32/f4/vector.c -- put the floating point initialization code into a function like in lpc43xx
Add object files to Makefile to include new modules into lib.
Correct typo in pwr.c
Add two new functions to timer.c to preset the counter (for use
with deadman style timeouts) and to identify an interrupt
source.
Also noticed lib/makefile.include didn't clean lib/stm32, which
isn't a target. Added a fix.
* simple polling of an injected channel
* timer triggered sampling of an injected channel
* timer triggered sampling and IRQ handling of an injected channel
* timer triggered sampling and IRQ handling of 4 injected channels
remove rcc_set_adc_clk - use rcc version
Added functions:
- adc_power_on
- adc_start_conversion_direct
- adc_set_dual_mode
- adc_eoc
- adc_eoc_injected
- adc_read_regular
- adc_read_injected
- adc_set_injected_offset
Tested dual mode scanned regular, but no tests of injected yet.
Changes: "discontinuous" was misspelled.
- adc_set_discontinuous_mode_regular - added "length" parameter
- adc_disable_discontinuous_mode_regular - name change
- adc_enable_discontinuous_mode_injected - name change
- adc_enable_automatic_injected_group_conversion - disable triggers
- adc_enable_jeoc_interrupt - name change to match common usage in lib
- adc_disable_jeoc_interrupt - ditto
- adc_enable_external_trigger_regular - remove incorrect test on parameter
- adc_enable_external_trigger_injected - ditto
- adc_set_sample_time - name change to match function's purpose
- adc_set_conversion_time_on_all_channels - ditto
- adc_set_injected_sequence - changed order of register loading (ref Barlow's issue)
- adc_enable_analog_watchdog_on_all_channels - flipped AWDSGL
- adc_enable_analog_watchdog_on_selected_channel - ditto
added aliases for expected commonly used functions to avoid sudden user code breakage
In adc.h, corrected errors in SQR names
added "deprecated" compiler warnings to adc_on and to aliases defined in adc.c
the cortex generic interrupts get moved to lib/cm3/vector.c, the
platorms' individual irq names, initialization and handler prototypes go
to platoform specific irq.h files.
as the vector.c file heavily depends on platoform specific headers, it
can't be built once-and-for-all in lib/cm3/, so there are inclusion
stubs in the various architecture dirs; this might be better solved with
Makefile / include path handling.
one particular file is lib/lpc43xx/vector.c; that platform's
initialization code contains an additional section to copy everything
from flash to ram (which probably performs better there). that code
still resides in the inclusion stub, and gets mashed in using defines.
would need a cleaner implementation together with the Makefile solution.
this commit contains some files of the upcoming efm32 branch, from which
it was cherry-picked.
the .bin files produced from before and after this commit only differ in
lpc43xx, where the startup sequence was subtly modified.
chrysn