----------------------------------------------------------------------------------- Fujitsu PRIMEPOWER flags/tunables description (Jun.30 2003) (Each section is sorted in case insensitive, alphabetical order) Table of Contents [1] Fujitsu Parallelnavi 1.0.2 compiler flag description [2] Fujitsu Parallelnavi 2.1 compiler flag description [3] Sun C, C++ and Fortran Sun ONE Studio 8 flag description [4] Environment Variables [5] Kernel Parameters (/etc/system) [6] Configuration file for large page manager (/etc/opt/FJSVpnrm/lpg.conf) [7] Commands for feedback control ----------------------------------------------------------------------------------- [1] Fujitsu Parallelnavi 1.0.2 compiler flag description (Jun.30 2003) Compiler options Remark ----------------------------------------------------------------------------------- -dy/-dn Specifies dynamic(-dy) or static(-dn) linkage of libraries. -dy is the default unless -Kfast_GP=n (n>=3) is specified and -Klargepage is not specified, in that case -dn is the default. -Kbcopy Convert memory copy loop to memmove or memcpy function. -Kcfunc This uses high speed mathematical functions and library functions (malloc,calloc,realloc,free) prepared by this compilation system. -Kfast_GP[={0|1|2|3|4|5}] This performs optimization for SPARC64 GP series. 0: This performs optimization suitable for SPARC64 GP. 1: This generates multiply and add instruction in addition to -Kfast_GP=0. (default) 2: This performs reordering of expression evaluation in addition to -Kfast_GP=1. 3: (C and C++ compiler) This generates crossfile optimization, and inter-procedural optimization in addition to -Kfast_GP=2. (Fortran compiler) This performs loop restructuring optimization in addition to -Kfast_GP=2. 4: (C and C++ compiler only) This generates advanced branch optimization in addition to -Kfast_GP=3. 5: (C and C++ compiler only) This generates global instruction scheduling optimization suit for scientific application in addition to -Kfast_GP=3. -KGREG The global registers g2 through g7 are subject to register allocation in the compile stage. -Klargepage Specifies to generate executable program which utilizes Parallelnavi largepage facility. This option expands the page size of virtual memory from original 8KB to 4MB. -Knounroll Prevents loop unrolling optimizations. -Kpg Generates instructions to produce a profile file for subsequent optimization (global instruction scheduling etc.). -Kpopt This option specifies the optimization data pointed to by pointers using a limited interpretation that the areas referred by pointers are only referred by pointers. -Kpu[=file] This performs optimization (global instruction scheduling, etc.) using program runtime profile information obtained by specifying -Kpg option. -Kprefetch[={1|2|3|4}] Generate prefetch instruction correspond to each prefetch level. 1: Basic level prefetch for array elements only inner-most loop. 2: In addition to the -Kprefetch=1, generates the prefetch instruction for array elements within the loop pre-header which access the first iteration in the loop. 3: In addition to the -Kprefetch=2, when the stride of access for array elements are larger than cache line size, compiler generates prefetch instruction for each cache line size access. 4: Maximum level for generating prefetch instruction. In addition to -Kprefetch=3, compiler generates the prefetch instruction for array elements which access in the outer loop. -Kstaticclump When this option is specified, all variables in the source file which are declared as static or global (except large arrays) are compiled together (renamed to avoid conflicts if necessary) into one big structure and can therefore be addressed with the base address of the structure and an index from this base address. -Kuse_rodata This option specifies whether string constant, floating point constant and initialization value of aggregate type local storage variable are allocated to read-only data section. -Kxi=N Inline expansion, instead of function calls, is performed. Expanded function is selected by result of profiler. N is the percentage that means increased object size. -O[level] Specifies the optimization level. 