中级

What’s New in the Intel Compiler

The list below summarizes new features in the Intel® C++ Compiler 15.0 and the Intel® Fortran Compiler 15.0. For more details about changes in the Intel compilers since the previous release, including a list of new options, please refer to the ‘What’s New’ section in the release notes (C++, Fortran).

  • 开发人员
  • Linux*
  • Microsoft Windows* (XP, Vista, 7)
  • Microsoft Windows* 8.x
  • C/C++
  • Fortran
  • 高级
  • 入门级
  • 中级
  • 英特尔® C++ 编译器
  • Intel® Fortran Compiler
  • OpenMP*
  • 开发工具
  • 英特尔® 酷睿™ 处理器
  • Intel® Many Integrated Core Architecture
  • 优化
  • 并行计算
  • 线程
  • 矢量化
  • Intel® C++ Compiler 15.0 for Linux Sets “-ansi-alias” by Default, Noncompliant Code May Fail

    Beginning with version 15.0, the Intel® C++ Compiler for Linux enables -ansi-alias by default at optimization level 2 (-O2). The -ansi-alias option indicates that the code complies with ANSI aliasing rules.

    If the application code adheres to ANSI aliasing rules, this change allows the compiler to make aggressive optimizations. The effect of this change on runtime performance depends on the application, the optimization methods supported by the compiler, and the compiler switches used.

  • 开发人员
  • Linux*
  • C/C++
  • 中级
  • 英特尔® C++ 编译器
  • Intel® Many Integrated Core Architecture
  • 优化
  • 矢量化
  • Using Intel® MPI Library 5.0 with MPICH based applications

    Why it is needed?

    Different MPI implementations have their specific benefits and advantages. So in the specific cluster environment the HPC application with the other MPI implementation can probably perform better.

     Intel® MPI Library has the following benefits:

  • 开发人员
  • 合作伙伴
  • 教授
  • 学生
  • Linux*
  • 服务器
  • 高级
  • 入门级
  • 中级
  • Intel® Cluster Toolkit
  • 英特尔® 跟踪分析器和跟踪采集器
  • Intel® MPI Library
  • Intel® Cluster Studio
  • Intel® Cluster Studio XE
  • Intel® Cluster Ready
  • 消息传递接口
  • 集群计算
  • 开发工具
  • Data transfer of non-contiguous array elements using the Intel® Language Extensions for Offload (LEO) for the Intel® Xeon Phi™ coprocessor

    The Intel® Parallel Studio XE 2015 Composer Editions for C++ Windows* and Linux* have a feature enhancement supporting data transfer for non-contiguous array elements with the Intel® Language Extensions for Offload (LEO) for the Intel® Xeon Phi™ coprocessor.

    The feature adds support under the LEO offload data marshalling model for transferring non-contiguous array elements within an array variable reference (variable-ref) in the data transfer clauses (i.e. in, out, inout, nocopy) of the #pragma offload/offload_transfer statement.

  • 开发人员
  • Linux*
  • Microsoft Windows* (XP, Vista, 7)
  • Microsoft Windows* 8.x
  • 服务器
  • C/C++
  • 中级
  • 英特尔® C++ 编译器
  • 英特尔® C++ Composer XE
  • 英特尔® Composer XE
  • 英特尔® Parallel Composer
  • 开发工具
  • Intel® Many Integrated Core Architecture
  • ELF executable binary file created for the Intel® Xeon Phi™ coprocessor offload image

    The Intel® Parallel Studio XE 2015 Composer Editions for Windows* and Linux* have an internal implementation feature enhancement to the Intel® Xeon Phi™ coprocessor binary image created when using the Intel® Language Extensions for Offload (LEO) or OpenMP* 4.0 target constructs.

    The enhancement enables the Intel® 15.0 compilers to create an Executable and Linkable Format (ELF) format executable binary file for Intel® Xeon Phi™ coprocessor offload image. In previous releases, the compiler created an ELF format shared library (.so) binary file.

  • 开发人员
  • Linux*
  • Microsoft Windows* (XP, Vista, 7)
  • Microsoft Windows* 8.x
  • C/C++
  • Fortran
  • 中级
  • 英特尔® C++ 编译器
  • 英特尔® C++ Composer XE
  • 英特尔® Composer XE
  • Intel® Fortran Compiler
  • 英特尔® Fortran Composer XE
  • 英特尔® Parallel Composer
  • 英特尔® Visual Fortran Composer XE
  • 开发工具
  • Intel® Many Integrated Core Architecture
  • 订阅 中级