The environment will first be available on the R-Car H3 as a proof of concept, then shift to the R-Car V3M and other R-Car SoCs on the Renesas Autonomy™ platform for advanced driver assistance systems (ADAS) and automated driving. The new framework is designed to support software development for the R-Car's latest image recognition IP, the IMP-X5, a multi-threaded core optimized for computer vision and cognitive processing. Codeplay will also provide ComputeCpp™ to R-Car, an implementation of the open standard SYCL™, which enables single-source C++ software for high-level, object-oriented programming. The result of this collaboration provides developers with standard software development tools and support for a wide range of open-source vision software or open-source deep learning software, such as the TensorFlow™ library. Developers will be able to focus on innovation while benefiting from the high-performance, low-power advantages of the R-Car SoCs. They can reuse their initial software assets or leverage the latest innovations in computer vision and cognitive processing, minimizing porting effort and accelerating time to market.
Key features of the ComputeAorta and ComputeCpp software environments for R-Car
: • High-performance computing from multi-threat environments.
The ComputeAorta environment provides an OpenCL hardware abstraction layer that allows Tier 1 suppliers and OEMs to leverage the full computing performance of Renesas' R-Car SoCs, especially the IMP-X5. ComputeCpp extends OpenCL with an open standard implementation of SYCL, providing an efficient programming environment for multithreaded execution across heterogeneous cores. By using these environments, users can fully leverage the IMP-X5's computing performance through custom multithreaded processing.
• Faster time to market by reducing development efforts.
Support for OpenCL and SYCL allows developers to use the standard C/C++ programming language, helping them reuse their software assets and work in a familiar development environment, without the additional effort of learning hardware details.
• Leveraging open-source ecosystem frameworks.
Developers and system manufacturers can now benefit from OpenCL and SYCL support with a broad support community for a range of open-source machine vision and cognitive processing software, accelerating the development of complex AI applications.
Renesas and Codeplay will provide a proof-of-concept implementation of OpenCL and SYCL in the R-Car H3 SoC in Q4 2017. The implementation in the R-Car V3M SoC will be available in Q1 2018.
