rusefi
/
rusefi-1
mirror of https://github.com/rusefi/rusefi-1.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity
12,614 Commits 1 Branch 0 Tags 1.2 GiB
Commit Graph

50 Commits

Author SHA1 Message Date
Matthew Kennedy 7277b04039
force a pwm cycle start (skip cycles) if late (#2097) 
* skip cycles if late

* fail at lower count

* start at -1
 2020-12-18 08:21:18 -05:00
rusefillc ca79f99be5 More explicit handling of CPU exhaustion #2093 2020-12-17 22:21:12 -05:00
rusefillc 99392d7951 More explicit handling of CPU exhaustion #2093 2020-12-17 21:15:28 -05:00
rusefillc ba33a758e8 avoid float -> int64 conversion #1977 
both cases have to be relatively small durations of time cherry picking...
 2020-12-03 13:30:17 -05:00
rusefillc 641959b9a3 Starter is engaged on start-up in pull-up configuration #1969 2020-11-26 17:48:09 -05:00
Matthew Kennedy f3b7a1d9da
Remove lockAnyContext, replace with CriticalSectionLocker (#1938) 
* switch to CriticalSectionLocker

* that's just about all

* clean up last usage

* include hpp for sim

* need the cpp wrappers in the makefile too

* include dir

* include
 2020-11-19 06:56:02 -05:00
Matthew Kennedy 6e4517be8c
adjust microsecond_timer api (#1952) 
* change timer api

* don't need all that
 2020-11-16 18:50:13 -05:00
Matthew Kennedy 7eeb60a8a0
trigger/scheduler perf improvements (#1885) 
* avoid 64b on hot path

* no tooth log when fast

* use crit section locker

* final

* only flip pins if not also self stim

* only trace if doing work

* slightly drop lateDelay

* trace if not bailing out
 2020-10-16 11:04:27 -04:00
Matthew Kennedy 018db0df94
Fix #1873 - compute dynamic PWM iterationLimit (#1874) 
* fix

* remove define

* add an assert

* typo

* don't do a 64bit divide

* oh that's a float
 2020-10-11 20:25:38 -04:00
Matthew Kennedy c885db32f0
use expected for event queue next event (#1859) 
* use expected for event queue next event

* minor cleanup

* tests
 2020-10-05 13:53:34 -04:00
Matthew Kennedy dd6b02d555
we don't need two loops (#1855) 2020-10-05 08:57:00 -04:00
Matthew Kennedy aad5d987a2
pwm perf (#1858) 
* pwm perf

* this trace is now redundant
 2020-10-05 08:23:30 -04:00
Matthew Kennedy c8b4fac456
move scheduling to its own file (#1842) 
* move scheduling to its own file

* header

* protip don't include a cpp file
 2020-10-01 17:55:03 -04:00
Matthew Kennedy 4527b7694a expose the problem 2020-07-31 14:41:29 -07:00
Matthew Kennedy cd62774be1 dead 2020-07-29 02:45:30 -07:00
rusefi 222655517b better unit test logging 2020-07-19 15:25:49 -04:00
rusefi 54c9bf21ca better unit test logging 2020-07-19 14:58:46 -04:00
rusefi fb5756fe3e nope, unit tests did not just fix themselves 2020-06-17 08:42:37 -04:00
Matthew Kennedy f6bae09ee7
spinwait for events in the near future (#1491) 
* add spinwait

* simplify and reduce lateDelay

* maybe probably fix tests

* comments

Co-authored-by: Matthew Kennedy <makenne@microsoft.com>
 2020-06-16 22:13:42 -04:00
rusefi 200575ec87 Trigger decoding issues (60-2) #1488 
100us is too much, 10us is more reasonable
 2020-06-15 00:12:50 -04:00
rusefi f3344753e8 Trigger decoding issues (60-2) #1488 
interesting stuff potentially with a not so great setting
 2020-06-14 23:41:39 -04:00
rusefi f9087fdf72 Trigger decoding issues (60-2) #1488 
very mild refactoring - better names & moving initializers
 2020-06-14 23:34:45 -04:00
Matthew Kennedy 3a7c6f242b
Hardware pwm (#1283) 
* hard pwm

