Script started on Wed 02 Oct 2002 01:18:31 PM PDT rohan% f90 -flags ________________________________________________________________________________ Items within [ ] are optional. Items within < > are variable parameters. Bar | indicates choice of literal values. -someoption[={yes|no}] implies -someoption is equivalent to -someoption=yes ________________________________________________________________________________ -a Collect data for tcov basic block profiling (old format) -aligncommon[=] Align common block elements to the specified boundary requirement; ={1|2|4|8|16} -ansi Report non-ANSI extensions. -autopar Enable automatic loop parallelization (requires WorkShop license) -Bdynamic Allow dynamic linking -Bstatic Require static linking -C Enable runtime subscript range checking -c Compile only; produce .o files but suppress linking -cg89 Generate code for generic SPARC V7 architecture -cg92 Generate code for SPARC V8 architecture -copyargs Allow assignment to constant arguments -D[=] Define preprocessor symbol with -dalign Expands to -xmemalign=8s -db Generate a compiler information (CIF) file -dbl_align_all[=] Force 8-byte alignment of all data; ={yes|no} -depend Analyze loops for data dependencies -dryrun Show compiler commands built by driver; no compilation -d{y|n} Allow or disallow dynamic libraries for the entire executable -e Accept extended (132 character) input source lines -erroff=[,...] Suppress specified warnings; ={%all|%none|} -errtags[=] Display messages with tags; ={yes|no} -explicitpar Parallelize loops explicitly marked with directives (requires WorkShop license) -ext_names=plain Use the Sun C compiler's naming conventions (no underscores) -ext_names=underscores Use the Sun FORTRAN 90 compiler's naming conventions (add underscores) -F Apply the preprocessor to .F and .F90 files, but do not compile -f Force non-standard data alignment on 8-byte boundaries -fast Optimize for speed using a selection of options -fixed Accept fixed-format input source files -flags Show this summary of compiler options -fnonstd Initialize floating-point hardware to non-standard preferences -fns[={yes|no}] Select non-standard floating point mode -fpover[=] Provide run-time overflow check during READ; ={yes|no} -fpp Expands source code macros and enables conditional compilation of code -free Accept free-format input source files -fround= Select the IEEE rounding mode in effect at startup -fsimple[=] Select floating-point optimization level; ={1|0|2} -ftrap=[,...] Select floating-point trapping mode in effect at startup -G Build a dynamic shared library -g Compile for debugging -h Assign to generated dynamic shared library -help Show this summary of compiler options -I Add to INCLUDE and #include search path -inline=[[,...]] Attempt inlining of specified user routines; ={%auto|[no%]} -KPIC Compile position independent code with 32-bit addresses -Kpic Compile position independent code -L Pass to linker to add to the library search paths -l Add lib.a or lib.so to linker search list -libmil Inline selected libm math routines for optimization -loopinfo Show loops that parallelized (requires WorkShop license) -M Add to module search path -mp=

Accept the sun/cray-style MP directives;

={%none|sun|cray|openmp} -mt Specify options needed when compiling multi-threaded code -native Same as -xtarget=native -noautopar Cancel -autopar on command line -nodepend Cancel -depend on command line -noexplicitpar Cancel -explicitpar on command line -nolib Do not link with default system libraries -nolibmil Cancel -libmil on command line -noqueue Disable queuing of compiler license requests -noreduction Cancel -reduction on command line -norunpath Do not build a runtime search path into the executable -O Same as -O3 and -xO3 -O1 Provide minimum statement-level optimizations -O2 Enable basic block-level optimizations -O3 Add loop unrolling and global optimizations -O4 Add automatic inlining of routines from same source file -O5 Attempt aggressive optimizations (with profile feedback) -o Set name of output file or executable to -onetrip Enable one-trip DO loops -openmp Accept the openMP directives -PIC Same as -KPIC -p Compile for profiling with prof -pad[=

] Pad variables for efficient use of cache;

={local|common} -parallel Parallelize loops with -autopar, -explicitpar, -depend -pg Compile for profiling with gprof -pic Same as -Kpic -Qoption

[,...] Pass option to compilation phase

-qp Same as -p -R

[:

