Khronos Finalizes OpenCL 2.0 Specification for Heterogeneous Computing

  Khronos Finalizes OpenCL 2.0 Specification for Heterogeneous Computing

   Industry feedback drives new generation open standard for cross-platform
parallel programming with increased flexibility, functionality and performance

SC13

Business Wire

DENVER -- November 18, 2013

The Khronos™ Group today announced the ratification and public release of the
finalized OpenCL™ 2.0 specification. OpenCL 2.0 is a significant evolution of
the open, royalty-free standard that simplifies cross-platform, parallel
programming. With an enhanced execution model and a subset of the C11 and
C++11 memory model, synchronization and atomic operations, OpenCL now enables
a significantly richer range of algorithms and programming patterns to be
easily accelerated with improved performance. Significant feedback from the
developer community was incorporated into the final specification, following
its provisional release in July. The OpenCL 2.0 specifications are available
at www.khronos.org/opencl/.

“Khronos received significant and thoughtful developer feedback from the
provisional release of OpenCL 2.0, much of which has been adopted, or will be
merged with emerging hardware capabilities as this state-of–the-art parallel
programming platform continues to evolve,” said Neil Trevett, chair of the
OpenCL working group, president of the Khronos Group and vice president of
mobile content at NVIDIA. “OpenCL continues to gather momentum on desktop,
mobile and embedded devices, including providing a unified programming
environment for dynamically balancing diverse CPU, GPU, DSP and hardware
resources in mobile SOCs for advanced use cases ranging from vision processing
for Augmented Reality to physics simulation for mobile gaming.”

OpenCL 2.0 updates and additions include:

Shared Virtual Memory

Host and device kernels can directly share complex, pointer-containing data
structures such as trees and linked lists, providing significant programming
flexibility and eliminating costly data transfers between host and devices.

Nested Parallelism

Device kernels can enqueue kernels to the same device with no host
interaction, enabling flexible work scheduling paradigms and avoiding the need
to transfer execution control and data between the device and host, often
significantly offloading host processor bottlenecks.

Generic Address Space

Functions can be written without specifying a named address space for
arguments, especially useful for those arguments that are declared to be a
pointer to a type, eliminating the need for multiple functions to be written
for each named address space used in an application.

Images

Improved image support including sRGB images and 3D image writes, the ability
for kernels to read from and write to the same image, and the creation of
OpenCL images from a mip-mapped or a multi-sampled OpenGL^® texture for
improved OpenGL interop.

C11 Atomics

A subset of C11 atomics and synchronization operations to enable assignments
in one work-item to be visible to other work-items in a work-group, across
work-groups executing on a device or for sharing data between the OpenCL
device and host.

Pipes

Pipes are memory objects that store data organized as a FIFO and OpenCL 2.0
provides built-in functions for kernels to read from or write to a pipe,
providing straightforward programming of pipe data structures that can be
highly optimized by OpenCL implementers.

Android Installable Client Driver Extension

Enables OpenCL implementations to be discovered and loaded as a shared object
on Android systems.

Industry Support

“Premiere Pro's support for OpenCL has proved to be a massive hit with our
customers; providing dramatic performance improvements whileallowing for
real-time editing and creativity. We're excited about the technological
developments in OpenCL 2.0 and look forward to discovering howit will enable
us to push the performance envelope even further,” said Al Mooney, senior
product manager, editing workflows at Adobe.

“The Khronos Group’s OpenCL 2.0 is the first key, foundational, programming
language to truly support the core capabilities of HSA enabled hardware. It is
going to be exciting to see where developers take this much richer programming
platform,” said Gregory Stoner, managing director and vice president of HSA
Foundation.

“It is impressive that OpenCL is supporting an increasingly diverse range of
heterogeneous computing units and accelerators,” said Zhenya Li,vice
president of 2012 Lab, Huawei Technologies. “We expect the OpenCL standard to
be widely adopted by the information and communications technology (ICT)
sector, and to be a key software standard used in Network Function
Virtualization (NFV) accelerators. Huawei will actively participate in and
contribute to OpenCL, and help it to provide an easy-to-use development
platform for future ICT virtualized applications.”

“As a long-time member of Khronos and a leading contributor to OpenCL
standards efforts, Imagination is delighted that Khronos continues to create
standards which make GPU compute programming easier for developers. With our
broad range of IP including PowerVR processors and MIPS CPUs, our customers
are creating innovative designs for mobile, consumer, automotive and more. GPU
compute is key to creating new applications within the power envelope of these
next generation devices,” said Peter McGuinness, director of multimedia
technology marketing, Imagination Technologies.

