OpenCL™ Developer Guide for Intel® Processor Graphics

Download PDF
ID 773088
Date 3/20/2019
Public

Visible to Intel only — GUID: GUID-A4A328F0-4984-467A-8CA3-D711A3FF8626

View Details

Document Table of Contents
Document Table of Contents x
OpenCL™ Optimization Guide for Visual Computing Systems
OpenCL™ Optimization Guide for Visual Computing Systems x
Legal Information Getting Help and Support Introduction Coding for the Intel® Processor Graphics Platform-Level Considerations Application-Level Optimizations Optimizing OpenCL™ Usage with Intel® Processor Graphics Check-list for OpenCL™ Optimizations Performance Debugging Using Multiple OpenCL™ Devices Coding for the Intel® CPU OpenCL™ Device OpenCL™ Kernel Development for Intel® CPU OpenCL™ device
Introduction x
About this Document Basic Concepts Using Data Parallelism Related Products
Coding for the Intel® Processor Graphics x
Execution of OpenCL™ Work-Items: the SIMD Machine Memory Hierarchy
Platform-Level Considerations x
Intel® Turbo Boost Technology Support Global Memory Size
Application-Level Optimizations x
Minimizing Data Copying Avoiding Needless Synchronization Reusing Compilation Results with clCreateProgramWithBinary Interoperability with Other APIs
Interoperability with Other APIs x
Interoperability between OpenCL and OpenGL Using Microsoft DirectX* Resources Aligning Pointers to Microsoft DirectX* Buffers Upon Mapping Note on Working with other APIs Note on Intel® Quick Sync Video
Optimizing OpenCL™ Usage with Intel® Processor Graphics x
Optimizing Utilization of Execution Units Work-Group Size Recommendations Summary Memory Access Considerations Using Loops
Memory Access Considerations x
Memory Access Overview Recommendations Kernel Memory Access Optimization Summary
Recommendations x
Granularity __global Memory and __constant Memory __private Memory __local Memory
Check-list for OpenCL™ Optimizations x
Mapping Memory Objects Using Buffers and Images Appropriately Using Floating Point for Calculations Using Compiler Options for Optimizations Using Built-In Functions Loading and Storing Data in Greatest Chunks Applying Shared Local Memory Using Specialization in Branching Considering native_ and half_ Versions of Math Built-Ins Using the Restrict Qualifier for Kernel Arguments Avoiding Handling Edge Conditions in Kernels
Performance Debugging x
Host-Side Timing Profiling Operations Using OpenCL™ Profiling Events Comparing OpenCL™ Kernel Performance with Performance of Native Code Getting Credible Performance Numbers Using Tools
Using Multiple OpenCL™ Devices x
Using Shared Context for Multiple OpenCL™ Devices Sharing Resources Efficiently Synchronization Caveats Writing to a Shared Resource Partitioning the Work Keeping Kernel Sources the Same Basic Frequency Considerations Eliminating Device Starvation Limitations of Shared Context with Respect to Extensions See Also
Coding for the Intel® CPU OpenCL™ Device x
Vectorization Basics for Intel® Architecture Processors Benefitting From Implicit Vectorization Vectorizer Knobs Using Vector Data Types Writing Kernels to Directly Target the Intel® Architecture Processors Work-Group Size Considerations Work-Group Level Parallelism
OpenCL™ Kernel Development for Intel® CPU OpenCL™ device x
Why Optimizing Kernel Code Is Important? Avoid Spurious Operations in Kernel Code Perform Initialization in a Separate Task Use Preprocessor for Constants Use Signed Integer Data Types Use Row-Wise Data Accesses Tips for Auto-Vectorization Local Memory Usage Avoid Extracting Vector Components Task-Parallel Programming Model Hints
OpenCL™ Optimization Guide for Visual Computing Systems

Limitations of Shared Context with Respect to Extensions

Level Two Title