0: No optimization. 1: Basic optimization. 2: Loop unrolling in addition to -O1. 3: Global instruction scheduling and restructuring of nested loop in addition to -O2. 4: Enhanced optimization of loop restructuring rather than -O3. ----------------------------------------------------------------------------------- [2] Fujitsu Parallelnavi 2.1 compiler flag description (Jun.30 2003) Compiler options Remark ----------------------------------------------------------------------------------- -Am Required if a source file contains modules which will be referenced by USE statements in other source files or if a source file contains USE statements that reference modules in another source file. -dy/-dn Specifies dynamic(-dy) or static(-dn) linkage of libraries. -dy is the default unless -Kfast_GP=n (n>=3) is specified and -Klargepage is not specified, in that case -dn is the default. -f omitmsg Set the level of diagnostic messages output and inhibit specific messages. omitmsg is one of i, w, or s, and/or a list of msgnum. i: All messages are output, this is the default. w: i level messages are not output. s: i and w level messages are not output. msgnum: Message number msgnum is inhibited. msgnum must be an i or w level message. -Fixed Specifies that Fortran source programs are written in fixed source form. -fs Do not print any warnings or diagnostic messages other than fatal errors. -Kalignc[=N] Adjust entry of global data alignment at n-byte boundary. N can be specified from 1 to 32768. -Kalignl[=N] Adjust entry of local data alignment at n-byte boundary. N can be specified from 1 to 32768. -Karraypad_const[=N] Insert padding elements after each row of an array whose size is declared with constants for efficient use of cache. -Kauto Local variables (without an initial value or the SAVE attribute) are allocated on the runtime stack. Their values are lost when the procedure ends. -Kcfunc This uses high speed mathematical functions and library functions (malloc,calloc,realloc,free) prepared by this compilation system. -Kcommonpad[=N] Insert padding elements in common blocks for efficient use of cache. N can be specified from 4 to 4096. -Kcrossfile This option specifies crossfile optimization. If program consists of several files, the compiler refers these files at one time, and analyzes data dependency and control relation across these files. -Kfast_GP2[={0|1|2}] This performs optimization for SPARC64 GP2 series. 0: This performs optimization suitable for SPARC64 GP2. 1: This generates multiply and add instruction in addition to -Kfast_GP2=0. (default) 2: This performs reordering of expression evaluation in addition to -Kfast_GP2=1. -KFMADD This option specifies use of the combined multiply- add/subtract floating-point instructions. -Kfrecipro This option specifies to convert a floating point division into multiplication by the reciprocal. -Kfuse Fuses neighboring loops. -KGREG_SYSTEM The global registers g5 through g7 are subject to register allocation in the compile stage. -Kgs Performs global instruction scheduling. -Kilfunc This option replaces several and double precision mathematical functions,sin,cos,log10,log and exp with compiler builtin functions. -Klargepage Specifies to generate executable program which utilizes Parallelnavi largepage facility. -KNOFMADD This option suppresses use of the combined multiply- add/subtract floating-point instructions. -Knoprefetch Suppresses use the prefetch instruction. -Knounroll Prevents loop unrolling optimizations. -Knovfunc Suppresses to change the intrinsic function (including power operation) to a multi-operation function. -Kpreex This option specifies the optimization by moving the evaluation of invariant expressions beyond branch. -Kpopt This option specifies the optimization data pointed to by pointers using a limited interpretation that the areas referred by pointers are only referred by pointers. -Kpg Generates instructions to produce a profile file for subsequent optimization (global instruction scheduling etc.). -Kprefetch[={1|2|3|4}] Generate prefetch instruction correspond to each prefetch level. 