The automotive landscape is undergoing a dramatic transformation, driven by the relentless pursuit of advanced driver-assistance systems (ADAS) and autonomous driving technologies. These innovations demand high-performance computing platforms capable of handling complex, data-intensive workloads with minimal latency and maximum efficiency. Traditional computing architectures, based on general-purpose processors, are increasingly unable to meet these growing demands for processing power and flexibility. The need for specialized processing capabilities tailored to specific tasks has led to the emergence of heterogeneous computing platforms, which integrate diverse processing elements.
Versal AI Edge Series Gen 2 and Versal Prime Series Gen 2 adaptive SoCs represent a leading solution to this challenge. These platforms combine adaptable programmable logic with high-performance processing systems, including multicore application processing units (APUs), advanced memory hierarchies, and AI Engines for machine learning. A pivotal component of these SoCs is the Arm Mali-G78AE GPU, specifically designed for demanding automotive applications.
The Mali-G78AE is Arm's latest 64-bit architecture GPU from the Automotive Enhanced line of IP. It meets ISO 26262 and IEC 61508 standards, making it suitable for applications requiring ASIL-B/SIL 2 safety levels. This GPU is engineered to handle both graphical and computational tasks with high efficiency. Its flexible partitioning capability allows the GPU to be divided into multiple, independent virtual GPUs, facilitating the concurrent execution of diverse workloads. This feature is particularly advantageous in automotive scenarios where multiple displays and computational tasks, such as ADAS functionality and 3D surround view systems, must operate simultaneously without interference. The Mali-G78AE features four shader cores configured in a two slice, two cores per slice arrangement, supporting one or two partitions. Additionally, it supports hardware virtualization, crucial for functional safety, with flexible core partitioning.
The integration of programmable logic with the GPU on the same SoC opens up new possibilities for advanced automotive use cases. Programmable logic can offload and accelerate specific tasks, such as sensor data processing and real-time decision-making algorithms, which are vital in autonomous driving. The synergy between the programmable logic and the GPU enables a highly flexible and powerful computing platform that can adapt to the evolving needs of automotive systems. This white paper delves into the capabilities of the Arm Mali-G78AE GPU within the Versal AI Edge Series Gen 2 and Versal Prime Series Gen 2 adaptive SoCs, exploring its architecture, functionality, and the synergistic advantages of this integration. It also outlines best practices for graphics rendering in tile-based rendering architectures and how to use efficient rendering passes in the rendering pipeline.
The system integrates a powerful suite of processing units for comprehensive visual and computational workloads. The Arm Mali-G78AE GPU handles high-performance graphics rendering and compute acceleration, while next-generation AI Engines (AIE-ML v2 tiles) provide dedicated hardware for AI inference. Complementing these are hardened video processing components, including image signal processors (ISPs) for capturing and preprocessing image data and video codec units (VCUs) for encoding and decoding video streams. This potent combination of processors enables a wide range of applications, from advanced driver-assistance systems and immersive in-cabin experiences to efficient data aggregation and analysis in real-time.