machine: Add nchanges param to time_pulse_us func. by IhorNehrutsa · Pull Request #16160 · micropython/micropython

IhorNehrutsa · 2024-11-05T15:32:33Z

time_pulse_us(pin, pulse_level, timeout_us=1000000, nchanges=2, /)

If nchanges is 3, if the pin is initially equal to pulse_level then first
waits until the pin input becomes different from pulse_level.
Then if the current input value of the pin is different to pulse_level,
the function first waits until the pin input becomes equal to pulse_level,
then times the duration that the pin is equal to pulse_level.

The advantage is that there is no need for additional synchronization before measuring the pulse duration.
A little bit longer, but with higher accuracy.

The jitter of the pulse measurement process is about 5-15 μs on the ESP32 board.

Idia from @robert-hh #16147 (comment)

This function is used in DRAFT PR for tests
PWM from 10845 and time_hardware_pulse_us() and test from 16147. #16161
The tests results show the functionality of the function over the entire ESP32 PWM frequency range from 1Hz to 40MHz.

github-actions · 2024-11-05T15:40:56Z

Code size report:

Reference:  esp32/machine_sdcard: Expose SDMMC host bus slot width in board config. [1a4df82]
Comparison: Update machine_pwm.py [merge of 9b147ab]
  mpy-cross:    +0 +0.000% 
   bare-arm:    +0 +0.000% 
minimal x86:    +0 +0.000% 
   unix x64:   +40 +0.005% standard
      stm32:    +8 +0.002% PYBV10
      esp32:   +68 +0.004% ESP32_GENERIC
     mimxrt:    +8 +0.002% TEENSY40
        rp2:   +24 +0.002% RPI_PICO_W
       samd:   +40 +0.014% ADAFRUIT_ITSYBITSY_M4_EXPRESS
  qemu rv32:    +0 +0.000% VIRT_RV32

robert-hh · 2024-11-05T15:59:47Z

Thanks for mentioning me. I considered creating that feature, but as optional argument to time_pulse_us() instead of a new method. That looks like a smaller change, even if the additional code size might be the same..

codecov · 2024-11-05T22:04:12Z

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 98.55%. Comparing base (49fe174) to head (9b147ab).
⚠️ Report is 5 commits behind head on master.

Additional details and impacted files

@@           Coverage Diff           @@
##           master   #16160   +/-   ##
=======================================
  Coverage   98.55%   98.55%           
=======================================
  Files         179      179           
  Lines       23244    23246    +2     
=======================================
+ Hits        22908    22910    +2     
  Misses        336      336

Flag	Coverage Δ
unix-coverage-32bit	`98.55% <100.00%> (-0.01%)`	⬇️
unix-coverage-64bit	`98.48% <100.00%> (+<0.01%)`	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

IhorNehrutsa · 2024-12-19T13:04:01Z

Remove additionnal synchronization time_pulse_us() from tests/extmod_hardware/machine_pwm.py

dpgeorge · 2025-05-23T03:30:42Z

After discussing with @projectgus , we feel that this is a useful enhancement but maybe not worth the cost in code size.

I think it would be possible to make this PR a lot simpler and less code size. I opened #17346 as a basis for that.

If #17346 is merged, then the additional argument added here could be generalised to a counter, eg:

machine.time_pulse_us(pin, pulse_level, timeout_us=1000000, num_levels=2)

With num_levels=2 it's the same behaviour as before. With num_levels=3 it waits for 3 level changes and that implements the behaviour in this PR. Higher values are possible, which allow more synchronisation.

robert-hh · 2025-05-23T10:08:11Z

then the additional argument added here

It's not (yet) visible in PR 17346.

dpgeorge · 2025-06-13T06:35:40Z

Now that #17346 is merged, this PR can be updated on that.

dpgeorge · 2026-05-05T01:29:19Z

+   The function returns -3 if there was timeout waiting for condition marked (***) above.
   The function will return -2 if there was timeout waiting for condition marked
-   (*) above, and -1 if there was timeout during the main measurement, marked (**)
+   (**) above, and -1 if there was timeout during the main measurement, marked (*)


It's quite tricky to document the nchanges parameter.