1: Basic level prefetch for array elements only inner-most loop. 2: In addition to the -Kprefetch=1, generates the prefetch instruction for array elements within the loop pre-header which access the first iteration in the loop. 3: In addition to the -Kprefetch=2, when the stride of access for array elements are larger than cache line size, compiler generates prefetch instruction for each cache line size access. 4: Maximum level for generating prefetch instruction. In addition to -Kprefetch=3, compiler generates the prefetch instruction for array elements which access in the outer loop. -Kprefetch_cache_level=[1/2/3] LEVEL-1: Generate prefetch instructions in order for the data to reside in the primary cache. LEVEL-2: Generate prefetch instructions in order for the data to reside only in the secondary cache. LEVEL-3: Generate both LEVEL-1 and -2 prefetch instructions. -Kprefetch_infer For data prefetch control, in cases where the stride of array accesses is not clear from static analysis of the source files, the compiler is told to use its internal heuristics for the addresses that have been determined by the compiler as prefetch addresses. -Kprefetch_iteration=N Generate the prefetch instruction of the data which is referred after N iterations. -Kprefetch_line=N Generate the prefetch instruction to get the data located N times bytes in a line of the primary data cache ahead of the address a neighboring load or store instruction points. -Kprefetch_line_L2=N Same to -Kprefetch_line=N, except the data reside only in the secondary cache. -Kpreload Moves load instructions across branches. -Kpreschedule_length[=N] When -O5 is used, the instruction scheduler works twice, before and after the register allocation, which are named pre-pass and post-pass scheduling respectively. -Kpreschedule_length[=N] can control how aggressively pre-pass scheduling works. The unit is the upper limit of the distance an instruction can move from its original place. -Kpu[=file] This performs optimization (global instruction scheduling, etc.) using program runtime profile information obtained by specifying -Kpg option. -Kunroll[=N] Performs loop unrolling. N means upper limit of unrolling expansion number, whose value should be from 2 to 9999. -O[level] Specifies the optimization level. 0: No optimization. 1: Basic optimization. 2: Loop unrolling in addition to -O1. 3: Global instruction scheduling and restructuring of nested loop in addition to -O2. 4: Enhanced optimization of loop restructuring rather than -O3. 5: Creates an object program by applying further optimizations of register allocation in addition to -O4. -SSL2 The whole set of routines from SSL II, SSL II Thread- Parallel Capabilities and BLAS/LAPACK becomes part of link-edit libraries. -x- Inline expansion, instead of function calls, is performed for all functions defined in the C source code. -x stm_no Applying optimization for inline expansion of user-defined external procedure having fewer than specified number of execution statements in the stm_no arguments. -x dir=directory_name Performs inline expansion of procedures defined in the files under the directory specified and in the file currently being compiled. ----------------------------------------------------------------------------------- [3] Sun C, C++ and Fortran Sun ONE Studio 8 flag description (Jun.30 2003) Compiler options Remark ----------------------------------------------------------------------------------- -array_pad_rows, Enable padding of arrays by n. (Fortran) cc Invoke the Sun ONE Studio 8 Compiler C (C compiler) CC Invoke the Sun ONE Studio 8 Compiler C++ (C++ compiler) -crit Enable optimization of critical control paths (optimizer) -dalign Assume data is naturally aligned. (C, C++, Fortran) -Dalloca=__builtin_alloca (Portability flag) Portability switch, used for 176.gcc: allow use of compiler's internal builtin alloca. -depend Synonym for -xdepend. (Fortran) -DHOST_WORDS_BIG_ENDIAN Portability switch, used for 176.gcc: (Portability flag) controls how bytes are numbered within a word. -D__MATHERR_ERRNO_DONTCARE (C) Allows the compiler to assume that your code does not rely on setting of the errno variable. -DSPEC_CPU2000_SOLARIS Portability switch, used for 253.perlbmk: (Portability flag) selects