STM32G070 Enters Hard Halt Mode During Protobuf Decoding - Seeking Assistance

[Pedro Kuusberg]

Hello NanoPB Community,

I'm dealing with an issue concerning protobuf decoding using NanoPB and am hoping to gain insights from this knowledgeable community.

Background Information:

• Microcontroller: STM32G070
• NanoPB Version: 3.14
• RTOS: CMSIS-RTOS2
• Processor: ARM Cortex M0+

Setup:

• Task 1: Engages in the creation and encoding of protobuf messages.
• Task 2: Designed to receive and decode messages.

Issue: The microcontroller exhibits a hard halt during decoding in Task 2, noticeably with the specified message:

• Message Example:
  • Status: 1
  • Length: 16
  • Content: 0x08 0x01 0x10 0x02 0x18 0x00 0x22 0x08 0x0A 0x06 0x0A 0x04 0x08 0x01 0x20 0x03
• Stack Size: 380

Observations:

• Task 1 proceeds without issues in message creation and encoding.
• The hard halt arises during decoding in Task 2.

Seeking Insights:

• Have similar hard halts been experienced by others using STM32G0 series microcontrollers during NanoPB decoding?
• Are there identifiable pitfalls in decoding messages within an RTOS environment?
• Can specific debugging tools or strategies be recommended for this scenario?

I truly appreciate any guidance or shared experiences that you might provide to assist in resolving this matter.

Warm regards, Pedro Kuusberg

[Petteri Aimonen]

Hi,

My first guess would be a stack overflow.
You could try increasing task 2 stack size to test whether that makes a difference.

Other than that, basic hardfault debugging should lead you to the cause.
First check backtrace and values after the crash to see if that gives any clues.
After that, set a breakpoint somewhere before the crash point and step instruction-by-instruction
until the crash, to see what is the faulting instruction (stepi and display /i $pc in gdb).

--
Petteri