Version: 2021.4
Last Updated: 09/27/2021
Public Content

Contents

Get Started with the
Intel® oneAPI
DPC++/C++
Compiler

The

Intel® oneAPI

DPC++/C++

Compiler

provides optimizations that help your applications to run faster on Intel® 64 and IA-32 (Windows* and Linux* only) architectures, with support for the latest C, C++, and DPC++ language standards (including C++17). This compiler produces optimized code that can run significantly faster by taking advantage of the ever-increasing core count and vector register width in Intel® Xeon® processors and compatible processors. The Intel® Compiler will help you boost application performance through superior optimizations and Single Instruction Multiple Data (SIMD) vectorization, integration with Intel® Performance Libraries, and by leveraging the OpenMP* 5.0/5.1 parallel programming model.

The

Intel® oneAPI

DPC++/C++

Compiler

compiles C++-based SYCL* source files for a wide range of compute accelerators.

The

Intel® oneAPI

DPC++/C++

Compiler

is part of the Intel® oneAPI Toolkits.

Before You Begin

Windows*

The compiler integrates into the following versions of Microsoft Visual Studio*:

Visual Studio 2019

Visual Studio 2017

For full functionality within Visual Studio, including debugging and development, Visual Studio Community Edition or higher is required. Visual Studio Express Edition allows only command-line builds. For all versions, Microsoft C++ support must be selected as part of the Visual Studio install. For Visual Studio 2015 and later, you must use a custom install to select this option.

You typically do not need to set the environment variables on Windows, as the compiler command-line window sets these variables for you automatically. If you need to set the environment variables, run the environment script as described in the suite-specific Get Started documentation.

<install_dir>

is the installation directory. By default, it is

C:\Program Files (x86)\IntelSWTools

Linux*

Before you can use the compiler, you must first set the environment variables by sourcing the environment script using the initialization utility to initialize all the tools in one step:

Determine your installation directory,

<install_dir>

If your compiler was installed in the default location by a root user or sudo user, the compiler will be installed under

/opt/intel/

. In this case,

<install_dir>

/opt/intel/

For non-root users, your home directory under

intel/

is used. In this case,

<install_dir>

will be

$HOME/intel/

For cluster or enterprise users, your admin team may have installed the compilers on a shared network file system. Check with your local admin staff for the location of installation (

<install_dir>

Source the environment-setting script for your shell:

bash:

source <install_dir>/bin/setvars.sh intel64

csh/tcsh:

source <install_dir>/bin/setvars.csh intel64

If you want to use the 32bit ia32 compiler instead of the default 64bit compiler, replace

intel64

in the source command above with

ia32

. For example:

bash:

source <install_dir>/bin/setvars.sh ia32

csh/tcsh:

source <install_dir>/bin/setvars.csh ia32

Compile and Execute SYCL Code

Use the steps below to compile and execute SYCL code.

A sample SYCL program is shown below and captured in the file

simple-sycl-app.cpp

#include <CL/sycl.hpp>

int main() {
  // Creating SYCL queue
  cl::sycl::queue Queue;

  // Creating buffer of 4 ints
  cl::sycl::buffer<cl::sycl::cl_int, 1> Buffer(4);

  // Size of index space for kernel
  cl::sycl::range<1> NumOfWorkItems{Buffer.get_count()};

  // Submitting command group to queue
  Queue.submit([&](cl::sycl::handler &cgh) {
    // Getting write only access to the buffer on a device
    auto Accessor = Buffer.get_access<cl::sycl::access::mode::write>(cgh);
    // Executing kernel
    cgh.parallel_for<class FillBuffer>(
        NumOfWorkItems, [=](cl::sycl::id<1> WIid) {
          // Fill buffer with indexes
          Accessor[WIid] = (cl::sycl::cl_int)WIid.get(0);
        });
  });

  // Getting read only access to the buffer on the host
  const auto HostAccessor = Buffer.get_access<cl::sycl::access::mode::read>();

  // Check that the results are correct
  bool MismatchFound = false;
  for (size_t I = 0; I < Buffer.get_count(); ++I) {
    if (HostAccessor[I] != I) {
      std::cout << "The result is incorrect for element: " << I
                << " , expected: " << I << " , got: " << HostAccessor[I]
                << std::endl;
      MismatchFound = true;
    }
  }

  if (!MismatchFound) {
    std::cout << "The results are correct!" << std::endl;
  }

  return MismatchFound;
}

To build the

simple-sycl-app

run the following command:

dpcpp simple-sycl-app.cpp -o simple-sycl-app

To run the

simple-sycl-app

, use:

./simple-sycl-app

You will see the following output:

The results are correct!

