Home

Fortran User's Guide - Physics at Oregon State University

1. Following are the restrictions no warnings are issued m Optimization must be 03 or greater a The source for the routine must be in the file being compiled unless xcrossfile is also specified m The compiler determines if actual inlining is profitable and safe Chapter 3 Fortran Compiler Options 69 70 The appearance of inline with 04 disables the automatic inlining that the compiler would normally perform unless auto is also specified With 04 the compilers normally try to inline all appropriate user written subroutines and functions Adding inline with 04 may degrade performance by restricting the optimizer s inlining to only those routines in the list In this case use the sauto suboption to enable automatic inlining at 04 and 05 demo 95 04 inline auto no zpoint f In the example above the user has enabled 04 s automatic inlining while disabling any possible inlining of the routine zpoint that the compiler might attempt Kpic Synonym for pic 77 95 KPIC Synonym for PIC 77 95 Ldir Add dir to list of directories to search for libraries 77 95 Adds dir to the front of the list of object library search directories A space between L and dir is optional This option is passed to the linker See also 1x on page 71 While building the executable file ld 1 searches dir for archive libraries a files and shared libraries so files 1
2. 32 Compiler Usage Tips The next sections suggest a number of ways to use the Sun Fortran compilers efficiently A complete compiler options reference follows in the next chapter Determining Hardware Platform Some compiler flags allow the user to tune code generation to a specific set of hardware platform options The utility command fpversion displays the hardware platform specifications for the native processor demo fpversion A SPARC based CPU is available CPU s clock rate appears to be approximately 467 1 MHz Kernel says CPU s clock rate is 480 0 MHz Kernel says main memory s clock rate is 120 0 MHz Sun 4 floating point controller version 0 found An UltraSPARC chip is available FPU s frequency appears to be approximately 492 7 MHz Use xtarget ultra2i xcache 16 32 1 2048 64 1 code generation option Hostid hardware_host_id The values printed depend on the load on the system at the moment fpversion is called See fpversion 1 and the Numerical Computation Guide for details Using Environment Variables You can specify options by setting the FFLAGS or OPTIONS variables Fortran User s Guide July 2001 Either FFLAGS or OPTIONS can be used explicitly in the command line When you are using the implicit compilation rules of make FFLAGS is used automatically by the make program Example Set FFLAGS C Shell demo setenv FFLAGS fast Xlist Example Use FFLAGS
3. copyargs Allow assignment to constant arguments 77 95 Allow a subprogram to change a dummy argument that is a constant This option is provided only to allow legacy code to compile and execute without a runtime error a Without copyargs if you pass a constant argument to a subroutine and then within the subroutine try to change that constant the run aborts a With copyargs if you pass a constant argument to a subroutine and then within the subroutine change that constant the run does not necessarily abort Code that aborts unless compiled with copyargs is of course not Fortran standard compliant Also such code is often unpredictable 50 Fortran User s Guide July 2001 Dname def Define symbol name for the preprocessor 77 95 This option only applies to F F90 and F95 source files Dname def Define name to have value def Dname Define name to be 1 On the command line this option will define name as if define name def had appears in the source file If no def specified the name name is defined as the value 1 The macro symbol name is passed on to the preprocessor fpp or cpp see the xpp option for expansion The predefined macro symbols have two leading underscores The Fortran syntax may not support the actual values of these macros they should appear only in fpp or cpp preprocessor directives m The product version is predefined in hex in ___SUNPRO_F77 ___SUNPRO_F9
4. 95 Fortran 95 Same action as f f95 only 90 free format for Fortran 77 or Same action as f Fortran 95 20 Fortran User s Guide July 2001 TABLE 2 1 File Name Suffixes Recognized by Sun Fortran Compilers Continued Suffix Language Action F Fortran 77 or Apply the Fortran or C preprocessor to the Fortran 77 source Fortran 95 file before compilation fixed format F95 Fortran 95 Apply the Fortran or C preprocessor to the Fortran 95 free F90 free format format source file before Fortran compiles it f95 only sS Assembler Assemble source files with the assembler SS Assembler Apply the C preprocessor to the assembler source file before assembling it FLE Inline Process template files for inline expansion The compiler will expansion use templates to expand inline calls to selected routines Template files are special assembler files see the inline 1 man page JO Object files Pass object files through to the linker a S Libraries Pass names of libraries to the linker a files are static libraries Mer so and so n files are dynamic libraries SO n Fortran 95 free format is described in Appendix C of this manual Source Files The Fortran compilers will accept multiple source files on the command line A single source file also called a compilation unit may contain any number of procedures main program subroutine function block data module and so on Applications may be config
5. WORKSHARE Directive SOMP WORKSHARE block of Fortran statements SOMP END WORKSHARE NOWAIT Divides the work of executing the enclosed code block into separate units of work and causes the threads of the team to share the work such that each unit is executed only once Appendix E Fortran Directives Summary 181 182 TABLE E 5 Summary of OpenMP Directives in Fortran 95 Continued PARALLEL DO Directive SOMP PARALLEL DO clause clause do_loop statements block SOMP END PARALLEL DO Shortcut for specifying a parallel region that contains a single DO loop a PARALLEL directive followed immediately by a DO directive clause can be any of the clauses accepted by the PARALLEL and DO directives PARALLEL SECTIONS Directive SOMP PARALLEL SECTIONS clause clause SOMP SECTION block of Fortran statements with no transfer in or out SOMP SECTION optional block of Fortran statements SOMP END PARALLEL SECTIONS Shortcut for specifying a parallel region that contains a single SECTIONS directive a PARALLEL directive followed by a SECTIONS directive clause can be any of the clauses accepted by the PARALLEL and SECTIONS directives PARALLEL WORKSHARE Directive SOMP PARALLEL WORKSHARE clause clause block of Fortran statements SOMP END PARALLEL WORKSHARE Provides a shortcut for specifying a parallel region that contains a single WORKSHARE directive clause can be one of the cl
6. Backward Compatibility and Legacy Options The following options are provided for backward compatibility with earlier compiler releases and certain Fortran legacy capabilities TABLE 3 5 Backward Compatibility Options Action Option Double default data sizes use xt ypemap instead r8 or dbl Allow assignment to constant arguments copyargs Treat hollerith constant as character or typelessin xhasc yes no call argument lists External names make external names without ext_names e underscores Nonstandard arithmetic allow nonstandard fnonstd arithmetic Optimize performance for the host system native Output use old style list directed output oldldo DO loops use one trip DO loops onetrip Compile for SPARC V7 architecture cg89 Compile for SPARC V8 architecture cg92 Use of these option flags is not recommended and should be avoided 44 Fortran User s Guide July 2001 Obsolescent Options The following options are no longer supported by the 77 and 95 compilers Their appearance on a compiler command does not cause an error and no action is taken they are ignored TABLE 3 6 Obsolescent Options Original Intention Option Compile for analysis by looptool Zlp Compile for Thread Analyzer Ztha Disable exception traps 95 fnonstop Options Reference This section shows all 77 and 95 compiler command line option flags including various risks restrictions caveats
7. DYNAMIC Scheduling SCHEDULE DYNAMIC chunk Iterations are broken into pieces of a size specified by chunk As each thread finishes a piece of the iteration space it dynamically obtains the next set of iterations GUIDED Scheduling SCHEDULE GUIDED chunk With GUIDED the chunk size is reduced in an exponentially decreasing manner with each dispatched piece of the iterations chunk specifies the minimum number of iterations to dispatch each time Default chunk size is 1 The size of the initial piece of the iterations is the number of iterations in the loop divided by the number of threads executing the loop RUNTIME Scheduling SCHEDULE RUNTIME Scheduling is deferred until runtime Schedule type and chunk size will be determined from the setting of the OMP_SCHEDULE environment variable Default is STATIC 186 Fortran User s Guide July 2001 OpenMP Library Routines OpenMP Fortran API library routines are external procedures In the following summary int_expr is a default scalar integer expression and logical_expr is a default scalar logical expression OMP_ functions returning INTEGER 4 and LOGICAL 4 are not intrinsic and must be declared properly otherwise the compiler will assume REAL Interface declarations for the OpenMP Fortran runtime library routines summarized below are provided by the Fortran include file omp_lib h and a Fortran 95 MODULE omp_lib as described in the Fortran
8. To Set Your MANPATH Environment Variable to Enable Access to Sun WorkShop Man Pages If you are using the C shell edit your home cshrc file If you are using the Bourne shell or Korn shell edit your home profile file Add the following to your MANPATH environment variable opt SUNWspro man Accessing Sun WorkShop Documentation You can access Sun WorkShop product documentation at the following locations The product documentation is available from the documentation index installed with the product on your local system or network Point your Netscape Communicator 4 0 or compatible Netscape version browser to the following file opt SUNWspro docs index html Before You Begin 5 If your product software is not installed in the opt directory ask your system administrator for the equivalent path on your system m Manuals are available from the docs sun com Web site The docs sun com Web site http docs sun com enables you to read print and buy Sun Microsystems manuals through the Internet If you cannot find a manual see the documentation index installed with the product on your local system or network Accessing Related Documentation The following table describes related documentation that is available through the docs sun com Web site Document Collection Forte for High Document Title Description Discusses issues relating to Performance Computing Collection Fortr
9. WEAK Directive CSPRAGMA SUN UNROLL n Advises the compiler that the following loop can be unrolled to a length n CSPRAGMA WEAK name name2 Declares name to be a weak symbol or an alias for name2 OPT Directive NOMEMDEP Directive CSPRAGMA SUN OPT n Set optimization level for a subprogram to n CSPRAGMA SUN NOMEMDEP Assert there are no memory dependencies in the following loop Requires parallel or explicitpar PIPELOOP Directive PREFETCH Directives Fortran User s Guide July 2001 CSPRAGMA SUN PIPELOOP n Assert dependency in loop between iterations n apart CSPRAGMA SPARC_PREFETCH_READ_ONCE name CSPRAGMA SPARC_PREFETCH_READ_MANY name CSPRAGMA SPARC_PREFETCH_WRITE_ONCE name CSPRAGMA SPARC_PREFETCH_WRITE_MANY name Request compiler generate prefetch instructions for references to name Requires xprefetch option Special Fortran 95 Directives The following directives are only available with 95 See Appendix C for details TABLE E 2 Special Fortran 95 Directives Format DIRS directive initial line IDIR amp continuation line With fixed format source C is also accepted as a directive indicator CDIRS directive the line must start in column 1 WIth free format source the line may be preceded by blanks FIXED FREE DIRS FREE Directives DIRS FIXED These directives specify the source format of the lines following the direct
10. fpp A Fortran source code preprocessor See fpp 1 m fsplit This utility splits one Fortran file of several routines into several files each with one routine per file Use sp1it on FORTRAN 77 or Fortran 95 source files See fsp1it 1 Debugging Utilities The following debugging utilities are available m error f77 only A utility to merge compiler error messages with the Fortran source file This utility is included if you do a developer install rather than an end user install of Solaris it is also included if you install the SUNWbtool package m Xlist A compiler option to check across routines for consistency of arguments COMMON blocks and so on a Sun WorkShop Provides a visual debugging environment based on dbx and includes a data visualizer and performance data collector Chapter1 Introduction 11 Sun Performance Library The Sun Performance Library is a library of optimized subroutines and functions for computational linear algebra and Fourier transforms It is based on the standard libraries LAPACK BLAS1 BLAS2 BLAS3 FFTPACK VFFTPACK and LINPACK generally available through Netlib www netlib org Each subprogram in the Sun Performance Library performs the same operation and has the same interface as the standard library versions but is generally much faster and accurate and can be used in a multiprocessing environment See the performance_library README file and the Sun Performa
11. g sb Produce table information for the Sun WorkShop source code browser 77 95 See Using Sun WorkShop for more information Note sb cannot be used on source files the compiler automatically passes through the fpp or cpp preprocessors that is files with F F90 or F95 extensions or used with the F option sbfast Produce only source code browser tables 77 95 Produce only table information for the Sun WorkShop source code browser and stop Do not assemble link or make object files Note sbfast cannot be used on source files the compiler automatically passes through the fpp or cpp preprocessors that is files with F F90 or F95 extensions or used with the F option Chapter 3 Fortran Compiler Options 87 88 silent Suppress compiler messages 77 95 Use this option to suppress non essential messages from the compiler error and warning messages are still issued The default is to show file and entry names as they are reached during the compilation stackvar Force all local variables to be allocated on the memory stack 77 95 Allocate on the memory stack all the local variables and arrays in routines unless otherwise specified This option makes these variables automatic rather than static and provides more freedom to the optimizer when parallelizing loops with calls to subprograms Use of stackvar is recommended with any of the parallelization options Variables a
12. or REWIND 138 RECL specifier is zero or negative 139 no record length specified for direct access fil 40 unexpected STATUS specifier 41 status is specified and not OLD for connected unit L42 STATUS specifier is not KEEP or DELETE L43 status KEEP specified for a scratch file 44 impossible status value L45 a file name has been specified for a scratch file 46 attempting to open a unit that is being read from or written to 47 attempting to close a unit that is being read from or written to 48 attempting to open a directory 49 status is OLD and the file is a dangling symbolic link 150 status is NEW and the file is a symbolic link 151 no free scratch file names 152 specifier ACCESS STREAM for default unit Fortran User s Guide July 2001 TABLE A 2 95 Runtime I O Messages Continued Error Message 1153 stream access to default unit 161 device does not support REWIND 1162 read permission required for BACKSPACE 163 BACKSPACE on direct access unit 1164 BACKSPACE on binary unit 165 nd of file seen while backspacing 1166 write permission required for ENDFILE 167 ENDFILE on direct access unit 168 stream access to sequential or direct access unit 1169 stream access to unconnected unit 170 direct access to stream access unit 171 incorrect value of POS specifier 172 unassociated ASYNCHRONOUS specifier 11713 unasso
13. procedure pg gprof 82 profiling xprofile 116 Q quick start 17 R range of subscripts 49 README file 14 105 RECURSIVE attribute 118 register usage 118 release history 137 reorder functions 104 rounding 63 64 S S 87 sb SourceBrowser 87 200 Fortran User s Guide July 2001 search modules 72 object library directories 70 set include path 67 shared libraries global offset table 82 shared library build G 66 disallow linking dn 55 name a shared library 67 position independent code 82 pure no relocations 123 shell limits 34 shell prompts 3 SIGFPE floating point exception 61 signal handler 126 silent options 45 size of compiled code 120 Solaris versions supported 3 source file preprocessing 21 source format mixing format of source lines f95 148 options f95 148 source lines extended 55 fixed format 61 free format 63 line length 147 preprocessor 113 preserve case 90 SourceBrowser 87 SPARC platform cache 99 chip 100 code address space 102 instruction set architecture 97 register usage xregs 118 specify target platform xtarget 120 xtarget expansions 169 stack increase stack size 89 overflow 89 stackvar 88 standard include files 68 standards conformance 9 identify non ANSI extensions ansi flag 47 statement profile by a and tcov 46 static binding 55 STOP statement return status 89 stric
14. 77 and 95 Fortran 95 New Features and Changes This section lists the new features and behavior changes introduced with this release of 95 and previous releases 95 New Features in Sun WorkShop 6 update 2 The following lists new and changed features in the Fortran 95 compiler released with Sun WorkShop 6 update 2 m ALLOCATABLE Attribute Extended Recent decisions by the Fortran 95 standards organizations have extended the data entities allowed for the ALLOCATABLE attribute Previously this attribute was limited to locally stored array variables It is now allowed on array components of structures dummy arrays array function results Allocatable entities remain forbidden in all places where they may be storage associated COMMON and EQUIVALE NCE components may appear in S prohibited from COMMON and EQUE EQUIVA NCE statements Allocatable array types but objects of such types are then ENCE See Appendix C page 157 137 138 a VALUE Attribute from Fortran 2000 95 recognizes the VALUE type declaration attribute Specifying a subprogram dummy input argument with this attribute indicates that the actual argument is passed by value See Appendix C page 157 a OpenMP 2 0 Fortran API Supported 95 now supports the OpenMP 2 0 API specifications for Fortran 95 Enhancements include WORKSHARE REDUCTION for arrays
15. 95 compiler Chapter 2 Using Sun Fortran Compilers 19 In the example below 95 is used to compile two source files to produce an executable file named growth with runtime debugging enabled demo 95 g o growth growth f fft f95 Note You can invoke the Sun WorkShop 6 Fortran 95 compiler with either the 95 or 90 command 90 is now an alias for 95 Compile Link Sequence In the previous example the compiler automatically generates the loader object files growth o and fft o and then invokes the system linker to create the executable program file growth After compilation the object files growth o and fft o will remain This convention permits easy relinking and recompilation of files If the compilation fails you will receive a message for each error No o files are generated for those source files with errors and no executable program file is written Command Line File Name Conventions The suffix extension attached to file names appearing on the command line determine how the compiler will process the file File names with a suffix extension other than one of those listed below or without an extension are passed to the linker TABLE 2 1 File Name Suffixes Recognized by Sun Fortran Compilers Suffix Language Action af Fortran 77 or Compile Fortran source files put object files in current Fortran 95 directory default name of object file is that of the source but fixed format with o suffix
16. A compiler directive is inserted into the source program as one or more lines of text Each line looks like a comment but has additional characters that identify it as more than a comment for this compiler For most other compilers it is treated as a comment so there is some code portability Sun style directives are the default with 95 and 77 To switch to Cray style directives use the mp cray compiler command line flag A complete summary of Fortran directives appears in Appendix E Form of Special 95 Directive Lines f95 recognizes its own special directives in addition to the general 95 77 directives described in Chapter 2 These have the following syntax DIRS d1 d2 Fixed Form Source Put CDIR or DIR in columns 1 through 5 Directives are listed in columns 7 and beyond Columns beyond 72 are ignored An initial directive line has a blank in column 6 A continuation directive line has a nonblank in column 6 Free Form Source m Put DIRS followed by a space anywhere in the line The DIRS characters are the first nonblank characters in the line actually non whitespace m Directives are listed after the space a An initial directive line has a blank tab or newline in the position immediately after the DIRS Fortran User s Guide July 2001 a A continuation directive line has a character other than a blank tab or newline in the position immediately after the DIRS Thus DIR in
17. All the threads in the team share the variables that appear in list and access the same storage area DEFAULT PRIVATE SHARED NONE Specify scoping attribute for all variables within a parallel region THREADPRIVATE variables are not affected by this clause If not specified DEFAULT SHARED is assumed FIRSTPRIVATE Clause FIRSTPRIVATE list Variables on list are PRIVATE In addition private copies of the variables are initialized from the original object existing before the construct LASTPRIVATE Clause LASTPRIVATE list Variables on the list are PRIVATE In addition when the LASTPRIVATE clause appears on a DO directive the thread that executes the sequentially last iteration updates the version of the variable before the construct On a SECTIONS directive the thread that executes the lexically last SECTION updates the version of the object it had before the construct REDUCTION Clause REDUCTION operator intrinsic list operator is one of AND OR EQV NEQV intrinsic is one of MAX MIN IAND IOR IEOR Variables in list must be named variables of intrinsic type The REDUCTION clause is intended to be used on a region in which the reduction variable is used only in reduction statements of the form shown previously for the ATOMIC directive Variables on list must be SHARED in the enclosing context A private copy of each variable is created for each thread as if it were PRIVATE At t
18. Fortran 95 does not accept the BYTE data type Fortran 95 does not allow non integers to be used as array subscripts 95 does not allow relational operators EQ and NE to be used with logical operands Appendix C Fortran 95 Features and Differences 163 164 a Programs Subroutines Functions Statements a The PROGRAM statement requires a name in Fortran 95 a The 95 maximum length for names is 31 characters a Functions in Fortran 95 cannot be called by a CALL statement as if they were subroutines a Functions in Fortran 95 must have their return value defined a While 77 allows mixed argument types to appear in some specific intrinsic functions 95 does not 95 does not recognize debugging comments comment lines with D in column one a Tab formatting in Fortran 95 does not allow source lines to extend beyond column 72 a 95 tab formatting will pad character strings to column 72 if they extend over a continuation line See page 148 Command line Options 95 does not recognize vax compiler options I O Compatibility 77 and 95 are generally I O compatible for binary I O since 95 links to the 77 compatibility library Such compatibility includes the following two situations m In the same program you can write some records in 95 then read them in 77 a An 95 program can write a file Then an 77 program can read it The numbers read back
19. SINGLE equivalent to Sun style SELF 1 CHUNKSIZE n equivalent to Sun style SELF n NUMCHUNKS m equivalent to Sun style SELF n m The default scheduling is Sun style STATIC for which there is no Cray style equivalent Interpretations of these scheduling qualifiers differ between Sun and Cray style Check the Fortran Programming Guide for details TASKCOMMON CMICS TASKCOMMON block_name Directive Declares variables in the named common block as thread private private to a thread but global within the thread Declaring a common block TASKCOMMON requires that this directive appear immediately after every common declaration of that block DOSERIAL Directive CMICS DOSERIAL Disables parallelization of the loop that follows DOSERIAL Directive CMICS DOSERIAL Disables parallelization of the loop nest that follows Appendix E Fortran Directives Summary 179 Fortran 95 OpenMP Directives The Sun Fortran 95 compiler supports the OpenMP 2 0 Fortran API The openmp compiler flag enables these directives See page 79 This section lists the OpenMP directives library routines and environment variables supported by 95 For details about parallel programming with OpenMP see the OpenMP 2 0 Fortran specification at http www openmp org The following table summarizes the OpenMP directives supported by 95 Items enclosed in square brackets are optional The compiler permits comments to follow an excl
20. executing in parallel and FALSE otherwise OMP_SET_DYNAMIC Subroutine SUBROUTINE OMP_SET_DYNAMIC logical_expr Enables or disables dynamic adjustment of the number of threads available for parallel execution of programs Dynamic adjustment is enabled by default OMP_GET_DYNAMIC Function LOGICAL 4 FUNCTION OMP_GET_DYNAMIC Returns TRUE if dynamic thread adjustment is enabled and returns FALSE otherwise OMP_SET_NESTED Subroutine SUBROUTINE OMP_SET_NESTED logical_expr Enables or disables nested parallelism Nested parallelism is disabled by default Nested parallelism is not supported OMP_GET_NESTED Function LOGICAL 4 FUNCTION OMP_GET_NESTED Returns TRUE if nested parallelism is enabled FALSE otherwise Nested parallelism is not supported this function will always return FALSE Lock Routines Two types of locks are supported simple locks and nestable locks Nestable locks may be locked multiple times by the same thread before being unlocked simple locks may not be locked if they are already in a locked state Simple lock variables may only be passed to simple lock routines and nested lock variables only to nested lock routines The lock variable var must be accessed only through these routines Use the parameters OMP_LOCK_KIND and OMP_NEST_LOCK_KIND defined in omp_lib h INCLUDE file and the omp_1lib MODULE for this purpose For example INTEGER KIND OMP_LOCK_KIND var
21. must be listed explicitly on the command line for the link step Typically programmers define one MODULE per file and assign the same name to the MODULE and the source file containing it However this is not a requirement The mod files cannot be stored into an archive file or concatenated into a single file Appendix C Fortran 95 Features and Differences 167 Example demo cat mod_one f90 ODULE one END MODULE demo cat mod_two f90 ODULE two END MODULE demo cat main 90 USE one USE two END demo 95 o main mod_one f 90 mod_two f90 main 90 In this example all the files are compiled at once The module source files appear first before their use in the main program Compilation creates the files main main o one mod mod_one o two mod mod_two o The next example compiles each unit separately and links them together 95 c mod_one f 90 mod_two f90 95 c main 90 95 o main main o mod_one o mod_two o When compiling main 90 the compiler searches the current directory for one mod and two mod These must be compiled before compiling any files that reference the modules on USE statements The link step requires the module implementation object files mod_one o and mod_two o appear along with all other object files to create the executable 168 Fortran User s Guide July 2001 APPENDIX D xtarget Platform Expansions This Appendix details the xtar
22. program unit is compiled with the option and another is not references to what should be the same location within the common block might reference different locations m If pad common is specified the declarations of common block variables in different program units must be the same except for the names of the variables The amount of padding inserted between variables in a common block depends on the declarations of those variables If the variables differ in size or rank in different program units even within the same file the locations of the variables might not be the same m If pad common is specified EQUIVALENCE declarations involving common block variables are flagged with a warning message and the block is not padded m Avoid overindexing arrays in common blocks with pad common specified The altered positioning of adjacent data in a padded common block will cause overindexing to fail in unpredictable ways parallel Parallelize with autopar explicitpar depend Fortran parallelization features require a Forte for HPC license 77 95 Parallelize loops chosen automatically by the compiler as well as explicitly specified by user supplied directives Optimization level is automatically raised to 03 if it is lower To improve performance also specify the stackvar option when using any of the parallelization options including autopar Use mp page 73 to select Sun Cray or 95 OpenMP style para
23. the x1 option forces the compiler to interpret it the VMS way This option enables the following VMS language features Unformatted record size in words rather than bytes x1 VMS style logical file names x1 Quote character introducing octal constants x1 Backslash as ordinary character within character constants x1 Nonstandard form of the PARAMETER statement x1 Alignment of structures as in VMS x1 Debugging lines as comment lines or Fortran statements x1d Use x1 to get VMS alignment if your program has some detailed knowledge of how VMS structures are implemented Use xld to cause compilation of debugging comments D or d in column one Without the x1d option they remain comments only There is no space between xl and d Programs that share structures with C routines should not use x1 See the Fortran Library Reference for information on the VMS libraries See also the chapter on VMS language extensions in the Fortran 77 Language Reference that the 77 compiler automatically recognizes xlang pl 95 Prepare for linking with runtime libraries for programming language pl For 95 only xlang 77 is allowed 95 xlang f 77 implies linking with the 77compat library and is a shorthand way for linking Fortran 95 object files with Fortran 77 object files that insures the proper runtime environment Chapter 3 Fortran Compiler Options 109 Use 95 xlang 77 when lin
24. 1 Appendix D xtarget Platform Expansions 171 TABLE D 1 xtarget Expansions Continued xtarget xarch xchip xcache ss600 512 v8 super 16 32 4 1024 32 1 ss600 514 v8 super 16 32 4 1024 32 1 ss600 61 v8 super 16 32 4 1024 32 1 ss600 612 v8 super 16 32 4 1024 32 1 sselc v7 old 64 32 1 ssipc v7 old 64 16 1 ssipx v7 old 64 32 1 sslc v8a micro 2 16 1 sslt v7 old 64 32 1 sslx v8a micro 2 16 1 sslx2 v8a micro2 8 16 1 ssslc v7 old 64 16 1 ssvyger v8a micro2 8 16 1 sun4 110 v7 old 2 16 1 sun4 15 v8a micro 2 16 1 sun4 150 v7 old 2 16 1 sun4 20 v7 old 64 16 1 sun4 25 v7 old 64 32 1 sun4 260 v7 old 128 16 1 sun4 280 v7 old 128 16 1 sun4 30 v8a micro 2 16 1 sun4 330 v7 old 128 16 1 sun4 370 v7 old 128 16 1 sun4 390 v7 old 128 16 1 sun4 40 v7 old 64 16 1 sun4 470 v7 old 128 32 1 sun4 490 v7 old 128 32 1 sun4 50 v7 old 64 32 1 sun4 60 v7 old 64 16 1 172 Fortran User s Guide July 2001 TABLE D 1 xtarget Expansions Continued xtarget xarch xchip xcache sun4 630 v7 old 64 32 1 sun4 65 v7 old 64 16 1 sun4 670 v7 old 64 32 1 sun4 690 v7 old 64 32 1 sun4 75 v7 old 64 32 1 ultra v8plusa ultra 16 32 1 512 64 ultral 140 v8plusa ultra 16 32 1 512 64 ultral 170 v8plusa ultra 16 32 1 512 64 1 ultral 200 v8plusa ultra 16 32 1 512 64 ultra2 v8plusa ultra2 16 32 1 512 64 ultra2 1170 v8plusa ultra 16 32 1 512 64 1 ultra2 1200 v8
25. 118 xs 119 xsafe mem 119 xsb 120 xsbfast 120 xspace 120 xtarget t table 169 xtarget t 120 xtime 121 xtypemap and db1 53 85 xtypemap 121 xunroll 122 xvector 122 xvpara 123 Z1p loop profiler obsolete 123 ztext 123 OPTIONS environment variable 32 order of functions 104 order of processing options 39 overflow 65 stack 89 P padding 80 parallelization and stackvar 88 automatic 48 automatic and explicit parallel 81 directives 77 29 explicit 57 loop information 72 messages 93 OpenMP 30 79 OpenMP directives summarized 180 reduction operations 86 select directives style 73 with multithreaded libraries 74 See also Fortran Programming Guide passes of the compiler 92 path include 67 dynamic libraries in executable 84 library search 70 modules search 72 to standard include files 68 PATH environment variable setting 3 performance optimization 59 Sun Performance Library 12 Index 199 performance library 110 PIPELOOP directive 28 pointee 153 pointer 153 position independent code 82 83 102 pragma See directives PREFETCH directive 29 114 preprocessor source file define symbol 51 force fpp 63 fpp cpp 21 specify with xpp p 113 undefine symbol 91 preserve case 90 print asa 11 processor specify target processor 100 processor specify target 120 prof p 80 profile by basic block 46
26. 22 2 0 All repeated values output explicitly 8 22 8 22 8 22 1 2 No trailing zero printing integer floating point 1 2 0 Floating point integers print with trailing zero 1 0 1 2 Value printed may not be the same value when read into a variable with the same type 0 1 when read in will print as 0 100000001 2 0 Prints the minimum number of digits required to ensure that a value written produces the same value when read back in 0 1 prints as 0 1 1 2 As required by the standard zero value prints in exponent form But 1 2 prints 0 E 0 2 0 Prints zero as 0 0E 0 1 2 Prints a space between the comma and the imaginary part of a complex value 1 0 E 0 2 0 No comma 1 0 0 0E 0 a NAMELIST Input Format 2 0 Allow the group name to be preceded by or amp on input The amp is the only form accepted by the Fortran 90 standard and is what is written by NAMELIST output 2 0 Accepts as the symbol terminating input except if the last data item in the group is CHARACTER in which case it is treated as input data 2 0 Allows NAMELIST input to start in the first column of a record PRINT no longer comma delimits output OPEN FORM BINARY permits I O of non standard raw text without record marks Opening a file with FORM BINARY has roughly the same effect as FORM UNFORMATTED except that no record lengths are embedded in the file Without this data there is no way to tell where one record begins or ends Thus
27. 512 v8 super 16 32 4 1024 32 1 ss10 514 v8 super 16 32 4 1024 32 1 ss10 61 v8 super 16 32 4 1024 32 1 ss10 612 v8 super 16 32 4 1024 32 1 ss10 71 v8 super2 16 32 4 1024 32 1 ss10 712 v8 super2 16 32 4 1024 32 1 ssl10 hs11 v8 hyper 256 64 ssl10 hs12 v8 hyper 256 64 ss10 hs14 v8 hyper 256 64 1 ssl10 hs21 v8 hyper 256 64 ssl10 hs22 v8 hyper 256 64 ss1000 v8 super 16 32 4 1024 32 1 sslplus v7 old 64 16 1 ss2 v7 old 64 32 1 170 Fortran User s Guide July 2001 TABLE D 1 xtarget Expansions Continued xtarget xarch xchip xcache ss20 v8 super 16 32 4 1024 32 1 ss20 151 v8 hyper 512 64 1 s20 152 v8 hyper 512 64 1 s20 50 v8 super 16 32 4 ss20 502 v8 super 16 32 4 ss20 51 v8 super 16 32 4 1024 32 1 ss20 512 v8 super 16 32 4 1024 32 1 ss20 514 v8 super 16 32 4 1024 32 1 ss20 61 v8 super 16 32 4 1024 32 1 ss20 612 v8 super 16 32 4 1024 32 1 ss20 71 v8 super2 16 32 4 1024 32 1 ss20 712 v8 super2 16 32 4 1024 32 1 ss20 hs11 v8 hyper 256 64 ss20 hs12 v8 hyper 256 64 ss20 hs14 v8 hyper 256 64 1 ss20 hs21 v8 hyper 256 64 ss20 hs22 v8 hyper 256 64 ss2p v7 powerup 64 32 1 ss4 v8a micro2 8 16 1 ss4 110 v8a micro2 8 16 1 ss4 85 v8a micro2 8 16 1 ss5 v8a micro2 8 16 1 ss5 110 v8a micro2 8 16 1 ss5 85 v8a micro2 8 16 1 ss600 120 v7 old 64 32 1 ss600 140 v7 old 64 32 1 ss600 41 v8 super 16 32 4 1024 32 1 ss600 412 v8 super 16 32 4 1024 32 1 ss600 51 v8 super 16 32 4 1024 32
