Lahey/Fujitsu Linux64 Fortran v8.1
Fortran 95/90/77/IV

Speed. Dependability. Great Tech Support. Lahey/Fujitsu Fortran 95. The standard for Fortran programming from the leader in Fortran language systems.

We've combined our 39 years of producing award-winning Fortran language systems with Fujitsu's compiler expertise and high-performance code generator to deliver the most-productive, best-supported Fortran 95 language system for the PC.

Whether you write new Fortran programs or downsize existing applications, you need speed, dependability, and great tech support!

• Intel EM64T and AMD AMD64 64-bit optimizations
• Unsurpassed global compile-time and runtime diagnostics
• Automatic Parallelization
• OpenMP Support
• Wisk, Winteracter Starter Kit - Graphics package
• Small and medium memory models
• BLAS and LAPACK v4.0 routines - thread-safe
• Fujitsu's SSL2 math library - thread-safe
• Fujitsu's FDB debugger
• Automake, automatic make utility
• COMPLEX constants (Fortran 2003) supported

New Features in LF64 v8.1

   *  COMPLEX constants (a Fortran 2003 feature) are now supported

        REAL, PARAMETER :: a = 1.0

        REAL, PARAMETER :: b = 2.0

        COMPLEX         :: c = (a,b)

   *  suboptions "f" and "o" are supported within --chk and --chkglobal option

      "f" checks whether a file is connected with two devices at the same time

      "o" checks overlapping dummy arguments

          checks when an INTENT(OUT) assumed-size array is referenced, whether the variable is defined

          checks when a variable with SAVE is referenced, whether the variable is defined forces "a" and "u" suboptions

   *  A new information message has been added:

      2003-I "a.h", line 1: 'xxx' is declared in INCLUDE but never referenced

      To suppress the message, you can specify:

      lfc --info -swm 2003 a.f95

   *  Updated User Guide and Language Reference on-line PDF documentation

LF64 v8.1 is available in two configurations, Express and PRO:

LF64 Express includes the powerful Lahey/Fujitsu Fortran 95 optimizing compiler, command line debugger, online documentation, and free e-mail support.

• LF64 Performance.
• LF64 Optimizations.
• Link GNU C and assembly object files.
• Legacy Fortran Support.
• ANSI/ISO-Compliant Fortran 95. 
• Free Technical Support. 

• Auto-Parallelization
• OpenMP compatibility
• Winteracter Starter Kit.
• BLAS and LAPACK
• Fujitsu Scientific Subroutine Library 2
• Automake
• Free Technical Support

LF64 Performance

Polyhedron Software (www.polyhedron.com) ran their suite of Fortran benchmarks on a Pentium D820 dual core 2.8 GHz processor, with 2 x 1MB L2 cache and 800 MHz FSB, and running SUSE 9.3 Linux. Specifying the switch --fast, the benchmarks ran an average of 21.8% faster than they did when built with LF95 Linux v6.2. Try LF64 v8.1 on your code today!  

LF64 Optimizations

Basic Optimization

• Constant folding
• Common subexpression elimination
• Copy propagation
• Strength Reduction
• Algebraic simplifications
• Dead code elimination
• Peephole optimization
• Loop invariant code motion
• Transform array element to simple variable
• Local Instruction scheduling
• Address calculation optimization

• Loop unrolling
• Loop interchange

• Inlining mathematical functions
• Stack optimization

• SSE2 and SSE3 instructions
• Prefetch instructions
• Using fast input/output libraries

