RISC-V MCU Debug

Overview

Use this skill to debug RISC-V MCU firmware systematically. Identify the exact core, SoC, debug module, toolchain, memory map, and trap state before changing startup code, linker scripts, flash settings, or machine-mode configuration.

When To Use

Use this skill when:

• The target is a RISC-V MCU or bare-metal RISC-V SoC.
• The user mentions OpenOCD, GDB, JTAG, RISC-V debug module, mstatus, mcause, mepc, mtvec, trap handlers, startup code, or linker scripts.
• Firmware cannot flash, stops before main, traps, hangs, or has no UART/log output.

Do not use this skill for Cortex-M, 8051, embedded Linux application debugging, or general C code review without target evidence.

First Questions

Ask for:

• Exact chip, board, and core if known, such as RV32IMAC, RV32E, RV64, F/D extension, or vendor core name.
• Toolchain: GCC, LLVM, vendor IDE, Zephyr, RT-Thread, or bare metal.
• Debug path: OpenOCD, vendor GDB server, J-Link, WCH-Link, onboard probe, or UART bootloader.
• Current symptom: cannot connect, cannot flash, no boot, trap, hang, no log, or peripheral failure.
• Available artifacts: ELF, MAP, linker script, startup file, OpenOCD config, UART log, trap dump, or GDB transcript.

Workflow

1. Confirm the target profile.

RISC-V MCU behavior depends on XLEN, extensions, privilege mode, vendor CSRs, debug module, and memory map.

1. Classify the failure phase.

Use connect, flash, reset/startup, trap, runtime hang, peripheral bring-up, or RTOS scheduling.

1. Establish a non-destructive debug session.

Prefer halt/reset-halt and symbol loading before flash erase or startup rewrites.

1. Collect trap and reset evidence.

Record pc, sp, mstatus, mcause, mepc, mtval, mtvec, and key memory addresses.

1. Check linker and startup alignment.

Confirm reset address, vector/trap base, RAM origin, stack top, .data copy, .bss zeroing, and main call path.

1. Ask before risky operations.

Mass erase, option bytes, boot mode, security locks, and vendor flash config changes require explicit approval.

OpenOCD And GDB Checks

Useful patterns:

openocd -f interface/<probe>.cfg -f target/<target>.cfg
riscv64-unknown-elf-gdb build/firmware.elf
riscv-none-elf-gdb build/firmware.elf

GDB baseline:

target extended-remote :3333
monitor reset halt
info registers
p/x $pc
p/x $sp
p/x $mcause
p/x $mepc
p/x $mtval
p/x $mtvec
bt

Check:

• pc points to a valid executable region.
• sp points inside RAM and is aligned.
• mtvec points to the intended trap handler.
• mepc maps to source or disassembly.
• OpenOCD target config matches the exact chip or debug module.

Trap Debugging

When a trap occurs:

1. Preserve state before reset.
2. Read mcause, mepc, mtval, mstatus, and mtvec.
3. Decode whether the cause is interrupt or exception.
4. Map mepc to a symbol and instruction.
5. Check whether mtval contains a bad address or instruction.

Common causes:

• Illegal instruction due to wrong ISA flags, such as compiling for rv32imac but running on a core without compressed instructions.
• Misaligned access or unsupported unaligned loads/stores.
• Executing from erased flash or wrong boot address.
• Stack pointer outside RAM.
• Trap vector not initialized or not aligned as required.
• Accessing peripheral addresses before clocks or bus bridges are ready.
• M-mode interrupt enable or PLIC/CLIC setup mismatch.

Startup And Linker Checks

If firmware fails before main():

• Confirm linker script flash/RAM origins match the exact SoC.
• Confirm reset entry address agrees with boot ROM or flash mapping.
• Confirm stack top is valid and aligned.
• Confirm global pointer gp and small data sections are initialized if used.
• Confirm .data and .bss initialization.
• Confirm trap handler is installed before enabling interrupts.
• Confirm compiler ABI, ISA string, and startup assembly agree.

Flashing Checks

If flashing fails:

• Separate connect, erase, program, and verify failures.
• Check probe wiring, target voltage, reset, and JTAG/SWD-equivalent pin mux.
• Lower adapter speed.
• Confirm flash algorithm and target config match the chip.
• Ask before mass erase or security unlock.

Verification

Before claiming progress:

• State chip/core, XLEN, ISA string if known, probe, debug server, and toolchain.
• Confirm whether GDB can halt and read registers.
• Report pc, sp, mcause, mepc, mtval, and mtvec when trap-related.
• Confirm flash erase/program/verify status separately if flashing was attempted.
• List destructive actions skipped or awaiting approval.

Common Failures

• Using the wrong -march or -mabi for the actual core.
• Assuming all RISC-V MCUs have the same interrupt controller.
• Resetting before reading trap CSRs.
• Debugging C code before confirming startup assembly and linker origins.
• Using a near-match OpenOCD target config without checking memory map and flash algorithm.

Example

User:

RISC-V 小板下载后进 trap，串口没有日志。

Agent:

1. Asks for chip, core/ISA, probe, OpenOCD config, ELF, and trap register dump.
2. Connects halt-first and reads pc, sp, mcause, mepc, mtval, and mtvec.
3. Maps mepc to source or disassembly.
4. Checks ISA flags, stack range, trap vector, and linker origins before editing code.