28. DOUBLE 62 Default Data Sizes and r8 Bytes 85 vax Suboptions 92 Xlist Suboptions 94 xarch ISA Keywords 95 Most General xarch Options on SPARC Platforms 96 xarch Values for SPARC Platforms 97 xcache Values 100 Valid xchip Values 101 xi xii TABLE A 1 TABLE A 2 TABLE C 1 TABLE C 2 TABLE C 3 TABLE D 1 TABLE E 1 TABLE E 2 TABLE E 3 TABLE E 4 TABLE E 5 TABLE E 6 TABLE E 7 TABLE E 8 77 Runtime I O Messages 127 95 Runtime I O Messages 130 F95 Source Form Command line options 148 Size Notation for Numeric Data Types 152 Nonstandard Intrinsics 162 xtarget Expansions 169 Summary of General Fortran Directives 175 Special Fortran 95 Directives 177 Sun Style Parallelization Directives Summary 177 Cray Parallelization Directives Summary 179 Summary of OpenMP Directives in Fortran 95 180 Summary of Fortran 95 OpenMP Library Routines 187 Summary of OpenMP Fortran Environment Variables 190 Environment variables not part of the OpenMP Fortran API 191 Fortran User s Guide July 2001 Before You Begin The Fortran User s Guide describes the compile time environment and command line options for the Sun WorkShop 6 Fortran compilers 77 FORTRAN 77 and 95 Fortran 95 This guide is intended for scientists engineers and programmers who have a working knowledge of the Fortran language and wish to learn how to use the Sun Fortran compilers effectively Familiarit
29. I O scheme has been proposed as part of the Fortran 2000 draft standard Stream I O access treats a data file as a continuous sequence of bytes addressable by a positive integer starting from 1 Declare a stream I O file with the ACCESS STREAM specifier on the OPEN statement File positioning to a byte address requires a POS scalar_integer_expression specifier on a READ or WRITE statement The INQUIRE statement accepts ACCESS STREAM a specifier STREAM scalar_character_variable and POS scalar_integer_variable STRUCTURE and UNION VAX Fortran To aid the migration of programs from 77 95 accepts VAX Fortran STRUCTURE and UNION statements a precursor to the derived types in Fortran 95 For syntax details see the FORTRAN 77 Language Reference manual The field declarations within a STRUCTURE can be one of the following a A substructure either another STRUCTURE declaration or a record that has been previously defined a A UNION declaration m A TYPE declaration which can include initial values a A derived type having the SEQUENCE attribute This is particular to 95 only As with 77 a POINTER statement cannot be used as a field declaration 95 also allows m Either or 3 can be used as a structure field dereference symbol struct field or struct field m Structures can appear in a formatted I O statement m Structure
30. LOGICAL 2 LOGICAL KIND 2 LOGICAL 2 Two byte logicals OGICAL 4 OGICAL KIND 4 LOGICAL 4 Four byte logicals REAL 4 REAL KIND 4 REAL 4 IEEE single precision floating point Four byte REAL 8 REAL KIND 8 REAL 8 IEEE double precision floating point Eight byte REAL 16 REAL KIND 16 REAL 16 IEEE quad precision floating point Sixteen byte Fortran User s Guide July 2001 TABLE C 2 Size Notation for Numeric Data Types Continued Nonstandard Declarator Short Form Meaning COMP LEX 8 COMP LEX KIND 4 COMP LEX 4 Single precision complex Four bytes each part COMP LEX 16 COMP LEX KIND 8 COMP LEX 8 Double precision complex Eight bytes each part COMP LEX 32 COMPLEX KIND 16 COMPLEX 16 Quad precision complex Sixteen bytes each part Cray Poin A Cray pointer is pointee ters a variable whose value is the address of another entity called the 95 supports Cray pointers Standard Fortran 95 does not Syntax The Cray POINTER statement has the following format POINTER p ointer_name pointee_name array_spec Where pointer_name pointee_name and array_spec are as follows pointer_name pointee_name array_spec Pointer to the corresponding pointee_name pointer_name contains the address of pointee_name Must be a scalar variable name but not a derived type Cannot be a
31. PDF file using the performance profiling tools prof gprof and tcov libraries and executables that can be redistributed under the terms of the End User License compiling for 64 bit Solaris operating environments overview of the Sun Performance Library The READMESs for all compilers are easily accessed by the xhelp readm command line option For example the command 95 xhelp readme will display the fortran_95 README file directly Fortran User s Guide July 2001 Command Line Help You can view very brief descriptions of the 77 and 90 command line options by invoking the compiler s help option as shown below f77 help or f95 help Items within are optional Items within lt gt are variable parameters Bar indicates choice of literal values For example someoption lt yes no gt implies someoption is someoption yes Sot Collect data for tcov basic block profiling old format ansi Report non ANSI extensions arg local Preserve actual arguments over ENTRY statements autopar Enable automatic loop parallelization requires WorkShop license Bdynamic Allow dynamic linking Bstatic Require static linking s Compile only produce o files suppress linking O Enable runtime subscript range checking cg89 Generate code for generic SPARC V7 architecture cg92 Generate code for SPARC V8 architecture copyargs Allow assignment to constant argume
32. as in fast nolibmil With libmil exceptions cannot be detected with errno or matherr 3m m The fsimple 2 option for aggressive floating point optimizations fsimple 2 is unsuitable if strict IEEE 754 standards compliance is required See page 64 This is a change from previous releases that set fsimple 1 with fast a The dalign option to generate double loads and stores for double and quad data in common blocks Using this option can generate nonstandard Fortran data alignment in common blocks m The xlibmopt option selects optimized math library routines a pad local inserts padding between local variables where appropriate to improve cache usage This was not set by fast in previous releases a xvector yes transforms certain math library calls within DO loops to single calls to a vectorized library equivalent routine with vector arguments This was not set by fast in previous releases a fns selects non standard SPARC floating point arithmetic exception handling and gradual underflow See page 62 m ftrap none to turn off all trapping for Fortran 77 Trapping on common floating point exceptions ft rap common is the used with Fortran 95 m xprefetch yes enables the compiler to generate hardware prefetch instructions where appropriate It is possible to add or subtract from this list by following the fast option with other options as in 95 fast fsimple 1 xnolibmopt which overrides
33. determination of the assumed latencies and therefore the result of a latx factor setting Chapter 3 Fortran Compiler Options 115 116 The lat x factor suboption is valid only when automatic prefetching is enabled That is latx factor is ignored unless it is used with yes or auto Warnings Explicit prefetching should only be used under special circumstances that are supported by measurements Because the compiler tunes the prefetch mechanism for optimal performance across a wide range of machines and applications you should only use xprefetch latx factor when the performance tests indicate there is a clear benefit The assumed prefetch latencies may change from release to release Therefore retesting the effect of the latency factor on performance whenever switching to a different release is highly recommended xprofile p Collect or optimize with runtime profiling data 77 95 p must be one of collect name use name or tcov Optimization level must be 02 or greater collect name Collect and save execution frequency data for later use by the optimizer with xprofile use The compiler generates code to measure statement execution frequency The name is the name of the program that is being analyzed This name is optional If name is not specified a out is assumed to be the name of the executable At runtime a program compiled with xprofile collect name will create by default the subdirectory name
34. every process that uses the library If a page of code in a shared library contains even a single non pic that is absolute memory reference the page becomes nonsharable and a copy of the page must be created each time a program using the library is executed The easiest way to tell whether or not a o file has been compiled with pic or PIC is with the nm command nm file o grep _GLOBAL_OFFSET_TABLE U _GLOBAL_OFFSET_TABLE A o file containing position independent code contains an unresolved external reference to _GLOBAL_OFFSET_TABLE_ as indicated by the letter U To determine whether to use pic or P IC use nm to identify the number of distinct global and static variables used or defined in the library If the size of _GLOBAL_OFFSET_TABLE_ is under 8 192 bytes you can use pic Otherwise you must use PIC When building shared dynamic libraries with xarch v9 or v9a or v9b in 64 bit Solaris environments the pic or PIC option or their xcode equivalents must be specified Compile position independent code but with 32 bit addresses 77 95 This option is similar to pic but it allows the global offset table to span the range of 32 bit addresses Use it in those rare cases where there are too many global data objects for pic Do not mix pic and PIC PIC is equivalent to xcode pic32 When building shared dynamic libraries with xarch v9 or v9a or v9b
35. executable files due to inlining The g option suppresses the 04 automatic inlining described above xcrossfile increases the scope of inlining with 04 Attempt aggressive optimizations Suitable only for that small fraction of a program that uses the largest fraction of compute time 05 s optimization algorithms take more compilation time and may also degrade performance when applied to too large a fraction of the source program Optimization at this level is more likely to improve performance if done with profile feedback See xprofile p o name Specify the name of the executable file to be written 77 95 There must be a blank between o and name Without this option the default is to write the executable file to a out When used with c o specifies the target o object file with G it specifies the target so library file 78 Fortran User s Guide July 2001 oldido Select an earlier list directed output style 77 Omit the blank that starts each record for list directed output This is a change from 77 releases 1 4 and earlier The default behavior is to provide that blank since the Fortran Standard requires it Note also the FORM PRINT option of OPEN You can compile parts of a program with oldldo and other parts without it onetrip Enable one trip DO loops 77 95 Compile DO loops so that they are executed at least once DO loops in standard Fortran are not performed at al
36. exploiting features of SPARC V8 and SPARC V9 including the UltraSPARC implementation These features are defined in the SPARC Architecture Manuals Version 8 ISBN 0 13 825001 4 and Version 9 ISBN 0 13 099227 5 published by Prentice Hall for SPARC International a In this document Standard means conforming to the versions of the standards listed above Non standard or Extension refers to features that go beyond these versions of these standards The responsible standards bodies may revise these standards from time to time The versions of the applicable standards to which these compilers conform may be revised or replaced resulting in features in future releases of the Sun Fortran compilers that create incompatibilities with earlier releases 10 Features of the Fortran Compilers Sun Fortran compilers provide the following features or extensions m Global program checking across routines for consistency of arguments commons parameters and the like Support for multiprocessor systems including automatic and explicit loop parallelization is integrated tightly with optimization Note Parallelization features of the Fortran compilers require a Forte for HPC license a 77 Many VAX VMS Fortran 5 0 extensions including NAMELIST DO WHILE Structures records unions maps Variable format expressions Recursion Pointers Double precision complex Quadruple precision real Quadruple precision co
37. extensions 47 arg local preserve ENTRY arguments 47 autopar parallelize automatically 48 Bdynamic 48 Bstatic 48 C check subscripts 49 c compile only 50 cg89 obsolete 50 cg92 obsolete 50 copyargs allow stores to literal arguments 50 dalign with fast 60 dalign 52 db 52 dbl and xtypemap 53 85 double default data sizes 53 dbl_align_all1 force data alignment 54 Index 197 depend data dependency analysis 54 with fast 60 dn 55 Dname define symbol 51 dryrun 55 dy 55 e extended source lines 55 erroff suppress warnings 56 errtags display message tag with warnings 56 explicitpar parallelize explicitly 56 ext_names externals without underscore 57 F 58 f align on 8 byte boundaries 58 fast 59 fixed 61 flags 61 fnonstd 61 fns with fast 60 fns 62 fpp Fortran preprocessor 63 free 63 fround r 63 fsimple simple floating point model 64 with fast 60 ftrap with fast 60 ftrap 65 G 66 g 66 help 67 hname 67 i2 short integers 68 i4 69 Idir 67 inline 69 KPIC 70 Kpic 70 Ldir 70 libmil with fast 60 libmil 71 llibrary 71 loopinfo show parallelization 72 Mdir 95 modules 72 167 misalign 73 198 Fortran User s Guide July 2001 mp cray Cray MP directives 73 mp openmp OpenMP directives 73 mp sun Sun MP directives 73 mt multithread safe libraries 74 na
38. f is equivalent to aligncommon 16 Note f may result in nonstandard alignment of data which could cause problems with variables in EQUIVALENCE or COMMON and may render the program non portable if f is required Using db1 77 with f aligns all 64 bit integer data on 8 byte boundaries as well Compiling any part of a program with f requires compiling all subprograms of that program with f Fortran User s Guide July 2001 By itself this option does not enable the compiler to generate faster multi word fetch store instructions on double and quad precision data The dalign option does this and invokes f as well Use of dalign is preferred over the older f See dalign page 52 Because dalign is part of the fast option so is f fast Select options that optimize execution performance 77 95 Note This option is defined as a particular selection of other options that is subject to change from one release to another and between compilers Also some of the options selected by fast might not be available on all platforms Compile with the v verbose flag to see the expansion of fast fast provides high performance for certain benchmark applications However the particular choice of options may or may not be appropriate for your application Use fast as a good starting point for compiling your application for best performance But additional tuning may still be required
39. in may or may not equal the numbers written out a Unformatted The numbers read back in do equal the numbers written out a Floating point formatted The numbers read back in can be different from the numbers written out This is caused by slightly different base conversion routines or by different conventions for uppercase lowercase spaces plus or minus signs and so forth Examples 1 0e12 1 0E12 1 0E 12 a List directed The numbers read back in can be different from the numbers written out This can be caused by various layout conventions with commas spaces zeros repeat factors and so forth Example 0 0 as compared to 0 Example 7 as compared to 7 Example 3 4 5 as compared to 3 4 5 Fortran User s Guide July 2001 Example 3 0 as compared to 0 0 0 The above results from integer v 3 0 0 0 print v Example 0 333333343 as compared to 0 333333 The above results from PRINT 1 0 3 0 Linking with 77 Compiled Routines a To mix 77 and 95 object binaries link with 95 and the 77 compatibility library Libf 77compat and not with 1ibF77 The xlang 77 option provides an easy way to do this Perform the link step with 95 even if the main program is an 77 program a Example 95 main and 77 subroutine demo cat m 95 CHARACTER 74 c This is a test CALL echol c END demo cat s f SUBROUTINE ech
40. one source program unit as a regular common block appears somewhere else on a TASK COMMON directive the program will stop with an error message indicating the first such inconsistency Example Missing TASKCOMMON directive in tc f demo cat tc f common x y 1000 do 1 i 1 1000 1 y i 1 call 257 end demo cat tz f subroutine z c common x h 1000 CSPAR TASKCOMMON X CSPAR DOALL do 1 i 1 1000 1 h i c h i return end demo 95 c O4 parallel xcommonchk tc f demo 95 c O4 parallel xcommonchk tz f demo 95 o tc O4 parallel xcommonchk tc o tz o demo te ERROR libmtsk inconsistent declaration of threadprivate taskcommon x_ not declared as threadprivate taskcommon at line 1 of tc f demo xcrossfile n Enable optimization and inlining across source files 77 95 If specified n may be 0 or 1 Normally the scope of the compiler s analysis is limited to each separate file on the command line For example 04 s automatic inlining is limited to subprograms defined and referenced within the same source file With xcrossfile the compiler analyzes all the files named on the command line as if they had been concatenated into a single source file xcrossfile is only effective when used with 04 or 05 Chapter 3 Fortran Compiler Options 103 104 Cross file inlining creates a possible source file interdependence that would not normally be there If any file in a set of f
41. or OpenMP explicit parallelization directives DO loops immediately preceded by parallelization directives will have threaded code generated for them Note This option should not be used to compile programs that already do their own multithreading with calls to the 1ibthread library To run a parallelized program in a multithreaded environment you must set the PARALLEL or OMP_NUM_THREADS environment variable prior to execution This tells the runtime system the maximum number of threads the program can create The default is 1 In general set the PARALLEL or OMP_NUM_THREADS variable to the available number of processors on the target platform If you use explicitpar and compile and link in one step then linking automatically includes the multithreading library and the thread safe Fortran runtime library If you use explicitpar and compile and link in separate steps then you must also link with explicitpar To improve performance also specify the stackvar option when using any of the parallelization options including explicitpar Use the mp option page 73 to select the style of parallelization directives enabled Sun Cray or OpenMP If the optimization level is not 03 or higher it is raised to 03 automatically For details see the Parallelization chapter in the Fortran Programming Guide ext_names e Create external names with or without trailing underscores 77 95 e must be either plain or und
42. particular the Fortran standard rule requiring compilers to honor explicit parentheses around subexpressions to control expression evaluation order may be broken with fsimple 2 This could result in numerical rounding differences with programs that depend on this rule For example with fsimple 2 the compiler may evaluate C A B as C A B breaking the standard s rule about explicit parentheses if the resulting code is better optimized The compiler might also replace repeated computations of x y with x z where z 1 y is computed once and saved in a temporary to eliminate the costly divide operations Programs that depend on particular properties of floating point arithmetic should not be compiled with fsimple 2 Even with fsimple 2 the optimizer still is not permitted to introduce a floating point exception in a program that otherwise produces none fast sets fsimple 2 ftrap i Set floating point trapping mode in effect at startup 77 95 t is a comma separated list that consists of one or more of the following all none common no Jinvalid no overflow no underflow no division no inexact ftrap common is a macro for ftrap invalid overflow underflow division Where the is shown it is a required character The 77 default is ft rap none The 95 default is ft rap common This option sets the IEEE 754 trapping modes that are established at program initialization Processing is left
43. portable wall clock timer OMP_GET_WTIME Function REAL 8 FUNCTION OMP_GET_WTIME Returns a double precision value equal to the elapsed wallclock time in seconds since some arbitrary time in the past OMP_GET_WTICK Function REAL 8 FUNCTION OMP_GET_WTICK Returns a double precision value equal to the number of seconds between successive clock ticks Appendix E Fortran Directives Summary 189 190 OpenMP Environment Variables TABLE E 7 and TABLE E 8 summarize the OpenMP Fortran API environment variables that control the execution of OpenMP programs TABLEE 7 Summary of OpenMP Fortran Environment Variables OMP_SCHEDULE Sets schedule type for DO and PARALLEL DO directives specified with schedule type RUNTIME If not defined a default value of STATIC is used Value is type chunk Example setenv OMP_SCHEDULE GUIDED 4 OMP_NUM_THREADS Sets the number of threads to use during execution unless set by a NUM_THREADS clause or a call to OMP_SET_NUM_THREADS subroutine If not set a default of 1 is used Value is a positive integer Current maximum is 128 Example setenv OMP_NUM_THREADS 16 OMP_DYNAMTC Enables or disables dynamic adjustment of the number of threads available for execution of parallel regions If not set a default value of TRUE is used Value is TRUE or FALSE Example setenv OMP_DYNAMIC FALSE OMP_NESTED Enables or disables nested parallelism Nested paralle
44. produce the fastest executable for the floating point options and platform currently being used For more information see the man pages 1ibm_single 3F and 1ibm_doub1le 3F Chapter 3 Fortran Compiler Options 71 loopinfo Show parallelization results The parallelization features of the Fortran compilers require a Forte for HPC license 77 95 Show which loops were and were not parallelized with the parallel autopar or explicitpar options Option loopinfo must appear with one of these parallelization options loopinfo displays a list of messages on standard error demo 95 o shalow fast parallel loopinfo shalow f Shalow f line 325 not parallelized not profitable inlined loop shalow f line 172 PARALLELIZED and serial version generated Shalow f line 173 not parallelized not profitable shalow f line 181 PARALLELIZED fused sShalow f line 182 not parallelized not profitable shalow f line 193 not parallelized not profitable shalow f line 199 PARALLELIZED and serial version generated shalow f line 200 not parallelized not profitable shalow f line 226 PARALLELIZED and serial version generated Shalow f line 227 not parallelized not profitable etc Use the error 1 utility with 77 compilations to merge this list with the source file to produce an annotated source listing with each loop tagged as parallelize
45. status old 1019 unknown system error This error should not happen but 1020 requires seek ability Attempted a seek on a file that does not allow it I O operation requiring a seek are direct access sequential unformatted I O and tabbing left 1021 illegal argument Certain arguments to open and related functions are checked for legitimacy Often only nondefault forms are checked 1022 negative repeat count The repeat count for list directed input must be a positive integer 1023 illegal operation for unit Attempted an I O operation that is not possible for the device associated with the logical unit You get this error if you try to read past end of tape or end of file 128 Fortran User s Guide July 2001 TABLE A 1 77 Runtime I O Messages Continued Error Message 1024 lt Not used gt 1025 incompatible specifiers in open Attempted to open a file with the new option and the access append option or some other invalid combination 1026 illegal input for namelist A namelist read encountered an invalid data item 1027 error in FILEOPT parameter The FILEOPT string in an OPEN statement has bad syntax 1028 WRITE to readonly file Attempt to write on a unit that was opened for reading only 1029 READ from writeonly file Attempt to read from a unit that was opened for writing only 1030 overflow converting numeric input Integer input data is too large for the corresponding input variable 1032 exponent overflo
46. suggests to the optimizer that it attempt to unroll loops n times Loop unrolling generally improves performance but will increase the size of the executable file For more information on this and other compiler optimizations see the Performance and Optimization chapter in the Fortran Programming Guide See also the discussion of the UNROLL directive on page 27 Show name and version of each compiler pass 77 95 This option prints the name and version of each pass as the compiler executes This information may be helpful when discussing problems with Sun service engineers Chapter 3 Fortran Compiler Options 91 v Verbose mode show details of each compiler pass 77 95 Like v shows the name of each pass as the compiler executes and details the options macro flag expansions and environment variables used by the driver Vax 0 Specify choice of VMS Fortran extensions enabled 77 v must be a comma separated list of at least one suboption Negatives may be constructed by prefixing each suboption keyword by no as in no logical_name The primary options are vax align and vax misalign vax align selects all the suboptions without allowing misaligned data This is the behavior of the x1 option prior to 77 release 3 0 1 vax misalign selects all the suboptions and allows misaligned data This is the behavior of the x1 option with 77 releases 3 0 1 4 0 4 2 5 0 and Sun WorkShop 6 The table be
47. to run out of memory or swap space Chapter 2 Using Sun Fortran Compilers 33 34 If the compiler runs out of memory try reducing the level of optimization or split multiple routine source files into files with one routine per file using fsp1lit 1 Swap Space Limits The SunOS operating system command swap s displays available swap space See swap 1M Example Use the swap command demo swap s total 40236k bytes allocated 7280k reserved 47516k used 1058708k available To determine the actual real memory demo usr sbin dmesg grep mem mem 655360K 0x28000000 avail mem 602476544 Increasing Swap Space Use mkfile 1M and swap 1M to increase the size of the swap space on a workstation You must become superuser to do this mkfile creates a file of a specific size and swap a adds the file to the system swap space demo mkfile v 90m home swapfile home swapfile 94317840 bytes demo usr sbin swap a home swapfile Control of Virtual Memory Compiling very large routines thousands of lines of code in a single procedure at optimization level 03 or higher may require additional memory that could degrade compile time performance You can control this by limiting the amount of virtual memory available to a single process In a sh shell use the ulimit command See sh 1 Fortran User s Guide July 2001 Example Limit virtual memory to 16 Mbytes demoS ulimit d 160
48. to right The common exceptions by definition are invalid division by zero and overflow For example ft rap overflow Example ftrap all no inexact means set all traps except inexact The meanings for ftrap t are the same as for ieee_flags except that m all turns on all the trapping modes and will cause trapping of spurious and expected exceptions Use common instead Chapter 3 Fortran Compiler Options 65 66 g m none the 77 default turns off all trapping modes a A nob prefix turns off that specific trapping mode To be effective compile the main program with this option For further information see the Floating Point Arithmetic chapter in the Fortran Programming Guide Build a dynamic shared library instead of an executable file 77 95 Direct the linker to build a shared dynamic library Without G the linker builds an executable file With G it builds a dynamic library Use o with G to specify the name of the file to be written See the Fortran Programming Guide chapter Libraries for details Compile for debugging and performance analysis 77 95 Produce additional symbol table information for debugging with dbx 1 or the Sun WorkShop debugging utility and for performance analysis with the Sun WorkShop Performance Analyzer Although some debugging is possible without specifying g the full capabilities of dbx and debugger are only available to those compilation units compiled wi
49. value is right justified it represents the right most bits of a word bits n through 31 The other bits are 0 Blanks are ignored Within an I O format specification the letter B indicates binary digits elsewhere it indicates octal digits Hexadecimal xX ddd or x ddd where d is any hexadecimal digit There can be 1 to 8 hexadecimal digits 0 through 9 A F Any of the letters can be uppercase or lowercase X x A F a f The digits must be enclosed in either apostrophes or quotes Blanks are ignored The hexadecimal digits may be preceded by a or sign 8 hexadecimal digits represent a full 32 bit word and the binary equivalents correspond to the contents of each bit position in the 32 bit word If less than 8 digits are present the value is right justified it represents the right most bits of a word bits n through 31 The other bits are 0 Fortran User s Guide July 2001 Hollerith Accepted forms for Hollerith data are nu Can H m H nL EE N de nR PCR Masts cise x Above is a string of characters and n is the character count a A Hollerith constant is type Boolean m If any character constant is in a bitwise logical expression the expression is evaluated as Hollerith a A Hollerith constant can have 1 to 4 characters Examples Octal and hexadecimal constants Boolean Constant Internal Octal for 32 bit word 0B 00000000000 77740B 00000077740 X ABE 00000005276
50. when debugging this option displays the commands and suboptions the compiler will invoke to perform the compilation d y n Allow or disallow dynamic libraries for the entire executable 77 95 m dy Yes allow dynamic shared libraries m dn No do not allow dynamic shared libraries The default if not specified is dy Unlike Bx this option applies to the whole executable and need appear only once on the command line dy dn are loader and linker options If you compile and link in separate steps with these options then you need the same option in the link step In a 64 bit Solaris environment many system libraries are not available only as shared dynamic libraries These include libm so and libc so libm a and libc a are not provided This means that dn and Bstatic may cause linking errors in 64 bit Solaris environments Applications must link with the dynamic libraries in these cases Accept extended length input source line 77 95 Accept source lines up to 132 characters long The compiler pads on the right with trailing blanks to column 132 If you use continuation lines while compiling with e then do not split character constants across lines otherwise unnecessary blanks may be inserted in the constants Chapter 3 Fortran Compiler Options 55 56 erroff taglist Suppress warning messages listed by tag name 77 95 Suppress the display of warning messages specified in th
51. with UltraSPARC extensions Adds to the SPARC V9 ISA the Visual Instruction Set VIS and extensions specific to UltraSPARC processors and enables the compiler to generate code for good performance on the V9 SPARC architecture e The resulting o object files are in ELF64 format and can only be linked with other SPARC V9 object files in the same format e The resulting executable can only be run on an UltraSPARC processor running a 64 bit enabled Solaris operating environment with the 64 bit kernel e xarch v9a is only available when compiling in a 64 bit enabled Solaris operating environment v9b Compile for the SPARC V9 ISA with UltraSPARC III extensions Adds UltraSPARC III extensions and VIS version 2 0 to the V9a version of the SPARC V9 ISA Compiling with this option uses the best instruction set for good performance in a Solaris UltraSPARC III environment e The resulting object code is in SPARC V9 ELF64 format and can only be linked with other SPARC V9 object files in the same format e The resulting executable can only be run on an UltraSPARC III processor running a 64 bit enabled Solaris operating environment with the 64 bit kernel e xarch v9b is only available when compiling in a 64 bit enabled Solaris operating environment xautopar Synonym for autopar 77 95 xcache c Define cache properties for the optimizer 77 95 c must be one of the following generic s1 11 al s1 11 a1 s2 12 a2
52. x 340 37777776300 X 1 2 3 00000000443 X FFFFFFFFFFFFFFFF STALI TENIA Examples Octal and hexadecimal in assignment statements i 1357B j X 28FF k X 5A Use of an octal or hexadecimal constant in an arithmetic expression can produce undefined results and do not generate syntax errors Appendix C Fortran 95 Features and Differences 151 152 Alternate Contexts of Boolean Constants 95 allows BOZ constants in the places other than DATA statements B bbb O 000 Z zzz B bbb O oo00 Z 277 If these are assigned to a real variable no type conversion occurs Standard Fortran allows these only in DATA statements Abbreviated Size Notation for Numeric Data Types 95 allows the following nonstandard type declaration forms in declaration statements function statements and IMPLICIT statements The form in column one is nonstandard Fortran 95 though in common use The kind numbers in column two can vary by vendor TABLE C 2 Size Notation for Numeric Data Types Nonstandard Declarator Short Form Meaning INTEGER 1 INTEGER KIND 1 INTEGER 1 One byte signed integers INTEGER 2 INTEGER KIND 2 INTEGER 2 Two byte signed integers INTEGER 4 INTEGER KIND 4 INTEGER 4 Four byte signed integers OGICAL 1 OGICAL KIND 1 LOGICAL 1 One byte logicals
53. 0 and __SUNPRO_F95 For example ___SUNPRO_F77 is 0x600 for the Sun WorkShop 6 release a The following macros are predefined on appropriate systems __ sparc unix sun SVR4 SunOS_5_6 SunOS_5_7 SunOS_5_8 For instance the value __sparc is defined on SPARC systems You can use these values in such preprocessor conditionals as the following ifdef _ sparc m The following are predefined with no underscores but they may be deleted in a future release sparc unix sun a On SPARC V9 systems the __sparcv9 macro is also defined The compilers use the fpp 1 preprocessor by default Like the C preprocessor cpp 1 fpp expands source code macros and enables conditional compilation of code Unlike cpp fpp understands Fortran syntax and is preferred as a Fortran preprocessor Use the xpp cpp flag to force the compiler to specifically use cpp rather than fpp Chapter 3 Fortran Compiler Options 51 52 dalign Align COMMON block data and generate faster multi word load stores 77 95 This flag changes the data layout in COMMON blocks and EQUIVALENCE classes and enables the compiler to generate faster multi word load stores for that data The data layout effect is that of the f flag double and quad precision data in COMMON blocks and EQUIVALENCE classes are laid out in memory along their natural alignment which is on 8 byte boundaries or on 16 byte boundaries for quad precision when compiling for 64
54. 00 In a csh shell use the limit command See csh 1 Example Limit virtual memory to 16 Mbytes demo limit datasize 16M Each of these command lines causes the optimizer to try to recover at 16 Mbytes of data space This limit cannot be greater than the system s total available swap space and in practice must be small enough to permit normal use of the system while a large compilation is in progress Be sure that no compilation consumes more than half the space Example With 32 Mbytes of swap space use the following commands In a sh shell demoS ulimit d 1600 In a csh shell demo limit datasize 16M The best setting depends on the degree of optimization requested and the amount of real and virtual memory available In 64 bit Solaris environments the soft limit for the size of an application data segment is 2 Gbytes If your application needs to allocate more space use the shell s limit or ulimit command to remove the limit For csh use demo limit datasize unlimited Chapter 2 Using Sun Fortran Compilers 35 For sh or ksh use demoS ulimit d unlimited See the Solaris 64 bit Developer s Guide for more information 36 Fortran User s Guide July 2001 CHAPTER 3 Fortran Compiler Options This chapter details the command line options for the 77 and 95 compilers a A description of the syntax used for compiler option flags starts on page 37 m Summar
55. 123 AaBbCc123 Meaning The names of commands files and directories on screen computer output What you type when contrasted with on screen computer output Book titles new words or terms words to be emphasized Command line placeholder text replace with a real name or value Examples Edit your login file Use 1s a to list all files o You have mail su Password Read Chapter 6 in the User s Guide These are called class options You must be superuser to do this To delete a file type rm filename m The symbol A stands for a blank space where a blank is significant AA36 001 a FORTRAN 77 examples appear in tab format while Fortran 95 examples appear in free format Examples common to both Fortran 77 and 95 use tab format except where indicated a The FORTRAN 77 standard uses an older convention of spelling the name FORTRAN capitalized Sun documentation uses both FORTRAN and Fortran The current convention is to use lower case Fortran 95 2 Fortran User s Guide July 2001 m References to online man pages appear with the topic name and section number For example a reference to GETENV will appear as getenv 3F implying that the man command to access this page would be man s 3F getenv m System Administrators may install the Sun WorkShop Fortran compilers and supporting material at lt install_point gt SUNWspro where lt install_point gt is usually o