...] Build runtime search path list into executable -r8const Promote single precision const to double precision. -reduction Recognize reduction operations in parallelized loops -S Compile and only generate assembly code (.s) -s Strip symbol table from the executable file -sb Compile for use with the WorkShop source browser -sbfast Generate only WorkShop source browser information, no compilation -silent Do not echo file and routine names during compilation -stackvar Allocate all local variables on the memory stack -stop_status[=] STOP statement returns integer status to shell; ={yes|no} -temp= Use as directory for temporary files -time Report the execution time for each compilation phase -U Recognize upper case and lower case in source files -U Cancel command line option -D for the same -u Report on use of undeclared variables -unroll= Enable unrolling loops to a depth > 0 where possible -V Report version number of each compilation phase -v Verbose mode -vpara Verbose parallelization warnings (requires WorkShop license) -w Suppress most compiler warning messages -w0 Show error messages -w1 Show errors and warnings (default) -w2 Show errors, warnings, and cautions -w3 Show errors, warnings, cautions, and notes -w4 Show errors, warnings, cautions, notes, and comments -Xlist Show global errors, cross reference, and listing -Xlistc Show only global errors and call-graph -XlistE Show only global errors -Xlisterr Suppress all -Xlist error messages -Xlisterr Suppress -Xlist error message -Xlistf For faster output, -Xlist but no compilation -Xlisth Halt compilation if global errors are detected -XlistI Analyze #include and INCLUDE files as well as source -XlistL Show only listing and global errors -Xlistl Set page length for listings to lines. Default is 66 -Xlisto Redirect output listings to the file .lst -Xlists Suppress unreferenced names from cross-reference -Xlistv Set level of checking; ={1|2|3|4}. Default is 2 -Xlistw Set width of output line to columns. Default is 79 -Xlistwar Suppress -Xlist warning message -XlistX Show only cross reference and errors -xa Same as -a -xarch= Specify target architecture instruction set -xautopar Same as -autopar -xcache= Define cache properties for use by optimizer -xcg89 Same as -cg89 -xcg92 Same as -cg92 -xchip= Specify the target processor for use by the optimizer -xcode= Generate different code for forming addresses -xcommonchk[=] Enable run-time taskcommon consistency check; ={yes|no} -xcrossfile[=] Enable optimization and inlining across source files, n={1|0} -xdepend Same as -depend -xexplicitpar Same as -explicitpar -xF Compile for later mapfile reordering -xhelp= Display on-line help information; ={flags|readme} -xia[=] Enable interval arithmetic with proper env; ={widestneed|strict} -xildoff Cancel -xildon -xildon Enable use of the incremental linker, ild -xinline=[[,...]] Same as -inline -xinterval[=] Enable interval arithmetic; ={no|widestneed|strict} -xipo[=] Perform interprocedural analysis and optimizations, ={0|1|2} -xlang=[,...] The set of languages used in the program; ={f77} -xlibmil Same as -libmil -xlibmopt Link with optimized math library -xlic_lib=[,...] Link with Sun supplied licensed libraries listed in -xlicinfo Show license server information -xloopinfo Same as -loopinfo -xmaxopt[=] Maximum optimization level allowed on #pragma opt; ={5|1|2|3|4} -xmemalign[=] Controls memory alignment; ={1|2|4|8|16}, ={f|i|s} -xnolib Same as -nolib -xnolibmil Same as -nolibmil -xnolibmopt Cancel -xlibmopt on command line -xO Same as -O -xopenmp Same as -openmp -xpad[=

] Same as -pad -xparallel Same as -parallel -xpg Same as -pg -xpp[=

] Select preprocessor;

={fpp|cpp} -xprefetch[=[,]] Enable UltraSPARC II and UltraSPARC III prefetch instructions, ={yes|no|[no%]auto|[no%]explicit|latx:.} -xprofile=collect[:] Compile for profile feedback data collection -xprofile=tcov Compile for tcov basic block profiling (new format) -xprofile=use[:] Optimize with runtime profile data feedback -xrecursive Allow recursive calls -xreduction Same as -reduction -xregs=[,...] Specify the usage of optional registers -xs Allow debugging without object (.o) files -xsafe=mem Assume that no memory based traps will occur -xsb Same as -sb -xsbfast Same as -sbfast -xspace Do not do optimizations that increase code size -xtarget= Specify target system for optimization -xtime Same as -time -xtypemap= Specify default data mappings -xunroll= Same as -unroll -xvector[=] Generate calls to vector library functions; ={yes|no} -Zlp Compile for looptool profiling -z Link editor option All other options are passed down to ld. Suffix 'a' Object library Suffix 'il' Inline expansion file Suffix 'o' Object file Suffix 'so' Shared object Suffix 's' Assembler source Suffix 'S' Assembler source for cpp Suffix 'for' f90 source (Fixed Form) Suffix 'ftn' f90 source (Fixed Form) Suffix 'f' f90 source (Fixed Form) Suffix 'F' not currently recognized Suffix 'f90' f90 source (Free Form) Suffix 'F90' f90 source (Free Form) Suffix 'f95' f95 source (Free Form) Suffix 'F95' f95 source (Free Form) Suffix 'lst' f90 listing Suffix 'fln' f90 Global Program Checking data Suffix 'T' Intermediate file For additional product information or the name of a Sun software reseller, please call 1-800-786-7638, or (+1) 512-434-1511 (outside the US), or fax your request to (+1) 512-218-3866. For technical support, please contact your Sun software reseller for the name of an Authorized Service Provider. For more information, please browse http://www.sun.com/workshop rohan% exit rohan%