samples: mesh_badge: conversion of k_work API #34299
Conversation
There was a problem hiding this comment.
Approving based on the expectation this has been tested; AFAICT it's unchanged from the rote conversion patch.
I think several of the calls could use schedule instead of reschedule, which would preserve deadlines better. With reschedule there's theoretically a potential for livelock: the trigger event occurs fast enough that the work is never executed. But that's not new behavior, just not taking advantage of new capabilities.
| k_delayed_work_init(&motion_work, motion_timeout); | ||
| k_delayed_work_submit(&motion_work, MOTION_TIMEOUT); | ||
| k_work_init_delayable(&motion_work, motion_timeout); | ||
| k_work_reschedule(&motion_work, MOTION_TIMEOUT); |
There was a problem hiding this comment.
This could be k_work_schedule() since there is no earlier timeout that might need to be overridden.
| @@ -93,7 +93,7 @@ static void motion_timeout(struct k_work *work) | |||
| printk("power save\n"); | |||
|
|
|||
| if (!mesh_is_initialized()) { | |||
| k_delayed_work_submit(&motion_work, MOTION_TIMEOUT); | |||
| k_work_reschedule(&motion_work, MOTION_TIMEOUT); | |||
There was a problem hiding this comment.
I think this too could be k_work_schedule(), in case a motion event was detected after the work was queued but before the work thread got to it.
Replace all existing deprecated API with the recommended alternative. Be aware that this does not address architectural errors in the use of the work API. Fixes zephyrproject-rtos#34097 Signed-off-by: Peter Bigot <peter.bigot@nordicsemi.no> Signed-off-by: Johann Fischer <johann.fischer@nordicsemi.no>
jfischer-no
force-pushed
the
pr-fix-34097
branch
from
4a55d40 to
8b763dd
Compare
Replace all existing deprecated API with the recommended alternative.
Fixes #34097