56. 5 now looks more like 77 and includes suboptions Xlistc Xlist Xlists Xlistvn and Xlistw n See Chapter 3 page 93 m Fortran Library Interface 95 recognizes the include file system inc for declaring the proper data types for the Fortran library Supply the statement INCLUDE system inc in every routine that references non intrinsic Fortran library routines to insure proper typing of return values See the Fortran Library Reference 95 New Features in Sun WorkShop 6 update 1 The following lists new and changed features in the Fortran 95 compiler released with Sun WorkShop 6 update 1 Fortran User s Guide July 2001 UltraSPARC III Support The xtarget and xchip options now accept ult ra3 and the compiler will generate optimized code for the UltraSPARC III processor See Chapter 3 page 120 Prefetch added to fast The xprefetch flag has been added to the fast option set fast automatically sets a number of optimization flags for best execution speed on the compiling platform Adding xprefetch takes advantage of the UltraSPARC II and III prefetch mechanism and can add a substantial performance gain in code with loops that process data See Chapter 3 page 59 Support for the int2 Intrinsic The Fortran 95 and Fortran 77 compilers now support the int2 intrinsic for conversion of data types to 2 byte integer Use of int2 as an intrinsic M int2 J appears in many legacy Fortran 77 codes and is implemented i
57. 7 subscript out of bounds in NAMELIST object name 1098 empty substring in NAMELIST object name 1099 substring out of range in NAMELIST object name 100 unexpected component name in NAMELIST object name 11 unassociated ACCESS specifier 1112 unassociated ACTION specifier 13 unassociated BINARY specifier 14 unassociated BLANK specifier T5 unassociated DELIM specifier 16 unassociated DIRECT specifier 17 unassociated FILE specifier 18 unassociated FMT specifier 1119 unassociated FORM specifier 120 unassociated FORMATTED specifier 121 unassociated NAME specifier 122 unassociated PAD specifier 123 unassociated POSITION specifier 124 unassociated READ specifier Appendix A Runtime Error Messages 133 134 TABLE A 2 95 Runtime I O Messages Continued Error Message 125 unassociated READWRITE specifier 126 unassociated SEQUENTIAL specifier 127 unassociated STATUS specifier 128 unassociated UNFORMATTED specifier 129 unassociated WRITE specifier 130 zero length file name 131 ACCESS specifier is not SEQUENTIAL or DIRECT 132 ACTION specifier is not READ WRITE or READWRITE 133 BLANK specifier is not ZERO or NULL 134 DELIM specifier is not APOSTROPHE QUOTE or NONE 135 unexpected FORM specifier 136 PAD specifier is not YES or NO 137 POSITION specifier is not APPEND ASIS
58. ARY treats the file as a sequential binary unformatted file with no record marks See the Fortran 77 Language Reference for details m Debugging Optimized Code Restrictions limiting use of g with other options has been relaxed allowing debugging parallelized and 04 or 05 optimized codes with dbx and the Sun WorkShop debugger a New Changed Command Line Options a aligncommon aligns COMMON block elements to specific byte boundaries a r8const promotes single precision data constants to REAL 8 xmemalign specifies general alignment in memory of data elements xprefetch for enabling UltraSPARC prefetch instructions has been expanded to include additional sub options a xtypemap has an expanded set of possible data type specifications a fast extended to set 05 fsimple 2 xvector yes and pad common m Use of 77 s parallelization features requires a Sun WorkShop HPC license a Hyper Linked Diagnostic Messages Sun WorkShop online help now interprets 77 error diagnostics in the Building window creating hypertext links from the error message to descriptive online help Features in 77 5 0 77 5 0 included the following new and changed features m New options a fpover detects floating point overflows in I O processing xcode code specifies the memory address model on SPARC platforms a xcommonchk enables runtime checking for inconsistent COMMON block declarations xmaxopt enable
59. EAD statement Data is read into the variables on the input list transferring as many bytes as required by the list Because there are no record marks on the file there will be no end of record error detection The only errors detected are end of file or abnormal system errors INQUIRE statement INQUIRE on a file opened with FORM BINARY returns FORM BINARY ACCESS SEQUENTIAL DIRECT NO FORMATTED NO UNFORMATTED YES RECL AND NEXTREC are undefined BACKSPACE statement Not allowed returns an error a ENDFILE statement Truncates file at current position as usual a REWIND statement Repositions file to beginning of data as usual Recursive I O possible on different units this is because the 95 I O library is MT Warm RECL 2147483646 2 2 is the default record length on sequential formatted list directed and namelist output CODE and DECODE are recognized and implemented as described in the FORTRAN 77 Language Reference Manual Naming of scratch files is the same as with 77 Non advancing I O is enabled with ADVANCE NO as in write a ADVANCE NO n read n Appendix C Fortran 95 Features and Differences 159 160 Directives A compiler directive directs the compiler to do some special action Directives are also called pragmas
60. ERED block of Fortran statements with no transfers in or out SOMP END ORDERED The enclosed block of statements are executed in the order that iterations would be executed in a sequential execution of the loop It can appear only in the dynamic extent of a DO or PARALLEL DO directive The ORDERED clause must be specified on the closest DO directive enclosing the block Data Environment Directives THREADPRIVATE Directive SOMP THREADPRIVATE list Makes the list of variables and named common blocks private to a thread but global within the thread Common block names must appear between slashes To make a common block THREADPRIVATE this directive must appear after every COMMON declaration of that block Data Scoping Clauses Several directives noted above accept clauses to control the scope attributes of variables enclosed by the directive If no data scope clause is specified for a directive the default scope for variables affected by the directive is SHARED list is a comma separated list of named variables or common blocks that are accessible in the scoping unit Common block names must appear within slashes for example ABLOCK PRIVATE Clause PRIVATE list Declares the variables in the comma separated list to be private to each thread in a team Fortran User s Guide July 2001 TABLE E 5 Summary of OpenMP Directives in Fortran 95 Continued SHARED Clause DEFAULT Clause SHARED list
61. Fortran 77 supported in 77 but not in 95 a I O see also page 158 and page 164 List directed output uses different formats Variable format expressions are not available in Fortran 95 You cannot open a file with ACCESS APPEND in Fortran 95 95 does not allow BACKSPACE or ENDFILE on a direct access file Fortran 95 requires explicit field width specifications in format edit descriptors For example FORMAT I is not allowed 95 does not recognize 77 escape sequences for example n t in output formats 95 does not recognize FILEOPT in OPEN statements 95 does not recognize the n form for specifying a record number in direct access I O READ 2 13 X Y Z m Data Types Declarations and Usage If it appears in a program unit the IMPLICIT statement in Fortran 95 must precede the first declarative statement in the unit 95 allows only 7 array subscripts f77 allows 20 LOGICAL and INTEGER variables cannot be used interchangeably in Fortran 95 Fortran 95 Cray pointers cannot appear in some intrinsic calls 77 style initializations using slashes on type declarations are not accepted in Fortran 95 Fortran 95 does not allow assigning Cray character pointers to non pointer variables to other Cray pointers that are not character pointers Fortran 95 does not allow the same Cray pointer to point to items of different type sizes for example REAL 8 and INTEGER 4
62. Guide or the SPARC Architecture Manual Version 9 for further information about prefetch instructions Parallelization Directives Parallelization directives explicitly request the compiler to attempt to parallelize the DO loop or the region of code that follows the directive The syntax differs from general directives Parallelization directives are only recognized when compilation options parallel or explicitpar are used Details regarding Fortran parallelization can be found in the Fortran Programming Guide Note Fortran parallelization features require a Forte for High Performance Computing HPC license Chapter 2 Using Sun Fortran Compilers 29 30 The Fortran compilers support three styles of parallelization directives Sun Cray and OpenMP Sun style parallelization directives are the default explicitly selected with the compiler option mp sun Sun directives have the directive sentinel SPAR Cray style parallelization directives enabled by the mp cray compiler option have the sentinel MICS Interpretations of similar directives differ between Sun and Cray styles See the chapter on parallelization in the Fortran Programming Guide for details Fortran 95 also accepts OpenMP parallelization directives described in the next section Sun Cray parallelization directives have the following syntax a The first character must be in column one m The first character can be any one of c C or m The next
63. INTEGER KIND OMP_NEST_LOCK_KIND nvar 188 Fortran User s Guide July 2001 TABLEE 6 Summary of Fortran 95 OpenMP Library Routines Continued OMP__INIT_LOCK Subroutine SUBROUTIN SUBROUTIN OMP_INIT_LOCK var OMP_INIT_NEST_LOCK nvar Initializes a lock associated with lock variable var for use in subsequent calls The initial state is unlocked OMP_DESTROY_LOCK Subroutine SUBROUTINE OMP_DESTROY_LOCK var SUBROUTINE OMP_DESTROY_NEST_LOCK nvar Disassociates the given lock variable var from any locks OMP_SET_LOCK Subroutine SUBROUTINE OMP_SET_LOCK var SUBROUTINE OMP_SET_NEST_LOCK nvar Forces the executing thread to wait until the specified lock is available The thread is granted ownership of the lock when it is available OMP_UNSET_LOCK Subroutine SUBROUTINE OMP_UNSET_LOCK var SUBROUTINE OMP_UNSET_NEST_LOCK nvar Releases the executing thread from ownership of the lock Behavior is undefined if the thread does not own that lock OMP_TEST_LOCK Function LOGICAL FUNCTION OMP_TEST_LOCK var INTEGER FUNCTION OMP_TEST_NEST_LOCK nvar Attempts to set the lock associated with lock variable Returns TRUE if the simple lock was set successfully FALSE otherwise OMP_TEST_NEST_LOCK returns the new nesting count if the lock associated with nvar was set successfully otherwise it returns 0 Timing Routines These two functions returning double precision REAL 8 support a
64. If your program behaves improperly when compiled with fast look closely at the individual options that make up fast and invoke only those appropriate to your program that preserve correct behavior Note also that a program compiled with fast may show good performance and accurate results with some data sets but not with others Avoid compiling with fast those programs that depend on particular properties of floating point arithmetic Because some of the options selected by fast have linking implications if you compile and link in separate steps be sure to link with fast also fast selects the following options dalign depend fns fsimple 2 ftrap Snone f77 or ftrap common 95 libmil xtarget nativ 095 xlibmopt pad local xvector yes xprefetch yes Chapter 3 Fortran Compiler Options 59 60 Details about the options selected by fast a The xtarget native hardware target If the program is intended to run on a different target than the compilation machine follow the fast with a code generator option For example 77 fast xtarget ultra m The 05 optimization level option This is a change from previous compiler releases that set 03 or 04 with fast m The depend option analyzes loops for data dependencies and possible restructuring m The libmil option for system supplied inline expansion templates For C functions that depend on exception handling follow fast by nolibmil
65. L SIZE SPACING SPREAD SUM TI UNPACK VERIFY TRANSFER TRANSPOSE UBOUN Fortran User s Guide July 2001 R ECTED REAL KIND SET_EXPONENT SHAPE Y D Forward Compatibility Future releases of 95 are intended to be source code compatible with this release Module information files generated by this release of 95 are not guaranteed to be compatible with future releases Mixing Languages On Solaris systems routines written in C can be combined with Fortran programs since these languages have common calling conventions Module Files Compiling a file containing a Fortran 95 MODULE generates a module interface file mod file for every MODULE encountered in the source The file name is derived from the name of the MODULE file xyz mod all lowercase will be created for MODULE xyz Compilation also generates a o module implementation object file for the source file containing the MODULE statements Link with the module implementation object file along with the all other object files to create an executable The compiler creates module interface files and implementation object files in the current working directory It looks in the current working directory for the interface files when compiling USE modulename statements The Mpath option allows you to give the compiler an additional path to search Module implementation object files
66. Note The command line examples in this chapter primarily show 77 usages Except where noted equivalent usages of 95 are similarly valid however the printed output may be slightly different The very basic steps to running a Fortran application involve using an editor to create a Fortran source file witha f for 90 95 F F90 or F95 filename suffix invoking the compiler to produce an executable and finally launching the program into execution by typing the name of the file Example This program displays a message on the screen demo cat greetings f PROGRAM GREETINGS PRINT Real programmers write Fortran END demos 77 greetings f greetings f MAIN greetings demo a out Real programmers write Fortran demo In this example 77 compiles source file greetings f and links the executable program onto the file a out by default To launch the program the name of the executable file a out is typed at the command prompt Traditionally UNIX compilers write executable output to the default file called a out It can be awkward to have each compilation write to the same file Moreover if such a file already exists it will be overwritten by the next run of the compiler Instead use the o compiler option to explicitly specify the name of the executable output file demos 77 o greetings greetings f greetings f MAIN greetings demo In the preceding example the o o
67. OpenMP 2 0 specifications Supply an INCLUDE omp_lib h statement or include omp_lib h preprocessor directive or a USE omp_lib statement in every program unit that references these library routines Compiling with Xlist will report any type mismatches TABLEE 6 Summary of Fortran 95 OpenMP Library Routines Execution Environment Routines OMP_SET_NUM_THREADS Subroutine SUBROUTINE OMP_SET_NUM_THREADS int_expr Sets the number of threads to use for the next parallel region OMP_GET_NUM_THREADS Function INTEGER 4 FUNCTION OMP_GET_NUM_THREADS Returns the number of threads currently in the team executing the parallel region from which it is called OMP_GET_MAX_THREADS Function INTEGER 4 FUNCTION OMP_GET_MAX_THREADS Returns the maximum value that can be returned by calls to the OMP_GET_NUM_THREADS function OMP_GET_THREAD_NUM Function INTEGER 4 FUNCTION OMP_GET_THREAD_NUM Returns the thread number within the team This is a number between 0 and OMP_GET_NUM_THREADS 1 The master thread is thread 0 OMP_GET_NUM_PROCS Function INTEGER 4 FUNCTION OMP_GET_NUM_PROCS Returns the number of processors that are available to the program Appendix E Fortran Directives Summary 187 TABLEE 6 Summary of Fortran 95 OpenMP Library Routines Continued OMP_IN_PARALLEL Function LOGICAL 4 FUNCTION OMP_IN_PARALLEL Returns TRUE if called from within the dynamic extent of a region
68. PC license Intrinsics 95 supports some intrinsic procedures that are extensions beyond the standard TABLE C 3 Nonstandard Intrinsics Name Definition Function Type Argument Types Arguments Notes COT Cotangent real real X x P E DDIM Positive difference double precision double precision X x Y y PE LEADZ Get the number of integer Boolean integer real I i NP I leading 0 bits or pointer POPCNT Get the number of set integer Boolean integer real I i NP I bits or pointer POPPAR Calculate bit integer Boolean integer real X x NP I population parity or pointer Notes on the above table P The name can be passed as an argument NP The name cannot be passed as an argument E External code for the intrinsic is called at run time I 95 generates inline code for the intrinsic procedure 162 Fortran User s Guide July 2001 Compatibility with FORTRAN 77 Standard conforming fixed format filename FORTRAN 77 source code is compatible with Fortran 95 Use of non standard extensions such as VMS Fortran features are not compatible and may not compile with 95 Incompatibility Issues Between 95 and 77 The following lists some of the known incompatibility issues that arise when compiling and testing 77 programs with this release of 95 These are due to either missing comparable features in 95 or differences in behavior These items are non standard extensions to
69. THREADPRIVATE for variables COPYPRIVATE for SINGLE directives See http www openmp org specs for the OpenMP 2 0 specifications See also Appendix E page 180 a OpenMP Library Interface The compiler now provides an include file omp_lib h and an interface module omp_1ib for defining the interfaces to the OpenMP Fortran library routines See Appendix E page 187 a Interprocedural Optimization xipo This new compiler flag performs whole program optimizations by invoking an interprocedural analysis pass Unlike xcrossfile xipo will perform optimizations across all object files in the link step and is not limited to just the source files on the compile command xipo is particularly useful when compiling and linking large multi file applications See Chapter 3 page 107 a VAX Fortran Structures To aid migration of programs from 77 95 accepts VAX Fortran STRUCTURE and UNION statements a precursor of Fortran 95 derived types See Appendix C page 158 m Stream I O Another feature proposed for Fortran 2000 is a new stream I O scheme which treats a data file as a continuous sequence of bytes addressable by a positive integer starting from 1 Enable stream I O by declaring a file with ACCESS STREAM Position files with READ or WRITE statements with the POS integer_expression specifier See Appendix C page 158 m Global Program Checking Invoked by the Xlist options GPC on 9
70. VE DATA EQUIVALENCE COMMON or PARAMETER statement A Cray pointee cannot be a dummy argument A Cray pointee cannot be a function value A Cray pointee cannot be a structure or a structure component A Cray pointee cannot be of a derived type Usage of Cray Pointers Cray pointers can be assigned values as follows Set to an absolute address Example q 0 Appendix C Fortran 95 Features and Differences 155 m Assigned to or from integer variables plus or minus expressions Example p q 100 a Cray pointers are not integers You cannot assign them to a real variable m The LOC function nonstandard can be used to define a Cray pointer Example p LOC x Example Use Cray pointers as described above SUBROUTINE sub n COMMON pool 100000 I R TEGER blk 128 word64 EAL a 1000 b n c 100000 n 1000 POINTER pblk blk ia a ib b amp ic c address word64 DATA address 64 pblk 0 ia LOC pool ib ia 4000 ic ib n Remarks about the above example word64 refers to the contents of absolute address 64 b1k is an array that occupies the first 128 words of memory a is an array of length 1000 located in blank common b follows a and is of length n c follows b a b and c are associated with pool word64 is the same as blk 17 because Cray pointers are byte address and the integer elements of b1k are each 4 bytes lo
71. a length n WEAK Directive CSPRAGMA WEAK name name2 Declares name to be a weak symbol or an alias for name2 OPT Directive CSPRAGMA SUN OPT n Set optimization level for a subprogram to n PIPELOOP Directive CSPRAGMA SUN PIPELOOP n Assert dependency in the following loop exists between iterations n apart NOMEMDEP Directive CSPRAGMA SUN NOMEMDEP Assert there are no memory dependencies in the following loop PREFETCH Directives CSPRAGMA SPARC_PREFETCH_READ_ONCE name CSPRAGMA SPARC_PREFETCH_READ_MANY name CSPRAGMA SPARC_PREFETCH_WRITE_ONCE name CSPRAGMA SPARC_PREFETCH_WRITE_MANY name Request compiler generate prefetch instructions for references to name Requires xprefetch option The C Directive The C directive specifies that its arguments are external functions It is equivalent to an EXT ERNAL declaration except that unlike ordinary external names the Fortran compiler will not append an underscore to these argument names See the C Fortran Interface chapter in the Fortran Programming Guide for more details The C directive for a particular function should appear before the first reference to that function in each subprogram that contains such a reference Fortran User s Guide July 2001 Example compiling ABC and XYZ for C EXTERNAL ABC XYZ CSPRAGMA C ABC XYZ The UNROLL Directive The UNROLL directive requ
72. ach compiler pass is displayed Recognize upper and lower case in source files 77 95 Do not treat uppercase letters as equivalent to lowercase The default is to treat uppercase as lowercase except within character string constants With this option the compiler treats Delta DELTA and delta as different symbols Portability and mixing Fortran with other languages may require use of U These are discussed in the Fortran Programming Guide Fortran User s Guide July 2001 Uname u Undefine preprocessor macro name 77 95 This option applies only to F and F95 source files that invoke the fpp or cpp pre processor It removes any initial definition of the preprocessor macro name created by Dname on the same command line including those implicitly placed there by the command line driver regardless of the order the options appear It has no effect on any macro definitions in source files Multiple Uname flags can appear on the command line There must be no space between U and the macro name Report undeclared variables 77 95 Make the default type for all variables be undeclared rather than using Fortran implicit typing This option warns of undeclared variables and does not override any IMPLICIT statements or explicit type statements unroll n Enable unrolling of DO loops where possible 77 95 n is a positive integer The choices are a n inhibits all loop unrolling m n gt 1
73. alignment of all data on 8 byte boundaries dbl_align_all Align COMMON block data on 8 byte boundaries f Specify memory alignment and behavior xmemalign ab Debugging Enable runtime subscript range checking C Compile for debugging g Compile for browsing with Sun WorkShop source browser sb sbfast Flag use of undeclared variables u Compile for Sun WorkShop Performance Analyzer xF Generate source listings Xlistx Enable debugging without object files XS Diagnostics Flag use of non standard extensions ansi Suppress specific error messages erroff list Display error tag names with error messages errtags Show summary of compiler options flags help Show version number of the compiler and its phases V Verbose messages v Verbose parallelization messages vpara 40 Fortran User s Guide July 2001 TABLE 3 3 Compiler Options Grouped by Functionality Continued Function Option Flag Show suppress warning messages wn Enable runtime task common check xcommonchk Display compiler README file xhelp readme Licensing Show license server information xlicinfo Linking and Libraries Allow require dynamic static libraries Bx Allow only dynamic static library linking dy dn Build a dynamic shared object library G Assign name to dynamic library hname Add directory to library search path Ldir Link with library libname a or libname so 1lname Build runtime library search path into executable Rdir D
74. allowed The logical unit was opened for direct access I O 1006 can t backspace file You cannot do a seek on the file associated with the logical unit therefore you cannot backspace The file may be a tty device or a pipe 1007 off beginning of record You tried to do a left tab to a position before the beginning of an internal input record Appendix A Runtime Error Messages 127 TABLE A 1 77 Runtime I O Messages Continued Error Message 1008 can t stat file The system cannot return status information about the file Perhaps the directory is unreadable 1009 no after repeat count Repeat counts in list directed I O must be followed by an with no blank spaces 1010 off end of record A formatted write tried to go beyond the logical end of record An unformatted read or write also causes this 1011 lt Not used gt 1012 incomprehensible list input List input has to be as specified in the declaration 1013 out of free space The library dynamically creates buffers for internal use You ran out of memory for them that is your program is too big 1014 unit not connected The logical unit was not open 1015 read unexpected character Certain format conversions cannot tolerate nonnumeric data 1016 illegal logical input field logical data must be T or F 1017 new file exists You tried to open an existing file with status new 1018 can t find old file You tried to open a nonexistent file with
75. amation mark on the same line as the directive When compiling with openmp the CPP FPP variable _OPENMP is defined and may be used for conditional compilation within ifdef _OPENMP and endif TABLE E 5 Summary of OpenMP Directives in Fortran 95 Directive Format Fixed CSOMP directive optional_clauses SOMP directive optional_clauses SOMP directive optional_clauses Must start in column one continuation lines must have a non blank or non zero character in column 6 Directive Format Free SOMP directive optional_clauses May appear anywhere preceded by whitespace an ampersand amp at the end of the line identifies a continued line Conditional Compilation Source lines beginning with C or in columns 1 and 2 fixed format or preceded by white space free format are compiled only when compiler option openmp or mp openmp is specified PARALLEL Directive SOMP PARALLEL clause clause block of Fortran statements with no transfer in or out of block SOMP END PARALLEL Defines a parallel region a block of code that is to be executed by multiple threads in parallel clause can be one of the following PRIVATE list SHARED list DEFAULT option FIRSTPRIVATE list REDUCTION list IF logical_expression COPY IN list NUM_THREADS integer_expression 180 Fortran User s Guide July 2001 TABLE E 5 Summary of OpenMP Directives in Fortran 95 Continued DO D
76. an Programming Guide Fortran Library Reference FORTRAN 77 Language input output libraries program analysis debugging and performance Provides details about the library routines supplied with the Fortran compilers Provides a complete language reference to Sun FORTRAN 77 Numerical Computation Guide Collection Solaris 8 Reference Manual Reference Numerical Computation Guide Describes issues regarding the numerical accuracy of floating point computations Provides information about the Collection See the titles of man page sections Solaris operating environment Solaris 8 Software Developer Collection Solaris 8 Software Linker and Libraries Guide Describes the operations of the Solaris link editor and runtime linker Covers the POSIX and Solaris Developer Collection Multithreaded Programming Guide threads APIs programming with synchronization objects compiling multithreaded programs and finding tools for multithreaded programs 6 Fortran User s Guide July 2001 Ordering Sun Documentation You can order product documentation directly from Sun through the docs sun com Web site or from Fatbrain com an Internet bookstore You can find the Sun Documentation Center on Fatbrain com at the following URL http www fatbrain com documentation sun Sending Your Comments Sun is interested in improving its documentation and welcomes your comments and suggesti
77. and erroff taglist stop_status no yes xcrossfile xlic_lib libs xpp fpp cpp xtypemap type spec m Changed options Options fround fsimple ftrap xprofile tcov xspace xunroll now available on Intel platforms xtarget xarch xchip expanded for SPARC Ultra and Intel platforms vax expanded to enable selection deselection of individual VAX VMS Fortran features Default sourcefile preprocessor is fpp 1 rather than cpp 1 FORTRAN 77 Upward Compatibility The FORTRAN 77 5 0 source is compatible with earlier releases except for minor changes due to operating system changes and bug fixes Appendix B_ Features Release History 145 Fortran 3 0 3 0 1 to 4 0 Executables a out libraries a and object files o compiled and linked in Fortran 3 0 3 0 1 under Solaris 2 are compatible with Fortran 5 0 under Solaris 2 BCP Running Applications from Solaris 1 You must install the Binary Compatibility Package for the executable to run Executables compiled and linked in Solaris 1 do run in Solaris 2 but they do not run as fast as when they are compiled and linked under the appropriate Solaris release Libraries a and object files o compiled and linked in Fortran 2 0 1 under Solaris 1 are not compatible with Fortran 5 0 146 Fortran User s Guide July 2001 APPENDIX C Fortran 95 Features and Differences This appendix shows some of the major features differences between m St
78. andard Fortran 95 and Sun Fortran 95 m FORTRAN 77 and Fortran 95 Features and Extensions Sun WorkShop 6 Fortran 95 provides the following features Continuation Line Limits 95 and 77 allow 99 continuation lines 1 initial and 99 continuation lines Standard Fortran 95 allows 19 for fixed form and 39 for free form Fixed Form Source Lines In fixed form source lines can be longer than 72 characters but everything beyond column 73 is ignored Standard Fortran 95 only allows 72 character lines Directives 95 allows directive lines starting with CDIRS DIRS CMIC CS PRAGMA or CSOMP in fixed format or DIRS MICS SPRAGMA or SOMP in either fixed or free format For a summary of directives see Appendix E Standard Fortran 95 does not consider directives 147 148 m Tabs in 95 force the rest of the line to be padded out to column 72 This may cause unexpected results if the tab appears within a character string that is continued onto the next line Source file Iprint Tab on next line Tithis continuation line starts with a tab Tend Running the code Tab on next line this continuation line starts with a tab Source Form Assumed The source form assumed by 95 depends on options directives and suffixes Files with a f or F suffix are assumed to be in fixed format Files with a 90 95 F90 or F95 suffix are assumed to be in free format TABLE C 1 F95 Sou
79. ar symbol separates B dynamic static arguments only one of which may be chosen Bstatic The colon like the comma is sometimes used Rdir dir to separate arguments R local libs U a The ellipsis indicates omission in a series xinline fl fn xinline alpha dos Brackets pipe and ellipsis are meta characters used in the descriptions of the options and are not part of the options themselves Some general guidelines for options are m 1x is the option to link with library libx a It is always safer to put 1x after the list of file names to insure the order libraries are searched 38 Fortran User s Guide July 2001 m In general processing of the compiler options is from left to right allowing selective overriding of macro options options that include other options a The above rule does not apply to linker options a However some options I L and R for example accumulate values rather than override previous values when repeated on the same command line Source files object files and libraries are compiled and linked in the order in which they appear on the command line Options Summary In this section the compiler options are grouped by function to provide an easy reference The details will be found on the pages in the following sections as indicated The following table summarizes the 77 and 95 compiler options by functionality The table does not include obsolete and legacy op
80. ariables REDUCTION list listed variables are reduction variables SCHEDTYPE type use scheduling type default is STATIC STATIC SELF nchunk FACTORING m css m CSPAR DOSERIAL Disables parallelization of the loop that follows DOSERIAL Directive Fortran User s Guide July 2001 CSPAR DOSERIAL Disables parallelization of the loop nest that follows Cray Parallelization Directives Cray style parallelization directives are detailed in the chapter on parallelization in the Fortran Programming Guide Requires mp cray compiler option TABLEE 4 Cray Parallelization Directives Summary Format CMICS directive qualifiers initial line CMIC amp more_qualifiers continuation line Fixed format Directive indicator may be C as shown here c or With 95 free format leading blanks can appear before MICS DOALL Directive CMIC DOALL SHARED list PRIVATE list more_qualifiers Parallelize loop that follows Qualifiers are Scoping qualifiers are required unless list is empty all variables in the loop must appear in a PRIVATE or SHARED clause PRIVATE list declare names on list PRIVATE SHARED list declare names on list SHARED AUTOSCOPE automatically determine scope of variables The following are optional MAXCPUS n use no more than n threads SAVELAST save last value of all private variables Only one scheduling qualifier may appear GUIDED equivalent to Sun style GSS 64
81. ariables declared without an explicit byte size as follows TABLE 3 7 Default Data Sizes and db1 Bytes With dbl Without dbl option ption Data Type default SPARC INTEGER 4 8 REAL 4 8 This option applies to variables parameters constants and functions Also LOGICAL is treated as INTEGER COMPLEX as two REALS and DOUBLI COMPLEX as two DOUBLEs Gl Compare db1 with r8 db1 and r8 can be expressed in terms of the more general xt ypemap option a dbl sameas xtypemap real 64 double 128 integer 64 r8 sameas xtypemap real 64 double 128 integer mixed These options promote default DOUBLE PRECISION data to QUAD PRECISION 128 bits This may be unwanted and may cause performance degradation It might be more appropriate to use xt ypemap real 64 double 64 integer 64 instead of db1 in these cases For all of the floating point data types db1 works the same as r8 using both r8 and db1 produces the same results as using only db1 m For INTEGER and LOGICAL data types db1 is different from r8 a db1 allocates 8 bytes and does 8 byte arithmetic a r8 allocates 8 bytes and does only 4 byte arithmetic mixed Chapter 3 Fortran Compiler Options 53 54 In general if you compile one subprogram with db1 then be sure to compile all subprograms of that program with db1 This is particularly important with pr
82. at 1068 unassociated input variable 1069 more I O list items than data edit descriptors 1070 zero stride in subscript triplet 1071 zero step in implied DO loop 1072 negative field width 1073 zero width field 1074 character string edit descriptor reached on input 1075 Hollerith edit descriptor reached on input 1076 no digits found in digit string 1077 no digits found in exponent 1078 scale factor out of range 1079 digit equals or exceeds radix 1080 unexpected character in integer field 1081 unexpected character in real field 1082 unexpected character in logical field 1083 unexpected character in integer value 1084 unexpected character in real value 1085 unexpected character in complex value 132 Fortran User s Guide July 2001 TABLE A 2 95 Runtime I O Messages Continued Error Message 1086 unexpected character in logical value 1087 unexpected character in character value 1088 unexpected character before NAMELIST group name 1089 NAMELIST group name does not match the name in the program 1090 unexpected character in NAMELIST item 1091 unmatched parenthesis in NAMELIST item name 1092 variable not in NAMELIST group 1093 too many subscripts in NAMELIST object name 1094 not enough subscripts in NAMELIST object name 1095 zero stride in NAMELIST object name 1096 empty section subscript in NAMELIST object name 109
83. at most 2 byte alignment Assume at most 4 byte alignment Assume at most 8 byte alignment 6 Assume at most 16 byte alignment Po Ae PH Allowed values for error behavior on accessing misaligned data lt b gt are i Interpret access and continue execution s Raise signal SIGBUS Raise signal SIGBUS only for alignments less or equal to 4 The defaults without xmemalign specified are m 4s for xarch generic v7 v8 v8a v8plus v8plusa m 8s for xarch v9 v9a for C and C m 8f for xarch v9 v9a for Fortran The default for xmemalign appearing without a value is 1i for all platforms The dalign page 52 and misalign page 73 options are macros dalign is a macro for xmemalign 8s aligncommon 16 misalign is a macro for xmemalign 1i aligncommon 1 xnolib Synonym for nolib 77 95 xnolibmil Synonym for nolibmil 77 95 xnolibmopt Do not use fast math library 77 95 Use with fast to override linking the optimized math library 77 fast xnolibmopt 112 Fortran User s Guide July 2001 x0n Synonym for On 77 95 xopenmp Synonym for openmp 95 xpad Synonym for pad 77 xparallel Synonym for parallel 77 95 XPpg Synonym for pg 77 95 xpp fpp cpp Select source file preprocessor 1fs 77 95 The default is xpp fpp The compilers use fpp 1 to preprocess F or 95 source files This preprocessor is appropri
