In a white paper released during a semiconductor design conference last week in Silicon Valley, researchers from the University of Wisconsin-Madison outlined the architecture of the open source GPU dubbed MIAOW, or Many-core Integrated Accelerator of Wisconsin. The prototype was essentially designed to demonstrate an open source GPGPU compatible with OpenCL. The researchers said their goal was to emulate a full system, not to compete with commercial designs.
The PowerVR Imaging Framework for Android comprises a set of extensions to the OpenCL and EGL Application Programming Interfaces (APIs) that enable efficient interoperability of software running on PowerVR GPUs with other components such as a CPU, ISP and VDE. These extensions enable the construction of shared memory allocations and software pipelines across multiple hardware components with no redundant memory copies (termed zero-copy). The framework is integrated at the library layer of the Android software stack, enabling efficient interoperability between APIs such as OpenCL, OpenGL ES and emerging APIs such as OpenVX.
The Khronos Group has posted the complete slide decks from SIGGRAPH 2015 BOFs online. Included in the slides are the many presentations from WebGL, along with slides from the 3D, OpenCL and OpenVX BOF.
AMD announced the release of APP SDK 3.0 supporting OpenCL 2.0. AMD APP SDK 3.0 adds support for Windows 10 as well as AMD’s latest 6th generation AMD A-series processors, Radeon R9 series graphics cards and FirePro W8100 and W9100 series graphics cards. AMD APP SDK 3.0 contains a complete set of sample code illustrating how to utilize each of the major new features of OpenCL 2.0. Some of these features are highlighted in the OpenCL 2.0 Demystified blog-series.
This Imagination Technologies article and a follow-up to be published next month introduce OpenCL programming for the PowerVR Rogue architecture. Starting with an overview of OpenCL programming fundamentals using a basic program, followed by an explanation of OpenCL execution on Rogue GPUs. This provides the background to understand the programming guidelines for the Rogue architecture which are illustrated by using a case study of an image filtering program.
GPU-STREAM allows you to measure memory transfer rates to and from global device memory on GPUs. This benchmark is similar in spirit, and based on, the STREAM benchmark for CPUs. Unlike other GPU memory bandwidth benchmarks this does not include the PCIe transfer time.
Advanced Simulation Library (ASL) is a free and open source multiphysics simulation software package. Its computational engine is based, among others, on the Lattice Boltzmann Methods and is written in OpenCL which enable extraordinarily efficient deployment on a variety of massively parallel architectures, ranging from inexpensive FPGAs, DSPs and GPUs up to heterogeneous clusters and supercomputers. ASL is distributed under the free GNU Affero General Public License (AGPL) with an optional commercial license.
The Khronos Group BOF Blitz is live streaming all BOFs. Starting with "Accelerating Vision Processing" at 1PM PT followed by "Heterogeneous Computing" at 2PM and "3D Web Graphics" at 3:45PM. Finally, at 5:30PM there will be the "3D Graphics API State of the Union" BOF which will include Vulkan:
MediaTek offers their drivers to all their customers. All phones using their latest chips have OpenCL support by default.
The Khronos Group SIGGRAPH schedule has been posted online. This year Khronos brings four BOFs to the BOF Blitz, a Chapter meetup and a course a party! Khronos is celebrating it's 15th Anniversary this year, and what a year it's turning out to be. BOFs include OpenVX & OpenCL, OpenCL, SYCL & SPIR-V, WebGL and glTF and Vulkan, OpenGL, OpenGL ES.