* Add udiv_32_16
* Apply udiv_32_16() where possible
* Convert udiv_32_16 to assembler
It's worth 20 loop/s
* Remove unused functions
* Remove degreesPeruSx2048 - unused
* Remove angleToTime - replace with direct calls
1. Drop angleToTime()
It's slow, only partially implemented and adds zero value
(and has MISRA violations)
2. Consistent function naming
3. Doxygen
* triggerPri_Nissan360 shouldn't set timePerDegree.
It will be overwritten every loop by doCrankSpeedCalcs()
* Use angleToTimeMicroSecPerDegree() instead of timePerDegree
No loss in performance
Increased injection open/close time accuracy (so unit test values must change)
Can remove timePerDegree global.
* Hide (encapsulate) crank math globals.
* Base all angle to time conversions on decoder computed variables.
This is within 2us of the revolution based method
and is much faster - which is essentially zero percent change.
* Performance: move calculation of degreesPeruSx32768
into decoders.
Remove doCrankSpeedCalcs() - it's doing nothing
at the moment.
* Apply libdivide to triggerSetEndTeeth functions.
Since triggerToothAngle is set once at initialization
time, we can generate the libdivide struct
once and reuse it many times.
* Remove lastToothCalcAdvance - unused
* Replace 16-bit division with shift
* Replace 32-bit divison with 16-bit division
* Avoid 32-bit division; use div100()
* inline percentage()
* Optimize div100()
* MISRA fixes
* Replace magic numbers with #defs
* Replace libdivide structs with inline constants
No perf or memory changes
* Use fixed types for PWM max count variables
* Accurate rounded integer division
* Formalise rounding behavior (DIV_ROUND_CORRECT)
* Apply DIV_ROUND_CORRECT to DIV_ROUND_CLOSEST(),
UDIV_ROUND_CLOSEST(), div100(), div360(),
percentage() & halfPercentage()
* Add, fix & improve unit tests
* Add udiv_32_16_closest()
* Perf: Limit percentage calculations to 16-bits
* MISRA fixes
* Add compare_executiontime() to encapsulate common perf testing code
* Signed to unsigned division
* Convert ignitionLimits() to an inline function.
Slight speed up, probably due to removing
multiple evaluations of macro arguments.
* Split unit tests up.
* udiv_32_16 - check for valid parameters
* Change X axis in memory orientation to match Y
* X & Y axes can use the same type (Ie.e class)
* Table3D axis iterator: replace reverse() with rbegin()
* Use iterators as part of unit tests
* Doxygen fixes/corrections
* Bi-linear interpolation - round towards nearest integer
Add the equivalent of 0.5 to the final calculation pre-rounding.
This will have the effect of rounding to the nearest integer, rather
than truncating. I.e. rounding down
* Unit tests: isolate table tests and check interpolation x/y bins
* Unit tests: add native table3d tests
* Unit tests: derive min/max from axis test values
* Unit tests: more detailed messages
* Unit tests: fix rounding & unit tests
* Performance
* Performance: use uint16_t instead of unsigned long
for the fixed point math: we only need the fractional
part. I.e. 1.16 not 16.16
* Use narrowest possible type
* Optimise bin position logic for performance
* Only promote to uint32_t when really required.
* Simplify bin checks - no zero width bins
* Save memory: use a single byte for the last bin caches
* Performance: increment pointers instead of
repeatedly dereferncing array by index.
* Comments
* Incorrect array underrun logic
* Rename type, comments
* Fix unit tests
* Use table_row_iterator_t to adjust ignitionTable
* Separate 2d & 3d table code, abstract page to table iterator conversion.
This is just moving code around in preparation for future changes.
* Reduce table RAM (.bss) usage
Generate a separate type for each possible
3d axis & size combination.
This turns what was runtime information into
compile time data.
* Save 1 byte per table.
Use a flag value (INT16_MAX) instead of a separate boolean
flag
* File renaming
table_iterator -> table3d_iterator.h
table3d.h -> table3d_interpolate.h
table3d.cpp -> table3d_interpolate.cpp
table3d_types.h -> table3d.h
* Optimize page.cpp: reduce code clutter, maintain performance
* Reduce flash usage
* Encapsulate table metadata
* Performance - hoist if statement
* Replace function with macro
* Use a packed enum as a type identifier
Use a packed enum as a type identifier
Slimmer data types
* Use table iterators for random access to table
values and axis.
* Centralize write buffer check
* Encapsulate 16-bit reference concept
* Performance: make table iterators proper classes
This allows us to chain calls on temporaries - not possible
with regular function calls.
* Performance: encapsulate EEPROM update
& address increment
* Save flash - don't duplicate function
* Performance: directly invalidate table cache
* Separate out iterator reversal
* Separate out entity mapping & per-byte access
Much faster, smaller code footprint & easier to understand
* Code quality fixes
* Separate out axis metadata
* Doxygen comments
* Separate int16_ref into separate file
* Separate out table axies & values into separate types
No need for metadata types & more localised code.
E.g. creating iterators is now alongside the data over
which they iterate.
* Doxygen
tx_haggis
Daniel Tobias