84. ate for Fortran Previous versions used the standard C preprocessor cpp To select cpp specify xpp cpp Chapter 3 Fortran Compiler Options 113 114 xprefetch al a Enable prefetch instructions on platforms that support prefetch such as UltraSPARC II 77 95 See page 29 for a description of the Fortran PREFETCH directives Enable prefetch instructions on those architectures that support prefetch such as UltraSPARC II xarch v8plus v8plusa v9plusb v9 v9a or v9b a must be one of the following values Value Meaning auto Enable automatic generation of prefetch instructions no auto Disable automatic generation of prefetch instructions explicit Enable explicit prefetch macros nosexplicit Disable explicit prefetch macros latx factor Adjust the compiler s assumed prefetch to load and prefetch to store latencies by the specified factor The factor must be a positive floating point or integer number yes xprefetch yes is the same as xprefetch auto explicit no xprefetch no is the same as xprefetch no auto notexplicit With xprefetch xprefetch auto and xprefetch yes the compiler is free to insert prefetch instructions into the code it generates This may result in a performance improvement on architectures that support prefetch If you are running computationally intensive codes on large multiprocessors you might find it advantageous to use xprefetch latx factor This optio
85. ation a program is compiled with the better runtime performance obtained However higher optimization levels may result in increased compilation time and larger executable files Debugging with g does not suppress On but on limits g in certain ways see the dbx documentation The 03 and 04 options reduce the utility of debugging such that you cannot display variables from dbx but you can still use the dbx where command to get a symbolic traceback If the optimizer runs out of memory it attempts to proceed over again at a lower level of optimization resuming compilation of subsequent routines at the original level For details on optimization see the Fortran Programming Guide chapters Performance Profiling and Performance and Optimization This is equivalent to 03 Provides a minimum of statement level optimizations Use if higher levels result in excessive compilation time or exceed available swap space Chapter 3 Fortran Compiler Options 77 Enables basic block level optimizations This level usually gives the smallest code size See also xspace 03 is preferred over 02 unless 03 results in unreasonably long compilation time exceeds swap space or generates excessively large executable files Adds loop unrolling and global optimizations at the function level Usually 03 generates larger executable files Adds automatic inlining of routines contained in the same file Usually 04 generates larger
86. ation on the analyzer Fortran User s Guide July 2001 xhasc yes no Treat Hollerith constant as a character string in an actual argument list 77 95 With xhasc yes the compiler treats Hollerith constants as character strings when they appear as an actual argument on a subroutine or function call This is the default and complies with the Fortran 77 standard The actual call list generated by the compiler contains hidden string lengths for each character string With xhasc no Hollerith constants are treated as typeless values in subprogram calls and only their addresses are put on the actual argument list No string length is generated on the actual call list passed to the subprogram Compile routines with xhasc no if they call a subprogram with a Hollerith constant and the called subprogram expects that argument as INTE other than CHARACTER Example demo cat hasc f call z 4habcd end subroutine z i integer 1 character S S 77 o hasi hasl string length demo xhasc no hasc f demo 7 abcdefg lt now correct length for s G ER or anything print string length len s return end demos 77 o hasO hasc f demo has0O string length 4 lt should be 7 Passing 4habcd to z is handled correctly by compiling with xhasc no This flag is provided to aid porting older Fortran programs xhelp Show summa
87. auses accepted by either the PARALLEL or WORKSHARE directive Synchronization Directives MASTER Directive SOMP MASTER block of Fortran statements with no transfers in or out SOMP END MASTER The block of statements enclosed by these directives is executed only by the master thread of the team The other threads skip this block and continue There is no implied barrier on entry to or exit from the master section Fortran User s Guide July 2001 TABLE E 5 Summary of OpenMP Directives in Fortran 95 Continued CRITICAL Directive SOMP CRITICAL name block of Fortran statements with no transfers in or out SOMP END CRITICAL name Restrict access to the statement block enclosed by these directives to only one thread at a time The optional name argument identifies the critical region All unnamed CRITICAL directives map to the same name Critical section names are global entities of the program If a name conflicts with any other entity the behavior of the program is undefined If name appears on the CRITICAL directive it must also appear on the END CRITICAL directive BARRIER Directive SOMP BARRIER Synchronizes all the threads in a team Each thread waits until all the others in the team have reached this point ATOMIC Directive SOMP ATOMIC Ensures that a specific memory location is to be updated atomically rather than exposing it to the possibility of multiple simultaneous writing th
88. bit environments with xarch v9 or v9a The default alignment of data in COMMON blocks is on 4 byte boundaries The compiler is also allowed to assume natural alignment and generate faster multi word load stores to reference the data Note dalign may result in nonstandard alignment of data which could cause problems with variables in EQUIVALENCE or COMMON and may render the program non portable if dalign is required dalign is a macro equivalent to xmemalign 8s aligncommon 16 See aligncommon page 46 and xmemalign page 111 Using both db1 77 and dalign also causes default INTEGER variables to be 8 byte aligned and 64 bits xt ypemap is preferred over db1 xtypemap real 64 double 128 integer 64 If you compile one subprogram with dalign compile all subprograms of the program with dalign This option is included in the fast option Generate optional CIF file 95 Generates an optional compiler information file CIF with the extension T This file is sometimes needed by the Sun WorkShop Source Browser Fortran User s Guide July 2001 Double the default size for REAL INTEGER DOUBLE and COMPLEX 77 Note This option and r8 is now considered obsolete and may be removed in future releases Use the more general xt ypemap option instead db1 promotes the default byte size for REAL INTEGER DOUBLE and COMPLEX v
89. brary in executable 76 position independent and pure 123 Sun Performance Library 12 110 vectorized math library libmvec 122 license information 110 limit command 35 196 Fortran User s Guide July 2001 stack size 89 limits Fortran 95 compiler 149 linear algebra routines 110 linking and parallelization with parallel 82 consistent compile and link 22 consistent with compilation 22 disable incremental linker 106 disable system libraries 75 enable dynamic linking shared libraries 55 explicit parallelization with explicitpar 57 linker Mmapfile option 104 mixed Fortran 77 and Fortran 90 compilations 23 separate from compilation 22 specifying libraries with 1 71 weak names 27 with automatic parallelization autopar 48 with compilation 20 list of directives 175 list of options 67 local variables allocate on memory stack 88 loop automatic parallelization 48 dependence analysis depend 54 executed once onetrip 79 explicit parallelization 56 parallelization messages 72 parallelizing a CALL ina loop 88 unrolling with unro11 91 loop unrolling directive 27 M man pages 12 man pages accessing 3 MANPATH environment variable setting 5 math library and Ldir option 70 optimized version 110 memory actual real memory display 34 limit virtual memory 35 optimizer out of memory 33 messages I O error 126 parallelization 72 93 runtime 125 signal handler 126 s
90. cences qui en restreignent l utilisation la copie la distribution et la d compilation Aucune partie de ce produit ou document ne peut tre reproduite sous aucune forme par quelque moyen que ce soit sans l autorisation pr alable et crite de Sun et de ses bailleurs de licence s il y en a Le logiciel d tenu par des tiers et qui comprend la technologie relative aux polices de caract res est prot g par un copyright et licenci par des fournisseurs de Sun Des parties de ce produit pourront tre d riv es des syst mes Berkeley BSD licenci s par l Universit de Californie UNIX est une marque d pos e aux Etats Unis et dans d autres pays et licenci e exclusivement par X Open Company Ltd La notice suivante est applicable a Netscape Netscape Navigator et the Netscape Communications Corporation logo Copyright 1995 Netscape Communications Corporation Tous droits r serv s Sun Sun Microsystems the Sun logo docs sun com AnswerBookz2 Solaris SunOS JavaScript SunExpress Sun WorkShop Sun WorkShop Professional Sun Performance Library Sun Performance WorkShop Sun Visual WorkShop et Forte sont des marques de fabrique ou des marques d pos es ou marques de service de Sun Microsystems Inc aux Etats Unis et dans d autres pays Toutes les marques SPARC sont utilis es sous licence et sont des marques de fabrique ou des marques d pos es de SPARC International Inc aux Etats Unis et dans d autres pays Les produi
91. ciated DECIMAL specifier 174 unassociated IOMSG specifier 175 unassociated ROUND specifier 1176 unassociated STREAM specifier Lay ASYNCHRONOUS specifier is not YES or NO 178 ROUND specifier is not UP DOWN ZERO NEAREST COMPATIBLE or PROCESSOR DEFINED 179 DECIMAL specifier is not POINT or COMMA 180 RECL specifier is not allowed in OPEN statement for stream access unit 181 attempting to allocate an allocated array 182 deallocating an unassociated pointer 1183 deallocating an unallocated allocatable array 184 deallocating an allocatable array through a pointer 185 deallocating an object not allocated by an ALLOCATE statement 186 deallocating a part of an object 187 deallocating a larger object than was allocated Appendix A Runtime Error Messages 135 136 TABLE A 2 Error 95 Runtime I O Messages Continued Message 191 192 193 194 195 196 197 2001 2002 2003 2004 2005 unallocated array passed to array intrinsic function illegal rank small source size zero array size negative elements in shape illegal kind nonconformable array invalid constant structure or component name handle not created character argument too short array argument too long or too short end of file record or directory stream Fortran User s Guide July 2001 APPENDIX B Features Release History This Appendix lists the new and changed features in this and previous release of
92. columns 1 through 5 works for both free form source and fixed form source FIXED and FREE Directives These directives specify the source form of lines following the directive line Scope They apply to the rest of the file in which they appear or until the next FREE or FIXED directive is encountered Uses m They allow you to switch source forms within a source file a They allow you to switch source forms for an INCLUDE file You insert the directive at the start of the INCLUDE file After the INCLUDE file has been processed the source form reverts back to the form being used prior to processing the INCLUDE file Restrictions The FREE FIXED directives m Each must appear alone on a compiler directive line not continued m Each can appear anywhere in your source code Other directives must appear within the program unit they affect Example A FREE directive DIRS FREE DO i a i END DO Appendix C Fortran 95 Features and Differences 161 Parallelization Directives A parallelization directive is a special comment that directs the compiler to attempt to parallelize the next DO loop These are summarized in Appendix E and described in the Fortran Programming Guide 95 recognizes both 77 Sun and Cray style parallelization directives as well as the OpenMP Fortran API directives Note Fortran parallelization features require a Forte H
93. constant a name of a structure an array or a function Pointee of the corresponding pointer_name Must be a variable name array declarator or array name If array_spec is present it must be explicit shape constant or non constant bounds or assumed size Example Declare Cray pointers to two pointees POINTER pb amp c Appendix C Fortran 95 Features and Differences 153 154 The above example declares Cray pointer p and its pointee b and Cray pointer q and its pointee c Example Declare a Cray pointer to an array POINTER ix x n O m The above example declares Cray pointer ix and its pointee x and declares x to be an array of dimensions n by m 1 Purpose of Cray Pointers You can use pointers to access user managed storage by dynamically associating variables to particular locations in a block of storage Cray pointers allow accessing absolute memory locations Cray Pointers and Fortran 95 Pointers Cray pointers are declared as follows POINTER pointer_name pointee_name array_spec Fortran 95 pointers are declared as follows POINTER object_name The two kinds of pointers cannot be mixed Features of Cray Pointers a Whenever the pointee is referenced 95 uses the current value of the pointer as the address of the pointee m The Cray pointer type statement declares both the pointer and the pointee a The Cray pointer is of type Cray pointer m T
94. d Tools Use the steps below to determine whether you need to change your PATH variable to access the Sun WorkShop compilers and tools To Determine If You Need to Set Your PATH Environment Variable 1 Display the current value of the PATH variable by typing echo PATH 2 Review the output for a string of paths containing opt SUNWspro bin If you find the path your PATH variable is already set to access Sun WorkShop development tools If you do not find the path set your PATH environment variable by following the instructions in the next section To Set Your PATH Environment Variable to Enable Access to Sun WorkShop Compilers and Tools 1 If you are using the C shell edit your home cshrc file If you are using the Bourne shell or Korn shell edit your home profile file 2 Add the following to your PATH environment variable opt SUNWspro bin 4 Fortran User s Guide July 2001 Accessing Sun WorkShop Man Pages Use the following steps to determine whether you need to change your MANPATH variable to access the Sun WorkShop man pages To Determine If You Need to Set Your MANPATH Environment Variable Request the workshop man page by typing man workshop Review the output if any If the workshop 1 man page cannot be found or if the man page displayed is not for the current version of the software installed follow the instructions in the next section for setting your MANPATH environment variable
95. d or not Example Passing standard error to the error utility demo 77 autopar loopinfo any f 2 gt amp 1 error options Be aware that error rewrites the input source file For details on error see the error 1 man page and the Fortran Programming Guide chapter on debugging Mdir Add dir to directories searched for Fortran 95 modules 95 Add dir to the list of directories to be searched for module files No space appears between the M and dir 72 Fortran User s Guide July 2001 The directories listed with M are searched after the current directory Compiling a source file containing a module generates a mod module file for each MODULE encountered See in Appendix C Module Files on page 167 for more information about modules in Fortran 95 misalign Allow misaligned data 77 The misalign option permits misaligned data in memory that would otherwise produce an error Particular uses of COMMON and EQUIVALENCE statements may cause data to be misaligned with a compiler diagnostic With misalign the compiler will allow intentional misalignment and will not add padding in COMMON blocks to insure proper data alignment However this seriously degrades performance recoding to eliminate the cause of data misalignment is a better alternative If used all routines in a program must be compiled with this option If you compile and link in separate steps compiling with the mi
96. d searches dir before searching the default directories See the Fortran Programming Guide chapter Libraries for information on library search order For the relative order between LD_LIBRARY_PATH and 1dir see Id 1 Note Specifying usr lib or usr ccs 1ib with Ldir may prevent linking the unbundled libm These directories are searched by default Fortran User s Guide July 2001 Example Use Ldir to specify library search directories demo 77 Ldirl Ldir2 any f Add library 1ibx a to linker s list of search libraries 77 95 Pass 1x to the linker to specify additional libraries for 1d to search for unresolved references ld links with object library libx If shared library 1ibx so is available and Bstatic or dn are not specified 1d uses it otherwise 1d uses static library libx a If it uses a shared library the name is built in to a out No space is allowed between 1 and x character strings Example Link with the library 1ibv77 demos 77 any f 1V77 Use ly again to link with more libraries Example Link with the libraries Liby and libz demos 77 any f ly 1z See also the Libraries chapter in the Fortran Programming Guide for information on library search paths and search order libmil Inline selected 1ibm library routines for optimization 77 95 There are inline templates for some of the 1ibm library routines This option selects those inline templates that
97. de chapter on Performance Profiling for details pad p Insert padding for efficient use of cache 77 95 This option inserts padding between arrays or character variables if they are static local and not initialized or if they are in common blocks The extra padding positions the data to make better use of cache In either case the arrays or character variables can not be equivalenced p if present must be either or both of local Add padding between adjacent local variables common Add padding between variables in common blocks Defaults for pad a Without the pad p option the compiler does no padding a With pad but without the p the compiler does both local and common padding The following are equivalent m 77 padany f m 77 pad local common any f m 77 pad common local any f The pad p option applies to items that satisfy the following criteria m The items are arrays or character variables a The items are static local or in common blocks For a definition of local or static variables see st ackvar page 88 Restrictions on pad common m Neither the arrays nor the character strings are equivalenced Fortran User s Guide July 2001 m If pad common is specified for compiling a file that references a common block it must be specified when compiling all files that reference that common block The option changes the spacing of variables within the common block If one
98. defined symbols but the interprocedural optimizations will be performed because it is a compile and link step Other important information about xipo m requires at least optimization level x04 m conflicts with xcrossfile if used together will result in a compilation error m objects compiled without xipo can be linked freely with objects compiled with x pO a The xipo option generates significantly larger object files due to the additional information needed to perform optimizations across files However this additional information does not become part of the final executable binary file Any increase in the size of the executable program will be due to the additional optimizations performed m In this release crossfile subprogram inlining is the only interprocedural optimization performed by xipo Fortran User s Guide July 2001 xl d Enable more VMS Fortran extensions 77 xl Enable the compiler to accept more VMS Fortran extensions This is a macro that is translated to vax misalign and provides the language features that are listed later in this description See the description of vax page 92 Although most VMS features are accepted automatically by 77 without any special options you must use the x1 option for a few VMS extensions In general you need the x1 option if a source statement can be interpreted as either a VMS feature or an 77 or 95 feature and you want the VMS feature In this case
99. do get a warning Example w still allows some warnings to get through demos 77 w fast silent 04 any f f77 Warning 04 overwrites previously set optimization level of 03 demo For 95 Individual levels from 0 to 4 can be specified w0 suppresses the least messages while w4 suppresses most warning w is equivalent to w0 Xlist x Produce listings and do global program checking GPC 77 95 Use this option to find potential programming bugs It invokes an extra compiler pass to check for consistency in subprogram call arguments common blocks and parameters across the global program The option also generates a line numbered Chapter 3 Fortran Compiler Options 93 listing of the source code including a cross reference table The error messages issued by the Xlist options are advisory warnings and do not prevent the program from being compiled and linked Note Be sure to correct all syntax errors in the source code before compiling with Xlist Unpredictable reports may result when run on a source code with syntax errors Example Check across routines for consistency demo 95 Xlist fil f The above example writes the following to the output file fil 1st a A line numbered source listing default m Error messages embedded in the listing for inconsistencies across routines Across reference table of the identifiers default By default the listings are w
100. e addresses Available only on 64 bit platforms xarch v9 v9a picl3 Generate position independent code small model Equivalent to pic Permits references to at most 2 11 unique external symbols on 32 bit platforms 2 10 on 64 bit platforms pic32 Generate position independent code large model Equivalent to PIC Permits references to at most 2 30 unique external symbols on 32 bit platforms 2 29 on 64 bit platforms The defaults not specifying xcode code explicitly are xcode abs32 on SPARC V8 and V7 platforms xcode abs64 on SPARC and UltraSPARC V9 xarch v9 or v9a When building shared dynamic libraries with xarch v9 or v9a and the 64 bit Solaris 7 environment xcode pic13 or xcode pic32 or pic or PIC must be specified xcommonchk no yes Enable runtime checking of common block inconsistencies 77 95 This option provides a debug check for common block inconsistencies in programs using TASK COMMON and parallelization See the discussion of the TASK COMMON directive in the Parallelization chapter in the Fortran Programming Guide The default is xcommonchk no runtime checking for common block inconsistencies is disabled because it will degrade performance Use xcommon yes only during program development and debugging and not for production quality programs Fortran User s Guide July 2001 Compiling with xcommonchk yes enables runtime checking If a common block declared in
101. e all units of a program with the same fsimple option If n is present it must be 0 1 or 2 The defaults are a Without the fsimple flag the compiler defaults to fsimple 0 a With fsimple alone the compiler defaults to fsimple 1 The different floating point simplification levels are fsimple 0 Permit no simplifying assumptions Preserve strict IEEE 754 conformance fsimple 1 Allow conservative simplifications The resulting code does not strictly conform to IEEE 754 but numeric results of most programs are unchanged With fsimple 1 the optimizer can assume the following m IEFE 754 default rounding trapping modes do not change after process initialization Computations producing no visible result other than potential floating point exceptions may be deleted Computations with Infinity or NaNs Not a Number as operands need not propagate NaNs to their results e g x 0 may be replaced by 0 Computations do not depend on sign of zero With fsimple 1 the optimizer is not allowed to optimize completely without regard to roundoff or exceptions In particular a floating point computation cannot be replaced by one that produces different results with rounding modes held constant at run time Fortran User s Guide July 2001 fsimple 2 Permit aggressive floating point optimizations This can cause some programs to produce different numeric results due to changes in the way expressions are evaluated In
102. e comma separated list of tag names taglist If taglist consists of none no warnings are suppressed If taglist consists of al1 all warnings are suppressed this is equivalent to the w option Example 77 erroff WDECL_LOCAL_NOTUSED ink f Use the errtags option to see the tag names associated with warning messages errtags yes no Display the message tag with each warning message 77 95 With errtags yes the compiler s internal error tag name will appear along with warning messages The default is not to display the tag errtags no demos 77 errtags ink f ink f MAIN ink line 11 Warning local variable i never used WDECL_LOCAL_NOTUSED lt The warning message s tag name errtags alone stands for errtags yes explicitpar Parallelize loops or regions explicitly marked by Sun Cray and or OpenMP directives Fortran parallelization features require a Forte for HPC license 77 95 The compiler will generate parallel code even if there are data dependencies in the DO loop that would cause the loop to generate incorrect results when run in parallel With explicit parallelization it is the user s responsibility to correctly analyze loops for data dependency problems before marking them with parallelization directives Parallelization is appropriate only on multiprocessor systems Fortran User s Guide July 2001 This option enables Sun Cray and
103. e for details Library options must be after the f and o files pg libraries are static If you compile and link in separate steps and you compile with pg then be sure to link with pg Compile position independent code for shared library 77 95 Use when compiling dynamic shared libraries Each reference to a global datum is generated as a dereference of a pointer in the global offset table Each function call is generated in program counter relative addressing mode through a procedure linkage table m The size of the global offset table is limited to 8Kb on SPARC Do not mix pic and PIC pic is equivalent to xcode pic13 There are two nominal performance costs with pic and PIC m A routine compiled with either pic or PIC executes a few extra instructions upon entry to set a register to point at the global offset table used for accessing a shared library s global or static variables m Each access to a global or static variable involves an extra indirect memory reference through the global offset table If the compile is done with PIC there are two additional instructions per global and static memory reference Fortran User s Guide July 2001 PIC When considering the above costs remember that the use of pic and PIC can significantly reduce system memory requirements due to the effect of library code sharing Every page of code in a shared library compiled pic or PIC can be shared by
104. e of some programs Where x does not cause total underflow x is a subnormal number if and only if x is in one of the ranges indicated TABLE 3 8 Subnormal REAL and DOUBLE Data Type Range REAL 0 0 lt Ix lt 1 17549435e 38 DOUBLE PRECISION 0 0 lt x lt 2 22507385072014e 308 See the Numerical Computation Guide for details on subnormal numbers and the Fortran Programming Guide chapter Floating Point Arithmetic for more information about this and similar options Some arithmeticians use the term denormalized number for subnormal number The standard initialization of floating point preferences is the default a IEEE 754 floating point arithmetic is nonstop do not abort on exception m Underflows are gradual To be effective the main program must be compiled with this option Fortran User s Guide July 2001 fpover yes no Detect floating point overflow in formatted input 77 95 With fpover yes specified the I O library will detect runtime floating point overflows in formatted input and return an error condition 1031 The default is no such overflow detection fpover no fpover is equivalent to fpover yes fpp Force preprocessing of input with fpp 95 Pass all the input source files listed on the 95 command line through the fpp preprocessor regardless of file extension Normally only files with F F90 or F95 extension are automatically preprocessed by f
105. e page 73 95 Library Interfaces and system inc The Fortran 95 compiler provides an include file system inc that defines the interfaces for most non intrinsic library routines Declare this include file to insure that functions you call and their arguments are properly typed especially when default data types are changed with xtypemap For example the following may produce an arithmetic exception because function getpid is not explicitly typed integer 4 mypid mypid getpid print mypid The getpid routine returns an integer value but the compiler assumes it returns a real value if no explicit type is declared for the function This value is further converted to integer most likely producing a floating point error To correct this you should explicitly type getuid and functions like it that you call integer 4 mypid getpid mypid getpid print mypid Problems like these can be diagnosed with the Xlist global program checking option The Fortran 95 include file system inc provides explicit interface definitions for these routines include system inc integer 4 mypid mypid getpid print mypid Chapter 2 Using Sun Fortran Compilers 31 Including system inc in program units calling routines in the Fortran library will automatically define the interfaces for you and help the compiler diagnose type mismatches See the Fortran Library Reference for more information
106. e time the TCOVDIR environment variable specifies the directory where the d and tcov files are located If this variable is not set then the d files remain in the same directory as the f files The xprofile tcov and the a options are compatible in a single executable That is you can link a program that contains some files which have been compiled with xprofile tcov and others with a You cannot compile a single file with both options If you compile and link in separate steps and you compile with a then be sure to link with a For details see the chapter Performance Profiling in the Fortran Programming Guide aligncommon n Specify the alignment of data in common blocks 77 95 n may be 1 2 4 8 or 16 and indicates the maximum alignment in bytes for data elements within common blocks For example aligncommon 4 would align all common block data elements with natural alignments of 4 bytes or more on 4 byte boundaries This option does not affect data with natural alignment smaller than the specified size Without aligncommon the compilers align common block data on at most 4 byte boundaries Fortran User s Guide July 2001 Specifying aligncommon without a value defaults to 1 on all platforms all common block data aligns on byte boundaries no padding between elements aligncommon 16 reverts to aligncommon 8 on platforms that are not 64 bit enabled platforms other tha
107. each of the xarch keywords on SPARC platforms TABLE 3 14 xarch Values for SPARC Platforms xarch generic generic64 native native64 vi v8a v8 Meaning Compile for good performance on most 32 bit systems This is the default This option uses the best instruction set for good performance on most processors without major performance degradation on any of them With each new release the definition of best instruction set may be adjusted if appropriate and is currently v7 Compile for good performance on most 64 bit enabled systems This option uses the best instruction set for good performance on most 64 bit enabled processors without major performance degradation on any of them With each new release the definition of best instruction set may be adjusted if appropriate and is currently interpreted as v9 Compile for good performance on this system This is the default for the fast option The compiler chooses the appropriate setting for the current system processor it is running on Compile for good performance in 64 bit mode on this system Like native compiler chooses the appropriate setting for 64 bit mode on the current system processor it is running on Compile for the SPARC V7 ISA Enables the compiler to generate code for good performance on the V7 ISA This is equivalent to using the best instruction set for good performance on the V8 ISA but without integer mul and div instructi
108. erscores The default is underscores ext_names plain Do not add trailing underscore ext_names underscores Add trailing underscore An external name is a name of a subroutine function block data subprogram or labeled common This option affects both the name of the routine s entry point and the name used in calls to it This option may be used to allow Fortran 77 routines to call and be called by other language routines Chapter 3 Fortran Compiler Options 57 58 f Invoke the source file preprocessor but do not compile 77 95 Apply the fpp preprocessor to F files and 95 files with 95 and write the processed result on a file with the same name but with suffix changed to f or 95 but do not compile Example 77 F source F writes the processed source file to source f fpp is the default preprocessor for Fortran The C preprocessor cpp can be selected instead by specifying xpp cpp Align data in COMMON blocks 77 95 Align double and quad precision data in COMMON blocks This flag changes the data layout in COMMON blocks and EQUIVALENCE classes double and quad precision data in COMMON blocks and EQUIVALENCE classes are laid out in memory along their natural alignment which is on 8 byte boundaries or on 16 byte boundaries for quad precision when compiling for 64 bit environments with xarch v9 or v9a The default alignment of data in COMMON blocks is on 4 byte boundaries
109. explicitly demo 95 SFFLAGS any f When using make if the FFLAGS variable is set as above and the makefile s compilation rules are implicit that is there is no explicit compiler command line then invoking make will result in a compilation equivalent to 77 fast Xlist files make is a very powerful program development tool that can easily be used with all Sun compilers See the make 1 man page and the Program Development chapter in the Fortran Programming Guide Note Default implicit rules assumed by make may not recognize files with extensions 95 and mod Fortran 95 Module files See the Fortran Programming Guide and the Fortran 95 readme file for details Memory Size A compilation may need to use a lot of memory This will depend on the optimization level chosen and the size and complexity of the files being compiled On SPARC platforms if the optimizer runs out of memory it tries to recover by retrying the current procedure at a lower level of optimization and resumes subsequent routines at the original level specified in the On option on the command line A workstation should have at least 24 megabytes of memory 32 megabytes are recommended Memory usage depends on the size of each procedure the level of optimization the limits set for virtual memory the size of the disk swap file and various other parameters Compiling a single source file containing many routines could cause the compiler
110. f any of the object files that make up a program were compiled with 95 then the final link step must be done with 95 Use 77 to produce the executable file only if none of the o object files were compiled with 95 See also Appendix C Compatibility with FORTRAN 77 on page 163 Unrecognized Command Line Arguments Any arguments on the command line that the compiler does not recognize are interpreted as being possibly linker options object program file names or library names The basic distinctions are m Unrecognized options with a generate warnings m Unrecognized non options no generate no warnings However they are passed to the linker and if the linker does not recognize them they generate linker error messages For example demo 95 bit move f lt bit is not a recognized 95 option 95 Warning Option bit passed to ld if ld is invoked ignored otherwise demo 95 fast move f lt The user meant to type fast ld fatal file fast cannot open file errno 2 ld fatal File processing errors No output written to a out Note that in the first example bit is not recognized by 95 and the option is passed on to the linker 1d who tries to interpret it Because single letter 1d options may be strung together the linker sees bit as b i t which are all legitimate ld options This may or may not be what the user expects or intended Chapter 2 Using Sun Fortran Compile