• Unlimited number of continuation lines in free or fixed source form 
• DO UNTIL statement 
• FIND statement 
• STRUCTURE and END STRUCTURE statements 
• UNION and END UNION statements 
• MAP and END MAP statements 
• RECORD statement 
• Non-standard POINTER statement 
• AUTOMATIC statement 
• STATIC statement 
• VALUE statement 
• BYTE statement 
• Hollerith constants 
• Alternative forms of binary, octal, and hexadecimal constants 
• Binary, octal, or hexadecimal constants in a DATA, declaration statement 
• Period structure component separator 
• IMPLICIT UNDEFINED statement 
• Namelist input/output on internal file 
• FORM = 'BINARY' 
• TOTALREC specifier 
• STATUS = 'SHR' 
• Gw, $, \, and R edit descriptors 
• LOC intrinsic function 
• The following service subroutines: ABORT, BIC, BIS, CLOCK, CLOCKM, CLOCKV, DATE, ERROR, ERRSAV, ERRSET, ERRSTR, ERRTRA, EXIT, FDATE, FLUSH, FREE, GETARG, GETCL, GETDAT, GETENV, GETLOG, GETPARM, GETTIM, GETTOD, GMTIME, IBTOD, IDATE, IETOM, IOSTAT_MSG, ITIME, IVALUE, LTIME, MTOIE, PERROR, PRECFILL, PRNSET, PROMPT, QSORT, REDLEN, SETBIT, SETRCD, SLEEP, SLITE, SLITET, TIMER 
• The following service functions: ACCESS, ALARM, BIT, CHDIR, CHMOD, CTIME, DRAND, DTIME, ETIME, FGETC, FORK, FPUTC, FSEEK, FSEEKO64, FSTAT, FSTAT64, FTELL, FTELLO64, GETC, GETCWD, GETFD, GETGID, GETPID, GETUID, HOSTNM, IARGC, IERRNO, INMAX, IOINIT, IRAND, ISATTY, JDATE, KILL, LINK, LNBLNK, LONG, LSTAT, LSTAT64, MALLOC, NARGS, PUTC, RAN, RAND, RENAME, RINDEX, RTC, SECNDS, SECOND, SH, SHORT, SIGNAL, STAT, STAT64, SYMLNK, SYSTEM, TCLOSE, TIME, TIMEF, TTYNAM, UNLINK, WAIT 

Free Technical Support
LF64 Linux Express includes e-mail technical support at no extra charge. 

Automatic Parallelization
The LF64 compiler automatically parallelizes DO loops and array operations without you having to make modifications to the program. This makes it easy to migrate source programs to other platforms (as long as the program conforms with the Fortran Standard). The effect is to save elapsed execution time by using two or more CPUs simultaneously. For instance, if a DO loop can be executed in parallel by dividing it in half, then, theoretically, the execution time of this DO loop may be cut in half. In practice, improving performance requires some care and some work on the part of the programmer. During compilation, the auto-parallel function will return information regarding which processes were (and which were not) parallelized and why. While certain loops can be analyzed sufficiently to be parallelized by the compiler without input from the programmer, many loops have data dependencies that prevent automatic parallelization because of the potential for incorrect results. For that reason, LF64 PRO also includes optimization control lines (OCLs) that provide information necessary for the compiler to parallelize these otherwise unparallelizable loops. The OCLs are Fortran comments in a particular format, for example:

!OCL PARALLEL

Note that programs with OCLs are standard-conforming and can be compiled with other compilers that do not support OCLs.

Four compiler switches control automatic parallelization: --parallel, --threads, --threadstack, and --ocl. Details of automatic parallelization (loop slicing, interchange, distribution, fusion, and reduction, as well as OCL syntax and specifiers) are documented in the LF64 User's Guide.

OpenMP v2.0 Compatibility
OpenMP specifies a set of compiler directives, library routines, and environment variables for shared-memory parallelism in Fortran and C/C++ programs. LF64 PRO v8.1 supports the OpenMP v2.0 specification for Fortran. Like automatic parallelization, OpenMP directives are used to parallelize a program that runs on a computer with more than one processor. With OpenMP you have more control over how code is parallelized, but also more coding to do.

The LF64 Linux PRO v8.1 includes the OpenMP v2.0 Fortran specification in PDF.

Winteracter Starter Kit
Use the Winteracter Starter Kit - WiSK - for creating true X/Windows programs with Fortran. WiSK is a subset of the X/Winteracter Library created by Interactive Software Services, Ltd. (X/Winteracter is available from Lahey.) X/Winteracter is a Fortran 95-callable, 64-bit, X/Windows, user-interface and graphics development kit. Derived from X/Winteracter, WiSK provides a library of subroutines for window management, input handling, dialog management, and high resolution graphics. Designed for use with X11R6 and Open Motif 2.2.

