メニュー
On-chip watchdog timer for TuyaOpen using the tkl_watchdog API: init with a timeout, then refresh ("feed") periodically so the MCU doesn't reset. 看门狗、watchdog、喂狗、复位保护、超时复位、tkl_watchdog、防死机。
デフォルトでは、最初にソースを確認する Prompt が選択されています。直接コマンドに切り替えるか、ローカルコピーをダウンロードすることもできます。
インストールを決める前に、SKILL.md と SkillsMP に表示されている付属ファイルをお読みください。
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
直接コマンドでは確認用 Prompt が省略されます。実行前にソースを確認してください。
npx skills add https://github.com/tuya/tuyaopen-ide-manifests --skill tuyaopen-onchip-watchdogコマンドは1行のまま表示されます。コピー前に横へスクロールして全体を確認してください。
ローカルで確認しますか？SkillsMP が現在取得できるファイルをダウンロードできます。
Build and compile TuyaOpen projects, select build configurations, edit Kconfig options, clean artifacts, and run Linux ELF binaries. Use when the user mentions compiling, building, tos.py build, config choice, menuconfig, Kconfig, build error, or running a project. 项目编译、构建、编译配置、清理编译、编译错误、menuconfig、Kconfig。
Agent-friendly non-blocking serial log capture for TuyaOpen devices. Runs tos.py monitor in the background, writes logs to file, and lets the agent tail or stop it at any time. Use when the agent needs to capture device logs without blocking, run a debug session, or monitor UART output during a build-flash-analyze loop. 后台串口日志、非阻塞监控、设备日志捕获、agent日志分析。
Automated build-flash-monitor-analyze development loop for TuyaOpen devices. Covers log analysis, error patterns, CLI testing, and iterative debugging. Use when the user mentions dev loop, automated testing, log analysis, debug cycle, iterative development, or CI loop. 开发闭环、自动化测试、日志分析、调试循环、迭代开发。
SOC 職業分類に基づく
SKILL.md を表示中
| name | tuyaopen/onchip-watchdog |
| description | On-chip watchdog timer for TuyaOpen using the tkl_watchdog API: init with a timeout, then refresh ("feed") periodically so the MCU doesn't reset. 看门狗、watchdog、喂狗、复位保护、超时复位、tkl_watchdog、防死机。 |
| when_to_use | Use when the user wants a hardware watchdog: auto-reset the MCU if the firmware hangs, by feeding it on a healthy heartbeat. |
| id | onchip-watchdog |
| surfaces | ["embedded"] |
| tags | ["watchdog","wdt","reset","on-chip","tkl_watchdog"] |
CONFIG_ENABLE_WATCHDOG. No app_default.config change.board_register_hardware(). Call tkl_watchdog_* directly. There is a single
system watchdog (no instance id / pins) — record as onchip:watchdog in
used-peripherals.json if you want it on the diagram.tkl_watchdog_init() returns the actual refresh interval the hardware granted
(may differ from what you asked). Refresh before that interval elapses, from a
point that only runs when the firmware is healthy.
#include "tal_api.h"
#include "tkl_watchdog.h"
#define WDT_TIMEOUT_MS 10000
void app_watchdog_start(void)
{
TUYA_WDOG_BASE_CFG_T cfg = { .interval_ms = WDT_TIMEOUT_MS };
uint32_t actual_ms = tkl_watchdog_init(&cfg); /* hardware-granted interval */
PR_NOTICE("watchdog interval = %u ms", actual_ms);
}
/* Call this well within the interval, on your healthy heartbeat (e.g. main loop). */
void app_watchdog_feed(void)
{
tkl_watchdog_refresh();
}
| Function | Description |
|---|---|
tkl_watchdog_init(&cfg) | cfg.interval_ms = requested timeout. Returns the actual granted interval (ms) |
tkl_watchdog_refresh() | Feed the dog — call before the interval expires |
tkl_watchdog_deinit() | Disable the watchdog |
| Symptom | Likely cause | Fix |
|---|---|---|
| Device keeps resetting | Not feeding in time | Refresh well within the returned interval, not the requested one |
| Watchdog never resets a hang | Feeding from a timer/ISR that runs regardless of health | Feed only from a path that stops when the app hangs (main loop) |
| Reset during long blocking op | Interval shorter than the operation | Increase interval_ms, or feed inside the long op at safe points |
SDK: examples/peripherals/watchdog/ (init with timeout, periodic refresh).