111. former Sun Performance WorkShop Personal Edition except that the Fortran compilers in that product no longer support the creation of automatically parallelized or explicit directive based parallel code This capability is still supported in the Fortran compilers in Forte for High Performance Computing We appreciate your continued use of our development products and hope that we can continue to fulfill your needs into the future Contents Before You Begin 1 How This Book Is Organized 1 Typographic Conventions 2 Shell Prompts 3 Supported Platforms 3 Accessing Sun WorkShop Development Tools and Man Pages 3 Accessing Sun WorkShop Documentation 5 Accessing Related Documentation 6 Ordering Sun Documentation 7 Sending YourComments 7 Introduction 9 Standards Conformance 9 Features of the Fortran Compilers 10 Other Fortran Utilities 11 Debugging Utilities 11 Sun Performance Library 12 Interval Arithmetic 12 Man Pages 12 READMEs 14 Command Line Help 15 2 Using Sun Fortran Compilers 17 A Quick Start 17 Invoking the Compiler 19 Compile Link Sequence 20 Command Line File Name Conventions 20 Source Files 21 Source File Preprocessors 21 Separate Compiling and Linking 22 Consistent Compiling and Linking 22 Linking Mixed Fortran 95 and Fortran 77 Compilations 23 Unrecognized Command Line Arguments 23 Modules Fortran 95 24 Directives 24 General Directives 25 Parallelization Directives 29 OpenMP Direc
112. four characters may be either PAR Sun style or MIC Cray style without blanks and in either upper or lower case m Next the directive keyword and qualifiers separated by blanks The explicit parallelization directive keywords are TASKCOMMON DOALL DOSERIAL and DOSERIAL Each parallelization directive has its own set of optional qualifiers that follow the keyword Example Specifying a loop with a shared variable CSPAR DOALL SHARED yvalue Sun style CMICS DOALL SHARED yvalue Cray style See Appendix E for a summary and the Fortran Programming Guide for details about parallelization and these directives OpenMP Directives The Sun WorkShop 6 Fortran 95 compiler recognizes the OpenMP Fortran shared memory multiprocessing API as specified by the OpenMP Architecture Review Board See the OpenMP website for details http www openmp org You must compile with the command line option openmp to enable OpenMP directives openmp is a macro flag that invokes the compiler options required by OpenMP see page 79 A summary of OpenMP directives appears in Appendix E Fortran User s Guide July 2001 OpenMP directives can be used in conjunction with either Sun or Cray style parallelization directives as long as these different directives are not nested within each other To enable OpenMP with Sun or Cray directives use mp openmp sun or mp openmp cray no spaces respectively Se
113. g A nonzero environment status is returned for testing in make files noreduction Disable reduction on command line 77 95 This option disables reduction norunpath Do not build a runtime shared library search path into the executable 77 95 The compiler normally builds into an executable a path that tells the runtime linker where to find the shared libraries it will need The path is installation dependent The norunpath option prevents that path from being built in to the executable 76 Fortran User s Guide July 2001 O n This option is helpful when libraries have been installed in some nonstandard location and you do not wish to make the loader search down those paths when the executable is run at another site Compare with Rpaths See the Fortran Programming Guide chapter on Libraries for more information Specify optimization level 77 95 n can be 1 2 3 4 or 5 No space is allowed between O and n If o n is not specified only a very basic level of optimization limited to local common subexpression elimination and dead code analysis is performed A program s performance may be significantly improved when compiled with an optimization level than without optimization Use of O which sets 03 or fast which sets 05 is recommended for most programs Each on level includes the optimizations performed at the levels below it Generally the higher the level of optimiz
114. ge 1010 record too long 1011 truncation failed 1012 incomprehensible list input 1013 out of free space 1014 unit not connected 1015 read unexpected character 1016 illegal logical input field 1017 new file exists 1018 can t find old file 1019 unknown system error 1020 requires seek ability 1021 illegal argument 1022 negative repeat count 1023 illegal operation for channel or device 1024 reentrant I O 1025 incompatible specifiers in open 1026 illegal input for namelist 1027 error in FILEOPT parameter 1028 writing not allowed 1029 reading not allowed 1030 integer overflow on input 1031 floating point overflow on input 1032 floating point underflow on input 1051 default input unit closed 1052 default output unit closed 1053 direct access READ from unconnected unit 1054 direct access WRITE to unconnected unit 1055 unassociated internal unit 1056 null reference to internal unit Appendix A Runtime Error Messages 131 TABLE A 2 95 Runtime I O Messages Continued Error Message 1057 empty internal file 1058 list directed I O on unformatted unit 1059 namelist I O on unformatted unit 1060 tried to write past end of internal file 1061 unassociated ADVANCE specifier 1062 ADVANCE specifier is not YES or NO 1063 EOR specifier present for advancing input 1064 SIZE specifier present for advancing input 1065 negative or zero record number 1066 record not in file 1067 corrupted form
115. get option platform system names and their expansions Each specific value for xtarget expands into a specific set of values for the xarch xchip and xcache options as shown in the following table Run fpversion 1 to determine the target definitions on any system For example xtarget sun4 15 means xarch v8a xchip micro xcache 2 16 1 TABLE D 1 xtarget Expansions xtarget xarch xchip xcache generic generic generic generic generic64 v9 generic generic cs6400 v8 super 16 32 4 2048 64 1 entr150 v8plusa ultra 16 32 1 512 64 1 entr2 v8plusa ultra 16 32 1 512 64 1 entr2 1170 v8plusa ultra 16 32 1 512 64 1 entr2 1200 v8plusa ultra 16 32 1 512 64 1 entr2 2170 v8plusa ultra 16 32 1 512 64 1 entr2 2200 v8plusa ultra 16 32 1 512 64 1 entr3000 v8plusa ultra 16 32 1 512 64 1 entr4000 v8plusa ultra 16 32 1 512 64 1 169 TABLE D 1 xtarget Expansions Continued xtarget xarch xchip xcache entr5000 v8plusa ultra 16 32 1 512 64 1 entr6000 v8plusa ultra 16 32 1 512 64 1 sc2000 v8 super 16 32 4 2048 64 1 solb5 v7 old 128 32 1 solb6 v8 super 16 32 4 1024 32 1 ssl v7 old 64 16 1 ss10 v8 super 16 32 4 ss10 20 v8 super 16 32 4 ss10 30 v8 super 16 32 4 ss10 40 v8 super 16 32 4 ss10 402 v8 super 16 32 4 ss10 41 v8 super 16 32 4 1024 32 1 ss10 412 v8 super 16 32 4 1024 32 1 ss10 50 v8 super 16 32 4 ss10 51 v8 super 16 32 4 1024 32 1 ss10
116. he Fortran Programming Guide chapter Libraries The hname option records the name name to the shared dynamic library being created as the internal name of the library A space between h and name is optional except if the library name is elp for which the space will be needed In general name must be the same as what follows the o Use of this option is meaningless without also specifying G Without the hname option no internal name is recorded in the library file If the library has an internal name whenever an executable program referencing the library is run the runtime linker will search for a library with the same internal name in any path the linker is searching With an internal name specified searching for the library at runtime linking is more flexible This option can also be used to specify versions of shared libraries If there is no internal name of a shared library then the linker uses a specific path for the shared library file instead help Display a summary list of compiler options 77 95 Displays a list of option summaries See also xhelp h on page 105 Idir Add dir to the INCLUDE file search path 77 95 Insert the directory dir at the start of the INCLUDE file search path No space is allowed between I and dir Invalid directories are ignored with no warning message Chapter 3 Fortran Compiler Options 67 68 i2 The include file search path is the list of directories sea
117. he end of the reduction the shared variable is updated by combining the original value with the final value of each of the private copies COPYIN Clause COPYIN list The COPYIN clause applies only to variables common blocks and variables in common blocks that are declared as THREADPRIVATE In a parallel region COPYIN specifies that the data in the master thread of the team be copied to the thread private copies of the common block at the beginning of the parallel region Appendix E Fortran Directives Summary 185 TABLE E 5 COPYPRIVATE Clause Summary of OpenMP Directives in Fortran 95 Continued COPYPRIVATE list Uses a private variable to broadcast a value or a pointer to a shared object from one member of a team to the other members Variables in list must not appear in a PRIVATE or FIRSTPRIVATE clause of the SINGLE construct specifying COPYPRIVATE Scheduling Clauses on DO and PARALLEL DO Directives SCHI EDULE Clause SCHEDULE type chunk Specifies how iterations of the DO loop are divided among the threads of the team type can be one of the following In the absence of a SCHEDULE clause STATIC scheduling is used STATIC Scheduling SCHEDULE STATIC chunk Iterations are divided into pieces of a size specified by chunk The pieces are statically assigned to threads in the team in a round robin fashion in the order of the thread number chunk must be a scalar integer expression
118. he value of a Cray pointer occupies one storage unit on 32 bit processors and two storage units on 64 bit SPARC V9 processors m The Cray pointer can appear in a COMMON list or as a dummy argument m The Cray pointee has no address until the value of the Cray pointer is defined m If an array is named as a pointee it is called a pointee array Its array declarator can appear in Fortran User s Guide July 2001 a A separate type statement a A separate DIMENSION statement a The pointer statement itself If the array declarator is in a subprogram the dimensioning can refer to a Variables in a common block or a Variables that are dummy arguments The size of each dimension is evaluated on entrance to the subprogram not when the pointee is referenced Restrictions on Cray Pointers pointee_name must not be a variable typed CHARACTER If pointee_name is an array declarator it must be explicit shape constant or non constant bounds or assumed size An array of Cray pointers is not allowed A Cray pointer cannot be a Pointed to by another Cray pointer or by a Fortran pointer a A component of a structure a Declared to be any other data type A Cray pointer cannot appear in A PARAMETER statement or in a type declaration statement that includes the PARAMETER attribute A DATA statement Restrictions on Cray Pointees A Cray pointee cannot appear in a SA
119. his option can be used alone it is part of the expansion of the xtarget option it is provided to allow overriding a xchip value implied by the a specific xtarget option Some effects of xchip c are m Instruction scheduling m The way branches are compiled a Choice between semantically equivalent alternatives The following table lists the valid xchip values TABLE 3 16 Valid xchip Values Value Optimize for generic most SPARC processors native this 32 bit host platform old pre SuperSPARC processors super the SuperSPARC processor super2 the SuperSPARC II processor micro the MicroSPARC processor micro2 the MicroSPARC II processor hyper the HyperSPARC processor hyper2 the HyperSPARC II processor powerup the Weitek PowerUp processor ultra the UltraSPARC processor ultra2 the UltraSPARC II processor ultra2e the UltraSPARC le processor ultra2i the UltraSPARC Ili processor ultra3 the UltraSPARC III processor Chapter 3 Fortran Compiler Options 101 102 xcode code Specify code address space on SPARC platforms 77 95 The values for code are abs32 Generate 32 bit absolute addresses Code data bss size is limited to 2 32 bytes This is the default on 32 bit platforms xarch generic v7 v8 v8a v8plus v8plusa abs44 Generate 44 bit absolute addresses Code data bss size is limited to 2 44 bytes Available only on 64 bit platforms xarch v9 v9a abs64 Generate 64 bit absolut
120. ibraries m Bstatic Require static linking no shared libraries Also note Fortran User s Guide July 2001 a If you specify static but the linker finds only a dynamic library then the library is not linked with a warning that the library was not found a If you specify dynamic but the linker finds only a static version then that library is linked with no warning You can toggle Bstatic and Bdynamic on the command line That is you can link some libraries statically and some dynamically by specifying Bstatic and Bdynamic any number of times on the command line as follows 77 prog f Bdynamic lwells Bstatic lsurface These are loader and linker options Compiling and linking in separate steps with Bx on the compile command will require it in the link step as well You cannot specify both Bdynamic and dn on the command line because dn disables linking of dynamic libraries In a 64 bit Solaris environment many system libraries are available only as shared dynamic libraries These include libm so and libc so libm a and libc a are not provided This means that Bstatic and dn may cause linking errors in 64 bit Solaris environments Applications must link with the dynamic libraries in these cases See the Fortran Programming Guide for more information on static and dynamic libraries Check array references for out of range subscripts 77 95 Subscripting arrays beyond their declared sizes may
121. ies of options arranged by functionality starts on page 39 m The complete reference detailing each compiler option flag starts on page 45 Some options are not available on both compilers 77 or 95 Check the reference section for availability Command Syntax The general syntax of the compiler command line is 77 options list_of_files additional_options 95 options list_of_files additional_options Items in square brackets indicate optional parameters The brackets are not part of the command The options are a list of option keywords prefixed by dash Some keyword options take the next item in the list as an argument The list_of_files is a list of source object or library file names separated by blanks Also there are some options that must appear after the list of source files and these could include additional lists of files for example B 1 and L 37 Options Syntax Typical compiler option formats are TABLE 3 1 Options Syntax Syntax Format Example flag 9 flagvalue Dnostep flag value xunroll 4 flag value o outfile The following typographical conventions are used when describing the individual options TABLE 3 2 Typographic Notations for Options Notation Meaning Example Text Instance Square brackets contain arguments that are O n optional 04 O Curly brackets contain a set of choices for a d y n required option dy The pipe or b
122. ile one file and link with others in separate steps demo 95 c filel f Make new object file demo 95 o prgrm filel o file2 o file3 o Make executable file Be sure that the link step lists all the object files needed to make the complete program If any object files are missing from this step the link will fail with undefined external reference errors missing routines Consistent Compiling and Linking Ensuring a consistent choice of compiling and linking options is critical whenever compilation and linking are done in separate steps Compiling any part of a program with any of the following options requires linking with the same options a autopar Bx fast G Lpath lname mt xmemalign nolib norunpath p pg xlibmopt xlic_lib name xprofile p Example Compiling sbr f with a and smain f without it then linking in separate steps a invokes tcov old style profiling demo 95 c a sbr f demo f95 c smain f demo 95 a sbr o smain o link step passes a to the linker Fortran User s Guide July 2001 Also a number of options require that all source files be compiled with that option including the link step These include autopar aligncommon dx dalign dbl explicitpar f misalign native parallel r8 xarch a xcache c xchip c xF xtarget t xtypemap ztext Linking Mixed Fortran 95 and Fortran 77 Compilations As a general rule i
123. iles compiled together with xcrossfile is changed then all files must be recompiled to insure that the new code is properly inlined See the discussion of inlining on page 69 The default without xcrossfile on the command line is xcrossfile 0 and no cross file optimizations are performed To enable cross file optimizations specify xcrossfile equivalent to xcrossfile 1 xdepend Synonym for depend 77 95 xexplicitpar Synonym for explicitpar 77 95 Allow function level reordering by the Sun WorkShop Performance Analyzer 77 95 Allow the reordering of functions subprograms in the core image using the compiler the performance analyzer and the linker If you compile with the xF option then run the analyzer you can generate a map file that optimizes the ordering of the functions in memory depending on how they are used together A subsequent link to build the executable file can be directed to use that map by using the linker Mmapfile option It places each function from the executable file into a separate section Reordering the subprograms in memory is useful only when the application text page fault time is consuming a large percentage of the application time Otherwise reordering may not improve the overall performance of the application The performance analyzer is part of the Sun WorkShop software See Using Sun WorkShop and Analyzing Program Performance with Sun WorkShop for further inform
124. imizations that increase code size xspace Generate calls to vector library functions automatically xvector yn Parallelization Note Fortran parallelization features require a Forte for High Performance Computing license Enable automatic parallelization of DO loops autopar Enable parallelization of loops explicitly marked with explicitpar directives Show loop parallelization information loopinfo Specify which style of directives to accept Sun Cray mp v OpenMP Compile for hand coded multithreaded programming mt Accept OpenMP API directives and set appropriate openmp environment macro Parallelize loops with autopar explicitpar depend parallel combination Recognize reduction operations in loops with automatic reduction parallelization Verbose parallelization messages vpara Source Code Define preprocessor symbol Dname val Undefine preprocessor symbol Uname 42 Fortran User s Guide July 2001 TABLE 3 3 Compiler Options Grouped by Functionality Continued Function Option Flag Accept extended 132 character source lines e Apply preprocessor to F and or F90 and F95 files but F do not compile Accept fixed format input 95 fixed Preprocess all source files with the fpp preprocessor fpp Accept free format input 95 free Add directory to include file search path Idir Add directory to module search path Mdir Recognize upper and lower case as disti
125. in 64 bit Solaris environments the pic or PIC option or their xcode equivalents must be specified Chapter 3 Fortran Compiler Options 83 84 Qoption pr Is qp R Is Pass options to compilation phase pr 77 95 Pass the suboption list s to the compilation phase pr There must be blanks separating Qoption pr and ls The Q can be uppercase or lowercase The list is a comma delimited list of suboptions with no blanks within the list Each suboption must be appropriate for that program phase and can begin with a minus sign This option is provided primarily for debugging the internals of the compiler by support staff Use the LD_OPTIONS environment variable to pass options to the linker See the chapter on linking and libraries in the Fortran Programming Guide Synonym for p 77 95 Build dynamic library search paths into the executable file 77 95 With this option the linker d 1 stores a list of dynamic library search paths into the executable file Is is a colon separated list of directories for library search paths The blank between R and Is is optional Multiple instances of this option are concatenated together with each list separated by a colon The list is used at runtime by the runtime linker 1d so At runtime dynamic libraries in the listed paths are scanned to satisfy any unresolved references Use this option to let users run shippable executables without a special path op
126. interactions examples and other details Each description indicates platform availability of the option The following table indicates availability of an option Legend Option Availability 77 only available with 77 95 only available with 95 77 95 available with both 77 and 95 Options that are not available for a compiler on a particular platform will still be accepted silently by the compiler That is the compiler will accept the option on the command line on that platform without issuing a warning but the option does nothing This options reference details each option flag Chapter 3 Fortran Compiler Options 45 46 a Profile by basic block using tcov old style 77 95 This is the old style of basic block profiling for tcov See xprofile tcov for information on the new style of profiling and the tcov 1 man page for more details Also see the manual Analyzing Program Performance with Sun WorkShop Insert code to count the times each basic block of statements is executed This invokes a runtime recording mechanism that creates one d file for every f file at normal program termination The d file accumulates execution data for the corresponding source file The t cov 1 utility can then be run on the source file s to generate statistics about the program The summary output produced by tcov is written to file tcov for each source file pg and gprof are complementary to a and tcov If set at compil
127. irective SOMP DO clause clause do_loop statements block SOMP END DO NOWAIT The DO directive specifies that the iterations of the DO loop that immediately follows must be executed in parallel This directive must appear within a parallel region clause can be one of the following PRIVATE list FIRSTPRIVATE Iist LASTPRIVATE list REDUCTION list SCHEDULE type ORDERED SECTIONS Directive SOMP SECTIONS clause clause SOMP SECTION block of Fortran statements with no transfer in or out SOMP SECTION optional block of Fortran statements SOMP END SECTIONS NOWAIT Encloses a non iterative section of code to be divided among threads in the team Each section is executed once by a thread in the team clause can be one of the following PRIVATE Iist FIRSTPRIVATE list LASTPRIVATE list REDUCTION Iist Each section is preceded by a SECTION directive which is optional for the first section SINGLE Directive SOMP SINGLE clause clause block of Fortran statements with no transfer in or out SOMP END SINGLE end modifier The statements enclosed by SINGLE is to be executed by only one thread in the team Threads in the team that are not executing the SINGLE block of statements wait at the END SINGLE directive unless NOWAIT is specified clause can be one of PRIVATE list FIRSTPRIVATE list end modifier is either COPYPRIVATE list COPYPRIVATE list or NOWAIT
128. ires that you specify SUN after CSPRAGMA The CSPRAGMA SUN UNROLL n directive instructs the compiler to unroll the following loop n times during its optimization pass The compiler will unroll a loop only when its analysis regards such unrolling as appropriate n is a positive integer The choices are a If n 1 the optimizer may not unroll any loops m If n gt 1 the optimizer may unroll loops n times If any loops are actually unrolled the executable file becomes larger For further information see the Fortran Programming Guide chapter on performance and optimization Example unrolling loops two times CSPRAGMA SUN UNROLL 2 The WEAK Directive The WEAK directive defines a symbol to have less precedence than an earlier definition of the same symbol This pragma is used mainly in sources files for building libraries The linker does not produce an error message if it is unable to resolve a weak symbol CSPRAGMA WEAK namel name2 WEAK name1 defines name1 to be a weak symbol The linker does not produce an error message if it does not find a definition for name1 WEAK name1 name2 defines name1 to be a weak symbol and an alias for name2 If your program calls but does not define name1 the linker uses the definition from the library However if your program defines its own version of name1 then the program s definition is used and the weak global definition of name in the library is Chapter 2 Using S
129. is performed with no conversion a If both operands are Boolean the operation is performed as if they were integers a No user specified function can generate a Boolean result although some nonstandard intrinsics can Boolean and logical types differ as follows a Variables arrays and functions can be of logical type but they cannot be Boolean type a There is a LOGICAL statement but no BOOLEAN statement Gl A logical variable constant or expression represents only two values TRUE or FALSE A Boolean variable constant or expression can represent any binary value a Logical entities are invalid in arithmetic relational or bitwise logical expressions Boolean entities are valid in all three Appendix C Fortran 95 Features and Differences 149 150 Alternate Forms of Boolean Constants 95 allows a Boolean constant octal hexadecimal or Hollerith in the following alternate forms no binary Variables cannot be declared Boolean Standard Fortran does not allow these forms Octal ddddddB where d is any octal digit You can use the letter B or b There can be 1 to 11 octal digits 0 through 7 11 octal digits represent a full 32 bit word with the leftmost digit allowed to be 0 1 2 or 3 Each octal digit specifies three bit values The last right most digit specifies the content of the right most three bit positions bits 29 30 and 31 If less than 11 digits are present the
130. isable use of incremental linker ild xildoff Link with optimized math library xlibmopt Link with Sun Performance Library xlic_lib sunperf Link editor option ZX Generate pure libraries with no relocations ztext Numerics and Floating Point Use non standard floating point preferences fnonstd Select SPARC non standard floating point fns Enable runtime floating point overflow during input fpover Select IEFE floating point rounding mode fpround r Select floating point optimization level fsimple n Select floating point trapping mode ftrap t Promote single precision constants to double precision r8const Enable interval arithmetic and set the appropriate floating xia e point environment includes xinterval Enable interval arithmetic extensions xinterval e Chapter 3 Fortran Compiler Options 41 TABLE 3 3 Compiler Options Grouped by Functionality Continued Function Option Flag Optimization and Performance Analyze loops for data dependencies depend Optimize using a selection of options fast Specify optimization level On Pad data layout for efficient use of cache pad p Allocate local variables on the memory stack stackvar Enable loop unrolling unroll m Enable optimization across source files xcrossfile n Invoke interprocedural optimizations pass xipo n Set highest optimization level for pragma OPT xmaxopt n Assert that no memory based traps will occur xsafe mem Do no opt
131. it is impossible to BACKSPACE a FORM BINARY file because there is no way of telling where to backspace to A READ ona BINARY file will read as much data as needed to fill the variables on the input list See Appendix C or the Fortran 77 Language Reference for details a Recursive I O possible on different units this is because the 90 I O library is MT Warm Appendix B Features Release History 141 142 RECL 2147483646 2 2 is the default record length on sequential formatted list directed and namelist output Default was 267 ENCODE and DECODE are recognized and implemented as described in the FORTRAN 77 Language Reference Manual Naming of scratch files is the same as with 77 Non advancing I O is enabled with ADVANCE NO as in write a ADVANCE NO n read n Handling of I O on internal files follows the Fortran 90 standard more closely than was the case with 90 1 2 Also calls to routines that do internal I O are allowed on I O lists This was not allowed with 1 2 or 77 a Operational Differences a Modules are handled differently Compiling a source code that contains one or more MODULE units now causes an information file name mod to be generated for each module The name of this information file is the name of the module in lower case with mod suffix A mod file must be available before the module can appear on a USE statement This mea
132. ive They apply to the rest of the source file in which they appear up to the next FREE or FIXED directive Sun Parallelization Directives Sun style parallelization directives are the default mp sun compiler option and are detailed in the chapter on parallelization in the Fortran Programming Guide TABLE E 3 Sun Style Parallelization Directives Summary Format CSPAR directive optional_qualifiers initial line C PAR amp more_qualifiers continuation line Fixed format the directive indicator may be C as shown c or Separate multiple qualifiers with commas Characters beyond column 72 ignored unless e compiler option specified TASKCOMMON CSPAR TASKCOMMON block_name Directive Declares variables in common block block_name as thread private private to a thread but global within the thread Declaring a common block TASKCOMMON requires that this directive appear after every common declaration of that block Appendix E Fortran Directives Summary 177 178 TABLE E 3 Sun Style Parallelization Directives Summary Continued DOALL Directive DOSERIAL Directive CSPAR DOALL qualifiers Parallelize DO loop that follows Qualifiers are PRIVATE list declare names on list PRIVATE SHARED list declare names on list SHARED MAXCPUS n use no more than n threads READONLY list listed variables not modified in loop SAVELAST save last value of all private variables STOREBACK list save last value of listed v
133. ization level 77 95 n has the value 1 through 5 and corresponds to the optimization levels of 01 through 05 If not specified the compiler uses 5 This option enables the CSPRAGMA SUN OPT n directive see page 28 when it appears in the source input Without this option the compiler treats these lines as comments If such a pragma directive appears with an optimization level greater than the maximum level on the xmaxopt flag the compiler uses the level set by xmaxopt xmemalign lt a gt lt b gt Specify maximum assumed memory alignment and behavior of misaligned data accesses 77 95 For memory accesses where the alignment is determinable at compile time the compiler will generate the appropriate load store instruction sequence for that data alignment For memory accesses where the alignment cannot be determined at compile time the compiler must assume an alignment to generate the needed load store sequence The xmemalign flag allows the user to specify the maximum memory alignment of data to be assumed by the compiler for those indeterminate situations It also specifies the error behavior at runtime when a misaligned memory access does take place The value specified consists of two parts a numeric alignment value lt a gt and an alphabetic behavior flag lt b gt Chapter 3 Fortran Compiler Options 111 Allowed values for alignment lt a gt are Assume at most 1 byte alignment Assume
134. ize performance for the host platform The compiler generates code optimized for the host platform It determines the available architecture chip and cache properties of the machine on which the compiler is running native6 4 Compile for native 64 bit environment Fortran User s Guide July 2001 Set the architecture chip and cache properties for the 64 bit environment on the machine on which the compiler is running generic Get the best performance for generic architecture chip and cache The compiler expands xtarget generic to xarch generic xchip generic xcache generic This is the default value generic64 Compile for generic 64 bit environment This expands to xarch v9 xcache generic xchip generic platform name Get the best performance for the specified platform Appendix D gives a complete list of current SPARC platform names accepted by the compilers For example xtarget ultra2i xtime Synonym for time 77 95 xtypemap spec Specify default data mappings 77 95 This option provides a flexible way to specify the byte sizes for default data types Use of this option is preferred over db1 and r8 and applies to both default size variables and constants The specification string spec may contain any or all of the following in a comma delimited list real size double size integer size The accepted data size values are 64 128 for real and double 32 64 a
135. king 95 and 77 compiled objects together into a single executable xlibmil Synonym for libmil 77 95 xlibmopt Use library of optimized math routines 77 95 Use selected math routines optimized for speed This option usually generates faster code It may produce slightly different results if so they usually differ in the last bit The order on the command line for this library option is not significant xlic_lib sunperf Link with the Sun Performance Library 77 95 For example 77 o pgx fast pgx f xlic_lib sunperf As with 1 this option should appear on the command line after all source and object file names This option must be used to link with the Sun Performance Library See the Sun Performance Library User s Guide xlicinfo Show license server information 77 95 Use this option to return license information about the licensing system in particular the name of the license server and the user ID for each of the users who have licenses checked out 110 Fortran User s Guide July 2001 Generally with this option no compilation takes place and a license is not checked out This option is normally used alone with no other options However if a conflicting option is used then the last one on the command line prevails and there is a warning xloopinfo Synonym for loopinfo 77 95 xmaxopt n Enable optimization pragma and set maximum optim
136. l if the upper limit is smaller than the lower limit unlike some legacy implementations of Fortran openmp Enable explicit parallelization with Fortran 95 OpenMP directives Fortran parallelization features require a Forte for HPC license 95 This option is a macro that combines these options mp openmp xplicitpar stackvar D_OPENMP 200011 OpenMP directives are summarized in Appendix E To run a parallelized program in a multithreaded environment you must set the PARALLEL or OMP_NUM_THREADS environment variable prior to execution This tells the runtime system the maximum number of threads the program can create The default is 1 In general set the PARALLEL or OMP_NUM_THREADS variable to the available number of processors on the target platform OpenMP requires the definition of the preprocessor symbol _OPENMP to have the decimal value YYYYMM where YYYY and MM are the year and month designations of the version of the OpenMP Fortran API that the implementation supports Chapter 3 Fortran Compiler Options 79 80 P Compile for profiling with the prof profiler 77 95 Prepare object files for profiling see prof 1 If you compile and link in separate steps and also compile with the p option then be sure to link with the p option p with prof is provided mostly for compatibility with older systems pg profiling with gprof is possibly a better alternative See the Fortran Programming Gui
137. limited descriptors 64 memorysize unlimited demo Example Set the main stack size to 64 Megabytes demo limit stacksize 65536 Example Set each thread stack size to 8 Megabytes demo setenv STACKSIZE 8192 For further information of the use of stackvar with parallelization see the Parallelization chapter in the Fortran Programming Guide See csh 1 for details on the limit command stop_status yn Permit STOP statement to return an integer status value 77 95 yn is either yes or no The default is no With stop_status yes a STOP statement may contain an integer constant That value will be passed to the environment as the program terminates Chapter 3 Fortran Compiler Options 89 90 STOP 123 The value must be in the range 0 to 255 Larger values are truncated and a run time message issued Note that STOP stop string is still accepted and returns a status value of 0 to the environment although a compiler warning message will be issued The environment status variable is status for the C shell csh and for the Bourne and Korn shells sh and ksh temp dir Define directory for temporary files 77 95 Set directory for temporary files used by the compiler to be dir No space is allowed within this option string Without this option the files are placed in the tmp directory time Time each compilation phase 77 95 The time spent and resources used in e
138. line 19 driver show commands with dryrun 54 55 frequently used options 39 show version 91 timing 90 verbose messages 92 compilers accessing 4 constant arguments copyargs 50 continuation lines 55 147 conventions file name suffixes 20 copy restore 47 cpp C preprocessor 21 51 58 Cray pointer 153 pointer and Fortran 95 pointer 154 cross reference table Xlist 94 D data alignment with db1_align_all 54 alignment with f 58 alignment with xmemalign 111 allow misaligned data misalign 73 COMMON alignment with aligncommon 46 default sizes and db1 53 default sizes and r8 85 interpret REAL as DOUBLE PRECISION 85 mappings with xtypemap 121 promote constants to REAL 8 86 data dependency depend 54 dbx compile with g option 66 194 Fortran User s Guide July 2001 faster initialization 119 debugging check array subscripts with c 49 cross reference table 93 g option 66 global program checking with Xlist 93 show compiler commands with dryrun 54 55 utilities 11 VMS D debugging statements 92 with optimization 66 without object files 119 Xlist 11 default include file paths 68 define symbol for cpp Dname 51 differences Sun WorkShop Fortran 95 147 directives Fortran 77 24 loop unrolling 27 OpenMP Fortran 95 30 180 optimization level 28 parallelization 29 parallelization 95 162 parallelization Cray Sun or OpenMP 73 sum
