Code

Stepper optimisations#25474

Code

Merged

thinkyhead merged 47 commits intoMarlinFirmware:bugfix-2.1.xMarlinFirmware/Marlin:bugfix-2.1.xfrom tombrazier:stepper_optimisationstombrazier/Marlin:stepper_optimisationsCopy head branch name to clipboardMar 18, 2023

+124-71

Commits

Commits on Mar 6, 2023

• When moving fast, do not waste CPU cycles applying an offset to improve the minimal speed case

  tombrazier

  committed
  27cca14

  Copy full SHA for 27cca14

• For the fastest step rates, do the least work in calc_timer_interval()

  tombrazier

  committed
  52771c5

  Copy full SHA for 52771c5

• Save 256 bytes of PROGMEM and gain cycles by saving the LERP gradient as a byte for speed_lookuptable_fast

  tombrazier

  committed
  e42ba84

  Copy full SHA for e42ba84

• Use hal_timer_t for timers in the stepper ISR

  tombrazier

  committed
  535a931

  Copy full SHA for 535a931

• The compiler is not smart enough to distribute the multiplication through the ternay function

  tombrazier

  committed
  8391f31

  Copy full SHA for 8391f31

• Removed redundant check against step_needed

  tombrazier

  committed
  bbf1d4e

  Copy full SHA for bbf1d4e

• Use the AxisFlags bitfield rather than the xyze_bool_t array for step_needed

  tombrazier

  committed
  a60777c

  Copy full SHA for a60777c

• Add 1 byte to RAM and save a lot of load/compare of 32 bit numbers

  tombrazier

  committed
  67ee043

  Copy full SHA for 67ee043

• Helped the compiler to keep bresenham logic in registers

  tombrazier

  committed
  848a638

  Copy full SHA for 848a638

• As well as not needing to disable interrupts on AVR, you do not need to re-enable them either

  tombrazier

  committed
  3469556

  Copy full SHA for 3469556

• Added multistepping debugging to max7219

  tombrazier

  committed
  6ecf510

  Copy full SHA for 6ecf510

• Not clean: added dynamic CPU control

  tombrazier

  committed
  62870e5

  Copy full SHA for 62870e5

• Merge branch 'bugfix-2.1.x' into stepper_optimisations

  tombrazier

  committed
  c537719

  Copy full SHA for c537719

• Update createSpeedLookupTable.py

  thinkyhead

  committed
  e34ce79

  Copy full SHA for e34ce79

• remove dead friends

  thinkyhead

  committed
  0c1057c

  Copy full SHA for 0c1057c

• sei() automatically invoked by AVR reti

  thinkyhead

  committed
  4a8ca11

  Copy full SHA for 4a8ca11

• style tweak

  thinkyhead

  committed
  6c8431a

  Copy full SHA for 6c8431a

• Minor calc_timer_interval optimizations

  thinkyhead

  committed
  ae934bf

  Copy full SHA for ae934bf

Commits on Mar 7, 2023

• style tweak

  thinkyhead

  committed
  ea948d6

  Copy full SHA for ea948d6

• Corrected logic for speed_lookuptable_fast to match values from 27cca14, matched formatting to speed_lookuptable.h and improved rounding

  tombrazier

  committed
  6c039bd

  Copy full SHA for 6c039bd

Commits on Mar 8, 2023

• Optimisation of dynamic multi-steppping

  tombrazier

  committed
  1a51ec8

  Copy full SHA for 1a51ec8

Commits on Mar 9, 2023

• Don't update max7219 from within stepper ISR

  tombrazier

  committed
  25bd2fb

  Copy full SHA for 25bd2fb

Commits on Mar 11, 2023

• Use steps_per_isr directly

  thinkyhead

  committed
  f5b69e4

  Copy full SHA for f5b69e4

• Actually, don't modify

  thinkyhead

  committed
  a648a3c

  Copy full SHA for a648a3c

• Drop MAX_STEP_ISR_FREQUENCY_*X

  thinkyhead

  committed
  3a4e3f6

  Copy full SHA for 3a4e3f6

• Merge branch 'bugfix-2.1.x' into pr/25474

  thinkyhead

  committed
  010ab01

  Copy full SHA for 010ab01

• cosmetic

  thinkyhead

  committed
  2fc197e

  Copy full SHA for 2fc197e

• Added MULTISTEPPING_LIMIT to config to limit the effect of multistepping. This replaces DISABLE_MULTI_STEPPING which was never visible in the config files. And removed the speed cap from calc_timer…

  Show description for ea1ee11
  tombrazier

  committed
  ea1ee11

  Copy full SHA for ea1ee11

• tweaks to allow CI test

  thinkyhead

  committed
  cb933fb

  Copy full SHA for cb933fb

• MULTISTEPPING_LIMIT must be simple

  thinkyhead

  committed
  776cfaa

  Copy full SHA for 776cfaa

• comment tweak

  thinkyhead

  committed
  e694a9c

  Copy full SHA for e694a9c

• comment tweak

  Show description for 3b1af15
  thinkyhead

  committed
  3b1af15

  Copy full SHA for 3b1af15

• move conditional

  thinkyhead

  committed
  96ed3d0

  Copy full SHA for 96ed3d0

• reorder calc_timer_interval for comparison

  thinkyhead

  committed
  216ba61

  Copy full SHA for 216ba61

• "calc_multistep_timer_interval"

  thinkyhead

  committed
  3644860

  Copy full SHA for 3644860

Commits on Mar 12, 2023

• filter by multistepping limit

  thinkyhead

  committed
  49d69aa

  Copy full SHA for 49d69aa

• Merge branch 'bugfix-2.1.x' into pr/25474

  thinkyhead

  committed
  3150048

  Copy full SHA for 3150048

• general cleanups from review

  thinkyhead

  committed
  6ec1f29

  Copy full SHA for 6ec1f29

• Signficant optimisation of input shaping

  tombrazier

  committed
  b8455e7

  Copy full SHA for b8455e7

• format tweak

  thinkyhead

  committed
  0a2df85

  Copy full SHA for 0a2df85

• branch-not-taken is faster…

  Show description for 3a6716c
  thinkyhead

  committed
  3a6716c

  Copy full SHA for 3a6716c

• Merge branch 'bugfix-2.1.x' into pr/25474

  thinkyhead

  committed
  4db7e8c

  Copy full SHA for 4db7e8c

Commits on Mar 13, 2023

• Merge branch 'bugfix-2.1.x' into pr/25474

  thinkyhead

  committed
  3cb8b2d

  Copy full SHA for 3cb8b2d

Commits on Mar 17, 2023

• remove Q

  thinkyhead

  committed
  fa0e4d8

  Copy full SHA for fa0e4d8

• Old adaptive multi-stepping option

  thinkyhead

  committed
  8c70c3e

  Copy full SHA for 8c70c3e

• tweak e motor direction

  thinkyhead

  committed
  e7a2bab

  Copy full SHA for e7a2bab

Commits on Mar 18, 2023

• Merge branch 'bugfix-2.1.x' into pr/25474

  thinkyhead

  committed
  679a587

  Copy full SHA for 679a587