Hi, I’m the Toradex AI Assistant.
I might make mistakes or occasionally get you wrong. Our support team will step in if needed.
I’d be glad to help with your CAN bus configuration for the Cortex-M7! To provide the most accurate guidance for your specific setup, could you let me know which OS and version you are running on your Verdin iMX8MP?
You can use the tdx-info command to get detailed information about your system configuration. For more details on this command, see: Getting Device Information with Tdx-Info | Toradex Developer Center
Key Information Needed:
- Operating system (Torizon OS, BSP, etc.) and version
- Which framework you’re planning to use for Cortex-M7 development (FreeRTOS, Zephyr, bare metal, etc.)
General Approach Overview
Based on the available documentation, enabling CAN1 on the Cortex-M7 involves several components:
1. Device Tree Configuration
Your approach to disable CAN1 on the Linux side is correct (1). For Heterogeneous Multicore Processing setups, Toradex provides device tree overlays to help with out-of-the-box experience (2).
2. RemoteProc Framework
For NXP i.MX platforms like your iMX8MP, the RemoteProc framework is used to manage the Cortex-M7 core (2). This framework allows you to:
- Load and execute firmware on the Cortex-M7
- Control the remote processor from the Linux side
- Handle inter-core communication
3. Resource Domain Controller (RDC)
Shared peripherals like CAN1 require proper RDC configuration to permit access by the M7 core (3). This ensures that the Linux (A53) side doesn’t conflict with M7 usage of the peripheral.
Once you provide your OS version and preferred development framework, I can give you more specific guidance on:
- The exact device tree overlay configuration needed
- How to set up the RemoteProc framework for your use case
- Firmware development considerations for CAN1 on Cortex-M7
- Resource sharing configuration between the cores
Best regards, Toradex AI Assistant.
If I answered your question, please tick the “Solution” checkbox.
For follow-up questions to me, please use @ToradexAI in your post.