I think the original description should remain unchanged. And then add a paragraph at the end stating something like "The above behaviour is for when nchanges=2. That behaviour can be generalised by passing in a larger value for nchanges which is the number of times the function waits for the pin to change. On error the return value can be down to -nchanges and indicates how many edges were missed." Or something like that.

dpgeorge · 2026-05-05T01:29:47Z

+        nchanges = mp_obj_get_int(args[3]);
+    }
+    mp_uint_t us = machine_time_pulse_us(pin, level, timeout_us, nchanges);
+    // May return -1 or -2 or -3 in case of timeout


This comment needs adjusting because the return value can be between -nchanges and -1 (on timeout).

dpgeorge · 2026-05-05T01:30:22Z

-            time_pulse_us(pulse_in, level, timeout)
            for _ in range(n_averaging):
-                t += time_pulse_us(pulse_in, level, timeout)
+                t += time_pulse_us(pulse_in, level, timeout, 3)


I think this code should stay as it was, because inside this loop you don't need to wait for an extra change.

nchanges=4 works, nchanges=2 don't works

dpgeorge · 2026-07-11T15:14:10Z

@IhorNehrutsa there are some review comments above that need to be addressed before this PR can move forward.

IhorNehrutsa · 2026-07-16T12:50:10Z

-            time_pulse_us(pulse_in, level, timeout)
            for _ in range(n_averaging):
-                t += time_pulse_us(pulse_in, level, timeout)
+                t += time_pulse_us(pulse_in, level, timeout, 4)


On esp32 generic board if nchanges=4 works, if nchanges=2 or 3 don't works.
@robert-hh
If you have time, could you test tests/extmod_hardware/machine_pwm.py on the mimxrt port?
Thanks.

Signed-off-by: IhorNehrutsa <[email protected]> Co-Authored-By: Robert Hammelrath <[email protected]>

IhorNehrutsa force-pushed the time_hardware_pulse_us branch from 7a9b45d to f59833b Compare November 5, 2024 15:42

IhorNehrutsa mentioned this pull request Nov 5, 2024

PWM from 10845 and time_hardware_pulse_us() and test from 16147. #16161

Closed

IhorNehrutsa force-pushed the time_hardware_pulse_us branch from f59833b to e29b760 Compare November 5, 2024 22:01

IhorNehrutsa force-pushed the time_hardware_pulse_us branch from f25ae18 to b9d6cf0 Compare November 7, 2024 06:22

IhorNehrutsa force-pushed the time_hardware_pulse_us branch from b9d6cf0 to aa1cba8 Compare December 5, 2024 22:55

IhorNehrutsa changed the title ~~machine: Add time_hardware_pulse_us function.~~ machine: Add wait_opposite param to time_pulse_us func. Dec 5, 2024

IhorNehrutsa force-pushed the time_hardware_pulse_us branch 3 times, most recently from 90f267b to 28906bc Compare December 5, 2024 23:24

IhorNehrutsa force-pushed the time_hardware_pulse_us branch 2 times, most recently from dd7b953 to b6bcf72 Compare December 19, 2024 12:50

IhorNehrutsa force-pushed the time_hardware_pulse_us branch from b6bcf72 to e2a1c3a Compare December 19, 2024 13:08

dpgeorge added the extmod Relates to extmod/ directory in source label Jan 23, 2025

thomasabbott mentioned this pull request Feb 6, 2025

nRF microbit time_pulse_us returns bad values half the time #16714

Closed

IhorNehrutsa force-pushed the time_hardware_pulse_us branch from e2a1c3a to 27486f2 Compare March 13, 2025 09:04

dpgeorge mentioned this pull request May 23, 2025

Optimise machine.time_pulse_us implementation for code size, and switch esp8266 port to use it #17346

