Ecosystem, Tooling, and Observability¶
The criterion. Writing the code is half the battle; debugging determines the timeline. Look for out-of-the-box trace/profiling integration (SEGGER SystemView, Percepio Tracealyzer) and execution hooks to track CPU load, stack usage, and runtime anomalies.
How Revera answers it¶
Revera has one instrumentation vocabulary at every scheduler / sync event, fanning out to two backends: a portable flight-recorder ring (SMP-safe, dumped over any console after a fault — the “last N events before the crash” black box) and SEGGER SystemView (live timeline over J-Link). Both are zero-cost when off. On top of that: post-mortem fault/panic decode, a ps/top-style snapshot, per-thread CPU accounting and stack high-water, and weak hooks (idle, stall, watchdog) the application overrides.
Feature · how to turn it on · how to use it · where it works¶
Feature |
Turn on with |
Use these APIs |
Where |
|---|---|---|---|
Flight-recorder trace ring |
|
|
all |
SEGGER SystemView |
|
(automatic — same call sites) |
ARM Cortex-M + J-Link |
Post-mortem / panic decode |
— |
fault handlers + trace dump |
all (see examples) |
Thread snapshot (ps/top) |
|
|
all |
Per-thread CPU accounting |
|
|
all |
Stack high-water / overflow |
|
|
all |
Execution hooks |
— |
|
all |
Full signatures and contracts: API reference — Threads / Tooling.
Limits & roadmap¶
Live-timeline integration is SystemView (ARM + J-Link); Percepio Tracealyzer is not integrated (the portable ring is the vendor-neutral alternative).
Build/packaging tooling from the project’s history — the
tkgenboard-config generator and an SDK packaging script — is not in 0.1.0 but is port-forward-ready.
Learn more¶
Run it:
examples/trace·examples/sysview·examples/panic/panic_decode·examples/hardfault·examples/stall_detect.Full API: API reference.