Press Briefing. GDC, March Neil Trevett Vice President Mobile Ecosystem, NVIDIA President Khronos. Copyright Khronos Group Page 1

Transcription

1 Copyright Khronos Group Page 1 Press Briefing GDC, March 2014 Neil Trevett Vice President Mobile Ecosystem, NVIDIA President Khronos

2 Copyright Khronos Group Page 2 Lots of Khronos News at GDC! OpenGL ES 3.1 Released - Compute shaders & enhanced rendering coming to over a billion mobile devices OpenGL Momentum - Developer tools and insights into low-overhead drivers EGL 1.5 Released - Enhanced rendering, interop and system portability OpenCL 2.0 Adopters Program Released - Full OpenCL 2.0 conformance tests available WebCL 1.0 Released - Web developers to get access to heterogeneous parallel computing SYCL 1.2 Provisional Released - Enabling high-level, C++ frameworks over OpenCL

3 Copyright Khronos Group Page 3 Speakers This Morning Neil Trevett - Vice President Mobile Ecosystem, NVIDIA - President, Khronos - Chair, OpenCL Working Group Tom Olson - Director of Graphics Research at ARM Media Processing Division - Chair, OpenGL ES Working Group Andrew Richards - CEO, Codeplay - Chair, SYCL Working group Jon Peddie - Founder and President, Jon Peddie Research David Cole - Founder and CEO, DFC Intelligence

4 Copyright Khronos Group Page 4 Khronos Connects Software to Silicon Open Consortium creating ROYALTY-FREE, OPEN STANDARD APIs for hardware acceleration Defining the roadmap for low-level silicon interfaces needed on every platform Graphics, compute, rich media, vision, sensor and camera processing Rigorous specifications AND conformance tests for crossvendor portability Acceleration APIs BY the Industry FOR the Industry Well over a BILLION people use Khronos APIs Every Day

5 Copyright Khronos Group Page 5 Khronos Standards 3D Asset Handling - 3D authoring asset interchange - 3D asset transmission format with compression Visual Computing - 3D Graphics - Heterogeneous Parallel Computing Camera Control API Sensor Processing - Vision Acceleration - Camera Control - Sensor Fusion Over 100 companies defining royalty-free APIs to connect software to silicon Acceleration in HTML5-3D in browser no Plug-in - Heterogeneous computing for JavaScript

6 Copyright Khronos Group Page 6 OpenGL ES 3.1 Tom Olson, Director of Graphics Research at ARM Media Processing Division Chair, OpenGL ES Working Group

7 OpenGL ES 3.1 Launched at GDC! OpenGL ES is used by over a BILLION users every day - Used in almost every mobile device phones, tablets, embedded, more Introducing OpenGL ES 3.1 at GDC! - A significant upgrade in functionality coming for a huge number of users Expecting rapid adoption - A driver upgrade for many SOCs - Backward compatible with 2.0/3.0 so apps can incrementally adopt features 2002 Working Group Formed Driver Update Silicon Update Silicon Update Driver Update Copyright Khronos Group Page 7

8 Copyright Khronos Group Page 8 OpenGL ES 3.1 Goals Bringing developer requested features from desktop OpenGL 4 to mobile - Advanced features, modern programming styles - Higher performance with lower overhead Headline features - Compute Shaders and Draw-Indirect - Compute shaders can create geometry or other rendering data - and also the draw commands needed to render them - Offload work from CPU to GPU critical for mobile perf and power Run on OpenGL ES 3.0 hardware expose hidden capabilities of shipping devices - Enable very rapid adoption across the industry Better looking, faster performing apps!

9 Copyright Khronos Group Page 9 Key Working Group Status and Participants API and shading language specifications ratified and released - Manual pages also available Conformance test is code complete - Expect to be accepting conformance submission within three months Widespread industry participation - Tool and Game Engine Developers - GPU Designers - SoC Vendors - Platform Owners - End Equipment Makers - Middleware ISVs Apple

10 Copyright Khronos Group Page 10 Other OpenGL ES 3.1 Key Features Separate shader objects - Vertex and fragment shaders treated as separate programs - Applications can mix and match shaders with compatible interfaces - Supports popular console / PC programming styles Texture gather - Read texture samples from a 2x2 pixel block - Supports fast PCF shadow filtering, other applications New texture types - Multi-sampled textures, packed depth/stencil textures Language features - Multidimensional arrays, bitfield operations, synchronization primitives

