Efficient software development with heterogeneous devices
2 likes708 views
The document discusses the advantages of heterogeneous devices in software development, highlighting the ARM architecture's capabilities for diverse computing needs. It addresses software challenges such as managing inter-processor communication and debugging across multiple operating systems while providing insights into solutions and tools for effective development. Key features include the integration of Cortex-A for performance and Cortex-M for real-time control, emphasizing the importance of standardization and productivity in embedded systems.
Efficient software development with heterogeneous devices
- 1. ©ARM 2017 Efficient software development with heterogeneous devices Stefano Cadario Embedded World 2017 – Nuremberg Product Manager – Keil MDK March 16th , 2017
- 2. ©ARM 20172 Agenda 1. Advantages of heterogenous devices 2. How to manage inter-processor communication 3. How tools can simplify development and debug 4. Summary and Q&A
- 3. ©ARM 20173 ARM architecture for diverse computing needs Cortex-A Highest performance Optimized for rich operating systems Cortex-R Fast response Optimized for high performance, hard real-time applications Cortex-M Smallest/lowest power Optimized for discrete processing and microcontrollers
- 4. ©ARM 20164 Thanks for reading For more on heterogeneous development visit arm.com Sign-up for the latest news and information from ARM
- 5. ©ARM 20175 Modern compute systems have diverse workloads Power Time Sleep mode Interactive mode Ambient mode
- 6. ©ARM 20176 Why heterogeneous devices? Cortex-A benefits § Feature rich OS § Complex applications § Sophisticated HMI Cortex-M benefits § Low I/O latency § Low power standby § Fast system start-up time Linux application RTOS system Inter-processor communication Combines best of both worlds: feature-rich Linux and MCU with deterministic RTOS
- 7. ©ARM 20177 Use cases of HMP systems in embedded Cortex-A Rich UI and OS, high performance Cortex-M Real-time control and monitoring Deterministic sensor control Real-time monitoring MedicalConsumerIndustrial
- 8. ©ARM 20178 Overview of software challenges Developer productivity Usability, portability, debugging Data sharing Is coherency necessary? Task partitioning How to optimally partition tasks?
- 9. ©ARM 20179 Overview of software challenges Developer productivity Usability, portability, debugging Data sharing Is coherency necessary? Task partitioning How to optimally partition tasks?
- 10. ©ARM 201710 Software development challenges § Running multiple operating systems § Debugging bare-metal code and Linux applications at the same time § Controlling communication between Linux application and bare-metal application Linux application MCU application Common peripherals Shared memory ARM Cortex-A7 Inter-processor communication ARM Cortex-M4 Linux OS Complexity increases with heterogeneous devices
- 11. ©ARM 201711 Software development challenges § Requires to write a Linux kernel module § Memory segmentation § Manage concurrency § Kernel mode to User mode communication § Bare-metal § Manage concurrency § Integration with RTOS § Handle interrupts and memory management Shared memory ARM Cortex-A7 ARM Cortex-M4 Linux OS Kernel module Linux application MCU application Manually manage communication between Cortex-A and Cortex-M Common peripherals IPC
- 12. ©ARM 201712 Software development challenges § CMSIS adopting OpenAMP § CMSIS - Cortex Microcontroller Software Interface Standard § Now open source on Github § OS support for HMP systems § Remote Processor Messaging (RPMsg) for inter- processor communication § Management framework using remoteproc Standardization of software interfaces Shared memory ARM Cortex-A7 ARM Cortex-M4 Linux OS RTX5 RPMsg module Linux application MCU application Common peripherals RPMsg library
- 13. ©ARM 201713 DS-MDK for heterogeneous systems Software development solution for Cortex-A and Cortex-MMDKToolsSoftwarePacks MDK-Core ARM C/C++ Compiler DS-MDK Device µVision IDE with Pack Management ARM Compiler 5 with Qualification Kit µVision Debugger with Streaming Trace Startup Device HAL CMSIS Driver CMSIS CMSIS-CORE CMSIS-DSP CMSIS-RTOS DS-5 IDE with Pack Management DS-5 Debugger with Streamline Middleware USB Host File System Graphics USB Device IPv6 NetworkIPv4 Network ARM Compiler 6 LLVMTechnology mbed TLS Encryption mbed Client IoT Connector Eclipse CDT based with CMSIS management provides access to documentation, CMSIS framework and software components. Software Packs for NXP's i.MX6 and i.MX7 are available. Support for other devices in the future.
- 14. ©ARM 201714 Multiple simultaneous debug connections Cortex-A Cortex-M RTOS systemLinux kernel Linux application JTAG TCP/IP CoreSight Microcontroller application Debug ü The Cortex-M application via ULINKpro and analyse with RTOS awareness and peripheral views. ü The Cortex-A Linux kernel and Linux kernel modules via ULINKpro and list kernel threads and processes. ü The Cortex-A Linux application via gdbserver on the running Linux system using Remote System Explorer. Complete visibility to all software applications in the heterogeneous system
- 15. ©ARM 201715 Efficient software development on HMP Combines best of both worlds Feature-rich Linux and MCU with deterministic RTOS on same chip Standardization of software interface Speed-up your development with CMSIS Multicore debug support Simplifies software development on heterogeneous devices
- 16. The trademarks featured in this presentation are registered and/or unregistered trademarks of ARM Limited (or its subsidiaries) in the EU and/or elsewhere. All rights reserved. All other marks featured may be trademarks of their respective owners. Copyright © 2016 ARM Limited ©ARM 2016