* autoscale m_config->timeConstant
* min/max afr
* more
* dwell voltage
* closed loop fuel
* knock
* map estimate
* applyNonlinearBelowPulse
* fix
* add a test while we're at it for un-covered code
* test fallback MAP since we touched that
* always compute fallback MAP even if sensor is OK
* cli registry: rework this hell, add FFF and FFFFF
* kinetis: no nodefaultlib
* revert back token helpers
* cli_registery: argument parser: do not eat qoutes
* Unit test for FFF
* cli_registry: remove debug
* less dependence on that logger
* comma
* refactor a bit, make it clear what this thing actually *does*
* static
* shrink the buffer a bit and make names useful
* dead util that's just subtraction lol
* vestigial config
* replace last functionality
* engine/board configs include
* files that didn't need to include this
* remove FSIO implementation
* stragglers
* random little bits
* s
* s
* ConfigDefinition
* jar
* s
* vss based dfco
* fix digits while we're here
* fix some other digits for fun
* don't use custom assert
* simplify test
* less than or equal, test vss behavior
* even better than comments are tooltips
* consumers and api
* dead test
* baro uses sensor model
* remove old (copy of?) map lookup logic
* I guess we don't need FastInterpolation any more?
* don't double init analog input
* hook up fuel trims
* pass all the params
* store per-cylinder fuel mass directly
* main trigger callback only touches per-cylinder, no banks!
* test test test test
* move UI to happy land
* changelog
* extract isSyncPoint
* test because why not
* check vvt resync nb2
* custom nb decoder
* test only resyncs once!
* good job valgrind, you found a bug!
* ignition controller detects rising edge on voltage
* update test
* comment
* ignore negative transients
* tweak
* test
* priming happens on ignition-on
* priming has its own scheduling
* config & UI
* dead config
* implementation
* look, the test caught a bug
* keep the watchdog happy
* bad merge
* changelog
* easier to read the test
* test naming
* ignition controller detects rising edge on voltage
* update test
* comment
* ignore negative transients
* tweak
* test
* priming happens on ignition-on
* priming has its own scheduling
* test
* dead adjacent line of code
* vss uses real values
* some defaults
* test and correct math
* km, not miles!
* comment
* tooltip
* that macro went away
* 100hz and default settings gives 9kph
* changelog
* order of operations safety
* make the test like the pwm test
* housekeeping
* this is why we need sensor automation
Co-authored-by: Matthew Kennedy <makenne@microsoft.com>
* fuel pump
* s
* call correct callback
* more dead
* test adjustment
* don't overflow
* don't need separate function
* java
* give it a name
* generated
* generated
* add to generate script
* import *
* this is so cheap we don't need a flag
* main relay
* plumbing
* dead fsio stuff
* test it
* by 100%, we mean 99.98%
* Drop SingleChannelStateSequence in favor of MultiChannelStateSequenceWithData
Most all the users were Multi* anyways, so just improve that:
1. Allow up to 8 waveforms to share one byte per timestamp. It could be better but this is simple
and gets most of the benefit.
2. Use a wrapper structure to handle reserving space for the arrays. Makes the interface simpler
and more rigid. Also saves 4 bytes per Multi*. Downside is access is now via -> and *,
not . and (nothing).
Saves 224 bytes of BSS, 1832 bytes of RAM4/CCM, 952 bytes of TEXT, and 103 bytes of RODATA
* Instantiate a base_t to make debugging easier.
Also fixes crash on real firmware by using &m_base instead - LTO optimization issue?
* No magical templated StaticAlloc thingy. Just virtual functions.
* data
* stub test
* use getTimeSinceSyncPoint
* s
* s
* no
* test
* stray ;
* those were 1/10 what they were supposed to be
* actually check something
* dead log line?
* Basic framework for high pressure fuel pump control
* Many changes
Move calculations to fast callback
Move main object into Engine
Respond to pin changes without requiring a reboot
* Use EngineModule for HpfpController
Schedule pin off after executing pin on so we are sure it ends, even if the motor stops.