11 Copyright Khronos Group Page 11 OpenGL ES 3.1 Optional Extensions Useful functionality that may be headed to future core releases Image atomics - Special shading language functions for atomic operations on images Multi-sampled array textures Advanced blending modes - Supports compositing APIs found in UI, web standards, and 2D art pipelines Sample shading - Fragment shader can run per-sample in multi-sampled rendering Stand-alone 8-bit stencil textures

12 Copyright Khronos Group Page 12 OpenGL ES 3.1 Desktop Compatibility ARB_ES_3_1_compatibility specification - Under development Will enable desktop drivers to be used for mobile development - OpenGL 4.4 drivers will be able to support OpenGL ES 3.1 context Adds features missing in OpenGL - New function MemoryBarrierByRegion() - Raise minimum SSBO size to 128 MB - Support for GLSL ES version ImageAtomicExchange() - Extend mix() to int, uint and bool - gl_helperinvocation - gl_maxsamples - Adds several gl_max*imageuniforms builtins - gl_maxcombinedshaderoutputresources

13 Copyright Khronos Group Page 13 OpenGL Ecosystem News Valve s VOGL OpenGL capture / playback debugger - OpenGL 3.3, OpenGL 4 in progress - Now on github! Valve s ToGL - Subset of Direct3D 9.0c -> OpenGL - API and DX bytecode - On github SIGGRAPH course Introduction to OpenGL programming - Free on youtube OpenTK updated to OpenGL Low-level C# library that wraps OpenGL and more

14 Copyright Khronos Group Page 14 OpenGL and Reducing Driver Overhead Because driver overhead == cost Costs - CPU cycles from app - CPU cache from app - Power / battery - GPU throughput Can we drive driver overhead to ZERO? - In OpenGL!

15 Copyright Khronos Group Page 15 OpenGL Fallacy: Old and Inefficient Immediate Mode Fixed Function Display Lists Evaluators Feedback Ancient crufty stuff Selectors Selection

16 Copyright Khronos Group Page 16 OpenGL Reality: Modern & Efficient Bindless ARB Multi-Draw Indirect GL4.3 Texture Arrays GL3.0 Buffer Storage GL4.4 SSBO GL4.3 UBO GL3.1

17 Copyright Khronos Group Page 17 Plus, OpenGL has all the features Compute Geometry Shaders Sparse Textures Tessellation Image Load/Store

18 Copyright Khronos Group Page 18 Classic OpenGL Model Memory indirect draw buffer object CPU cmd cmd cmd cmd GPU buffer object texture object buffer object buffer object texture object buffer object buffer object buffer object Direct Drawing Commands (via the command fifo) render target buffer object

19 Efficient OpenGL Model CPU CPU CPU CPU Memory indirect draw buffer object indirect draw buffer object texture object buffer object indirect draw buffer object texture object buffer object buffer object buffer object Memory access mediated through OpenGL fences GPU CPU Writes Memory multi-threaded (no API)! render target buffer object GPU Writes Commands to Memory Reads Commands from Memory No API Minimal CPU Involvement Copyright Khronos Group Page 19

20 Copyright Khronos Group Page 20 Results OpenGL enables scalable multi-threading with no new API - CPU and GPU Cores just write to memory - GPU work creation - builds buffers, constructs MDI commands Integer multiple speedups ~5x ~15x (not a typo) - On driver limited cases, obviously Works TODAY on existing drivers! - Mostly OpenGL Extensions are at least EXT Does not require a new object model - Does not require breaking existing applications Approaching Zero Driver Overhead in OpenGL NVIDIA, AMD, Intel presenting Thursday 1PM, Room 2004, West Hall

21 Copyright Khronos Group Page 21 EGL 1.5 Released EGL 1.5 brings functionality from multiple extensions into core - Increased reliability and portability EGLImages - Sharing textures and renderbuffers Context Robustness - Defending against malicious code EGLSync objects - Improved OpenGL /OpenCL interop Platform extensions - Standardized interactions for multiple OS e.g. Android and 64-bit platforms srgb colorspace rendering API Interop EGL provides efficient transfer of data and events between Khronos APIs Applications OS and Display Platforms Application Portability EGL abstracts graphics context management, surface and buffer binding and rendering synchronization

