This blog will give a quick run through of the SYCL profiling features that have been developed in the latest version of LPGPU2 CodeXL. LPGPU2 CodeXL is not yet available to the public but it was made available to the LPGPU2 consortium during February 2018. It is the aim to make a version of CodeXL with SYCL profiling features available when the project is completed.
VeriSilicon today announced significant milestones have been achieved for its versatile and highly scalable neural network inference engine family VIP8000. The fully programmable VIP8000 processors reach the performance and memory efficiency of dedicated fixed-function logic with the customizability and future proofing of full programmability in OpenCL, OpenVX, and a wide range of NN frameworks including NNEF. “The biggest thing to happen in the computer industry since the PC is AI and machine learning, it will truly revolutionize, empower, and improve our lives. It can be done in giant machines from IBM and Google, and in tiny chips made with VeriSilicon’s neural network processors,” said Dr. Jon Peddie, president Jon Peddie Research. “By 2020 we will wonder how we ever lived without our AI assistants,” he added.
Khronos member Renesas Electronics has outlined their plans for ADAS and self-driving cars. Renesas is working with Codeplay Software Ltd., experts in high-performance compilers and software optimization for multi-core processing. The collaboration allows programs already written in CUDA for Nvidia’s SoC to be brought to R-Car SoCs, using Codeplay’s OpenCL open standard-based software framework. The framework, first made available on R-Car H3 as a proof of concept, is now coming to the R-Car V3M and other R-Car SoCs of Renesas’ autonomous platform for both ADAS and automated driving.
Codeplay has written up a detailed run through of how they how they ensure C++ fundamental types are translated correctly from SYCL code through to OpenCL, retaining their correct size and signedness.
If you’re an application developer, this will help you learn a little about how SYCL works under the hood. If you’re looking to implement SYCL, this will help you find a way to get the compiler to do your lifting for you.
Imagination Technologies announces the PowerVR CLDNN SDK for developing neural network applications on PowerVR GPUs. The neural network SDK makes it easy for developers to create Convolutional Neural Networks (CNNs) using PowerVR hardware. CLDNN sits on top of OpenCL making use of OpenCL constructs so it can be used alongside other custom OpenCL code. It uses standard OpenCL memory, so it can be used alongside standard OpenGL ES contexts. Learn more about CLDNN and download the SDK today.
The Khronos recently announced SYCL 1.2.1. The SYCL 1.2.1 specification has improved on the existing 1.2 standard by introducing new features which allow for better integration with existing machine learning and OpenCL-based frameworks such as TensorFlow as well as various improvements based on user feedback. This is an exciting piece of news for Codeplay is it enables them to work towards full compliance of SYCL 1.2.1 for ComputeCpp, and their v0.5 release which is now available to download. Read more about how Codeplay is using SYCL 1.2.1 with ComputeCpp v.0.5.0.
Developed by Khronos member Continental in collaboration with Khronos member Xilinx, the Assisted & Automated Driving Control Unit will enable Continental’s customers to get to market faster by building upon the Open Computing Language (OpenCL) framework.
Codeplay has a very good write-up today on machine alternatives that don’t use Neural Networks. The included code, SYCL-ML was developed as a proof of concept to show what a machine learning application using heterogeneous computing can look like and has been published as an open source project. The project was developed using SYCL and ComputeCpp, which is an implementation of SYCL developed by Codeplay.
Portable Computing Language (pocl) 1.0 has been released. One of the bigger highlights of this release is that most of the OpenCL 1.2 standard conformance tests pass with the CPU backend. There are some caveats though to this listed in the documentation. Pocl is a portable open source (MIT-licensed) implementation of the OpenCL standard (1.2 with some 2.0 features supported).
Neil Trevett, Khronos Group President and Radhakrishna Giduthuri, Software Architecture and Compute Performance Acceleration at AMD, spoke at two Khronos related events this past week. Neils presented was an update on the Khronos Standards for Vision and Machine Learning which covered Khronos Standards OpenVX, NNEF, OpenCL, SYCL and Vulkan. Radhakrishna presented Standards for Neural Networks Acceleration and Deployment covered Khronos Standards OpenVX and NNEF. The slides from both presentations are now online.