X/Winteracter offers a wide range of powerful GUI capabilities to the Fortran 9x developer under X Windows, including:

• Multiple windows.
• Memory bitmap manipulation & bitmap viewer windows.
• Text editor windows, with optional command lines.
• Event handling.
• Text based menus, including floating menus (toolbars have still to be implemented).
• Dialog handling, including tabbed dialogs and nearly all control types (grid controls are the current exception).
• Common dialogs, e.g. file selector, message box, etc.
• Presentation graphics.
• and more!

Below are examples of WiSK's capabilities.



 

BLAS and LAPACK
BLAS is a library for vector and matrix operations. The BLAS thread-safe version is based on BLAS provided on Netlib. BLAS includes 57 functions. The total number of routines for all precision types amounts to approximately 170.

BLAS thread-safe version provides the following routines:
Level 1 BLAS : Vector operations
Level 2 BLAS : Matrix and vector operations
Level 3 BLAS : Matrix and matrix operations
Sparse-BLAS : Sparse vector operations

The thread-safe implementation of BLAS has exactly the same subroutine names and calling parameters as those of the Netlib baseline version.

Differences include:

• the thread-safe version can be used in the environment of SMP (Symmetric Multiple Processing)
• subroutines of the thread-safe version can be called from an OpenMP Fortran program

• Linear equations
• Linear least squares problems
• Eigenvalue problems
• Singular value decomposition

Fujitsu Scientific Subroutine Library 2
The Fujitsu Scientific Subroutine Library 2 (SSL2) has been in use for years in Japan on Fujitsu mainframe and workstation hardware. SSL2 offers over 250 optimized thread-safe routines in the following areas:

Linear Algebra
Matrix Storage Mode Conversion
Matrix Manipulation
Linear Equations and Matrix Inversion (Direct Method)
Least Squares Solution

Eigenvalues and Eigenvectors
Eigenvalues and Eigenvectors of a Real Matrix
Eigenvalues and Eigenvectors of a Complex Matrix
Eigenvalues and Eigenvectors of a Real Symmetric Matrix
Eigenvalues and Eigenvectors of a Hermitian Matrix
Eigenvalues and Eigenvectors of a Real Symmetric Band Matrix
Eigenvalues and Eigenvectors of a Real Symmetric Generalized Eigenproblem
Eigenvalues and Eigenvectors of a Real Symmetric Band Generalized Eigenproblem

Nonlinear Equations
Polynomial Equations
Transcendental Equations
Nonlinear Simultaneous Equations

Extrema
Minimization of Function with a Variable
Unconstrained Minimization of Multivariable Function
Unconstrained Minimization of Sum of Squares of Functions (Nonlinear Least Squares Solution)
Linear Programming
Nonlinear Programming (Constrained Minimization of Multivariable Function)

Interpolation and Approximation
Interpolation
Approximation
Smoothing
Series

Transforms
Discrete Real Fourier Transforms
Discrete Cosine Transforms
Discrete Sine Transforms
Discrete Complex Fourier Transforms 
Laplace Transform

Numerical Differentiation and Quadrature

Differential Equations

Special Functions
Elliptic Integrals
Exponential Integral
Sine and Cosine Integrals
Fresnel Integrals
Gamma Functions
Error Functions
Bessel Functions 
Normal Distribution Functions

Pseudo Random Numbers
Pseudo Random Generation
Pseudo Random Testing  

• Intel© EM64T or AMD© AMD64 64-bit processor.
• 32 MB of RAM.
• 70 MB of available hard disk space for LF64 Linux PRO; 40 MB for LF64 Linux Express.

• X-Windows to useWiSK and view the online PDF documentation.
• 64-bit version of as, the GNU assembler.
• 64-bit version of ld, the GNU linker.
• 64-bit versions of C startup and support object files crt1.o, crti.o, crtn.o, crtbegin. o, and crtend.o.
• 64-bit versions of C runtime and support libraries libc, libm, libpthread, librt, libgcc, libgcc_eh, libgcc_s, and libelf.
• A compatible version of the Linux operating system. Table 1 shows the versions of Linux that are known to be compatible with LF64. Other Linux variants might be compatible if they include kernel version 2.6.9 or later and libc version 2.3.4 or later.