22 Copyright Khronos Group Page 22 OpenCL: Portable Heterogeneous Computing Portable Heterogeneous programming of diverse compute resources - Targeting supercomputers -> embedded systems -> mobile devices One code tree can be executed on CPUs, GPUs, DSPs and hardware - Dynamically interrogate system load and balance work across available processors OpenCL = Two APIs and C-based Kernel language - Kernel language - Subset of ISO C99 + language extensions C Platform API To query, select and initialize compute devices GPU DSP HW OpenCL Kernel OpenCL Code Kernel OpenCL Code Kernel OpenCL C Code Kernel Code CPU CPU C Runtime API To build and execute kernels across multiple devices

23 Copyright Khronos Group Page 23 OpenCL 2.0 Adopters Program Launched Official conformance test suite for the OpenCL 2.0 specification - Implementers can certify that their implementations are officially conformant Released a set of header files for OpenCL Available on Updated OpenCL 2.0 specification - Clarifications and corrections to the specification first released in November 2013 First conformant implementations of OpenCL 2.0 expected to be available to developers in the first half of 2014

24 Copyright Khronos Group Page 24 WebCL: Heterogeneous Computing for the Web WebCL 1.0 specification officially finalized today at GDC! - WebCL defines JavaScript binding to the OpenCL APIs - Enables initiation of Kernels written in OpenCL C within the browser Typical Use Cases - 3D asset codecs, video codecs and processing, imaging and vision processing - Physics for WebGL games, Online data visualization, Augmented Reality JavaScript Platform API To query, select and initialize compute devices GPU DSP HW OpenCL Kernel OpenCL Code Kernel OpenCL Code Kernel OpenCL C Code Kernel Code CPU CPU JavaScript Runtime API To build and execute kernels across multiple devices

25 Copyright Khronos Group Page 25 WebGL/WebCL Ecosystem Content downloaded from the Web Middleware can make WebGL and WebCL accessible to non-expert programmers E.g. three.js library: used by majority of WebGL content Content JavaScript, HTML, CSS,... JavaScript Middleware Low-level APIs provide a powerful foundation for a rich JavaScript middleware ecosystem Browser provides WebGL and WebCL Alongside other HTML5 technologies No plug-in required JavaScript HTML5 / CSS OS Provided Drivers WebGL uses OpenGL ES 2.0 or Angle for OpenGL ES 2.0 over DX9 WebCL uses OpenCL 1.X

26 Copyright Khronos Group Page 26 WebCL Architectural Security Designed-in Leverages OpenCL 1.2 robustness/security extensions - Context Termination: to prevent DoS from long running kernels - Memory Initialization: no leakage from out of bounds memory access WebCL Kernel Validator - Open source - provided as a library API for easy integration into browsers - Parses and validates kernel code against specification - Initializes local/private memory if underlying OpenCL implementation does not - Tracks memory allocations and traces valid ranges for reads and writes - Run time checks to make all memory accesses safe API and Language Restrictions - Not supported: structures as Kernel arguments, Kernel names>256 characters, mapping of CL memory objects into host memory, program binaries, some OpenCL API calls and built-ins

27 Open Source Implementations and Resources WebCL Conformance Framework and Test Suite (contributed by Samsung) - Nokia - Firefox build with integrated WebCL - Firefox extension, open sourced May 2011 (Mozilla Public License 2.0) - Samsung - uses WebKit, open sourced June 2011 (BSD) - Motorola Mobility - uses Node.js, open sourced April 2012 (BSD) - AMD uses Chromium (open source) - Based on Apple QJulia Based on Iñigo Quilez, Shader Toy Based on Iñigo Quilez, Shader Toy Copyright Khronos Group Page 27

28 Copyright Khronos Group Page 28 WebCL Parallel Computing for Web Acceleration

29 Copyright Khronos Group Page 29 OpenCL as Parallel Compute Foundation 100+ tool chains and languages leveraging OpenCL - Heterogeneous solutions emerging for the most popular programming languages C++ AMP Shevlin Park Uses Clang and LLVM Halide Image Processing Language WebCL JavaScript binding to OpenCL Aparapi Java language extensions for parallelism River Trail Language extensions to JavaScript PyOpenCL Python wrapper around OpenCL Harlan High level language for GPU programming Compiler directives for Fortran C and C++ OpenCL provides vendor optimized, cross-platform, cross-vendor access to heterogeneous compute resources Device X Device Y Device Z

