Intel® - Fortran Compiler, Professional and Standard Editions, for Mac OS*

	Intel® Software Development Products for Intel® Platforms and Technologies

Intel® Fortran Compiler 10.1, Professional and Standard Editions for Mac OS* X

Overview	Features in Depth	Print
Features	Compatability
New in This Release	System Requirements

Overview

Intel® Fortran Compiler 10.0 for Mac OS*

Intel® Fortran Compiler Professional Edition offers the best support for creating multi-threaded applications. Only the Professional Edition offers the breadth of advanced optimization, multi-threading, and processor support that includes automatic processor dispatch, vectorization, auto-parallelization, OpenMP*, data prefetching, loop unrolling, substantial Fortran 2003 support, and an optimized math processing library.

The Professional Edition combines a high performance compiler with Intel® Math Kernel Library (Intel® MKL). While this library is available separately, the Professional Edition creates a strong foundation for building robust, high performance parallel code at significant price savings.

The Standard Edition compiler has the same performance and features as the Professional Edition compiler, but does not include Intel MKL.

Product Brief [PDF 812KB]

Features

The Intel® Fortran Compiler for Mac OS* X delivers rapid development and winning performance for the full range of Intel-based Mac* systems. The Intel Fortran Compiler fully supports the Fortran 95 language standard, as well as the previous standards Fortran 90, Fortran 77 and Fortran IV. It also includes many features from the Fortran 2003 language standard, as well as numerous popular language extensions.

Automatically optimize and parallelize software to take best advantage of multi-core Intel® processors, including dual-core mobile, desktop, and enterprise platforms.

Mac* and Multi-Core Capabilities

The following features specifically address the architecture capabilities of the new Intel®-based Macs. Each is explained briefly here, with a link to a more complete description.

	Multi-Threaded Application Support, including OpenMP* and auto-parallelization, allows you to take full advantage of multi-core technology like the Intel® Core™ microarchitecture.
	*Xcode Integration** allows developers to continue working with this familiar environment while benefiting from the advanced capabilities of the Intel® Compiler.

Performance

Intel® Fortran Compiler Professional Edition lets you choose the tools that get most out of multi-core processors by combining the Fortran compiler and its built-in optimization, threading and security capabilities with a highly optimized math library that simplifies the introduction of robust, scalable, multi-threaded math functions.

Advanced Optimization Features

Mac hardware based on Intel® processors can also benefit from advanced optimization features, a few of which are explained briefly here, with links to more complete descriptions.

	Multithreaded Application Support, including OpenMP and auto-parallelization for simple and efficient software threading.
	Auto-vectorization parallelizes code to utilize the Streaming SIMD Extensions (SSE) instruction set architectures (SSE, SSE2, SSE3, SSSE3, and SSE4) of our latest processors.
	High-Performance Parallel Optimizer (HPO)restructures and optimizes loops to ensure that auto-vectorization, OpenMP, or auto-parallelization best utilizes the processor’s capabilities for cache and memory accesses, SIMD instruction sets, and for multiple cores. This revolutionary capability, new in Version 10, combines vectorization, parallelization and loop transformations into a single pass which is faster, more effective and more reliable than prior discrete phases.
	Interprocedural Optimization (IPO)dramatically improves performance of small- or medium-sized functions that are used frequently, especially programs that contain calls within loops. The analysis capabilities of this optimizer can also give feedback on vulnerabilities and coding errors, such as uninitialized variables or OpenMP API issues, which cannot be detected as well by compilers which rely strictly on analysis by a compiler front-end.
	Profile-guided Optimization (PGO) improves application performance by reducing instruction-cache thrashing, reorganizing code layout, shrinking code size, and reducing branch mispredictions.
	Optimized Code Debugging with the Intel® Debugger improves the efficiency of the debugging process on code that has been optimized for Intel® architecture.

New in This Release

The Intel Fortran Compiler for Mac OS X builds on a winning foundation. Position yourself to create next-generation software, for next-generation hardware. The following features are new since Version 9 of the compiler.

What's new

Benefit to you

More Fortran 2003 Features, including Stream I/O

C Interoperability features make it easier to develop mixed-language applications. Asynchronous I/O enhances performance of applications which read and write large files. See the compiler Release Notes for a full list of supported Fortran 2003 features.

Improved Performance and Threading

New Parallel/Loop Optimizer (HPO)

Better application performance for computationally intensive applications such as graphics/digital media, financial modeling, and high-performance computing for threaded and non-threaded applications. Our new High Performance Parallel Optimizer, HPO, offers an improved ability to analyze, optimize, and parallelize more loop nests.

Security Checking and Diagnostics

	GNU Mudflap
	Static Verifier for buffer overflow
	OpenMP* API verification

Ability to create code that is less susceptible to security vulnerabilities, such as buffer overflow. The diagnostics are very helpful for novice and expert users for catching common coding errors, from unitialized variables to mismatched dummy and actual arguments to OpenMP API coding issues.

64-bit Mac OS* X Support

Ability to create applications that take advantage of the addressing and performance capabilities enabled by Intel® 64 architecture-based processors used in the latest Apple Mac OS X systems.