“We are very excited about the user benefits of OpenCL 2.0’s new features,”
said Simon McIntosh-Smith, Head of theMicroelectronics Research Group at the
University of Bristol. “These latest evolutions in OpenCL will enable us
toefficiently solve a much wider range of parallel processing problems than
ever before, and across a growing range ofembedded and HPC hardware
platforms. The new shared virtual memory (SVM) feature will make it easier for
programmers to develop heterogeneous parallel programs, while support for
dynamic parallelism will enable more efficient solutions for a much wider
range of applications.”

“Vivante is pleased to support and contribute to the OpenCL 2.0 specification.
The latest industry standard will expand adoption of the technology in mobile,
home entertainment, and automotive products, creating the next wave of
innovative compute use cases and consumer experiences,” said Wei-Jin Dai,
President and CEO of Vivante. “As the OpenCL ecosystem rapidly grows and more
applications come to market, SoCs using our latest Vega GPUs will be ready to
support the latest 2.0 specification and take advantage of platform level
optimizations built into our architecture.”

See OpenCL at SC13, Denver, CO

Visit the Khronos Booth #4137 to meet with OpenCL experts and get a free
OpenCL reference card.

OpenCL: A Hands-On Introduction

Monday, November 18^th, 8:30 - 17:00, Room 403
Tim Mattson, Alice Koniges, Simon McIntosh-Smith

OpenCL BOF: Version 2.0 and Beyond

Including announcement of IWOCL 2014 (2nd International Workshop on OpenCL)
Wednesday, November 20^th, 17:30 - 19:00, Room 405/406/407
Tim Mattson, Ben Bergen, Simon McIntosh-Smith

Exhibitor Forum OpenCL 2.0: Unlocking the Power of Your Heterogeneous Platform

Thursday, November 21^st, 11:30 - 12:00, Room 501/502, Tim Mattson

Structured Parallel Programming with Patterns

Sunday, Nov. 17, 8:30am-5pm, Room 302
Michael McCool, James Reinders, Arch Robison, Michael Hebenstreit

See OpenCL at SIGGRAPH Asia, Hong Kong

Visit the Khronos Booth #F07 to meet with OpenCL experts and get a free OpenCL
reference card.

Khronos DevU, Wednesday November 20^th, Room S226

13:30-13:45   Neil Trevett, NVIDIA     Introduction to OpenCL
13:45-14:30     Allen Hux, Intel            OpenCL 2.0 Overview
14:30-14:50     Tomasz Bednarz, CSIRO       Accelerated Science – use of
                                            OpenCL in Land Down Under

OpenCL 2.0 Reference Cards

Laminated OpenCL 2.0 Reference cards are now available at www.amazon.com
(search for ‘Khronos reference card’).

About The Khronos Group

The Khronos Group is an industry consortium creating open standards to enable
the authoring and acceleration of parallel computing, graphics, vision, sensor
processing and dynamic media on a wide variety of platforms and devices.
Khronos standards include OpenGL^®, OpenGL^® ES, WebGL™, OpenCL™, WebCL™,
OpenVX™, OpenMAX™, OpenVG™, OpenSL ES™, StreamInput™ and COLLADA™. All Khronos
members are enabled to contribute to the development of Khronos
specifications, are empowered to vote at various stages before public
deployment, and are able to accelerate the delivery of their cutting-edge
media platforms and applications through early access to specification drafts
and conformance tests. More information is available at www.khronos.org.

Khronos, DevU, StreamInput, WebGL, WebCL, COLLADA, OpenKODE, OpenVG, OpenVX,
OpenSL ES and OpenMAX are trademarks of the Khronos Group Inc. ASTC is a
trademark of ARM Holdings PLC, OpenCL is a trademark of Apple Inc. and OpenGL
is a registered trademark and the OpenGL ES and OpenGL SC logos are trademarks
of Silicon Graphics International used under license by Khronos. All other
product names, trademarks, and/or company names are used solely for
identification and belong to their respective owners.

Contact:

Horizon PR
Jonathan Hirshon, Principal, +1-415-952-3001
jh@horizonpr.com
 
Press spacebar to pause and continue. Press esc to stop.