The SYCL host device is not fully supported.

How to Specify a SYCL Device (optional)

To specify the device, SYCL provides the abstract

cl::sycl::device_selector

class, which you can subclass to define how the runtime selects the device. The method operator() of the SYCL

device_selector

is an abstract member function, which takes a reference to a SYCL device and returns an integer score. This abstract member function can be implemented in a derived class and provide a logic for selecting a SYCL device. The SYCL runtime uses the device with the highest returned score. This object can be passed to the

cl::sycl::queue

and

cl::sycl::device

constructors.

This example illustrates how to use the

device_selector

to create a device and queue objects that are bound to a GPU:

#include <CL/sycl.hpp>

int main() {
  class NEOGPUDeviceSelector : public cl::sycl::device_selector {
  public:
    int operator()(const cl::sycl::device &Device) const override {
      using namespace cl::sycl::info;

      const std::string DeviceName = Device.get_info<device::name>();
      const std::string DeviceVendor = Device.get_info<device::vendor>();

      return Device.is_gpu() && (DeviceName.find("HD Graphics NEO") != std::string::npos);
    }
  };

  NEOGPUDeviceSelector Selector;
  try {
    cl::sycl::queue Queue(Selector);
    cl::sycl::device Device(Selector);
  } catch (cl::sycl::invalid_parameter_error &E) {
    std::cout << E.what() << std::endl;
  }
}

You can specify the SYCL device used for execution with one of the following device selectors:

cl::sycl::intel::fpga_selector

cl::sycl::intel::fpga_emulator_selector

cl::sycl::cpu_selector

cl::sycl::gpu_selector

Use the Command Line on Windows

Follow the steps below to invoke the compiler using the command line from within Microsoft Visual Studio.

You must have a version of Microsoft Visual Studio installed to use the compiler.

Step 1: Open a command prompt.

Step 2: Invoke the compiler.

Use the commands below to invoke the compiler from the command line:

For C/C++ source files:

icx [options] file1 [file2...] [/link link_options]

For DPC++ source files:

dpcpp-cl [options] file1 [file2...] [/link link_options]

For DPC++ source files (Intel® oneAPI HPC Toolkit only):

icx -fsycl [options] file1 [file2...] [/link link_options]

To enable unnamed SYCL lambda kernel support with Intel® oneAPI HPC Toolkit, add the

-fsycl-unnamed-lambda

option.

Intel® C++ Compiler Classic is included as part of the Intel® oneAPI HPC and IoT Toolkits. Use the command below to invoke the classic compiler from the command line:

For C/C++ source files:

icl [options] file1 [file2...] [/link link_options]

For more information about invoking the Intel® C++ Classic compiler, see Invoking the Compiler

Use the Command Line on Linux

Use the commands below to invoke the compiler from the command line on Linux:

For C/C++ source files:

{icx|icpx} [options] file1 [file2...]

For DPC++ source files:

dpcpp [options] file1 [file2...]

For DPC++ source files (Intel® oneAPI HPC Toolkit only):

icpx -fsycl [options] file1 [file2...]

To enable unnamed SYCL lambda kernel support with Intel® oneAPI HPC Toolkit, add the

-fsycl-unnamed-lambda

option.

Intel® C++ Compiler Classic is included as part of the Intel® oneAPI HPC and IoT Toolkits. Use the command below to invoke the classic compiler from the command line:

For C/C++ source files:

{icc|icpc} [options] file1 [file2...]

For more information about invoking the Intel® C++ Classic compiler, see Invoking the Compiler

Use Microsoft Visual Studio on Windows

Follow the steps below to invoke the compiler from within Microsoft Visual Studio.

Step 1: Build a binary:

Launch Microsoft Visual Studio

Open an existing project or solution

Right click on Project
in Solution Explorer > Intel Compiler > Use Intel oneAPI DPC++/C++ Compiler

Select
OK

Select Build > Rebuild Solution

Step 2: Set build configurations.

Right click on Project
in Solution Explorer > Properties

Locate C/C++ in the list and expand the heading

Walk through the available properties to select your configuration

Switch to the Intel® oneAPI DPC++/C++ Compiler from the Microsoft Visual Studio C++ Compiler

Launch Microsoft Visual Studio and open a solution or project

Select Project > Intel Compiler > Use Intel oneAPI DPC++/C++

Select OK

Select Build > Rebuild Solution

Switch to the Microsoft Visual Studio C++ Compiler from the Intel® oneAPI DPC++/C++ Compiler

This action updates the solution file to use the Microsoft Visual Studio C++ compiler. All configurations of affected projects are automatically cleaned unless you select Do not clean project(s)
. If you choose not to clean projects, you will need to rebuild updated projects to ensure all source files are compiled with the new compiler.