139. lism is not supported Value is TRUE or FALSE The default if not set is FALSE Example setenv OMP_NESTED TRUE Fortran User s Guide July 2001 TABLEE 8 Environment variables not part of the OpenMP Fortran API SUNW_MP_WARN Controls warning messages issued by the runtime library If set TRUE the runtime library issues warning messages to stderr FALSE disables warning messages The default is FALSE Example setenv SUNW_MP_WARN TRUE SUNW_MP_THR_IDLE Controls the end of task status of each thread executing the parallel part of a program You can set the value to spin sleep ns or sleep mms The default is SPIN a thread should spin or busy wait after completing a parallel task until a new parallel task arrives Choosing SLEEP time specifies the amount of time a thread should spin wait after completing a parallel task If while a thread is spinning a new task arrives for the thread the tread executes the new task immediately Otherwise the thread goes to sleep and is awakened when a new task arrives time may be specified in seconds ns or just n or milliseconds nms SLEEP with no argument puts the thread to sleep immediately after completing a parallel task SLEEP SLEEP 0 SLEEP 0s and SLEEP Oms are all equivalent Example setenv SUNW_MP_THR_IDLE 50ms STACKSIZE Sets the thread stack size The value is in kilobytes Example setenv STACKSIZE 8192 sets the thread stack size to 8Mb Appendix E For
140. llelization directives Avoid parallel if you do your own thread management See the discussion of mt on page 74 Parallelization options like parallel are intended to produce executable programs to be run on multiprocessor systems On a single processor system parallelization generally degrades performance To run a parallelized program in a multithreaded environment you must set the PARALLEL or OMP_NUM_THREADS environment variable prior to execution This tells the runtime system the maximum number of threads the program can create The default is 1 In general set the PARALLEL or OMP_NUM_THREADS variable to the available number of processors on the target platform Chapter 3 Fortran Compiler Options 81 82 pic If you use parallel and compile and link in one step then linking automatically includes the multithreading library and the thread safe Fortran runtime library If you use parallel and compile and link in separate steps then you must also link with parallel See the Fortran Programming Guide chapter Parallelization for further information Compile for profiling with the gprof profiler 77 95 Compile self profiling code in the manner of p but invoke a runtime recording mechanism that keeps more extensive statistics and produces a gmon out file when the program terminates normally Generate an execution profile by running gprof See the gprof 1 man page and the Fortran Programming Guid
141. lly used with both G and pic With ztext if 1d finds an incomplete relocation in the text segment then it does not build the library If it finds one in the data segment then it generally builds the library anyway the data segment is writable Without ztext 1d builds the library relocations or not A typical use is to make a library from both source files and object files where you do not know if the object files were made with pic Chapter 3 Fortran Compiler Options 123 124 Example Make library from both source and object files demo 77 G pic ztext o MyLib hMyLib a f b f x o y o An alternate use is to ask if the code is position independent already compile without pic but ask if it is pure text Example Ask if it is pure text already even without pic demos 77 G ztext o MyLib hMyLib a f b f x o y o If you compile with ztext and 1d does not build the library then you can recompile without ztext and 1d will build the library The failure to build with ztext means that one or more components of the library cannot be shared however maybe some of the other components can be shared This raises questions of performance that are best left to you the programmer Fortran User s Guide July 2001 APPENDIX A Runtime Error Messages This appendix describes the error messages generated by the Fortran I O library signal handler and operating system Opera
142. low lists suboptions that can be individually selected TABLE 3 10 vax Suboptions vax Affect blank_zero Treat blank in a numeric field as zero bslash Allow backslash in character constants debug Allow VMS Fortran D debugging statements logical_name Allow VMS Fortran style logical file names oct_const Allow double quote character to signify octal constants param Allow non standard form of PARAMETER statement rsize Allow unformatted record size in words rather than bytes struct_align Align structures as in VMS Fortran all and none can also be used to select all or none of these suboptions Sub options accumulate from left to right For example to enable all but one feature vax all no rsize See also xl and misalign 92 Fortran User s Guide July 2001 vpara Show verbose parallelization messages 77 95 As the compiler analyzes loops explicitly marked for parallelization with directives it issues warning messages about certain data dependencies it detects but the loop will still be parallelized Example vpara for verbose parallelization warnings demos 77 explicitpar vpara any f any f MAIN any any f line 11 Warning the loop may have parallelization inhibiting reference w Suppress warning messages 77 95 This option suppresses most warning messages However if one option overrides all or part of an option earlier on the command line you
143. mary of all directives 175 weak linking 27 directory temporary files 90 DOALL directive 30 documentation index 5 documentation accessing 5 DOSERIAL directive 30 dynamic library build G 66 name a dynamic library 67 E environment program terminations by STOP 89 environment variables usage 32 error messages 90 130 I O 126 message tags 56 suppress with erroff 56 with error 11 error error message display 11 exceptions floating point 64 trapping 65 executable file built in path to dynamic libraries 84 name 78 strip symbol table from 87 explicit typing 91 explicit parallelization directives 29 extensions non ANSI ansi flag 47 VAX VMS Fortran features with x1 109 extensions and features 10 external C functions 26 external names 57 F 77 90 command line 19 37 features Fortran 95 147 release history 137 features and extensions 10 FFLAGS environment variable 33 file M See module files executable 20 object 20 size too big 33 file names recognized by the compiler 20 recognized by the compiler f95 148 FIXED directive 161 fixed format source 61 flags See options floating point fpversion displays hardware platform 32 interval arithmetic 107 non standard 62 preferences fsimple 64 rounding 63 trapping mode 65 See also the Numerical Computation Guide Fortran features and extensions 10 mixing 77 and 90 compilations 23 preprocess
144. mote both constants and variables use xt ypemap page 121 reduction Recognize reduction operations in loops 77 95 Analyze loops for reduction operations during automatic parallelization There is potential for roundoff error with the reduction A reduction operation accumulates the elements of an array into a single scalar value For example summing the elements of a vector is a typical reduction operation Although these operations violate the criteria for parallelizability the compiler can recognize them and parallelize them as special cases when reduction is specified See the Fortran Programming Guide chapter Parallelization for information on reduction operations recognized by the compilers This option is usable only with the automatic parallelization options autopar or parallel It is ignored otherwise Explicitly parallelized loops are not analyzed for reduction operations Example Automatically parallelize with reduction demos 77 parallel reduction any f Fortran User s Guide July 2001 Compile and only generate assembly code 77 95 Compile the named programs and leave the assembly language output on corresponding files suffixed with s No o file is created S Strip the symbol table out of the executable file 77 95 This option makes the executable file smaller and more difficult to reverse engineer However this option inhibits debugging with dbx or other tools and overrides
145. mplex a Cray style parallelization directives including TASKCOMMON with extensions for 95 m OpenMP parallelization directives accepted by 95 m Global peephole and potential parallelization optimizations produce high performance applications Benchmarks show that optimized applications can run significantly faster when compared to unoptimized code Common calling conventions on Solaris systems permit routines written in C or C to be combined with Fortran programs Support for 64 bit enabled Solaris environments on UltraSPARC platforms a Call by value sVAL implemented in both 77 and 95 a Interoperability between Fortran 77 and Fortran 95 programs and object binaries m Interval Arithmetic expressions in 95 Fortran User s Guide July 2001 See Appendix B for details on new and extended features added to the compilers with each software release Other Fortran Utilities The following utilities provide assistance in the development of software programs in Fortran a Sun WorkShop Performance Analyzer In depth performance analysis tool for single threaded and multi threaded applications See analyzer 1 m asa This Solaris utility is a Fortran output filter for printing files that have Fortran carriage control characters in column one Use asa to transform files formatted with Fortran carriage control conventions into files formatted according to UNIX line printer conventions See asa 1
146. n instructs the code generator to adjust the default latency time between a prefetch and its associated load or store by the specified factor The prefetch latency is the hardware delay between the execution of a prefetch instruction and the time the data being prefetched is available in the cache The compiler assumes a prefetch latency value when determining how far apart to place a prefetch instruction and the load or store instruction that uses the prefetched data Note The assumed latency between a prefetch and a load may not be the same as the assumed latency between a prefetch and a store Fortran User s Guide July 2001 The compiler tunes the prefetch mechanism for optimal performance across a wide range of machines and applications This tuning may not always be optimal For memory intensive applications especially applications intended to run on large multiprocessors you may be able to obtain better performance by increasing the prefetch latency values To increase the values use a factor that is greater than 1 A value between 5 and 2 0 will most likely provide the maximum performance For applications with datasets that reside entirely within the external cache you may be able to obtain better performance by decreasing the prefetch latency values To decrease the values use a factor that is less than 1 To use the xprefetch latx factor option start with a factor value near 1 0 and run performance tests against
147. n the Fortran 95 compiler for compatibility int is the preferred Fortran 95 standard intrinsic M int J 2 Mixed Language Linking with xlang The new xlang option provides an easy way to link object files and libraries compiled by 77 with 95 object files The proper runtime environment is insured when using xlang See the CC 1 man page and Chapter 3 page 109 95 New Features in Sun WorkShop 6 The following lists the new and changed features in the Fortran 95 compiler released with Sun Performance WorkShop 6 Compliance The 95 is fully compliant with the Fortran 95 standard New Command The Fortran 95 compiler can be invoked by either the 95 or 90 command Debugging Optimized Code Restrictions limiting use of g with other options has been relaxed allowing debugging parallelized and 04 or 05 optimized codes with dbx and the Sun WorkShop debugger Source Filename Extensions The compiler will accept source files with 95 and 90 filename extensions as well as F95 and F90 Interval Arithmetic This release implements a number of extensions that enable interval arithmetic computations See the Interval Arithmetic Programming Reference and interval_arithmetic README for details Enhanced Array Optimizations The compiler now performs aggressive array optimizations at levels 04 and 05 Hyper Linked Diagnostic Messages Sun WorkShop online help now interprets 95 error diagnostics in the Building wi
148. n v9 v9a or v9b ansi Identify many nonstandard extensions 77 95 Warning messages are issued for any uses of non standard Fortran 77 or Fortran 95 extensions in the source code arg local Preserve actual arguments over 77 ENTRY statements When you compile a subprogram with alternate entry points with this option 77 uses copy restore to preserve the association of dummy and actual arguments For example the following program would require compilation with arg local to insure proper execution A SETUP ALPHA BETA GAMMA ZORK RYE E FXGAMMA F B RETURN END FXGAMMA GCONST FUNCTION SETUP A1 A2 A3 XGAMMA F F GAMMA Without this option there is no guarantee that the correct values of the actual arguments from the S ETUP call will be referenced when the routine is entered through FXGAMMA Code that relies on arg local is nonstandard Chapter 3 Fortran Compiler Options 47 48 autopar Enable automatic loop parallelization Fortran parallelization features require a Forte for HPC license 77 95 Finds and parallelizes appropriate loops for running in parallel on multiple processors Analyzes loops for inter iteration data dependencies and loop restructuring If the optimization level is not specified 03 or higher it will automatically be raised to 03 To improve performance also specify
149. nce Library User s Guide for Fortran and C for details Man pages for the performance library routines are in section 3P Interval Arithmetic The Fortran 95 compiler provides the compiler flags xia and xinterval to enable new language extensions and generate the appropriate code to implement interval arithmetic computations See the Fortran 95 Interval Arithmetic Programming Reference for details 12 Man Pages Online manual man pages provide immediate documentation about a command function subroutine or collection of such things See the Preface for the proper setting of the MANPATH environment variable for accessing Sun WorkShop man pages You can display a man page by running the command demo man topic Fortran User s Guide July 2001 Throughout the Fortran documentation man page references appear with the topic name and man section number f95 1 is accessed with man 95 Other sections denoted by icee_flags 3M for example are accessed using the s option on the man command demo man s 3M ieee_flags The Fortran library routines are documented in the man page section 3F The following lists man pages of interest to Fortran users 77 1 and 95 1 analyzer 1 asa 1 dbx 1 fpp 1 cpp 1 fsplit 1 ieee_flags 3M ieee_handler 3M matherr 3M ild 1 1a 1 The Fortran compilers command line options Sun WorkShop Performance Analyzer Fortran carriage control print output pos
150. nct U Select preprocessor cpp or fpp to use xpp fppl cpp Allow recursive subprogram calls xrecursive Target Platform Optimize for the host system native Specify target platform instruction set for the optimizer xarch a Specify target cache properties for optimizer xcache a Specify target processor for the optimizer xchip a Specify target platform for the optimizer xtarget a Commonly Used Options The Sun Fortran compilers have many features that are selectable by optional command line parameters The short list below of commonly used options is a good place to start TABLE 3 4 Commonly Used Options Action Option Debug global program checking across routines for Xlist consistency of arguments commons and so on Debug produce additional symbol table information to enable g the dox and Sun WorkShop debugging Performance invoke the optimizer to produce faster running O n programs Chapter 3 Fortran Compiler Options 43 TABLE 3 4 Commonly Used Options Continued Action Option Performance Produce efficient compilation and run times for fast the native platform using a set of predetermined options Dynamic Bdynamic or static Bstatic library binding Bx Compile only Suppress linking make a o file for each source c file Output file Name the executable output file nm instead of o nm a out Source code Compile fixed format Fortran 77 code with 95 fixed
151. nd arrays are local unless they are m Arguments in a SUBROUTINE or FUNCTION statement already on stack Global items in a COMMON SAVE or STATIC statement m Items initialized in a type statement or DATA statement such as REAL X 8 0 or DATA X 8 0 f77 only Initializing a local variable in a DATA statement after an executable reference to that variable is an extension to 77 and is flagged as an error when stackvar is used demo cat stak f real x x 1 t 0 print t data x 3 0 print x t end demo 77 o stak stackvar stak f stak f MAIN Stak line 5 Error attempt to initialize an automatic variable x Fortran User s Guide July 2001 Putting large arrays onto the stack with stackvar can overflow the stack causing segmentation faults Increasing the stack size may be required The initial thread executing the program has a main stack while each helper thread of a multithreaded program has its own thread stack The default stack size is about 8 Megabytes for the main stack and 1 Megabyte 2 Megabytes on SPARC V9 platforms for each thread stack The limit command with no parameters shows the current main stack size If you get a segmentation fault using stackvar try increasing the main and thread stack sizes Example Show the current main stack size demo limit cputime unlimited filesize unlimited datasize 523256 kbytes stacksize 8192 kbytes a coredumpsize un
152. nd mixed for integer For example xtypemap real 64 double 128 integer 64 This option applies to all variables declared with default specifications without explicit byte sizes as in REAL XYZ resulting in a 64 bit XYZ Also all single precision REAL constants are promoted to REAL 8 Chapter 3 Fortran Compiler Options 121 122 The allowable combinations on each platform are m real 32 m real 64 m double 64 m double 128 m integer 32 m integer 64 a integer mixed f77 only The integer mixed mapping specifies 8 byte data but only 4 byte arithmetic and is only available with 77 Preferred is integer 64 The 77 flags dbl and r8 options have their xt ypemap equivalents m dbl sameas xtypemap real 64 double 128 integer 64 m r8 sameas xtypemap real 64 double 128 integer mixed There are two additional possibilities on m xtypemap real 64 double 64 integer mixed m xtypemap real 64 double 64 integer 64 which map both default REAL and DOUBLE to 8 bytes and may be preferable over the use of db1 or r8 because they do not promote DOUBLE PRECISION to QUAD PRECISION Note that INTEGER and LOGICAL are treated the same and COMPLEX is mapped as two REALs Also DOUBLE COMPLEX will be treated the way DOUBLE is mapped xunroll n Synonym for unroll n 77 95 xvector yes no Enable automatic calls t
153. nd not with 1ibF77 For example perform the link step with 90 files 1 77compat even if the main program is an 77 program a The structure of 90 COMMON is now compatible with 77 90 scalar pointers are compatible with C pointers Fortran 77 New Features and Changes This section lists the new features and behavior changes in 77 introduced with this and previous releases 77 New Features in Sun WorkShop 6 update 2 No new features were introduced in 77 with the release of Sun WorkShop 6 update 2 77 New Features in Sun WorkShop 6 update 1 The release of Sun WorkShop 6 update 1 introduced the following new or changed features in 77 m UltraSPARC III Support The xtarget and xchip options now accept ult ra3 and the compiler will generate optimized code for the UltraSPARC III processor See Chapter 3 page 120 m Prefetch added to fast The xprefetch flag has been added to the fast option set fast automatically sets a number of optimization flags for best execution speed on the compiling platform Adding xprefetch takes advantage of the UltraSPARC II and III prefetch mechanism and can add a substantial performance gain in code with loops that process data See Chapter 3 page 59 Appendix B_ Features Release History 143 77 New Features in Sun WorkShop 6 Sun WorkShop 6 Fortran 77 includes the following new and changed features a I O Extension Opening a file with OPEN FORM BIN
154. ndow creating hypertext links from the error message to descriptive online help OpenMP The compiler accepts OpenMP explicit parallelization directives The OpenMP specifications can be viewed at http www openmp org Appendix B_ Features Release History 139 140 m AUTOSCOPE added to Cray style DOALL parallelization directive a New Changed Command Line Options aligncommon aligns COMMON block elements to specific byte boundaries r8const promotes single precision data constants to REAL 8 xinterval and xia enable interval arithmetic extensions xmemalign specifies general alignment in memory of data elements mp openmp and openmp enable native compilation of OpenMP explicit parallelization directives xprefetch for enabling UltraSPARC prefetch instructions has been expanded to include additional sub options xrecursive allows recursive calls from subprograms without the RECURSIVE attribute xtypemap has an expanded set of possible data type specifications fast extended to set 05 fsimple 2 xvector yes and pad common m Use of 95 s parallelization features requires a Sun WorkShop HPC license New Features Released In 90 2 0 The following new and changed features appeared in the 90 2 0 compiler released with Sun WorkShop 5 0 over the earlier 90 1 2 release m New options Most 77 options now recognized by 90 fpover detects floating point overflows in I O
155. ng 156 Fortran User s Guide July 2001 Other Language Extensions Extended ALLOCATABLE Attribute Recent decisions by the Fortran 95 standards organizations have extended the data entities allowed for the ALLOCATABLE attribute Previously this attribute was limited to locally stored array variables It is now allowed with m array components of structures a dummy arrays m array function results Allocatable entities remain forbidden in all places where they may be storage associated COMMON blocks and EQUIVALENCE statements Allocatable array components may appear in SEQUENCE types but objects of such types are then prohibited from COMMON and EQUIVALENCE VALUE Attribute Fortran 2000 The 95 compiler recognizes the VALUE type declaration attribute This attribute has been proposed for the Fortran 2000 standard Specifying a subprogram dummy input argument with this attribute indicates that the actual argument is passed by value The following example demonstrates the use of the VALUE attribute with a C main program calling a Fortran 95 subprogram with a literal value as an argument C code include lt stdlib h gt int main int ac char av to_fortran 2 Fortran code subroutine to_fortran i integer value i print i end Appendix C Fortran 95 Features and Differences 157 158 Stream I O Fortran 2000 A new stream
156. nly for programs in which such faults cannot occur Because few programs incur memory based traps you can safely use this option for most programs Do not use this option with programs that explicitly depend on memory based traps to handle exceptional conditions Chapter 3 Fortran Compiler Options 119 xsb Synonym for sb 77 95 xsbfast Synonym for sbfast 77 95 xspace Do not allow optimizations to increase code size 77 95 120 Do no optimizations that increase the code size Example Do not unroll or parallelize loops if it increases code size xtarget t Specify target platform for optimization 77 95 Specify the target platform for the instruction set and optimization t must be one of native native 4 generic generic64 platform name The xtarget option permits a quick and easy specification of the xarch xchip and xcache combinations that occur on real platforms The only meaning of xtarget is in its expansion The performance of some programs may benefit by providing the compiler with an accurate description of the target computer hardware When program performance is critical the proper specification of the target hardware could be very important This is especially true when running on the newer SPARC processors However for most programs and older SPARC processors the performance gain is negligible and a generic specification is sufficient native Optim
157. not by itself allocate local variables on the memory stack xreduction Synonym for reduction 77 95 xregs r Specify register usage 77 95 r is a comma separated list that consists of one or more of the following no appl no float Where the is shown it is a required character Example xregs appl no float m appl Allow using the application registers On SPARC systems certain registers are described as application registers Using these registers can increase performance because fewer load and store instructions are needed However such use can conflict with some old library programs written in assembly code The set of application registers depends on the SPARC platform a xarch v8 or v8a registers g2 3g3 and g4 a xarch v8 or v8a registers g2 3g3 and g4 a xarch v8plus or v8plusa registers g2 g3 and g4 a xarch v9 or v9a registers g2 and g3 m no appl Do not use the app registers a float Allow using the floating point registers as specified in the SPARC ABI You can use these registers even if the program contains no floating point code 118 Fortran User s Guide July 2001 XS m no float Do not use the floating point registers With this option a source program cannot contain any floating point code The default is xregs appl float Allow debugging by dbx without object o files 77 95 With xs if you move executables to another direct
158. ns that all MODULE files must be compiled and the module information files created before compiling any file referencing a MODULE in a USE statement ftrap common is the default trapping mode Routines from the Sun Performance Library are automatically linked to perform array operations m New Language Elements a Some Fortran 95 elements are implemented The attributes PURE and ELEMENTAL The enhanced forms of MAXVAL and MINVAL New data types are recognized COMPLEX 32 REAL 16 INTEGER 8 also 1 2 LOGICAL 8 also 1 2 Some data representations have changed from 90 1 2 INTEGER 2 is now 2 bytes not 4 INTEGER 1 is now 1 byte not 4 OGICAL 2 is now 2 bytes not 4 OGICAL 1 is now 1 byte not 4 This will affect programs that read binary data files containing these data items that were written with 90 programs compiled with the 1 2 compiler A workaround would be to change the declarations to be INTEGER 4 or LOGICAL 4 instead of 1 or 2 when compiling with 2 0 Fortran User s Guide July 2001 a Call by value VAL is implemented in the same manner as 77 The only difference is that 90 2 0 allows REAL 8 and REAL 16 to be passed to C routines as doubles and long doubles a 77 and C Interoperability with 90 2 0 a To mix 77 and 90 object binaries link with the 77 compatibility library libf 77compat a
159. nts eeCl Cs Chapter1 Introduction 15 16 Fortran User s Guide July 2001 CHAPTER 2 Using Sun Fortran Compilers This chapter describes how to use the Fortran 77 and Fortran 95 compilers The principal use of any compiler is to transform a program written in a procedural language like Fortran into a data file that is executable by the target computer hardware As part of its job the compiler may also automatically invoke a system linker to generate the executable file The Sun Fortran 77 and Fortran 95 compilers can also be used to a Generate a parallelized executable file for multiple processors parallel m Analyze program consistency across source files and subroutines and generate a report Xlist a Transform source files into a Relocatable binary 0 files to be linked later into an executable file or static library a file a A dynamic shared library so file G m Link files into an executable file Compile an executable file with runtime debugging enabled g Compile with runtime statement or procedure level profiling pg Compile an executable file with runtime parallelized loop profiling Z1p m Check source code for ANSI standards conformance ansi A Quick Start This section provides a quick overview of how to use the Sun Fortran compilers to compile and run Fortran programs A full reference to command line options appears in the next chapter 18
160. o the SPARC vector library functions 77 95 With xvector yes the compiler is permitted to transform certain math library calls within DO loops into single calls to the equivalent vectorized library routine whenever possible This could result in a performance improvement for loops with large loop counts The compiler defaults to xvector no Specifying xvector by itself defaults to xvector yes Fortran User s Guide July 2001 This option also triggers depend Follow xvector with nodepend on the command line to cancel the dependency analysis The compiler will automatically notify the linker to include the libmvec and libc libraries in the load step if xvector appears However to compile and link in separate steps requires specifying xvector on the link step as well to correctly select these necessary libraries xvpara Synonym for vpara 77 Z1lp Obsolete Compile for loop performance profiling by looptool 77 95 This option and looptool are no longer supported Use the Sun WorkShop Performance Analyzer instead Refer to Analyzing Program Performance With Sun WorkShop and the analyzer 1 man page for more information ztext Generate only pure libraries with no relocations 77 95 Do not make the library if relocations remain The general purpose of ztext is to verify that a generated library is pure text instructions are all position independent code Therefore it is genera
161. ograms communicating through files with unformatted I O if one program is compiled with db1 then the other program must similarly be compiled Be also aware that this option alters the default data size of function names including calls to library functions unless the function name is typed explicitly with a data size dbl_align_all yes no Force alignment of data on 8 byte boundaries 77 95 The value is either yes or no If yes all variables will be aligned on 8 byte boundaries Default is dbl_align_all no When compiling for 64 bit environments with xarch v9 or v9a this flag will align quad precision data on 16 byte boundaries This flag does not alter the layout of data in COMMON blocks or user defined structures On SPARC use with dalign to enable added efficiency with multi word load stores If used all routines must be compiled with this flag depend Analyze loops for data dependencies 77 95 Analyze loops for data dependencies and do loop restructuring This option will raise the optimization level to 03 if no optimization level is specified or if it is specified less than 03 depend is also included with fast autopar and parallel See the Fortran Programming Guide Disallow dynamic libraries See d y n page 55 77 95 Fortran User s Guide July 2001 dryrun Show commands built by 77 or 95 command line driver but do not compile 77 95 Useful
162. ol a CHARACTER 74 a PRINT a RETURN END demo 77 c silent s f demo 95 xlang f77 m f95 s o demo a out This is a test demo m The FORTRAN 77 library is generally compatible with 95 Appendix C Fortran 95 Features and Differences 165 166 Example 95 main calls a routine from the FORTRAN 77 library demo cat tdtime f95 REAL e dtime t 2 e dtime t DO i 1 100000 as as cos sqrt float i END DO e dtime t PRINT elapsed e user t 1 sys END demo 95 tdtime f95 demo a out elapsed 0 14 user 0 14 sys 0 0E 0 demo rhp 2 See dt ime 3F Intrinsics The Fortran 95 standard supports the following intrinsic functions that FORTRAN 77 does not have If you use one of these names in your program you must add an EXTERNAL statement to make 95 use your function rather than the intrinsic one Fortran 95 intrinsics ADJUSTL ADJUSTR ALL ALLOCATED ANY BIT_SIZE COUNT CSHIFT T DIGITS DOT_PRODUCT EOSHIFT EPSILON EXPONENT HUGE KIND LBOUND LEN_TRIM MA MAXEXPONENT MAXLOC MAXVAL MERG MINEXPONENT MINLOC MINVAL NEAREST PACK PRECISION PRES U RODUCT RADIX RANGE REPEAT EI ESHAPE RRSPACING SCALE SCAN I P SELECTED_INT_KIND SE
163. ompiler regarding special optimization or parallelization choices Compiler directives are also sometimes called pragmas The compilers recognize a set of general directives and parallelization directives Fortran 95 also processes OpenMP shared memory multiprocessing directives Directives unique to 95 are described in Appendix C A complete summary of all the directives recognized by 77 and 95 appears in Appendix E Note Directives are not part of the Fortran standard Fortran User s Guide July 2001 General Directives The various forms of a general Sun Fortran directive are CSPRAGMA keyword a a keyword a a CSPRAGMA SUN keyword a a keyword a a CSPRAGMA SPARC keyword a a keyword a a The variable keyword identifies the specific directive Additional arguments or suboptions may also be allowed Some directives require the additional keyword SUN or SPARC as shown above A general directive has the following syntax In column one any of the comment indicator characters c C or For 95 free format is the only comment indicator recognized PRAGMA The examples in this chapter assume fixed format The next seven characters are PRAGMA no blanks in either uppercase or lowercase With 77 directives using the comment indicator character may appear in any position on the line With 95 this is only possible for free format source
164. ons Email your comments to Sun at this address docfeedback sun com Before You Begin 7 8 Fortran User s Guide July 2001 CHAPTER 1 Introduction The Sun Fortran compilers 77 and 95 described in this book and the companion Sun WorkShop Fortran Programming Guide are available under the Solaris operating environment on SPARC and UltraSPARC platforms The compilers themselves conform to published Fortran language standards and provide many extended features including multiprocessor parallelization sophisticated optimized code compilation and mixed C Fortran language support The Fortran compilers are components of the Forte for High Performance Computing Sun Performance WorkShop software The Fortran 90 compiler 90 has been renamed Fortran 95 95 The 90 command is now an alias for 95 both invoke the Fortran 95 compiler Standards Conformance m 77 was designed to be compatible with the ANSI X3 9 1978 Fortran standard and the corresponding International Organization for Standardization ISO 1539 1980 as well as standards FIPS 69 1 BS 6832 and MIL STD 1753 m 95 was designed to be compatible with the ANSI X3 198 1992 ISO IEC 1539 1991 and ISO IEC 1539 1997 standards documents a Floating point arithmetic for both compilers is based on IEEE standard 754 1985 and international standard IEC 60559 1989 m On SPARC platforms both compilers provide support for the optimization
165. ons and the fsmuld instruction Examples SPARCstation 1 SPARCstation 2 Compile for the V8a version of the SPARC V8 ISA By definition V8a means the V8 ISA but without the fsmuld instruction This option enables the compiler to generate code for good performance on the V8a ISA Example Any system based on the microSPARC I chip architecture Compile for the SPARC V8 ISA Enables the compiler to generate code for good performance on the V8 architecture Example SPARCstation 10 Chapter 3 Fortran Compiler Options 97 TABLE 3 14 xarch Values for SPARC Platforms Continued xarch Meaning v8plus Compile for the V8plus version of the SPARC V9 ISA By definition V8plus means the V9 ISA but limited to the 32 bit subset defined by the V8plus ISA specification without the Visual Instruction Set VIS and without other implementation specific ISA extensions e This option enables the compiler to generate code for good performance on the V8plus ISA e The resulting object code is in SPARC V8 ELF32 format and only executes in a Solaris UltraSPARC environment it does not run on a V7 or V8 processor Example Any system based on the UltraSPARC chip architecture v8plusa Compile for the V8plusa version of the SPARC V9 ISA By definition V8plusa means the V8plus architecture plus the Visual Instruction Set VIS version 1 0 and with UltraSPARC extensions e This option enables the compiler to generate code for good perfo