Merged

IhorNehrutsa force-pushed the time_hardware_pulse_us branch 2 times, most recently from 1cdca1b to c3b0cfc Compare May 4, 2026 18:44

IhorNehrutsa changed the title ~~machine: Add wait_opposite param to time_pulse_us func.~~ machine: Add nchanges param to time_pulse_us func. May 4, 2026

dpgeorge reviewed May 5, 2026

View reviewed changes

IhorNehrutsa force-pushed the time_hardware_pulse_us branch 2 times, most recently from 0682017 to d887a4f Compare July 13, 2026 10:39

IhorNehrutsa force-pushed the time_hardware_pulse_us branch 3 times, most recently from 3b28814 to eefa846 Compare July 15, 2026 13:36

IhorNehrutsa commented Jul 16, 2026

View reviewed changes

IhorNehrutsa and others added 5 commits July 22, 2026 12:19

machine: Add nchanges param to time_pulse_us func.

b68ebf1

Signed-off-by: IhorNehrutsa <[email protected]> Co-Authored-By: Robert Hammelrath <[email protected]>

Update machine.rst

2d185ce

Update machine_pwm.py

0c122f3

Update machine_pulse.c

51d62d6

Update machine_pulse.py

e99a890

IhorNehrutsa force-pushed the time_hardware_pulse_us branch from eefa846 to da6c42a Compare July 22, 2026 09:20

Update machine_pwm.py

9b147ab

IhorNehrutsa force-pushed the time_hardware_pulse_us branch from da6c42a to 9b147ab Compare July 22, 2026 09:36

Navigation Menu

Repository navigation

machine: Add nchanges param to time_pulse_us func. - #16160

machine: Add nchanges param to time_pulse_us func.#16160
IhorNehrutsa wants to merge 6 commits into
micropython:masterfrom
IhorNehrutsa:time_hardware_pulse_us

IhorNehrutsa commented Nov 5, 2024 •

edited

Loading

Uh oh!

github-actions Bot commented Nov 5, 2024 •

edited

Loading

Uh oh!

robert-hh commented Nov 5, 2024

Uh oh!

codecov Bot commented Nov 5, 2024 •

edited

Loading

Uh oh!

IhorNehrutsa commented Dec 19, 2024

Uh oh!

dpgeorge commented May 23, 2025

Uh oh!

robert-hh commented May 23, 2025

Uh oh!

dpgeorge commented Jun 13, 2025

Uh oh!

dpgeorge May 5, 2026

Uh oh!

dpgeorge May 5, 2026

Uh oh!

dpgeorge May 5, 2026

Uh oh!

IhorNehrutsa Jul 13, 2026 •

edited

Loading

Uh oh!

dpgeorge commented Jul 11, 2026

Uh oh!

IhorNehrutsa Jul 16, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

Conversation

IhorNehrutsa commented Nov 5, 2024 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

github-actions Bot commented Nov 5, 2024 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

robert-hh commented Nov 5, 2024

Uh oh!

codecov Bot commented Nov 5, 2024 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Codecov Report

Uh oh!

IhorNehrutsa commented Dec 19, 2024

Uh oh!

dpgeorge commented May 23, 2025

Uh oh!

robert-hh commented May 23, 2025

Uh oh!

dpgeorge commented Jun 13, 2025

Uh oh!

dpgeorge May 5, 2026

Choose a reason for hiding this comment

Uh oh!

dpgeorge May 5, 2026

Choose a reason for hiding this comment

Uh oh!

dpgeorge May 5, 2026

Choose a reason for hiding this comment

Uh oh!

IhorNehrutsa Jul 13, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

dpgeorge commented Jul 11, 2026

Uh oh!

IhorNehrutsa Jul 16, 2026

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

IhorNehrutsa commented Nov 5, 2024 •

edited

Loading

github-actions Bot commented Nov 5, 2024 •

edited

Loading

codecov Bot commented Nov 5, 2024 •

edited

Loading

IhorNehrutsa Jul 13, 2026 •

edited

Loading