Launch Microsoft Visual Studio and open a solution or project

Select Project > Intel Compiler > Use Visual C++

To change the compiler version in Microsoft Visual Studio, navigate to Tools > Options > Intel Compilers and Libraries > C++ > Compilers

Use the Eclipse* CDT on Linux

Follow the steps below to invoke the compiler from within the Eclipse* CDT.

Step 1: Install the Intel® Compiler Eclipse CDT plugin.

Start Eclipse

Select Help > Install New Software

Select Add
to open the Add Site dialog

Select Archive
, browse to the directory

<install_dir>/compiler/<version>/linux/ide_support

, select the .zip file that starts with

com.intel.dpcpp.compiler

, then select OK

Select the options beginning with Intel, select Next
, then follow the installation instructions

When asked if you want to restart Eclipse*, select Yes

Step 2: Build a new project or open an existing project.

Open Existing Project or Create New Project on Eclipse

Right click on Project > Properties > C/C++ Build > Tool chain Editor

Select
Intel DPC++/C++ Compiler
from the right panel

Step 3: Set build configurations.

Open Existing Project on Eclipse

Right click on Project > Properties > C/C++ Build > Settings

Create or manage build configurations in the right panel

Next Steps

Use the latest oneAPI Code Samples and follow along with the Intel® oneAPI Training Resources.

Explore the

Intel® oneAPI

DPC++/C++

Compiler

Developer Guide and Reference on the Intel® Developer Zone.

Find More

Content	Description and Links
Release Notes	Visit the Release Notes page for known issues and the most up-to-date information.
Intel® oneAPI Programming Guide	Provides details on the Intel® oneAPI DPC++/C++ Compiler programming model, including details about DPC++ , programming for various target accelerators, and introductions to the Intel® oneAPI libraries.
Intel® oneAPI DPC++/C++ Compiler Developer Guide and Reference	Explore Intel® oneAPI DPC++/C++ Compiler features and setup and get more detailed information about compiler options, attributes, and more.
SYCL Specification Version 1.2.1 PDF	The SYCL Specification PDF, explains how SYCL integrates OpenCL devices with modern C++: https://www.khronos.org/registry/SYCL/specs/sycl-1.2.1.pdf
SYCL Overview site	An overview of SYCL: https://www.khronos.org/sycl/
The GNU* C++ Library	Using dual ABI: https://gcc.gnu.org/onlinedocs/libstdc++/manual/using_dual_abi.html

Notices and Disclaimers

Intel technologies may require enabled hardware, software or service activation.

No product or component can be absolutely secure.

Your costs and results may vary.

No license (express or implied, by estoppel or otherwise) to any intellectual property rights is granted by this document.

The products described may contain design defects or errors known as errata which may cause the product to deviate from published specifications. Current characterized errata are available on request.

Intel disclaims all express and implied warranties, including without limitation, the implied warranties of merchantability, fitness for a particular purpose, and non-infringement, as well as any warranty arising from course of performance, course of dealing, or usage in trade.

In This Topic

Product and Performance Information

Performance varies by use, configuration and other factors. Learn more at www.Intel.com/PerformanceIndex.

Select Your Region

Asia Pacific

Europe

Latin America

Middle East/Africa

North America

Top Results

Using Intel.com Search

Quick Links

Recent Searches

Get Started with the
Intel® oneAPI
DPC++/C++
Compiler

Before You Begin

Compile and Execute SYCL Code

How to Specify a SYCL Device (optional)

Use the Command Line on Windows

Use the Command Line on Linux

Use Microsoft Visual Studio on Windows

Use the Eclipse* CDT on Linux

Next Steps

Find More

Notices and Disclaimers

Product and Performance Information

Sign In

Top Results

Using Intel.com Search

Quick Links

Recent Searches

Get Started with the Intel® oneAPI DPC++/C++ Compiler

Before You Begin

Compile and Execute SYCL Code

How to Specify a SYCL Device (optional)

Use the Command Line on Windows

Use the Command Line on Linux

Use Microsoft Visual Studio on Windows

Use the Eclipse* CDT on Linux

Next Steps

Find More

Notices and Disclaimers

Product and Performance Information

Get Started with the
Intel® oneAPI
DPC++/C++
Compiler