• Home
  • Features
  • Pricing
  • Docs
  • Announcements
  • Sign In

dangernoodle-io / TaipanMiner / 25644454287 / 3
90%
main: 90%

Build:
DEFAULT BRANCH: main
Ran 11 May 2026 12:49AM UTC
Files 66
Run time 1s
Badge
Embed ▾
README BADGES
x

If you need to use a raster PNG badge, change the '.svg' to '.png' in the link

Markdown

Textile

RDoc

HTML

Rst

11 May 2026 12:44AM UTC coverage: 91.171%. Remained the same
25644454287.3

push

github

web-flow
feat: bb v0.26.0 + use bb_diag abnormal-reset counter + fix asic WDT spin (#385)

* chore(breadboard): bump pin to v0.26.0

Pulls in:
- feat(bb_diag): NVS-backed abnormal-reset counter + DELETE /api/diag/abnormal-resets
- feat(bb_diag): /api/diag/heap and /api/diag/heap/check
- feat(bb_diag): /api/diag/tasks (gated on CONFIG_FREERTOS_USE_TRACE_FACILITY)

Counter migration follows in next commit; the new observability routes
self-register and need no call-site changes here.

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>

* refactor: drop local wdt_resets NVS counter; use bb_diag_abnormal_reset_count()

bb_diag now owns the abnormal-reset counter (NVS-backed) and exposes
DELETE /api/diag/abnormal-resets to clear it. Drop the duplicated
read/bump/cache logic in TM and call the bb_diag getter for the
/api/info JSON field.

Migration: deployed devices' taipanminer/wdt_resets NVS key becomes
orphaned (data preserved but unread); bb_diag's bb_diag/reset_count
key starts at 0 on first boot of this firmware. Cosmetic loss — the
count is operational telemetry, not load-bearing.

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>

* fix(asic): pet WDT at top of loop to survive empty-queue spin

asic_mining_task registers with the task watchdog at startup but only
calls esp_task_wdt_reset() after successfully processing a work item.
When the work queue stays empty for >5s (e.g. stratum disconnect, pool
switch, network glitch) the loop spins on xQueuePeek's 100ms timeout
without petting the WDT, and the task starves the watchdog -> device
reboots with task_wdt.

Hoist esp_task_wdt_reset() to the top of the main loop so every
iteration pets the WDT, regardless of queue state.

Verified end-to-end against a captured task_wdt coredump from v0.19.0:
asic task was blocked in xQueuePeek at the spin point while stratum
task was mid-vTaskDelay(4000). Cumulative wdt_resets count on the
affected device was 21 historical hits.

No host test added: asic_task is... (continued)

1365 of 1631 branches covered (83.69%)

Branch coverage included in aggregate %.

2373 of 2469 relevant lines covered (96.11%)

27.32 hits per line

Source Files on job webui - 25644454287.3
  • Tree
  • List 66
  • Changed 0
  • Source Changed 0
  • Coverage Changed 0
Coverage ∆ File Lines Relevant Covered Missed Hits/Line Branch Hits Branch Misses
  • Back to Build 25644454287
  • f7975466 on github
  • Prev Job for on main (#25633132593.2)
  • Next Job for on main (#25647217387.2)
STATUS · Troubleshooting · Open an Issue · Sales · Support · CAREERS · ENTERPRISE · START FREE · SCHEDULE DEMO
ANNOUNCEMENTS · TWITTER · TOS & SLA · Supported CI Services · What's a CI service? · Automated Testing

© 2026 Coveralls, Inc