unix: Wake sleeping operations immediately on scheduled callbacks. by andrewleech · Pull Request #9 · andrewleech/micropython

andrewleech · 2026-02-24T23:53:51Z

Summary

On the unix port, time.sleep() and MICROPY_INTERNAL_WFE() were unresponsive to scheduled callbacks — time.sleep() used select() or sleep() with no wakeup mechanism, and WFE busy-polled with a 500us delay.

This adds a signal-based notification so that when something is scheduled (via MICROPY_SCHED_HOOK_SCHEDULED), a signal is sent to the process. A sig_atomic_t flag is set by both mp_hal_signal_event() and the signal handler. mp_unix_sched_sleep() uses pselect() to atomically unblock the signal and enter the wait, avoiding the TOCTOU race where signals for already-queued callbacks could be consumed before the sleep. The signal is blocked process-wide at init so only the thread inside pselect() can receive it, and pthread_sigmask is used on threaded builds for POSIX correctness.

time.sleep() is reworked to loop over mp_unix_sched_sleep() with elapsed-time tracking, processing pending callbacks each iteration. Negative values now raise ValueError to match CPython. WFE uses the same sleep primitive so it blocks properly and wakes on events rather than spinning.

Uses SIGRTMIN+7 where real-time signals are available, falling back to SIGURG.

Testing

Tested on the unix port (Linux), both standard and coverage variants. Windows codepath (_WIN32) is left unchanged — it keeps the existing select()/delay_us behaviour.

Trade-offs and Alternatives

Could have used a pipe/eventfd self-pipe pattern instead of signals, which would avoid any signal-number collision concerns. Signals are simpler and don't require managing an fd, and the chosen signal numbers avoid known conflicts with GC (SIGRTMIN+5) and thread terminate (SIGRTMIN+6).

The time.sleep() rewrite replaces the MICROPY_SELECT_REMAINING_TIME Linux-specific assumption about select() modifying the timeout struct with explicit mp_hal_ticks_ms() tracking, which is portable. Precision drops from microseconds to milliseconds which is acceptable for time.sleep().

Generative AI

I used generative AI tools when creating this PR, but a human has checked the code and is responsible for the description above.

andrewleech · 2026-02-25T00:01:45Z

/review

mpy-reviewer

The signal-based approach using pselect is the right design here — the atomic unblock-and-wait correctly eliminates the TOCTOU race. A few issues to address before merge.

mpy-reviewer · 2026-02-25T00:11:35Z

+}
+
+void mp_unix_init_sched_signal(void) {
+    struct sigaction sa;


sa is a stack variable; sa_restorer (and any padding) is uninitialised. Use:

Suggested change

struct sigaction sa;

struct sigaction sa = {0};

mpy-reviewer · 2026-02-25T00:11:35Z

+
+void mp_unix_deinit_sched_signal(void) {
+    MP_SIGMASK(SIG_UNBLOCK, &sched_signal_mask, NULL);
+    signal(MP_SCHED_SIGNAL, SIG_DFL);


Use sigaction consistently rather than mixing with signal:

Suggested change

signal(MP_SCHED_SIGNAL, SIG_DFL);

struct sigaction sa = {0};

sa.sa_handler = SIG_DFL;

sigemptyset(&sa.sa_mask);

sigaction(MP_SCHED_SIGNAL, &sa, NULL);

mpy-reviewer · 2026-02-25T00:11:35Z

-// Note that we don't delay for the full TIMEOUT_MS, as execution
-// can't be woken from the delay.
+// Wait for an event (scheduled callback) or timeout. A signal from
+// mp_hal_signal_event() causes select() to return EINTR, waking the wait.


"select()" → "pselect()".

mpy-reviewer · 2026-02-25T00:11:35Z

+        mp_raise_ValueError(MP_ERROR_TEXT("sleep length must be non-negative"));
    }
-    RAISE_ERRNO(res, errno);
+    uint64_t total_ms = (uint64_t)(val * MICROPY_FLOAT_CONST(1000.0));


This truncates to milliseconds, so time.sleep(0.0001) becomes a no-op (total_ms=0 → loop exits immediately without sleeping). The old code had microsecond precision. At minimum, round up:

Suggested change

uint64_t total_ms = (uint64_t)(val * MICROPY_FLOAT_CONST(1000.0));

uint64_t total_ms = (uint64_t)MICROPY_FLOAT_C_FUN(ceil)(val * MICROPY_FLOAT_CONST(1000.0));

github-actions · 2026-02-25T00:46:04Z

Code size report:

Reference:  zephyr/mpconfigport: Remove duplicate builtins.open definition. [1ab9b66]
Comparison: unix: Use sched signal in MICROPY_INTERNAL_WFE. [merge of 4443673]
  mpy-cross:    +0 +0.000% 
   bare-arm:    +0 +0.000% 
minimal x86:    +0 +0.000% 
   unix x64:  +827 +0.097% standard[incl +16(data)]
      stm32:    +0 +0.000% PYBV10
      esp32:    +0 +0.000% ESP32_GENERIC
     mimxrt:    +0 +0.000% TEENSY40
        rp2:    +0 +0.000% RPI_PICO_W
       samd:    +0 +0.000% ADAFRUIT_ITSYBITSY_M4_EXPRESS
  qemu rv32:    +0 +0.000% VIRT_RV32

