2. Tools
  3. Intel® oneAPI(Beta)
  4. / Tools / ...

Data Parallel C++(Beta)

A Standards-Based, Cross-Architecture Language

An Evolution of C++ for Direct Programming

Data Parallel C++ (DPC++) is an open, standards-based alternative to single-architecture proprietary languages. It allows developers to reuse code across hardware targets (CPUs and accelerators such as GPUs and FPGAs) and also perform custom tuning for a specific accelerator.

  • Delivers the productivity benefits of ISO C++ using familiar C and C++ constructs.
  • Incorporates standard SYCL* from The Khronos Group* to support data parallelism and heterogeneous programming.
  • Includes a compiler to deploy across diverse hardware targets and a compatibility tool for migrating legacy CUDA* applications to a multiplatform program in DPC++.
  • Additional language enhancements are driven through a community open source project for cooperative development at oneAPI.com. These are submitted to The Khronos Group as proposed additions to the SYCL standard.
  • The Intel® oneAPI Base Toolkit includes the following DPC++ elements:
    • Intel® oneAPI DPC++ Compiler to deploy applications across diverse hardware targets.
    • Intel® oneAPI DPC++ Library to maximize productivity and performance across CPUs, GPUs, and FPGAs.
    • Intel® DPC++ Compatibility Tool to migrate CUDA* applications to a multi-platform program in DPC++.
DPC++ = 
  ISO C++ and 
  Khronos SYCL and 
  community extensions

Explore DPC++ Language through Intel Samples

oneAPI Programming Guide

Develop in the Cloud

Get what you need to build, test, and optimize your oneAPI projects for free. With an Intel® DevCloud account, you get 120 days of access to the latest Intel® hardware—CPUs, GPUs, FPGAs—and Intel oneAPI tools and frameworks.

Get Access

 

Download the Toolkit

Data Parallel C++ is included as part of the Intel® oneAPI Base Toolkit. This download includes the Intel® oneAPI DPC++ Compiler(Beta), Intel® oneAPI DPC++ Library, and Intel® DPC++ Compatibility Tool(Beta).

Get it Now

View Full Size

Intel® oneAPI DPC++ Compiler

  • Targets CPUs and accelerators through single-source code while permitting custom tuning
  • Builds on the Intel history of compiler performance leadership
  • Is built to the latest standards, including C++17 and SYCL*, to ensure portability and support of features like generic lambda expressions and variable templates
  • Interoperates with oneAPI performance, productivity, and threading libraries. For example, the Intel oneAPI DPC++ Library extends the C++ Standard Library to better support data parallelism productivity. 1

Learn More about the Intel® oneAPI DPC++ Library

View Full Size

FPGA Emulation and Compilation

For experienced FPGA developers, the Intel® oneAPI Data Parallel C++ Compiler enables you to target your acceleration workload to field-programmable platforms. Simply select the optional Intel® FPGA Add-On for oneAPI Base Toolkit, available when you download the Intel® oneAPI Base Toolkit.

Learn More

View Full Size

CUDA* Source Code Migration

The Intel DPC++ Compatibility Tool is a migration engine that transforms CUDA applications into a standards-based DPC++ code.

  • One-time-only migration ports both kernels and API calls.
  • Inline comments guide developers to produce output that can be most easily compiled with the Intel oneAPI DPC++ Compiler.
  • The process is further streamlined by command-line tools and IDE plug-ins.

Key Specifications

Intel oneAPI DPC++ Compiler

 

Processors:

  • Intel® Core™ processors Gen6 and newer
  • Intel® Xeon® processors

GPUs:

  • Intel® Processor Graphics Gen9

FPGAs:

  • Intel® Programmable Acceleration Card with Intel®Arria® 10 GX FPGA

 

 

Host and target operating systems:

  • Windows*
  • Linux*

Languages:

  • Data Parallel C++ (DPC++)
  • C++

 

 

Development environments:

  • Microsoft Visual Studio*
  • Eclipse* IDE

Threading:

  • Intel® oneAPI Threading Building Blocks
  • Native threads

For more information, see the system requirements.

Intel® DPC++ Compatibility Tool

 

Operating system for development:

  • Linux
  • Windows

 

 

Software tool requirements:

  • Certain CUDA header files
  • Eclipse (optional)
  • Microsoft Visual Studio* (optional)

For additional details, please see the compatibility tool system requirements.

Get Help

Your success is our success. Access these support resources whenever you get stuck.

 

Product and Performance Information

1Intel oneAPI DPC++ Compiler Stack supports multiple target devices through the DPC++ runtime and LLVM backend.
2

Intel's compilers may or may not optimize to the same degree for non-Intel microprocessors for optimizations that are not unique to Intel microprocessors. These optimizations include SSE2, SSE3, and SSSE3 instruction sets and other optimizations. Intel does not guarantee the availability, functionality, or effectiveness of any optimization on microprocessors not manufactured by Intel. Microprocessor-dependent optimizations in this product are intended for use with Intel microprocessors. Certain optimizations not specific to Intel microarchitecture are reserved for Intel microprocessors. Please refer to the applicable product User and Reference Guides for more information regarding the specific instruction sets covered by this notice.

Notice revision #20110804

Give Feedback