Optimization Reports

More detailed optimization diagnostics for users who want to use our advanced optimizations to help the compiler do a better job at tuning their applications.

Options to enable more advanced optimizations for loop unrolling and streaming stores

Improved application performance.

Support for the Latest Multi-Core Processors
The Intel Fortran Compiler provides optimization support for the very latest multi-core processors, including:

	Intel® Core™2 Duo processor
	Intel® Core™2 Quad processor
	Quad-Core Intel® Xeon® processor 5300 series
	Dual-Core Intel® Xeon® processor 3000 series
	Dual-Core Intel® Xeon® processor 5000 series
	Dual-Core Intel® Xeon® processor 7000 series
	Dual-Core Intel® Itanium 2 processor

Intel® compilers future-proof your investment with assurance of world-class support for each successive generation of processors. That's a key advantage in a world where new hardware platforms come to market with awesome speed.

Support for auto-parallelization and OpenMP enable you to create optimized, multithreaded applications that take full advantage of multi-core processing features to deliver outstanding performance.

Professional Edition

Includes not only the advanced capabilities of the compiler, but also the Intel® Math Kernel Library with highly optimized functions for math processing.

Advanced Optimization Features in Depth

This section gives detailed descriptions of the compiler’s advanced optimization features.

Multi-Threaded Application Support

OpenMP and auto-parallelization help convert serial applications into parallel applications, allowing you to take full advantage of multi-core technology like the Intel® Core™ Duo processor and Dual-Core Intel® Itanium® 2 processor, as well as symmetric multi-processing systems:

	OpenMP* is the industry standard for portable multithreaded application development. It is effective at fine-grain (loop-level) and large-grain (function-level) threading. OpenMP directives are an easy and powerful way to convert serial applications into parallel applications, enabling potentially big performance gains from parallel execution on multi-core and symmetric multiprocessor systems.
	Auto Parallelization improves application performance on multiprocessor systems by means of automatic threading of loops. This option detects parallel loops capable of being executed safely in parallel and automatically generates multithreaded code. Automatic parallelization relieves the user from having to deal with the low-level details of iteration partitioning, data sharing, thread scheduling, and synchronizations. It also provides the performance benefits available from multiprocessor systems and systems that support Hyper-Threading Technology (HT Technology).
For more information on multi-threaded application support, visit Intel's Threading Developer Center.

High Performance, Parallel Optimizer (HPO)

This revolutionary capability, new in Version 10, combines automatic vectorization, automatic parallelization and loop transformations into a single pass which is faster, more effective and more reliable than prior discrete phases.

HPO optimizes and restructures program loops to ensure that auto-parallelization, OpenMP, and auto-vectorization occur smoothly in conjunction with each other. HPO’s optimization technology utilizes a unique cost-benefit analysis to make the right optimization decisions for the given program and loop structure. It will perform many transformations such as loop unrolling, peeling, interchange, splitting, etc., as well as other optimizations to ensure the processor’s cache architecture, SIMD instruction set, and multiple cores are well utilized.

Automatic Vectorizer

Vectorization automatically parallelizes code to maximize underlying processor capabilities. This advanced optimization analyzes loops and determines when it is safe and effective to execute several iterations of the loop in parallel by utilizing MMX™, SSE, SSE2, SSE3, SSSE3, and SSE4 instructions. Figure 1 is a graphical representation of a vectorized loop that shows four iterations computed with one SSE2 operation.		Figure 1. The Vectorizer in action
Use vectorization to optimize your application code and take advantage of these new extensions when running on Intel® processors. Features include support for advanced, dynamic data alignment strategies, including loop peeling to generate aligned loads and loop unrolling to match the prefetch of a full cache line.

Interprocedural Optimization (IPO)

Interprocedural optimization (IPO) can dramatically improve application performance in programs that contain many small- or medium-sized functions that are frequently used, especially for programs that contain calls within loops. This set of techniques, which can be enabled for automatic operation in the Intel® compilers, uses multiple files or whole programs to detect and perform optimizations, rather than focusing within individual functions.

Figure 1. The interprocedural optimization process

Figure 2. The interprocedural optimization process

The IPO process, shown in Figure 2, first requires that source files are compiled with the IPO option, creating object (.o) files that contain the intermediate language (IL) used by the compiler. Upon linking, the compiler combines all of the IL information and analyzes it for optimization opportunities. Typical optimizations made as part of the IPO process include procedure inlining and re-ordering, eliminating dead (unreachable) code, and constant propagation, or the substitution of known values for constants. IPO enables more aggressive optimization than what is available at the intra-procedural level, since the added context of multiple procedures makes those more-aggressive optimizations safe.

The analysis capabilities of IPO can also give feedback on vulnerabilities and coding errors, such as uninitialized variables, which cannot be detected as well by compilers which rely strictly on analysis by a compiler front-end.