Install an empty signal handler (without SA_RESTART) for a dedicated signal so that select() calls return EINTR when the scheduler queues a callback. mp_hal_signal_event() sends this signal via kill(getpid()), and MICROPY_SCHED_HOOK_SCHEDULED calls it from mp_sched_schedule(). This replaces the need for a self-pipe mechanism while remaining async-signal-safe. Signed-off-by: Andrew Leech <andrew.leech@planetinnovation.com.au>

Add a drain loop before the sleep to process any already-pending callbacks, subtracting the time taken from the requested duration. The existing EINTR/MICROPY_SELECT_REMAINING_TIME loop handles new callbacks scheduled during the sleep via the signal mechanism. Also add ValueError for negative sleep values (matching CPython) and a test exercising the drain loop and error handling. Signed-off-by: Andrew Leech <andrew.leech@planetinnovation.com.au>

Replace the fixed 500us delay in MICROPY_INTERNAL_WFE with mp_unix_sched_sleep(), which sleeps for the full requested timeout but wakes immediately on EINTR from the scheduler signal. Signed-off-by: Andrew Leech <andrew.leech@planetinnovation.com.au>

codecov-commenter · 2026-03-21T22:46:01Z

⚠️ Please install the to ensure uploads and comments are reliably processed by Codecov.

Codecov Report

✅ All modified and coverable lines are covered by tests.
⚠️ Please upload report for BASE (review/unix-sleep-process-pending@1ab9b66). Learn more about missing BASE report.
❗ Your organization needs to install the Codecov GitHub app to enable full functionality.

Additional details and impacted files

@@                         Coverage Diff                          @@
##             review/unix-sleep-process-pending       #9   +/-   ##
====================================================================
  Coverage                                     ?   98.42%           
====================================================================
  Files                                        ?      174           
  Lines                                        ?    22339           
  Branches                                     ?        0           
====================================================================
  Hits                                         ?    21988           
  Misses                                       ?      351           
  Partials                                     ?        0

☔ View full report in Codecov by Sentry.
📢 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.

andrewleech closed this Feb 25, 2026

andrewleech reopened this Feb 25, 2026

mpy-reviewer bot reviewed Feb 25, 2026

View reviewed changes

pi-anl added 3 commits March 22, 2026 09:24

andrewleech force-pushed the unix-sleep-process-pending branch from 3bd023c to 4443673 Compare March 21, 2026 22:28

andrewleech deleted the branch review/unix-sleep-process-pending March 21, 2026 22:35

andrewleech closed this Mar 21, 2026

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

unix: Wake sleeping operations immediately on scheduled callbacks.#9

unix: Wake sleeping operations immediately on scheduled callbacks.#9
andrewleech wants to merge 3 commits intoreview/unix-sleep-process-pendingfrom
unix-sleep-process-pending

andrewleech commented Feb 24, 2026

Uh oh!

andrewleech commented Feb 25, 2026

Uh oh!

mpy-reviewer bot left a comment

Uh oh!

mpy-reviewer bot Feb 25, 2026

Uh oh!

mpy-reviewer bot Feb 25, 2026

Uh oh!

mpy-reviewer bot Feb 25, 2026

Uh oh!

mpy-reviewer bot Feb 25, 2026

Uh oh!

github-actions bot commented Feb 25, 2026 •

edited

Loading

Uh oh!

codecov-commenter commented Mar 21, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

-    signal(MP_SCHED_SIGNAL, SIG_DFL);
+    struct sigaction sa = {0};
+    sa.sa_handler = SIG_DFL;
+    sigemptyset(&sa.sa_mask);
+    sigaction(MP_SCHED_SIGNAL, &sa, NULL);

	uint64_t total_ms = (uint64_t)(val * MICROPY_FLOAT_CONST(1000.0));
	uint64_t total_ms = (uint64_t)MICROPY_FLOAT_C_FUN(ceil)(val * MICROPY_FLOAT_CONST(1000.0));

Conversation

andrewleech commented Feb 24, 2026

Summary

Testing

Trade-offs and Alternatives

Generative AI

Uh oh!

andrewleech commented Feb 25, 2026

Uh oh!

mpy-reviewer bot left a comment

Choose a reason for hiding this comment

Uh oh!

mpy-reviewer bot Feb 25, 2026

Choose a reason for hiding this comment

Uh oh!

mpy-reviewer bot Feb 25, 2026

Choose a reason for hiding this comment

Uh oh!

mpy-reviewer bot Feb 25, 2026

Choose a reason for hiding this comment

Uh oh!

mpy-reviewer bot Feb 25, 2026

Choose a reason for hiding this comment

Uh oh!

github-actions bot commented Feb 25, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

codecov-commenter commented Mar 21, 2026

Codecov Report

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

github-actions bot commented Feb 25, 2026 •

edited

Loading