The reference cards for OpenCL 1.1, 1.2, 2.0 and 2.1 have been updated. What changed: Most Preprocessor directives & macros in OpenCL begin with a double-underscore (__), however the “CL_VERSION_X” ones do not. These revised reference cards remove the double underscore from those. The reference cards on Lulu for purchase are also updated.
Qualcomm has introduced the new Qualcomm Snapdragon 675 Mobile Platform. The Snapdragon 675 offers outstanding gaming, a leap in artificial intelligence (AI) capability and a cutting-edge camera. Premium features in the Snapdragon 675 are enabled by the Qualcomm AI Engine, Qualcomm Spectra ISP, Qualcomm Kryo CPU and Qualcomm Adreno GPU. A number of specific games and game engines have been optimized including Unity, Unreal, Messiah, and NeoX. Qualcomm Technologies also supports popular tools and APIs, including Vulkan, OpenGL 3.2, OpenCL, and Snapdragon profiler.
NXP delivers a wide range of processing solutions on which machine-learning (ML) applications can run. Developers will need the associated software and tools to make them work and this is where eIQ framework and development tools come into play. The eIQ framework is designed to work with hardware abstraction layers like OpenCL, OpenVX, and the Arm Compute Library, as well as inference engines like the Arm NN (neural net), Android NN, GLOW, and OpenCV.
Intel’s driver team has published a new graphics driver for all recent GPUs (Intel 6th, 7th and 8th Gen processors) on Windows 10. This new driver (version 188.8.131.5223) brings support for Windows 10 October 2018 Update (version 1809 with WDDM 2.5) and HDR for embedded laptop panels. Hardware support now included up to Vulkan 1.1.83, OpenCL 2.1 and OpenGL 4.5. Complete details are in the Intel Release Notes PDF.
European Union-funded researchers have today released a tool suite which enables developers to deliver longer battery life in mobile devices, while ensuring high quality and performance. The LPGPU2 tool-suite helps programmers develop power-efficient code for GPUs by identifying bottlenecks relating to performance (for example in terms of frames-per-second) and power (for example in terms of energy per instruction). The LPGPU2 tool suite has benefited from the expertise of a range of academic and industrial partners including Khronos members Samsung, who designed and implemented the data collection frameworks and feedback engine; Think Silicon validated it on their four-core NEMA GPU system and Codeplay extended AMD’s CodeXL tool, allowing programmers to profile their SYCL applications. Download the tool suite from the GitHub repository.
Codeplay are pleased to announce that Andrew Richards, CEO & Founder, Charles Macfarlane, VP Marketing & Rod Burns, Developer Relations Manager, will be at this year’s AutoSens in Brussels, exhibiting from 17th to 20th September. Codeplay will be there to discuss the use of open standards such as OpenCL and SYCL on heterogeneous processor solutions to enable the latest vision and machine learning applications for automotive.
Recent work from Boston University has shown that with key optimizations that leverage OpenCL on Arria 10 FPGAs for 3D fast fourier transforms (FFTs), a common HPC workload, the performance can beat out FFT specific IP cores as well as GPU and CPU implementations of the same problem.
Apple has published an official list of all of their devices that use OpenGL and OpenCL, including which version is supported. The earliest machines date back to 2007. The most recent machines to support both OpenCL and OpenGL are the 2018 MacBook Pro 13 and 15 inch. Both machines support OpenGL 4.1 and OpenCL 1.2.
The Khronos Group announces the availability of the SYCL Adopters Program for the C++-based programming framework for parallel programming. Under the Adopters Program, implementers of SYCL 1.2.1 can access an extensive conformance test suite, and then upload their test results to Khronos for review and the opportunity to become officially conformant. Together with the SYCL Adopters Program, Khronos also announces the release of a maintenance update for SYCL 1.2.1, delivering specification clarifications that enable enhanced run-time optimizations.
The OpenCL support by NNVM & TVM session from Linaro Connect 2018 in Hong Kong is now online. Abstract: To use mobile GPU to accelerate deep learning inference on ARM platforms in device side, OpenCL support seems a proper and promising fit. NNVM is an open compiler for AI frameworks with graph IR implementation, and TVM is an open source end-to-end Tensor IR/DSL stack. NNVM together with TVM provides a flexible architecture to support different frameworks and backends. OpenCL is one of the supported backends by NNVM & TVM now, the latest status and some how-tos will be discussed in this session.