* fix, comments

* PC6-9 and PD12-15

* PB6-9

* dynamic frequency

* f7 mcuconf

* format

* relocate to common, support variable frequency

* error handling

* oops

* fix test build

* noop kinetis

* better guards

* guard for bl

* guards don't work if they have typos

* more mapping

* oops

* don't use for slow pwm

* enable pwm

* guard properly

* save a little memory

* use less mem

* correct timers

* all your ram are belong to us

* update limit & comment

Co-authored-by: Matthew Kennedy <makenne@microsoft.com>
 2020-05-28 16:02:41 -04:00
rusefi a4329169e9 better file name 2020-05-25 13:02:05 -04:00
Matthew Kennedy 96a800acb9
fix gppwm on-off mode (#1406) 
* fix

* put it back

* do it with PWM instead

* comments

Co-authored-by: Matthew Kennedy <makenne@microsoft.com>
 2020-05-09 09:25:45 -04:00
rusefi 4ea0a3de26 nicer error messages 2020-05-09 01:49:17 -04:00
rusefi 1b84406e77 dead code 2020-04-26 17:40:12 -04:00
Matthew Kennedy f128b33694
General purpose PWM: implementation & tests (#1366) 
* config

* significant digits

* renumber enum, no need for a "none"

* ui

* impl base

* error handle MAP

* init & update

* don't need arg

* don't lie about sensor

* fix test build

* test stub

* initialize

* null check

* fix clamping

* test output

* types & enums

* don't need param

* test getOutput

* fix

* output pin instead of brain pin

* default config
 2020-04-26 14:06:28 -04:00
Matthew Kennedy f704bf22cb
use the linker! (#1363) 2020-04-24 21:21:04 -04:00
rusEfi 4854d123bb PWM name into warning message 2020-04-19 21:48:14 -04:00
Matthew Kennedy 2349d74f0b
Default to applyPinState (#1219) 
* make applyPinState default

* consumers

* format
 2020-03-25 18:14:09 -04:00
Matthew Kennedy 986b80671b
warnings (#1169) 2020-03-03 08:37:02 -05:00
rusefi 0c3f73c2b4 code style 2020-01-21 01:47:58 -05:00
rusefi 83b1ca00a4 code style 2020-01-21 01:40:11 -05:00
Matthew Kennedy 09dc2f2390 don't use us for pwm calculation (#1106) 2020-01-19 07:02:49 -05:00
rusefi 7da4196f9c NT not US not MS 
thank you Matt
 2020-01-19 03:49:35 -05:00
rusefi 85a0490796 one step back 2020-01-19 02:58:54 -05:00
rusefi 2ae53d1681 better constraint validation 2020-01-19 00:28:58 -05:00
rusefi bb217b01dd happy new year unification 2020-01-13 21:57:43 -05:00
rusefi 645f51a038 better method names 2020-01-10 23:17:58 -05:00
Matthew Kennedy a6a1335adb Require tooth reference time for scheduleByAngle (#1091) 
* injection

* injectors

* add edge timestamp to ShaftPositionListener

* scheduleByAngle require edgeTimestamp

* schedule with nt not us

* oops, these were missing from this branch
 2020-01-09 23:45:13 +03:00
rusefi b83a5310be happy new year 2020-01-08 00:02:40 -05:00
Matthew Kennedy 95a7a9a8e1 Strongly typed action constructor (#1087) 
* strong typing

* maybe we needed that one
 2020-01-07 18:10:31 -05:00
Matthew Kennedy 5c85d53e16 Pass action_s instead of callback/param (#1084) 
* change action

* consumers

* fix test

* didn't mean to add those

* simplify

* fix simulator

* fix sim for real

* oy

* maybe this time the simulator will actually be fixed, for real

* don't rely on undefined behavior
 2020-01-07 00:41:18 -05:00
rusefi fde9a449cf random refactoring: hopefully not changing byte size of any variables but clarifying/fixing type between ticks, US and MS 2019-12-21 21:11:09 -05:00
Matthew Kennedy db3ae5b03f DRAFT simplify dequeue logic (#1050) 
* simplify dequeue logic

* unnecessary

* format

* doc
 2019-12-14 18:00:39 -05:00
rusefi 9403b2a8fc fixing build is my most favourite time! 2019-12-13 18:02:24 -05:00
rusefi 5055797f6b The Big Refactoring of 2019: folder structure #723 2019-12-04 01:15:52 -05:00
rusefi 813b6aa6ce The Big Refactoring of 2019: folder structure #723 2019-12-04 01:11:10 -05:00
rusefi f6a50b7375 The Big Refactoring of 2019: folder structure #723 2019-12-04 00:35:05 -05:00