166. or 51 invoking with F 58 utilities 11 Fortran 95 case 149 directives 160 features 147 I O extensions 158 incompatibilities with Fortran 77 163 linking with Fortran 77 165 modules 167 fpp Fortran preprocessor 21 51 58 63 fpversion show floating point platform information 32 FREE directive 161 free format 2 free format source 63 fsplit Fortran utility 11 function external C 26 function level reordering 104 G global offset table 82 global program checking Xlist 93 global symbols weak 27 gproft pg profile by procedure 82 H hardware architecture 95 100 120 help command line 15 README information 105 hexadecimal 150 Index 195 Hollerith 151 l impatient user s guide 17 INCLUDE files 67 information files 14 inline 60 templates libmil 71 inlining automatic with 04 78 with inline 69 input output compatibility f77 f95 164 error messages 126 installation 14 installation path 68 integer size four and eight bytes 69 interval arithmetic xia option 106 xinterval option 107 intrinsic procedures extensions 162 invalid floating point 65 ISA instruction set architecture 95 L large files 33 legacy compiler options 44 libm searched by default 70 library build G 66 disable system libraries 75 dynamic search path in executable 84 linking with 1 71 multithread save 74 name a shared library 67 path to shared li
167. ory then you can use dbx and ignore the object o files Use this option when you cannot keep the o files m The compiler passes s to the assembler and then the linker places all symbol tables for dbx in the executable file a This way of handling symbol tables is the older way It is sometimes called no auto read m The linker links more slowly and dbx initializes more slowly Without xs if you move the executables you must move both the source files and the object 0 files or set the path with either the dbx pathmap or use command a This way of handling symbol tables is the newer and default way of loading symbol tables It is sometimes called auto read m The symbol tables are distributed in the o files so that dbx loads the symbol table information only if and when it is needed Hence the linker links faster and dbx initializes faster xsafe mem Allow the compiler to assume that no memory protection violations occur 77 95 Using this option allows the compiler to assume no memory based traps occur It grants permission to use the speculative load instruction on the SPARC V9 platforms This option is effective only when used with optimization level 05 one one of the following architectures xarch v8plus v8plusa v8plusb v9 v9a or v9b Warning m Because non faulting loads do not cause a trap when a fault such as address misalignment or segmentation violation occurs you should use this option o
168. parallel or explicitpar options The PIPELOOP n Directive The PIPELOOP n directive requires that you specify SUN after CSPRAGMA This directive must appear immediately before a DO loop n is a positive integer constant or zero and asserts to the optimizer a dependence between loop iterations A value of zero indicates that the loop has no inter iteration loop carried 28 Fortran User s Guide July 2001 dependencies and can be freely pipelined by the optimizer A positive n value implies that the I th iteration of the loop has a dependency on the I n th iteration and can be pipelined at best for only n iterations at a time We know that the value of K is such that there can be no G cross iteration dependencies E g K gt N CSPRAGMA SUN PIPELOOP 0 DO I 1 N A I A I K D T B I B I A I END DO For more information on optimization see the Fortran Programming Guide The PREFETCH Directives The xprefetch option flag page 114 enables a set of PREFETCH directives that advise the compiler to generate prefetch instructions for the specified data element Prefetch instructions are only available on UltraSPARC platforms CSPRAGMA SPARC_PRE CH_READ_ONCE name CH_READ_MANY name CH_WRITE_ONCE name CH_WRITE_MANY name CSPRAGMA SPARC_PRE CSPRAGMA SPARC_PRE CSPRAGMA SPARC_PRE HY AY AY OY Howe See also the C User s
169. pecified without a value xipo 1 is used When compiling and linking are performed in separate steps xipo must be specified in both steps to be effective Example using xipo in a single compile link step demo 95 xipo x04 o prog partl f part2 f part3 f Chapter 3 Fortran Compiler Options 107 108 The optimizer performs crossfile inlining across all three source files This is done in the final link step so the compilation of the source files need not all take place in a single compilation and could be over a number of separate compilations each specifying xipo Example using xipo in separate compile link steps 95 xipo x04 c part1 f part2 f 95 xipo x04 c part3 f f95 xipo x04 o prog partl o part2 o part3 o The object files created in the compile steps have additional analysis information compiled within them to permit crossfile optimizations to take place at the link step A restriction is that libraries even if compiled with xipo do not participate in crossfile interprocedural analysis as shown in this example demo 95 xipo x04 one f two f three f demo ar r mylib a one o two o three o demo 95 xipo x04 o myprog main f four f mylib a Here interprocedural optimizations will be performed between one f two f and three f and between main f and four f but not between main f or four f and the routines on mylib a The first compilation may generate warnings about un
170. plusa ultra 16 32 1 1024 64 1 ultra2 1300 v8plusa ultra2 16 32 1 2048 64 1 ultra2 2170 v8plusa ultra 16 32 1 512 64 1 ultra2 2200 v8plusa ultra 16 32 1 1024 64 1 ultra2 2300 v8plusa ultra2 16 32 1 2048 64 1 ultra2e v8plusa ultra2e 16 32 1 256 64 4 ultra2i v8plusa ultra2i 16 32 1 512 64 1 ultra3 v8plusa ultra3 64 32 4 8192 512 1 Appendix D xtarget Platform Expansions 173 174 Fortran User s Guide July 2001 APPENDIX E Fortran Directives Summary This appendix summarizes the directives recognized by the f77 and f95 Fortran compilers General Fortran Directives Sun Parallelization Directives Cray Parallelization Directives OpenMP Fortran 95 Directives Library Routines and Environment Note Fortran parallelization features require a Sun WorkShop HPC license General Fortran Directives General directives accepted by both 77 and 95 are described in Chapter 2 TABLE E 1 Summary of General Fortran Directives Format CSPRAGMA keyword a a keyword al a m pe CSPRAGMA SUN keyword a a keyword al a pa CSPRAGMA SPARC keyword a a keyword al a pa Comment indicator in column 1 may be c C or Weuse C in these examples 95 free format must use C Directive CSPRAGMA C list Declares a list of names of external functions as C language routines 175 176 TABLE E 1 Summary of General Fortran Directives Continued UNROLL Directive
171. pp See also xpp page 113 free Specify free format source input files 95 All source files on the command line will be interpreted as 95 free format regardless of filename extension Normally 95 interprets f files as fixed format 95 as free format fround r Set the IEEE rounding mode in effect at startup 77 95 r must be one of nearest tozero negative positive The default is fround nearest To be effective compile the main program with this option This option sets the IEEE 754 rounding mode that m Can be used by the compiler in evaluating constant expressions m Is established at runtime during the program initialization When r is tozero negative or positive the option sets the rounding direction to round to zero round to negative infinity or round to positive infinity respectively when the program begins execution When fround is not specified Chapter 3 Fortran Compiler Options 63 64 fround nearest is used as the default and the rounding direction is round to nearest The meanings are the same as those for the icee_flags function See the Floating Point Arithmetic chapter of the Fortran Programming Guide fsimple n Select floating point optimization preferences 77 95 Allow the optimizer to make simplifying assumptions concerning floating point arithmetic See the Floating Point Arithmetic chapter of the Fortran Programming Guide For consistent results compil
172. processing xcode code specifies the memory address model on SPARC platforms xcommonchk enables runtime checking for inconsistent COMMON block declarations xprefetch allows the compiler to generate prefetch instructions on UltraSPARC II platforms xvector allows the compiler to replace certain math library calls within DO loops with single calls to a vectorized math routine m Changed options xcrossfile n optional level number added fns yes no optional yes no added Ztha option now ignored m New Features Compile for the 64 bit Solaris 7 environment on 64 bit SPARC platforms with xarch v9 or v9a Support in the I O library for large files larger than 2 Gigabytes Support for large arrays on 64 bit Solaris operating environments Accepts Sun style directives by default The REDUCTION directive accepts arrays in the list of variables SPARC A TASKCOMMON directive declares variables in COMMON to be private Fortran User s Guide July 2001 a New optimization pragma allows setting the compilers optimization level on a routine by routine basis m I O Differences Comparing 90 2 0 against the 1 2 release a NAMELIST Output Format 1 2 All variables in a single print statement written to a single line without line breaks 2 0 Each variable printed to a separate line 1 2 Comma used to separate values 2 0 Single blank separates values 1 2 Repeated values output using the r form 3 8
173. profile to hold the runtime feedback information The program writes its runtime profile data to the file feedback in this subdirectory If you run the program several times the execution frequency data accumulates in the feedback file that is output from prior runs is not lost You can set the environment variables SUN_PROFDATA and SUN_PROFDATA_DIR to control the file and directory where a program compiled with xprofile collect writes its runtime profile data With these variables set the program compiled with xprofile collect writes its profile data to SUN_PROFDATA_DIR SUN_PROFDATA These environment variables similarly control the path and names of the profile data files written by tcov as described in the tcov 1 man page Fortran User s Guide July 2001 use nm Use execution frequency data to optimize strategically As with collect nm the nm is optional and may be used to specify the name of the program The program is optimized by using the execution frequency data previously generated and saved in the feedback files written by a previous execution of the program compiled with xprofile collect The source files and other compiler options must be exactly the same as used for the compilation that created the compiled program that generated the feedback file If compiled with xprofile collect nm the same program name nm must appear in the optimizing compilation xprofile use nm tcov Basic block coverage analysi
174. programs Observe the following restrictions After the first eight characters blanks are ignored and uppercase and lowercase are equivalent as in Fortran text Because it is a comment a directive cannot be continued but you can have many CSPRAGMA lines one after the other as needed If a comment satisfies the above syntax it is expected to contain one or more directives recognized by the compiler if it does not a warning is issued The C preprocessor cpp will expand macro symbol definitions within a comment or directive line the Fortran preprocessor fpp will not expand macros in comment lines fpp will recognize legitimate 77 and 95 directives and allow limited substitution outside directive keywords However be careful with directives requiring the keyword SUN cpp will replace lower case sun with a predefined value Also if you define a cpp macro SUN it might interfere with the SUN directive keyword A general rule would be to spell those pragmas in mixed case if the source will be processed by cpp or fpp as in CSPRAGMA Sun UNROLL 3 Chapter 2 Using Sun Fortran Compilers 25 26 The Fortran compilers recognize the following general directives TABLE 2 2 Summary of General Fortran Directives C Directive CSPRAGMA C list Declares a list of names of external functions as C language routines UNROLL Directive CSPRAGMA SUN UNROLL n Advises the compiler that the following loop can be unrolled to
175. pt for a standard install This is the location assumed in this book Shell Prompts Shell Prompt C shell Bourne shell and Korn shell S C shell Bourne shell and Korn shell superuser Supported Platforms This Sun WorkShop release of the Fortran compilers supports only versions 2 6 7 and 8 of the Solaris SPARC Platform Edition Accessing Sun WorkShop Development Tools and Man Pages The Sun WorkShop product components and man pages are not installed into the standard usr bin and usr share man directories To access the Sun WorkShop compilers and tools you must have the Sun WorkShop component directory in your PATH environment variable To access the Sun WorkShop man pages you must have the Sun WorkShop man page directory in your MANPATH environment variable Before You Begin 3 For more information about the PATH variable see the csh 1 sh 1 and ksh 1 man pages For more information about the MANPATH variable see the man 1 man page For more information about setting your PATH and MANPATH variables to access this release see the Sun WorkShop 6 update 2 Installation Guide or your system administrator Note The information in this section assumes that your Sun WorkShop 6 update 2 products are installed in the opt directory If your product software is not installed in the opt directory ask your system administrator for the equivalent path on your system Accessing Sun WorkShop Compilers an
176. ption tells the compiler to write the executable code to the file greetings By convention executable files usually are given the same name as the main source file but without an extension Fortran User s Guide July 2001 Alternatively the default a out file could be renamed via the mv command after each compilation Either way run the program by typing the name of the executable file demo greetings Real programmers write Fortran demo Here is the same example using f95 demo cat greetings f95 program greetings print Real programmers write Fortran 95 end demo 95 o greetings greetings f95 demo greetings Real programmers write Fortran 95 demo The next sections of this chapter discuss the conventions used by the 77 and 95 commands compiler source line directives and other issues concerning the use of these compilers The next chapter describes the command line syntax and all the options in detail Invoking the Compiler The syntax of a simple compiler command invoked at a shell prompt is 77 options files invokes the Fortran 77 compiler 95 options files invokes the Fortran 95 compiler Here files is one or more Fortran source file names ending in f F 90 95 F90 F95 or for options is one or more of the compiler option flags Files with names ending ina 90 or 95 extension are free format Fortran 95 source files recognized only by the
177. r s Guide July 2001
178. rade performance It is generally recommended to declare specific variables INTEGER 2 rather than use i2 Fortran User s Guide July 2001 i4 Set the default integer size to four bytes 77 Set the default size to 4 bytes for integer and logical constants and variables declared without an explicit size INTEGER n Y still declares Y to be n bytes regardless of the i4 Although 4 bytes is the default size for INTEGER and LOGICAL this option can be used for overriding settings made by options like db1 and r8 which set these defaults to 8 demos 77 dbl i4 f Command line warning i4 overrides integer part of dbl inline auto no f1 no fn Enable or disable inlining of specified routines 77 95 Request the optimizer to inline the user written routines named in the 1 fn list Prefixing a routine name with no disables inlining of that routine Inlining is an optimization technique whereby the compiler effectively replaces a subprogram reference such as a CALL or function call with the actual subprogram code itself Inlining often provides the optimizer more opportunities to produce efficient code The lists are a comma separated list of functions and subroutines To inhibit inlining of a function prefix its name with nos Example Inline the routines xbar zbar vpoint demo 95 03 inline xbar zbar vpoint f
179. rce Form Command line options Option Action fixed Interpret all source files as Fortran fixed form free Interpret all source files as Fortran free form If the free or fixed option is used it overrides the file name suffix If either a DIR FREE or DIR FIXED directive is used it overrides the option and file name suffix Mixing Forms Some mixing of source forms is allowed m Inthe same 95 command some source files can be fixed form some free m Inthe same file free form can be mixed with fixed form by using DIR FREE and DIRS FIXED directives Fortran User s Guide July 2001 Case Sun Fortran 95 is case insensitive by default That means that a variable AbcDeF is treated as if it were spelled abcdef Compile with the U option to have the compiler treat upper and lower case as unique Known Limits A single Fortran 95 program unit can define up to 65 535 derived types and 16 777 215 distinct constants Boolean Type 95 supports constants and expressions of Boolean type There are no Boolean variables or arrays and there is no Boolean type statement Miscellaneous Rules Governing Boolean Type m Masking A bitwise logical expression has a Boolean result each of its bits is the result of one or more logical operations on the corresponding bits of the operands m For binary arithmetic operators and for relational operators a If one operand is Boolean the operation
180. rched for INCLUDE files file names appearing on preprocessor include directives or Fortran INCLUDE statements Example Search for INCLUDE files in usr app include demo f95 I usr app include growth F Multiple Idir options may appear on the command line Each adds to the top of the search path list first path searched The search order for relative path on INCLUDE or include is 1 The directory that contains the source file 2 The directories that are named in the Idir options 3 The directories in the default list The default list for Idir depends on the installation directory for the compiler In a standard install compiler software packages reside in the opt directory however systems administrators may decide to install packages in other locations The default search paths for INCLUDE files used by the compilers are m for 77 lt install_dir gt SUNWspro lt release gt include 77 usr include m for 95 lt install_dir gt SUNWspro lt release gt include 90 usr include where lt install_dir gt is the path to the installed packages typically opt in anormal install and lt release gt is a path that varies with each software release Set the default integer size to two bytes 77 Set the default size to 2 bytes for integer and logical constants and variables declared without an explicit size INTEGER n Y still declares Y to be n bytes regardless of the i2 This option may deg
181. read management for example by calling the Libthread library compiling with mt prevents conflicts Use mt if you mix Fortran with multithreaded C code that calls the libthread library See also the Solaris Multithreaded Programming Guide mt is implied automatically when using the autopar explicitpar or parallel options Note the following m A function subprogram that does I O should not itself be referenced as part of an I O statement Such recursive I O may cause the program to deadlock with mt m In general do not compile your own multithreaded code with autopar explicitpar or parallel The compiler generated calls to the threads library and the program s own calls may conflict causing unexpected results m Ona single processor system performance may be degraded with the mt option native Optimize performance for the host system Obsolete 77 95 This option is a synonym for xtarget native The fast option sets xtarget native 74 Fortran User s Guide July 2001 noautopar Disable automatic parallelization 77 95 Disables automatic parallelization invoked by autopar earlier on the command line nodepend Cancel depend in command line 77 95 Cancel any depend appearing earlier on the command line noexplicitpar Disable explicit parallelization 77 95 Disables explicit parallelization invoked by explicitpar earlier on the command line nolib Di
182. reads The directive applies only to the immediately following statement which must be one of these forms x x operator expression x expression operator x x intrinsic x expression x intrinsic expression x where e x is a scalar of intrinsic type e expression is a scalar expression that does not reference x e intrinsic is one of MAX MIN IAND IOR or IEOR e operator is one of AND OR EQV NEQV This implementation replaces all ATOMIC directives by enclosing the target statement in a critical section Appendix E Fortran Directives Summary 183 184 TABLE E 5 Summary of OpenMP Directives in Fortran 95 Continued FLUSH Directive SOMP FLUSH list Thread visible variables are written back to memory at the point at which this directive appears The FLUSH directive only provides consistency between operations within the executing thread and global memory The optional list consists of a comma separated list of variables that need to be flushed The FLUSH directive is implied for the following directives BARRIER CRITICAL ENDCRITICAL ENDDO END SECTIONS ENDSINGLE ENDWORKSHARE ORDERED ENDORDERED PARALLEL ENDPARALLEL PARALLEL ENDPARALLELDO PARALLELSECTIONS ENDPARALLELSECTIONS PARALLELWORKSHARE ENDPARALLELWORKSHARE FLUSH is not implied if NOWAIT is specified It is not implied by Do IASTER ENDMASTER SECTIONS SINGLE and WORKSHARE ORDERED Directive SOMP ORD
183. result in unexpected results including segmentation faults The C option checks for possible array subscript violations in the source code and during execution Specifying C may make the executable file larger If the C option is used array subscript violations are treated as an error If an array subscript range violation is detected in the source code during compilation it is treated as a compilation error If an array subscript violation can only be determined at runtime the compiler generates range checking code into the executable program This may cause an increase in execution time As a result it is appropriate to enable full array subscript checking while developing and debugging a program then recompiling the final production executable without subscript checking Chapter 3 Fortran Compiler Options 49 C Compile only produce object o files but suppress linking 77 95 Suppress linking Compile a o file for each source file If only a single source file is being compiled the o option can be used to specify the name of the o file written cg89 Compile for generic SPARC architecture Obsolete 77 95 This option is a macro for xarch v7 xchip old xcache 64 32 1 which is equivalent to xtarget ss2 cg92 Compile for SPARC V8 architecture Obsolete 77 95 This option is a macro for xarch v8 xchip super xcache 16 32 4 1024 32 1 which is equivalent to xtarget ss1000
184. ritten to the file name 1st where name is taken from the first listed source file on the command line A number of sub options provide further flexibility in the selection of actions These are specified by suffixes to the main Xlist option as shown in the following table TABLE 3 11 Xlist Suboptions Option Feature Xlist Show errors listing and cross reference table Xlistc Show call graphs and errors XlistE Show errors Xlisterr nnn Suppress error nnn messages Xlistf Show errors listing and cross references but no object files Xlistflndir Put fln files in directory dir which must already exist 77 only Xlisth Terminate compilation if errors detected XlistI Analyze include and INCLUDE files as well as source files XlistL Show listing and errors only Xlistln Set page length to n lines Xlisto name Rename report file to name 1st Xlists Suppress unreferenced names from the cross reference table Fortran User s Guide July 2001 xa TABLE 3 11 Xlist Suboptions Continued Option Feature Xlistvn Set checking level to n 1 2 3 or 4 default is 2 Xlistwlnnn Set width of output line to nnn columns default is 79 Xlistwar nnn Suppress warning nnn messages Xlistx Show cross reference table and errors Option Xlistflndir is not available with 95 See the Fortran Programming Guide chapter Program Analysis and Debugging for details Synonym for a 77 95
185. rmance on the UltraSPARC architecture but limited to the 32 bit subset defined by the V8plus specification e The resulting object code is in SPARC V8 ELF32 format and only executes in a Solaris UltraSPARC environment it does not run on a V7 or V8 processor Example Any system based on the UltraSPARC chip architecture v8plusb Compile for the V8plusb version of the SPARC V8plus ISA with UltraSPARC III extensions Enables the compiler to generate object code for the UltraSPARC architecture plus the Visual Instruction Set VIS version 2 0 and with UltraSPARC III extensions e The resulting object code is in SPARC V8 ELF32 format and executes only in a Solaris UltraSPARC III environment e Compiling with this option uses the best instruction set for good performance on the UltraSPARC III architecture v9 Compile for the SPARC V9 ISA Enables the compiler to generate code for good performance on the V9 SPARC architecture e The resulting o object files are in ELF64 format and can only be linked with other SPARC V9 object files in the same format e The resulting executable can only be run on an UltraSPARC processor running a 64 bit enabled Solaris operating environment with the 64 bit kernel e xarch v9 is only available when compiling in a 64 bit enabled Solaris environment 98 Fortran User s Guide July 2001 TABLE 3 14 xarch Values for SPARC Platforms Continued xarch Meaning v9a Compile for the SPARC V9 ISA
186. rocessors in 32 bit mode e no execution on other platforms v9a e runs Optimally on UltraSPARC II processors in 64 bit mode e no execution on other platforms v9b e runs Optimally on UltraSPARC III processors in 64 bit mode e no execution on other platforms Also note the following m SPARC instruction set architectures V7 V8 and V8a are all binary compatible a Object binary files o compiled with v8plus and v8plusa can be linked and can execute together but only on a SPARC V8plusa compatible platform a Object binary files o compiled with v8plus v8plusa and v8plusb can be linked and can execute together but only on a SPARC V8plusb compatible platform a xarch values v9 v9a and v9b are only available on UltraSPARC 64 bit Solaris environments m Object binary files o compiled with v9 and v9a can be linked and can execute together but will run only on a SPARC V9a compatible platform a Object binary files o compiled with v9 v9a and v9b can be linked and can execute together but will run only on a SPARC V9b compatible platform For any particular choice the generated executable may run much more slowly on earlier architectures Also although quad precision REAL 16 and long double floating point instructions are available in many of these instruction set architectures the compiler does not use these instructions in the code it generates Fortran User s Guide July 2001 TABLE 3 14 gives details for
187. rs 23 In the second example the user intended to type the 77 95 option fast but neglected the leading dash The compiler again passes the argument to the linker which in turn interprets it as a file name These examples indicate that extreme care should be observed when composing compiler command lines Modules Fortran 95 95 automatically creates module interface files for each MODULE declaration encountered in the source files and searches for modules referenced by a USE statement For each module encountered MODULE module_name the compiler generates a corresponding file module_name mod in the current directory For example 95 generates the module information file List mod for the MODULE list unit found on file mysrc 95 The compiler searches the current directory for module files referenced in USE statements Module files must be compiled before compiling any source file referencing a MODULE in a USE statement Directories can be added to the search path with the M command line option However individual mod files cannot be specified directly on the command line The 95 compiler also creates an object file filename o for every source file containing MODULE statement and these implementation object files must be included when linking to create an executable See page 167 24 Directives Use a source code directive a form of Fortran comment to pass specific information to the c
188. rvice marks of Sun Microsystems Inc in the U S and other countries All SPARC trademarks are used under license and are trademarks or registered trademarks of SPARC International Inc in the U S and other countries Products bearing SPARC trademarks are based upon an architecture developed by Sun Microsystems Inc The OPEN LOOK and Sun Graphical User Interface was developed by Sun Microsystems Inc for its users and licensees Sun acknowledges the pioneering efforts of Xerox in researching and developing the concept of visual or graphical user interfaces for the computer industry Sun holds a non exclusive license from Xerox to the Xerox Graphical User Interface which license also covers Sun s licensees who implement OPEN LOOK GUIs and otherwise comply with Sun s written license agreements Sun 90 95 is derived from Cray CF90 a product of Cray Inc Federal Acquisitions Commercial Software Government Users Subject to Standard License Terms and Conditions DOCUMENTATION IS PROVIDED AS IS AND ALL EXPRESS OR IMPLIED CONDITIONS REPRESENTATIONS AND WARRANTIES INCLUDING ANY IMPLIED WARRANTY OF MERCHANTABILITY FITNESS FOR A PARTICULAR PURPOSE OR NON INFRINGEMENT ARE DISCLAIMED EXCEPT TO THE EXTENT THAT SUCH DISCLAIMERS ARE HELD TO BE LEGALLY INVALID Copyright 2001 Sun Microsystems Inc 901 San Antonio Road Palo Alto CA 94303 4900 Etats Unis Tous droits r serv s Ce produit ou document est distribu avec des li
189. ry help information on options or README file 77 95 The h is either readme or flags Chapter 3 Fortran Compiler Options 105 xhelp readm Show the online README file for this release of the compiler xhelp flags Show the compiler flags options and is same as help xia v Enable interval arithmetic extensions and set a suitable floating point environment 95 v can be one of either widestneed or strict The default if not specified is widestneed Fortran 95 extensions for interval arithmetic calculations are detailed in the Interval Arithmetic Programming Reference See also xinterval page 107 The xia flag is a macro that expands as follows xia or xinterval widestneed ftrap non fns no fsimple 0 xla widestneed xia strict xinterval strict ftrap non fns no fsimple 0 xild off on Enable disable the Incremental Linker 77 95 xildoff disables the use of the incremental linker ild The standard linker 14d is used instead xildon enables use of ild instead of 1d xildoff is the default if you do not use the g option It is also the default if you use G or name any source file on the command line xildon is the default if you use g and do not use G and no source files appear on the command line just object files and or libraries See the section on ild in the C User s Guide xinline list Synonym for inline 77 95 106 For
190. s can be initialized in a PARAMETER statement the format is the same as a derived type initialization a Structures can appear as components in a derived type but the derived type must be declared with the SEQUENCE attribute I O Extensions Some I O extensions that appear in Sun Fortran 77 have been added to the Fortran 95 compiler a NAMELIST Input Format The group name may be preceded by or amp on input The amp is the only form accepted by the Fortran 95 standard and is what is written by NAMELIST output Fortran User s Guide July 2001 Accepts as the symbol terminating input except if the last data item in the group is CHARACTER data in which case the is treated as input data Allows NAMELIST input to start in the first column of a record OPEN FORM BINARY treats the file as binary data without record marks Opening a file with FORM BINARY has roughly the same effect as FORM UNFORMATTED except that no record lengths are embedded in the file Without this data there is no way to tell where one record begins or ends Thus it is impossible to BACKSPACE a FORM BINARY file because there is no way of telling where to backspace to A READ ona BINARY file will read as much data as needed to fill the variables on the input list a WRITE statement Data is written to the file in binary with as many bytes transferred as specified by the output list a R
191. s the OPT n pragma and controls the maximum optimization level allowed by OPT pragmas in the source code a xprefetch allows the compiler to generate prefetch instructions on UltraSPARC II platforms a xvector allows the compiler to replace certain math library calls within DO loops with single calls to a vectorized math routine m Changed options a xcrossfile n optional level number added a fns yes no optional yes no added Ztha option now ignored m New Features 144 Fortran User s Guide July 2001 Compile for the 64 bit Solaris 7 environment on 64 bit SPARC platforms with xarch v9 or v9a Support in the I O library for large files larger than 2 Gigabytes Support for large arrays on 64 bit Solaris 7 environments Dynamic arrays local arrays with dynamic size implemented see FORTRAN 77 Language Reference Manual The REDUCTION directive accepts arrays in the list of variables SPARC A TASKCOMMON directive declares variables in COMMON to be private Fortran 90 style constants that allows specification of byte size for example 12345678_8 for a 64 bit 8 byte constant New optimization pragma allows setting the compilers optimization level on a routine by routine basis Year 2000 safe date_and_time library routine Features in 77 4 2 77 4 2 included the following features that were new or changed since the 4 0 release New options dbl_align_all errtags yes no
192. s using new style tcov Code instrumentation is similar to that of a but d files are no longer generated for each source file Instead a single file is generated whose name is based on the name of the final executable For example if stuff is the executable file then stuff profile tcovd is the data file When running tcov you must pass it the x option to make it use the new style of data If not tcov uses the old d files if any by default for data and produces unexpected output Unlike a the TCOVDIR environment variable has no effect at compile time However its value is used at program runtime to identify where to create the profile subdirectory See the tcov 1 man page the Performance Profiling chapter of the Fortran Programming Guide and the Analyzing Program Performance with Sun WorkShop manual for more details Note The report produced by tcov can be unreliable if there is inlining of subprograms due to 04 or inline Coverage of calls to routines that have been inlined is not recorded Chapter 3 Fortran Compiler Options 117 xrecursive Allow routines without RECURSIVE attribute call themselves recursively 95 Only subprograms defined with the RECURSIVE attribute can call themselves recursively unless they are compiled with xrecursive However compiling with xrecursive may cause performance degradations Also consider using stackvar with xrecursive since xrecursive does
193. s1 11 a1 s2 12 a2 s3 13 a3 The si li ai are defined as follows si The size of the data cache at level i in kilobytes Chapter 3 Fortran Compiler Options 99 li The line size of the data cache at level i in bytes ai The associativity of the data cache at level i This option specifies the cache properties that the optimizer can use It does not guarantee that any particular cache property is used Although this option can be used alone it is part of the expansion of the xtarget option it is provided to allow overriding an xcache value implied by a specific xtarget option TABLE 3 15 xcache Values Value Meaning generic Define the cache properties for good performance on most SPARC processors without any major performance degradation This is the default s1 11 al Define level 1 cache properties s1 11 a1 s2 12 a2 Define levels 1 and 2 cache properties s1 11 a1 s2 12 a2 s3 13 a3 Define levels 1 2 and 3 cache properties Example xcache 16 32 4 1024 32 1 specifies the following A Level 1 cache has 16K bytes 32 byte line size 4 way associativity A Level 2 cache has 1024K bytes 32 byte line size direct mapping associativity xcg89 Synonym for cg89 77 95 xcg92 Synonym for cg92 77 95 xchip c Specify target processor for the optimizer 77 95 This option specifies timing properties by specifying the target processor 100 Fortran User s Guide July 2001 Although t
194. sS amp SUN microsystems Fortran User s Guide Sun Microsystems Inc 901 San Antonio Road Palo Alto CA 94303 U S A 650 960 1300 Part No 806 7988 10 July 2001 Revision A Send comments about this document to docfeedback sun com Forte Developer 6 update 2 Sun WorkShop 6 update 2 Copyright 2001 Sun Microsystems Inc 901 San Antonio Road Palo Alto CA 94303 4900 USA All rights reserved This product or document is distributed under licenses restricting its use copying distribution and decompilation No part of this product or document may be reproduced in any form by any means without prior written authorization of Sun and its licensors if any Third party software including font technology is copyrighted and licensed from Sun suppliers Parts of the product may be derived from Berkeley BSD systems licensed from the University of California UNIX is a registered trademark in the U S and other countries exclusively licensed through X Open Company Ltd For Netscape Netscape Navigator and the Netscape Communications Corporation logo the following notice applies Copyright 1995 Netscape Communications Corporation All rights reserved Sun Sun Microsystems the Sun logo docs sun com AnswerBookz2 Solaris SunOS JavaScript SunExpress Sun WorkShop Sun WorkShop Professional Sun Performance Library Sun Performance WorkShop Sun Visual WorkShop and Forte are trademarks registered trademarks or se