30 Widening OpenCL Ecosystem OpenCL C Kernel Source Alternative Alternative Alternative Language for Language for Language Kernels for Kernels Kernels High-level High-level Apps Frameworks and Frameworks Frameworks SPIR Generator (e.g. patched Clang) SPIR Standard Portable Intermediate Representation SPIR 1.2 Released January 2014 OpenCL run-time can consume SPIR OpenCL C Runtime SYCL Programming abstraction that combines portability and efficiency of OpenCL with ease of use and flexibility of C++ SPIR 1.2 Released here at GDC! Device X Device Y Device Z Copyright Khronos Group Page 30

31 Copyright Khronos Group Page 31 SPIR and LLVM SPIR Portable non-source encoding for OpenCL 1.2 device programs LLVM is an optimizing compiler toolkit - Open source platform for innovation that is portable, flexible, well understood - SPIR based on LLVM 3.2 with open consultation with LLVM community Consumption API for target hardware - cl_khr_spir extension to OpenCL runtime API Example SPIR generator - Open source patch to Clang translates OpenCL C to SPIR IR - If you can do it in OpenCL C You can do it in SPIR

32 Copyright Khronos Group Page 32 SYCL for OpenCL Andrew Richards, CEO Codeplay Chair, SYCL Working group GDC, March 2014

33 Copyright Khronos Group Page 33 Where is OpenCL today? OpenCL: supported by a very wide range of platforms - Huge industry adoption Provides a C-based kernel language NEW: SPIR provides ability to build other languages on top of OpenCL run-time Now, we need to provide additional languages and libraries Topic for today: C++ OpenCL C Kernels OpenCL Runtime Other Language Kernels Device X Device Y Device Z

34 Copyright Khronos Group Page 34 SYCL for OpenCL Pronounced sickle - To go with spear (SPIR) Royalty-free, cross-platform C++ programming layer - Builds on portability and efficiency of OpenCL - Ease of use and flexibility of C++ Single-source C++ development - C++ template functions can contain host & device code - Construct complex reusable algorithm templates that use OpenCL for acceleration OpenCL C Kernels High-level High-level C++ Based Frameworks Apps Frameworks & Frameworks OpenCL Runtime Device X Device Y Device Z

35 Copyright Khronos Group Page 35 Enabling C++ within OpenCL Ecosystem Want C++ code to be portable to OpenCL - C++ libraries supported on OpenCL - C++ tools supported on OpenCL Aim to achieve long-term support for OpenCL features with C++ - Good performance of C++ software on OpenCL Multiple sources of implementations and enable future innovation - Allows innovators in C++ for heterogeneous devices to leverage an open standard - Example of what can be done now OpenCL supports multiple languages with SPIR Developers can now use OpenCL as the basis for a whole range of innovations in software for heterogeneous systems

36 Copyright Khronos Group Page 36 Simple Example #include <CL/sycl.hpp> int main () { int result; // this is where we will write our result { // by sticking all the SYCL work in a {} block, we ensure // all SYCL tasks must complete before exiting the block // create a queue to work on cl::sycl::queue myqueue; Does everything * expected of an OpenCL program: compilation, startup, shutdown, host fall-back, queue-based parallelism, efficient data movement. * (this sample doesn t catch exceptions) // wrap our result variable in a buffer cl::sycl::buffer<int> resultbuf (&result, 1); // create some commands for our queue cl::sycl::command_group (myqueue, [&] () { // request access to our buffer auto writeresult = resultbuf.access<cl::sycl::access:write_only> (); // enqueue a single, simple task single_task(kernel_lambda<class simple_test>([=] () { writeresult [0] = 1234; } }); // end of our commands for this queue } // end scope, so we wait for the queue to complete } printf ( Result = %d\n, result);

37 Copyright Khronos Group Page 37 What Now? We are releasing this provisional specification to get feedback from developers - So please give feedback! - Khronos forums are the best place - Next steps - Full specification, based on feedback - Conformance test suite to ensure compatibility between implementations Release of implementations - Codeplay is working on an implementation - Anyone can implement it - it s an open, royalty-free standard! Roadmap - OpenCL working group considering C++ as kernel language - Complementary to SPIR - SPIR 2.0 and SYCL 2.0 for use with OpenCL 2.0