Profile-Guided Optimization (PGO)

The Profile-guided optimization (PGO) compilation process enables the Intel Fortran Compiler to take better advantage of the processor microarchitecture, more effectively use instruction paging and cache memory, and make better branch predictions. It improves application performance by reorganizing code layout to reduce instruction-cache thrashing, shrinking code size, and reducing branch mispredictions.

PGO is a three-stage process, as shown in Figure 3. Those steps include 1) a compile of the application with instrumentation added, 2) a profile-generation phase, where the application is executed and monitored, and 3) a recompile where the data collected during the first run aids optimization. A description of several code size influencing profile-guided optimizations follows:

	Basic block and function ordering — Place frequently-executed blocks and functions together to take advantage of instruction-cache locality.
	Aid inlining decisions — Inline frequently-executed functions so the increase in code size is paid in areas of highest performance impact.
	Aid vectorization decisions — Vectorize high trip count and frequently-executed loops so the increase in code size is mitigated by the increase in performance.

Figure 3. Profile-guided optimization

Optimized Code Debugging with the Intel® Debugger

The Intel® Debugger enables optimized code debugging (i.e., debugging code that has been significantly transformed for optimal execution on a specific hardware architecture). The Intel compilers produce standards-compliant debug information for optimized code debugging that is available to all debuggers that support Intel compilers. The Intel Debugger supports multi-core architectures by enabling debugging of multithreaded applications, providing the following related capabilities:

	An all-stop/all-go execution model (i.e., all threads are stopped when one is stopped, and all threads are resumed when one is resumed).
	List all created threads.
	Switch focus between threads.
	Examine detailed thread state.
	Set breakpoints (including all stop, trace, and watch variations) and display a back-trace of the stack for all threads or for a subset of threads.
	The built-in GUI provides a Thread panel (on the Current Source pane) that activates when a thread is created, and that allows an operator to select thread focus and display related details.
The recently enhanced GNU Project Debugger (GDB debugger) can also be used for parallel applications. For additional information, please refer to the Intel Debugger Technical White Paper (PDF 210KB).

Compatibility and Flexibility

Mac and Multi-Core Capabilities in Depth

This section gives detailed descriptions about the features described at a high level in the section above called "Overview of Mac and Multi-Core Capabilities."

Xcode* Integration

The Intel Fortran Compilers, Standard and Professional Editions for Mac OS X are compatible with Xcode*, enabling developers to work with popular IDE while also taking advantage of advanced optimization features from Intel.

Xcode Integration Generate C/Fortran Universal Binaries from the Xcode environment using the Intel Fortran Compilers for Mac OS X and GCC for PowerPC*, retaining compatibility with GCC 4.0. Universal Binaries are designed to ease the transition between PowerPC and Intel® architecture by combining native code for both architectures in a single compiled package.

High-level code that contains no processor dependencies requires few if any changes to create a Universal Binary. Lower-level code that contains hardware dependencies presents a greater challenge.

Support for Apple Frameworks

Apple frameworks are a special type of bundle used to distribute shared resources, including library code, resource files, header files, and reference documentation. They offer flexibility that is often preferable to using dynamic shared libraries.

For more information on Apple frameworks, please visit Apple’s website*.

Standards Compliance and Broad Compatibility

The Intel Fortran Compiler fully supports the Fortran 95 language standard, as well as the previous standards Fortran 90, Fortran 77 and Fortran IV. It also includes many features from the Fortran 2003 language standard, as well as numerous popular language extensions. Significant supported language extensions include:
	Quadruple precision REAL data type REAL(16)
	STRUCTURE, RECORD, UNION, MAP syntax for user-defined types
	Directives and functions to enhance mixed-language application development
	Binary stream I/O
For a complete list of language features, see the product documentation.
The Intel Fortran Compiler also enhances programmer productivity with features such as:
	Run-time array and string bounds checking
	Cross-file procedure interface checking
	Run-time uninitialized variable detection
	Error traceback with file name and line number

System Requirements

Hardware

Minimum Requirements

	Intel® Core™ microarchitecture-based Mac* system
	512MB RAM minimum, 1GB RAM recommended
	100 MB of disk space, plus an additional 200 MB during installation for the download and temporary files

Software

Minimum Requirements

	Mac OS* X developers tools including Xcode* 2.4.1 or 2.5
	Mac OS* X 10.4.10
	GCC 4.0

Intel provides both the tools and support to enhance the performance, functionality, and efficiency of software applications.

Compatible with leading Windows* and Linux* development environments, Intel® Software Development Products are the fastest and easiest way to take advantage of the latest features of Intel processors. Intel Software Development Products are designed for use in the full development cycle, and include Intel® Performance Libraries, Intel® Compilers (C++, Fortran for Windows, Linux, and Mac OS* X), Intel® VTune™ Analyzer, Intel® Threading Tools and Intel® Cluster Tools.

The Intel® Premier Support Web site provides expert technical support for all Intel software products, product updates and related downloads. For additional product information visit: www.intel.com/software/products.

Intel, the Intel logo, and VTune are trademarks or registered trademarks of Intel Corporation or its subsidiaries in the United States and other countries.

*Other brands and names may be claimed as the property of others.