195. sable linking with system libraries 77 95 Do not automatically link with any system or language library that is do not pass any default 1x options on to 1d The normal behavior is to link system libraries into the executables automatically without the user specifying them on the command line The nolib option makes it easier to link one of these libraries statically The system and language libraries are required for final execution It is your responsibility to link them in manually This option provides you with complete control For example consider a program linked dynamically with 1ibF77 that fails on a remote system because it has no 1ibF77 With this option you can link the library into your program statically Link 1ibF77 statically and link libc dynamically with 77 demos 77 nolib any f Bstatic 1F77 Bdynamic lm lc Chapter 3 Fortran Compiler Options 75 Link libm statically and libc dynamically with 95 demo 95 nolib any f95 Bstatic lm Bdynamic lc The order for the 1x options is important Follow the order shown in the examples nolibmil Cancel libmil on command line 77 95 Use this option after the fast option to disable inlining of 1ibm math routines demo 77 fast nolibmil noqueue Disable license queueing 77 95 With this option if no software license is available to run the compiler it returns without queueing your request and without compilin
196. salign option requires the option on the link step as well misalign is a macro equivalent to xmemalign 1i aligncommon 1 See xmemalign page 111 mp none sun cray openmp Select the style for parallelization directives Fortran parallelization features require a Forte for HPC license 77 95 The default without specifying mp is none mp sun Accept Sun style directives CSPAR or SPAR prefix mp cray Accept Cray style directives CMIC or MICS prefix mp openmp Accept OpenMP Fortran directives Available with 95 only mp none Ignore all parallelization directives You can combine OpenMP directives with Sun or Cray directives in the same compilation unit But both Sun and Cray directives cannot both be active in the same compilation unit For example mp sun openmp and mp cray openmp are permitted but mp sun cray is not Chapter 3 Fortran Compiler Options 73 You must also specify explicitpar or parallel to enable parallelization For correctness also specify stackvar explicitpar stackvar mp openmp When compiling for OpenMP use the openmp flag which includes mp openmp along with other flags required by OpenMP See page 79 A summary of these 77 95 directives appears in Appendix E in this manual See the Fortran Programming Guide for details Require thread safe libraries 77 95 Require linking to thread safe libraries If you do your own low level th
197. t interval arithmetic 107 strip executable of symbol table s 87 suffix of file names recognized by compiler 20 of file names recognized by compiler f95 148 suppress blank in listed directed output 79 implicit typing 91 license queue 76 linking 50 warnings 93 warnings by tag name erroff 56 swap command 34 swap space display actual swap space 34 increasing 34 limit amount of disk swap space 33 symbol table for dbx 66 119 syntax compiler command line 37 77 90 commands 37 77 90 commands 19 options on compiler command line 38 system inc 31 T T file 52 tab format 2 tcov a profile by statement 46 new style with xprofile 117 templates inline 71 temporary files directory for 90 trapping floating point exceptions 65 on memory 119 type declaration alternate form 152 typographic conventions 2 U ulimit command 34 underflow 65 gradual 62 underscore 57 do not append to external names 26 unrecognized options 23 UNROLL directive 27 usage compiler 19 utilities 11 V variables undeclared 91 VAX VMS Fortran features with vax 92 features with x1 109 version id of each compiler pass 91 WwW warnings message tags 56 suppress messages 93 suppress with erroff 56 undeclared variables 91 use of non standard extensions 47 WEAK directive 27 weak linker symbols 27 widestneed interval arithmetic 107 Index 201 202 Fortran Use
198. t be more appropriate to use xt ypemap real 64 double 64 integer 64 instead of r8 in these cases a For all of the floating point data types db1 works the same as r8 using both r8 and db1 produces the same results as using only db1 m For INTEGER and LOGICAL data types db1 differs from r8 a db1 allocates 8 bytes and does 8 byte arithmetic a r8 allocates 8 bytes and does only 4 byte arithmetic mixed In general if you compile one subprogram with r8 then be sure to compile all subprograms of that program with r8 This also important with programs communicating through unformatted I O files if one program is compiled with Chapter 3 Fortran Compiler Options 85 86 r8 then the other program must be similarly compiled with r8 Be also aware that this option alters the default data size of function names including calls to library functions unless the function name is typed explicitly with a data size The impact on runtime performance may be great With r8 an expression like float 15 0d0 float is evaluated in quadruple precision due to the declaration of the constant If you select both r8 and i2 the results are unpredictable r8const Promote single precision constants to REAL 8 constants 77 95 All single precision REAL constants are promoted to REAL 8 Double precision REAL 8 constants are not changed This option only applies to constants To pro
199. t processor Command line interactive debugger Fortran source code pre processor C source code pre processor Pre processor splits Fortran 77 routines into single files Examine set or clear floating point exception bits Handle floating point exceptions Math library error handling routine Incremental link editor for object files Link editor for object files Chapter1 Introduction 13 14 READMES The READMEs directory contains files that describe new features software incompatibilities bugs and information that was discovered after the manuals were printed The location of this directory depends on where your software was installed The path is install_directory SUNWspro READMEs In a normal install install_directory is opt TABLE 1 1 READMEs of Interest README File fortran_77 fortran_95 fpp_readme interval_arithmetic math_libraries omp_directives pdf profiling_tools runtime_libraries 64bit_Compilers performance_library Describes new and changed features known limitations documentation errata for this release of the FORTRAN 77 compiler 77 new and changed features known limitations documentation errata for this release of the Fortran 95 compiler 95 overview of fpp features and capabilities overview of the interval arithmetic features in 95 optimized and specialized math libraries available summarizes OpenMP directives accepted by 95 This is a
200. t set up error handlers I O Error Messages f77 The error messages in this section are generated by the Fortran 77 I O library The error numbers are returned in the IOSTAT variable if the ERR return is taken 126 Fortran User s Guide July 2001 For example the following program tries to do an unformatted write to a file opened for formatted output and produces error messages like the following sue 1003 unformatted io not allowed logical unit 6 named stdout lately writing sequential unformatted external IO The following error messages are generated These same messages are also documented at the end of the man page perror 3F If the error number is less than 1000 then it is a system error See int ro 2 TABLE A 1 77 Runtime I O Messages Error Message 1000 error in format Read the error message output for the location of the error in the format It can be caused by more than 10 levels of nested parentheses or an extremely long format statement 1001 illegal unit number It is illegal to close logical unit 0 Negative unit numbers are not allowed The upper limit is 2 1 1002 formatted io not allowed The logical unit was opened for unformatted I O 1003 unformatted io not allowed The logical unit was opened for formatted I O 1004 direct io not allowed The logical unit was opened for sequential access or the logical record length was specified as 0 1005 sequential io not
201. th g Some capabilities of other options specified along with g may be limited See the dbx documentation for details The g option makes xildon the default incremental linker option when o object files appear on the command line see page 106 That is with g the compiler default behavior is to automatically invoke ild in place of 1d unless the G option is present or any source file is named on the command line To use the full capabilities of the Forte Developer 6 Sun WorkShop 6 Performance Analyzer compile with g While some performance analysis features do not require g you must compile with g to view annotated source some function level information and compiler commentary messages See the analyzer 1 man page and Analyzing Program Performance With Sun WorkShop The commentary messages generated with g describe the optimizations and transformations the compiler made while compiling your program The messages interleaved with the source code can be displayed by the er_src 1 command Fortran User s Guide July 2001 Note that commentary messages only appear if the compiler actually performed any optimizations You are more likely to see commentary messages when you request high optimization levels such as with x04 or fast hname Specify the name of the generated dynamic shared library 77 95 This option is passed on to the linker For details see the Solaris Linker and Libraries Guide and t
202. the fsimple 2 option and disables the xlibmopt selected by fast Because fast invokes dalign fns fsimple 2 programs compiled with fast can result in nonstandard floating point arithmetic nonstandard alignment of data and nonstandard ordering of expression evaluation These selections might not be appropriate for most programs Fortran User s Guide July 2001 fixed Specify fixed format Fortran 95 source input files f95 All source files on the command line will be interpreted as 77 fixed format regardless of filename extension Normally 95 interprets only files as fixed format 95 as free format flags Synonym for he1p 77 95 fnonstd Initialize floating point hardware to non standard preferences 77 95 This option is a synonym for the combination of the following option flags m fns ftrap common Specifying fnonstd is approximately equivalent to the following two calls at the beginning of a Fortran main program i i _handler set common SIGFPE_ABORT call nonstandard_arithmetic The nonstandard_arithmetic routine replaces the obsolete abrupt_underflow routine of earlier releases To be effective the main program must be compiled with this option Using this option initializes the floating point hardware to Abort trap on floating point exceptions a Flush underflow results to zero if it will improve speed rather than produce a subnormal n
203. the stackvar option when using any of the parallelization options including autopar Avoid autopar if the program already contains explicit calls to the libthread threads library See note with mt on page 74 The autopar option is not appropriate on a single processor system and the compiled code will generally run slower To run a parallelized program in a multithreaded environment you must set the PARALLEL or OMP_NUM_THREADS environment variable prior to execution This tells the runtime system the maximum number of threads the program can create The default is 1 In general set the PARALLEL or OMP_NUM_THREADS variable to the available number of processors on the target platform If you use autopar and compile and link in one step the multithreading library and the thread safe Fortran runtime library will automatically be linked If you use autopar and compile and link in separate steps then you must also link with autopar to insure linking the appropriate libraries The reduction option may also be useful with autopar Other parallelization options are parallel and explicitpar Refer to the Fortran Programming Guide for more information on parallelization B static dynamic Prefer dynamic or require static library linking 77 95 No space is allowed between B and dynamic or static The default without B specified is Bdynamic m Bdynamic Prefer dynamic linking try for shared l
204. the application Then increase or decrease the factor as appropriate and run the performance tests again Continue adjusting the factor and running the performance tests until you achieve optimum performance When you increase or decrease the factor in small steps you will see no performance difference for a few steps then a sudden difference then it will level off again Defaults If xprefetch is not specified xprefetch no auto explicit is assumed If only xprefetch is specified xprefetch auto explicit is assumed The default of no auto is assumed unless explicitly overridden with the use of xprefetch without any arguments or with an argument of auto or yes For example xprefetch explicit is the same as xprefetch explicit no auto The default of explicit is assumed unless explicitly overridden with an argument of no explicit or an argument of no For example xprefetch auto is the same as xprefetch auto explicit If automatic prefetching is enabled such as with xprefetch or xprefetch yes but a latency factor is not specified then xprefetch latx 1 0 is assumed Interactions With xprefetch explicit the compiler will recognize the directives SPRAGMA SPARC_PREFETCH_READ_ONCE name SPRAGMA SPARC_PREFETCH_READ_MANY name SPRAGMA SPARC_PREFETCH_WRITE_ONCE name SPRAGMA SPARC_PREFETCH_WRITE_MANY name The xchip setting effects the
205. ting System Error Messages Operating system error messages include system call failures C library errors and shell diagnostics The system call error messages are found in int ro 2 System calls made through the Fortran library do not produce error messages directly The following system routine in the Fortran library calls C library routines which produce an error message integer system status status system cp afile bfile print status status end The following message is displayed cp cannot access afile status 512 125 Signal Handler Error Messages 77 Before beginning execution of a program the Fortran 77 library sets up a signal handler sigdie for signals that can cause termination of the program sigdie prints a message that describes the signal flushes any pending output and generates a core image Presently the only arithmetic exception that produces an error message is the INTEGER 2 division with a denominator of zero All other arithmetic exceptions are ignored A signal handler error example follows where the subroutine SUB tries to access parameters that are not passed to it CALL SUB D UBROUTINE SUB I J K J K ETURN D ADHM B Ea The following error message results when compiled with 77 and run xx TERMINATING sub77 xxx Received signal 11 SIGSEGV Segmentation Fault The Fortran 95 compiler does no
206. tion to find needed dynamic libraries Building an executable file using Rpaths adds directory paths to a default path that is always searched last Standard Default Path opt SUNWspro lib For more information see the Libraries chapter in the Fortran Programming Guide and the Solaris Linker and Libraries Guide Fortran User s Guide July 2001 Double default byte size for REAL INTEGER DOUBLE and COMPLEX 77 Note r8 and db1 are now considered obsolete and may be removed in future releases Use the more general xt ypemap option instead r8 promotes the default byte size for REAL INTEGER DOUBLE and COMPLEX variables declared without an explicit byte size as follows TABLE 3 9 Default Data Sizes and r8 Bytes With r8 Without r8 option ption Data Type default SPARC INTEGER 4 8 REAL 4 8 This option applies to variables parameters constants and functions Also LOGICAL is treated as INTEGER COMPLEX as two REALS and DOUBLI COMPLEX as two DOUBLEs Gl db1 and r8 can be expressed in terms of the more general xt ypemap option a dbl sameas xtypemap real 64 double 128 integer 64 r8 sameas xtypemap real 64 double 128 integer mixed These options promote default DOUBLE PRECISION data to QUAD PRECISION 128 bits This may be unwanted and may cause performance degradation It migh
207. tion flags Some flags serve more than one purpose and appear more than once TABLE 3 3 Compiler Options Grouped by Functionality Function Option Flag Compilation Mode Compile only do not produce an executable file g Show commands built by the driver but do not compile dryrun Specify name of object library or executable file to write o filename Compile and generate only assembly code S Strip symbol table from executable s Suppress compiler messages except error messages silent Define path to directory for temporary files temp directory Show elapsed time for each compilation phase time Show version number of compiler and its phases V Verbose messages Vv Compiled Code Add suppress trailing underscores on external names ext_names x Inline specified user functions inline list Chapter 3 Fortran Compiler Options 39 TABLE 3 3 Compiler Options Grouped by Functionality Continued Function Option Flag Compile position independent code KPIC kpic Inline certain math library routines libmil STOP returns integer status value to shell stop_status yn Specify code address space xcode x Enable UltraSPARC prefetch instructions xprefetch x Specify use of optional registers xregs x Specify default data mappings xtypemap x Data Alignment Specify alignment of data in COMMON blocks aligncommon n Force COMMON block data alignment to allow double dalign word fetch store Force
208. tive with fast 60 native 74 noautopar 75 nodepend 75 noexplicitpar 75 nolib 75 nolibmil 76 noqueue 76 noreduction 76 norunpath 76 o output file 78 oldldo 79 On with fast 60 with g 77 On 77 onetrip 79 openmp 79 p profile by procedure 80 pad p 80 parallel parallelize loops 81 pg profile by procedure 82 PIC 83 pic 82 Qoption 84 R list 84 r8 85 r8const 86 reduction 86 s 87 sbfast 87 silent 88 stop_status 89 temp 90 time 90 u 91 u do not convert to lowercase 90 Uname undefine preprocessor macro 91 unrol1 unroll loops 91 v 91 v 92 vax v 92 109 vpara 93 w 93 xa 95 xarch isa 95 xautopar 99 xcache c 99 xcg 89 92 100 xchip c 100 xcode c 102 xcommoncheck 102 xcrossfile 103 xdepend 104 xexplicitpar 104 xF 104 xhasc Hollerith as character 105 xhelp h 105 xia interval arithmetic 106 xildoff 106 xinline 106 xinterval v for interval arithmetic 107 xipo interprocedural optimizations 107 xlibmil 110 xlibmopt with fast 60 xlibmopt 110 xlic_lib sunperf 110 xlicinfo 110 Xlist suboptions 94 Xlist global program checking 93 xloopinfo 111 xmaxopt 111 xmemalign 111 xnolib 112 xnolibmopt 112 xOn 113 xparallel 113 xpg 113 xpp p 113 xprefetch PREFETCH directive 29 xprefetch 114 xprofile p 116 xrecursive 118 xreduction 118 xregs r
209. tives 30 95 Library Interfacesand system inc 31 Compiler Usage Tips 32 Determining Hardware Platform 32 Memory Size 33 3 Fortran Compiler Options 37 Command Syntax 37 Options Syntax 38 Options Summary 39 Commonly Used Options 43 Backward Compatibility and Legacy Options 44 vi Fortran User s Guide July 2001 Obsolescent Options 45 Options Reference 45 Runtime Error Messages 125 Operating System Error Messages 125 Signal Handler Error Messages f77 126 I O Error Messages 77 126 I O Error Messages f95 130 Features Release History 137 Fortran 95 New Features and Changes 137 95 New Features in Sun WorkShop 6 update 2 137 95 New Features in Sun WorkShop 6 update 1 138 95 New Features in Sun WorkShop 6 139 New Features Released In 90 2 0 140 Fortran 77 New Features and Changes 143 77 New Features in Sun WorkShop 6 update 2 143 77 New Features in Sun WorkShop 6 update 1 143 77 New Features in Sun WorkShop 6 144 Features in 77 5 0 144 Features in 77 4 2 145 FORTRAN 77 Upward Compatibility 145 Fortran 3 0 3 0 1 to 4 0 146 BCP Running Applications from Solaris 1 146 Fortran 95 Features and Differences 147 Features and Extensions 147 Continuation Line Limits 147 Fixed Form Source Lines 147 Contents vii viii Directives 147 Source Form Assumed 148 Known Limits 149 Boolean Type 149 Abbreviated Size Notation for Numeric Data Types 152 Cray Pointers 153 Other Language E
210. tran Directives Summary 191 192 Fortran User s Guide July 2001 Index SYMBOLS DIRS in directives 160 ifdef 21 include 21 A blank character 2 A abrupt_underflow 61 align dalign 52 data in COMMON with aligncommon 46 structures as in VMS Fortran 109 See also data analyzer compile option xF 104 application registers SPARC 118 arithmetic See floating point 61 array bounds checking 49 asa Fortran print utility 11 assembly code 87 automatic variables 88 auto read dbx disable 119 B backslash in character constants 92 109 backward compatibility options 44 basic block profile by a 46 binding dynamic 55 Boolean constant alternate forms 150 type constants 149 browser 87 C C directive 26 cache padding for 80 specify hardware cache 99 CALL in a loop parallelize 88 inlining subprogram calls with inline 69 preserving arguments over ENTRY statements 47 case preserve upper and lower case 90 CDIR in directives 160 CIF compiler information file 52 code size 120 command line help 15 unrecognized options 23 comments as directives 160 COMMON alignment 46 consistency checking with xcommoncheck 102 Index 193 padding 80 compatibility between compiler releases 145 Fortran 95 vs Fortran 77 163 with C 167 compile and link 20 22 and B 49 build a dynamic shared library 66 compile only 50 dynamic shared libraries 55 compiler command
211. tran User s Guide July 2001 xinterval v Enable interval arithmetic extensions 95 v can be one of either no widestneed or strict The default if not specified is widestneed no Interval arithmetic extensions not enabled widestneed Promotes all non interval variables and literals in any mixed mode expression to the widest interval data type in the expression strict Prohibits mixed type or mixed length interval expressions All interval type and length conversions must be explicit Fortran 95 extensions for interval arithmetic calculations are detailed in the Interval Arithmetic Programming Reference See also xia page 106 xipo 0 1 Perform interprocedural optimizations 77 95 Performs whole program optimizations by invoking an interprocedural analysis pass Unlike xcrossfile xipo will perform optimizations across all object files in the link step and is not limited to just the source files on the compile command xipo is particularly useful when compiling and linking large mullti file applications Object files compiled with this flag have analysis information compiled within them that enables interprocedural analysis across source and pre compiled program files However analysis and optimization is limited to the object files compiled with xipo and does not extend to object files on libraries xipo 0 disables and xipo 1 enables interprocedural analysis The default is xipo 0 and if xipo is s
212. ts portant les marques SPARC sont bas s sur une architecture d velopp e par Sun Microsystems Inc L interface d utilisation graphique OPEN LOOK et Sun a t d velopp e par Sun Microsystems Inc pour ses utilisateurs et licenci s Sun reconna t les efforts de pionniers de Xerox pour la recherche et le d veloppement du concept des interfaces d utilisation visuelle ou graphique pour l industrie de l informatique Sun d tient une licence non exclusive de Xerox sur l interface d utilisation graphique Xerox cette licence couvrant galement les licenci s de Sun qui mettent en place l interface d utilisation graphique OPEN LOOK et qui en outre se conforment aux licences crites de Sun Sun f90 f95 est deriv de CRAY CF90 un produit de Cray Inc CETTE PUBLICATION EST FOURNIE EN L ETAT ET AUCUNE GARANTIE EXPRESSE OU IMPLICITE N EST ACCORDEE Y COMPRIS DES GARANTIES CONCERNANT LA VALEUR MARCHANDE L APTITUDE DE LA PUBLICATION A REPONDRE A UNE UTILISATION PARTICULIERE OU LE FAIT QU ELLE NE SOIT PAS CONTREFAISANTE DE PRODUIT DE TIERS CE DENI DE GARANTIE NE S APPLIQUERAIT PAS DANS LA MESURE OU IL SERAIT TENU JURIDIQUEMENT NUL ET NON AVENU GS tem Ca Adobe PostScript Important Note on New Product Names As part of Sun s new developer product strategy we have changed the names of our development tools from Sun WorkShop to Forte Developer products The products as you can see are the same high quality products
213. umber as the IEEE standard requires See fns for more information about gradual underflow and subnormal numbers The fnonstd option allows hardware traps to be enabled for floating point overflow division by zero and invalid operation exceptions These are converted into SIGFPE signals and if the program has no SIGFPE handler it terminates with a dump of memory Chapter 3 Fortran Compiler Options 61 62 For more information see the ieee_handler 3m and ieee_functions 3m man pages the Numerical Computation Guide and the Fortran Programming Guide fns no yes Select the SPARC nonstandard floating point mode 77 95 The default is the SPARC standard floating point mode fns no See the Floating Point Arithmetic chapter of the Fortran Programming Guide Optional use of yes or no provides a way of toggling the fns flag following some other macro flag that includes it such as fast fns is the same as fns yes This option flag enables nonstandard floating point mode when the program begins execution On some SPARC systems specifying nonstandard floating point mode disables gradual underflow causing tiny results to be flushed to zero rather than producing subnormal numbers It also causes subnormal operands to be silently replaced by zero On those SPARC systems that do not support gradual underflow and subnormal numbers in hardware use of this option can significantly improve the performanc
214. un Fortran Compilers 27 not used If the program directly calls name2 the definition from library is used a duplicate definition of name2 causes an error See the Solaris Linker and Libraries Guide for more information The OPT Directive The OPT directive requires that you specify SUN after CSPRAGMA The OPT directive sets the optimization level for a subprogram overriding the level specified on the compilation command line The directive must appear immediately before the target subprogram and only applies to that subprogram For example CSPRAGMA SUN OPT 2 SUBROUTINE smart a b c d e etc When the above is compiled with an 77 command that specifies 04 the directive will override this level and compile the subroutine at 02 Unless there is another directive following this routine the next subprogram will be compiled at 04 The routine must also be compiled with the xmaxopt n option for the directive to be recognized This compiler option specifies a maximum optimization value for PRAGMA OPT directives ifa PRAGMA OPT specifies an optimization level greater than the xmaxopt level the xmaxopt level is used The NOMEMDEP Directive The NOMEMDEP directive requires that you specify SUN after CSPRAGMA This directive must appear immediately before a DO loop It asserts to the optimizer that there are no memory based dependencies within an iteration of the loop to inhibit parallelization Requires
215. uppress with silent 88 verbose 92 misaligned data specifying behavior 111 mod file module file 167 modules creating and using 24 mod file 167 search path 72 multithreading See parallelization multithread safe libraries 74 N name argument do not append underscore 26 object executable file 78 nonstandard_arithmetic 61 O object files compile only 50 name 78 object library search directories 70 obsolescent options 45 octal 150 one trip DO loops 79 OpenMP 30 73 directives summary 180 environment variables summarized 190 library routines summarized 187 OPT directive 28 xmaxopt option 111 optimization across source files 103 107 floating point 64 inline user written routines 69 interprocedural 107 levels 77 loop unrolling 91 loop unrolling by directive 27 math library 110 OPT directive 28 111 PIPELOOP directive 28 prefetch 114 PREFETCH directive 29 specify cache 99 specify instruction set architecture 95 specify processor 100 target hardware 74 120 vector library transformations with xvector 122 with debugging 66 with fast 59 options commonly used 43 grouped by function 39 legacy 44 obsolete 45 order of processing 39 pass option to compilation phase 84 show list of help 67 silent 45 syntax on command line 38 unrecognized 23 Reference to all option flags 45 to 124 a 46 aligncommon 46 ansi
216. ured with one source code procedure per file or by gathering procedures that work together into single files The Fortran Programming Guide describes the advantages and disadvantages of these configurations Source File Preprocessors Both 77 and 95 support two source file preprocessors fpp and cpp Either can be invoked by the compiler to expand source code macros and symbolic definitions prior to compilation The compilers will use fpp by default the xpp cpp option changes the default from fpp to cpp See also the discussion of the Dname option fpp is a Fortran specific source preprocessor See the fpp 1 man page and the fpp README for details It is invoked by default by 77 on files with a F extension and by 95 on files with a F F90 or F95 extension Chapter 2 Using Sun Fortran Compilers 21 22 The source code for fpp is available from the Netlib web site at http www netlib org fortran See cpp 1 for information on the standard Unix C language preprocessor Use of fpp over cpp is recommended on Fortran source files Separate Compiling and Linking You can compile and link in separate steps The c option compiles source files and generates o object files but does not create an executable Without the c option the compiler will invoke the linker By splitting the compile and link steps in this manner a complete recompilation is not needed just to fix one file as shown in the following example Comp
217. w on numeric input The floating point input data is too large to be represented by the corresponding input variable Appendix A Runtime Error Messages 129 130 I O Error Messages 95 Error messages generated by Fortran 95 programs are different than those generated by Fortran 77 programs Here is the previous example compiled and run with Fortran 95 demo cat wf f WRITE 6 1 END demo 95 o wf wf f demo wf xx x x FORTRAN RUN TIME SYSTEM Error 1003 unformatted I O on formatted unit Location the WRITE statement at line 1 of wf f Unit 6 File standard output Abort Because the 95 message contains references to the originating source code filename and line number application developers should consider using the ERR clause in I O statements to softly trap runtime I O errors TABLE A 2 lists the runtime I O messages issued by 95 TABLE A 2 95 Runtime I O Messages Error Message 1000 format error 1001 illegal unit number 1002 formatted I O on unformatted unit 1003 unformatted I O on formatted unit 1004 direct access I O on sequential access unit 1005 sequential access I O on direct access unit 1006 device does not support BACKSPACE 1007 off beginning of record 1008 can t stat file 1009 no after repeat count Fortran User s Guide July 2001 TABLE A 2 95 Runtime I O Messages Continued Error Messa
218. xarch isa Specify instruction set architecture ISA Architectures that are accepted by xarch keyword isa are shown in TABLE 3 12 TABLE 3 12 xarch ISA Keywords Platform Valid xarch Keywords SPARC generic generic64 native native64 v7 v8a v8 v8plus v8plusa v8plusb v9 v9a vIb Note that although xarch can be used alone it is part of the expansion of the xtarget option and may be used to override the xarch value that is set by a specific xtarget option For example 95 xtarget ultra2 xarch v8plusb overrides the xarch v8 set by xtarget ultra2 This option limits the code generated by the compiler to the instructions of the specified instruction set architecture by allowing only the specified set of instructions This option does not guarantee use of any target specific instructions If this option is used with optimization the appropriate choice can provide good performance of the executable on the specified architecture An inappropriate choice results in a binary program that is not executable on the intended target platform Chapter 3 Fortran Compiler Options 95 96 TABLE 3 13 summarizes the most general xarch options TABLE 3 13 Most General xarch Options on SPARC Platforms xarch Performance generic e runs adequately on all platforms v8plusa e runs Optimally on UltraSPARC II processors in 32 bit mode e no execution on other platforms v8plusb e runs Optimally on UltraSPARC III p
219. xtensions 157 I O Extensions 158 Directives 160 Form of Special f95 Directive Lines 160 FIXED and FREE Directives 161 Parallelization Directives 162 Intrinsics 162 Compatibility with FORTRAN 77 163 Incompatibility Issues Between 95 and 77 163 I O Compatibility 164 Linking with 77 Compiled Routines 165 Intrinsics 166 Forward Compatibility 167 Mixing Languages 167 Module Files 167 xtarget Platform Expansions 169 Fortran Directives Summary 175 General Fortran Directives 175 Special Fortran 95 Directives 177 Sun Parallelization Directives 177 Fortran User s Guide July 2001 Cray Parallelization Directives 179 Fortran 95 OpenMP Directives 180 OpenMP Library Routines 187 OpenMP Environment Variables 190 Index 193 Contents ix x Fortran User s Guide July 2001 TABLE 1 1 TABLE 2 1 TABLE 2 2 TABLE 3 1 TABLE 3 2 TABLE 3 3 TABLE 3 4 TABLE 3 5 TABLE 3 6 TABLE 3 7 TABLE 3 8 TABLE 3 9 TABLE 3 10 TABLE 3 11 TABLE 3 12 TABLE 3 13 TABLE 3 14 TABLE 3 15 TABLE 3 16 Tables READMEs of Interest 14 File Name Suffixes Recognized by Sun Fortran Compilers 20 Summary of General Fortran Directives 26 Options Syntax 38 Typographic Notations for Options 38 Compiler Options Grouped by Functionality 39 Commonly Used Options 43 Backward Compatibility Options 44 Obsolescent Options 45 Default Data Sizes and db1 Bytes 53 Subnormal REAL and
220. y with the Solaris operating environment or UNIX in general is also assumed Discussion of Fortran programming issues on Solaris operating environments including input output application development library creating and use program analysis porting optimization and parallelization can be found in the companion Sun WorkShop Fortran Programming Guide Other Fortran manuals in this collection include the Fortran Library Reference and the FORTRAN 77 Language Reference See Accessing Related Documentation on page 6 How This Book Is Organized Chapter 1 briefly describes the features of the compilers Chapter 2 discusses the compiler environments Chapter 3 gives detailed descriptions of all the compile time command line options and flags Appendix A lists error messages issued by the Fortran runtime library and operating environment Appendix B notes new features of the compilers and changes in recent releases Appendix C describes the differences between the Sun 95 compiler and the Fortran 95 standard and incompatibilities with 77 programs Appendix D lists all the platform system names accepted by the compiler xtarget option Appendix E summarizes the directives accepted by the compilers including parallelization and OpenMP directives Typographic Conventions The following table and notes describe the typographical conventions used in the manual Typeface AaBbCc123 AaBbCc123 AaBbCc
221. you have come to expect from Sun the only thing that has changed is the name We believe that the Forte name blends the traditional quality and focus of Sun s core programming tools with the multi platform business application deployment focus of the Forte tools such as Forte Fusion and Forte for Java The new Forte organization delivers a complete array of tools for end to end application development and deployment For users of the Sun WorkShop tools the following is a simple mapping of the old product names in WorkShop 5 0 to the new names in Forte Developer 6 Old Product Name New Product Name Sun Visual WorkShop C Forte C Enterprise Edition 6 Sun Visual WorkShop C Personal Forte C Personal Edition 6 Edition Sun Performance WorkShop Fortran Forte for High Performance Computing 6 Sun Performance WorkShop Fortran Forte Fortran Desktop Edition 6 Personal Edition Sun WorkShop Professional C Forte C 6 Sun WorkShop University Edition Forte Developer University Edition 6 In addition to the name changes there have been major changes to two of the products m Forte for High Performance Computing contains all the tools formerly found in Sun Performance WorkShop Fortran and now includes the C compiler so High Performance Computing users need to purchase only one product for all their development needs a Forte Fortran Desktop Edition is identical to the

Fortran User's Guide - Physics at Oregon State University

Contents

Download Pdf Manuals

Related Search

Related Contents