Software ecosystem
Debugging tools
CHERI debugging extends conventional source, register and memory inspection with capability metadata and capability-specific faults.
Why it matters
Good tools turn a hardware exception into a precise explanation of the violated authority assumption.
In brief
CHERI debugging extends conventional source, register and memory inspection with capability metadata and capability-specific faults. Developers need to see validity, address, bounds, permissions and provenance-related behaviour at the point of failure.
Key considerations
1. Compiler diagnostics
Treat pointer conversion, size and bounds warnings as design evidence rather than noise.
2. Debugger
Use CHERI-aware GDB or platform tools to inspect capability registers, memory and fault causes.
3. Tracing
Capture logs, core data and minimal reproductions with the exact binary, ABI and platform revision.
Evidence and caveats
Tool support differs by architecture and release. A debugger that can run a binary may not expose every capability field or fault accurately.
Treat build, test, integration, upstream and maintained support as distinct states. Name the target and last verified date.