* start with radar interface
* car interface
* panda safety rx check
* check panda safety inits
* check NO_RADAR_SLEEP in all radar interfaces
* sonata 2019 and kia optima should use hyundai legacy safety
* check radar errors
* check radar can errors
* real fingerprint
* run in CI
* it shoud pass now
* ignore old openpilot msgs
* ensure safety mode is set
* sort can msgs
* filter out openpilot msgs
old-commit-hash: 99106fd9c4
* lgtm in readme, and mac nui fixes
* call super inits in radars
* unused imports, dup lines
* more radars, more unused imports
* pass CP into RadarInterfaceBase
* more fixups
* unused imports
* delete unused lines
* ugh, new unused import
Co-authored-by: George Hotz <geohot@gmail.com>
old-commit-hash: 1295cfe06c
* don't need individual prev state vars since whole CS is copied
* don't need a whole function since toyota alerts were simplified
* better here
* abstract pcmEnable/pcmDisable events
* the flippening came to NEOS
* delete
* unused
* Revert "the flippening came to NEOS"
This reverts commit 91cfb73b29.
old-commit-hash: 8d791d78dd
* min_steer_speed alert (with adjusted hysteresis) for all cars, abstracted gas pre_enable event
* move lines a bit
* more GM cleanup
* typo
* move espDisabled to carState packet
* brought back min steer speed alert only in hyundai. It's a bit tricky and behavior should be changed
* ops, wrong change
* bug
* update cereal
* update ref
old-commit-hash: 9685a5a743
* too many if
* unused
* whitespace
* key
* sefldrive/car/*
* no more gctx
* lower
* start abstracting common events
* all cars
* start small
* all cars
* reverse gear
* wrongCarMode
* wrongGear
* espDisabled
* steerUnvailable
* make linter happy
* c isn't used
* fix esp_disabled in VW
* update ref
* more red
* more cleanup
* fix subaru
* update ref
old-commit-hash: e8cb6ea06a
* Before abstraction, adding speed init from VW as well
* strting to abstract carstate class
* fix bug and update lock?
* revert pipfile change
* another bug
* fix linter
* bug fix
* remove a bunch of diplicated kf code
* better to not have class vars. will abstract __init__ anyway later
* abstract common instance vars in carstate init and a generic gear parser static method
* abstract gear parser for chrysler
* abstract gm gear parser too
* remove unnecessary random vars
* Chrysler: carstate returns capnp struct directly
* revert ref commit
* test ref
* WIP
* more WIP
* ops, missed this conflict
* ford as well
* not sure why this got deleted
* no need to copy
* remove copy
* remove copy import
* remove unnecessary intermediate variable
* remove obsolete comments
* GM: have carstate returning capnp struct directly
* Honda carstate also outputing capnp struct
* hyundai too now returns capnp from carstate
* ops, not meant this
* Subaru carstate also returning capnp
* Toyota: capnp struct as output of carstate
* fix bool
* minor simplififcation in Honda
* no need to negate
* VW carstate returning capnp struct (#1118)
* VW carstate also returning capnp struct
* fixed typo
* Remove unused blinker button (#1119)
* remove unused blinker button
* ops, this wasn't meant
* remove blinker button for VW as well
* update ref
Co-authored-by: Willem Melching <willem.melching@gmail.com>
old-commit-hash: 0c67143c92
* Before abstraction, adding speed init from VW as well
* strting to abstract carstate class
* fix bug and update lock?
* revert pipfile change
* another bug
* fix linter
* bug fix
* remove a bunch of diplicated kf code
* better to not have class vars. will abstract __init__ anyway later
* Abstract gear parser static method (#1107)
* abstract common instance vars in carstate init and a generic gear parser static method
* update opendbc (#1105)
* abstract gear parser for chrysler
* update opendbc (#1106)
* abstract gm gear parser too
* remove unnecessary random vars
old-commit-hash: 5c52be27ca
* Resume Required Alert should be triggered off of ECM state
On GM cars the ECM goes into cruise state 4 (standstill) and in order
for the ECM to accept cruise inputs, the resume button must be pressed
to drive it to state 1 (active).
Previously rather than testing for if the ECM was in state 4, the wheel
speed sensors were checked that the vehicle was going less than 0.5 m/s.
In some cases this could cause the resume required alert to be displayed
before the car actually came to a stop as far as the ECM was concerned.
If the alert is triggered before state 4 and OP demands acceleration,
its possible for OP to acelerate while the Resume Required Alert is
displayed created a very confusing UX.
* After more review, the problem was actually with the 'if at_full_stop' check was inside an else statement preventing entering the stopped mode if braking is 0
old-commit-hash: 4b12aca941