Test scheduling.
Less RAM use by only having one event and reordering fields.
* Make scheduling test actually useful - need non-0 activation angle.
Co-authored-by: rusefillc <48498823+rusefillc@users.noreply.github.com>
* fuel pump
* s
* call correct callback
* more dead
* test adjustment
* don't overflow
* don't need separate function
* java
* give it a name
* generated
* generated
* add to generate script
* import *
* this is so cheap we don't need a flag
engineModules.get<FOO>(). becomes modules<FOO>()->
I believe the new API is more conducive to supporting arrays, by doing std::array<T, N> or
std::array<Mockable<T>, N>, with the support of a helper class.
Fix some uses of ENGINE() not detected by regex due to parenthesis.
Remove now empty engine_ptr.h
Don't worry about EFI_* not being defined, the compiler defaults to 0 if they aren't.
* Create a base class EngineModule for that contains various useful callbacks.
This cleans up the API by not requiring the notifiers to know about who wants callbacks. The only
place you need to update to add a module is in one place.
* Add mockability.
* Convert InjectorModel to a Mockable EngineModule
* Programmatically remove EnginePtr/inject, EXPAND_Engine, and WITH_ENGINE_TEST_HELPER*
for i in ": public EnginePtr " ", public EnginePtr" "EXPAND_Engine;" "EXPAND_Engine"; do
git grep -l "$i" | xargs sed -i "s/$i//g"
done
git grep -l "inject" | xargs sed -i "/inject[(][)]/d"
for i in WITH_ENGINE_TEST_HELPER_SENS WITH_ENGINE_TEST_HELPER_BOARD_CALLBACK WITH_ENGINE_TEST_HELPER; do
git grep -l "$i" | xargs sed -i "s/$i/EngineTestHelper eth/g"
done
git checkout firmware/controllers/core/engine_ptr.h
git checkout unit_tests/global.h
* Review fixups.
* Start the process of using a global engine ptr for unit tests.
Passing around the engine, config, and persistent state wasn't buying anything. Instead, use
EngineTestHelper(Base) to correctly set and clear the global variables. Add a dummy check in
case some test tries to set them manually.
* Fix OSX build
* Adapt PR to recently added code.
* static functions with (void)
* more
* Revert "more"
This reverts commit 246e53441f935451437df186ac92d7df26b62fb6.
* s
Co-authored-by: Matthew Kennedy <makenne@microsoft.com>
Better than reaching into members that should be private. I didn't feel like actually making
them private though, as one user validates pinStates isn't NULL.
Memory seems more valuable than CPU; use the O(lg n) lookup by angle. A side effect is that it
seems to fix trigger lookup for the second phase of TRIGGERTYPE 53 544 TT_TRI _TACH 0.00
Having to remove the precompiled header when switching from firmware to simulator to unit_tests and
back again is annoying. Use a directory to store the precompiled header output. Turn off
precompiled header warnings as GCC still complains even if it finds a valid one eventually.
The last version is already much better than the original, mostly because it doesn't call
CriticalSectionLocker, which has a ton of overhead due to debug tracking. But this version is
another 4 instructions / 12 bytes shorter. Does as much match in 32-bit land as possible, and
avoids math operations that span 64-bits (i.e. either operate on the lower half or upper half, but
not both). The result is only 3 instructions not including the necessary 4 loads (ptr to now, now,
ptr to upper, upper), 1 store (upper), and 1 return/branch.
Disallow scaled_channel for float. I can't think of a reason to allow it, and it gets in the way
of rounding.
Add separate template param to setTable; let the compiler sort out whether assignment can happen
between TElement and VElement without forcing them to be the same at function call time.
* scale map Y axis
* allow different row/col types
* scaled channel detector
* interpolation
* looks like this actually works
* tests, no manual scaling
* comment
Matthew Kennedy
rusefillc
rusefillc
Andrey
Andrey G
Andreika
Andrei
Scott Smith