header files and code paths compatible with Solaris. -DSUN Portability switch, used for 186.crafty: (Portability flag) selects header files and code paths compatible with Solaris. -DSYS_HAS_CALLOC_PROTO Portability switch, used for 254.gap: (Portability flag) allows use of the designated prototype. -DSYS_HAS_IOCTL_PROTO Portability switch, used for 254.gap: (Portability flag) allows use of the designated prototype. -DSYS_HAS_SIGNAL_PROTO Portability switch, used for 254.gap: (Portability flag) allows use of the designated prototype. -DSYS_HAS_TIME_PROTO Portability switch, used for 254.gap: (Portability flag) allows use of the designated prototype. -DSYS_IS_USG Portability switch, used for 254.gap: (Portability flag) selects code compatible with USG-based systems. -e Portability switch, used for 178.galgel: (Portability, Fortran) allows source lines to be up to 132 characters long. f90 Invoke the Sun ONE Studio 8 Compiler Fortran 90 (Fortran compiler) -fast A convenience option, this switch selects the (C) following switches that are defined elsewhere in this page: -D__MATHERR_ERRNO_DONTCARE -dalign -fns -fsimple=2 -fsingle -ftrap=%none -xalias_level=basic -xbuiltin=%all -xdepend -xlibmil -xO5 -xprefetch=auto,explicit -xtarget=native -fast A convenience option, this switch selects the (C++) following switches that are defined elsewhere in this page: -dalign -fns -fsimple=2 -ftrap=%none -xbuiltin=%all -xlibmil -xlibmopt -xO5 -xtarget=native -fast A convenience option, this switch selects the (Fortran) following switches that are defined elsewhere in this page: -dalign -depend -fns -fsimple=2 -ftrap=common -xlibmil -xlibmopt -xO5 -xpad=local -xprefetch=auto,explicit -xtarget=native -xvector=yes -fixed Portability switch, used for 178.galgel: (Portability, Fortran) assume fixed-format source input. -fns Selects faster (but nonstandard) handling of (C, C++, Fortran) floating point arithmetic exceptions and gradual underflow. -fsimple= Controls simplifying assumptions for (C, C++, Fortran) floating point arithmetic: -fsimple=0 Permits no simplifying assumptions. Preserves strict IEEE 754 conformance. -fsimple=1 Allows the optimizer to assume: The IEEE 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 as operands need not propagate NaNs to their results. For example, x*0 may be replaced by 0. Computations do not depend on sign of zero. -fsimple=2 Permits more aggressive floating point optimizations that may cause programs to produce different numeric results due to changes in rounding. Even with -fsimple=2, the optimizer still is not permitted to introduce a floating point exception in a program that otherwise produces none. -fsingle Evaluate float expressions as single precision. (C) -ftrap=common Sets the IEEE 754 trapping mode to common exceptions (C, C++, Fortran) (invalid, division by zero, and overflow). -ftrap=%none Turns off all IEEE 754 trapping modes. (C, C++, Fortran) -library=iostream Portability switch, used for 252.eon: (Portability, C++) allow use of the classic iostream library. -ll2amm Include a library containing chip specific (linker) memory routines. -lm Include the math library. (linker) -lmopt Include the optimized math library. This option (linker) usually generates faster code, but may produce slightly different results. Usually these results will differ only in the last bit. -lprism32 (linker) Library to enable Intimate Shared Memory (ISM) (linker) (4MB page) usage. -noex Do not allow C++ exceptions. A throw specification (C++) on a function is accepted but ignored; the compiler does not generate exception code. -O A synomym for -xO3. (Fortran) -Qoption Pass flags along to compiler phase: f90comp Fortran first pass iropt Global optimizer cg Code Genetator -Qoption cg See -Wc, below. (The code generator (code generator) phase is addressed via -Qoption cg in Fortran and C++; and via -Wc in C.) -Qoption cg -Qeps:enabled=1 (code generator) See -Wc,-Qeps:enabled=1 -Qoption cg -Qeps:ws= (code generator) See -Wc,-Qeps:ws= -Qoption cg -Qgsched-T (code generator) See -Wc,-Qgsched-T -Qoption cg -Qgsched-trace_late=1 (code generator) See -Wc,-Qgsched-trace_late=1 -Qoption iropt See -W2, below. (The optimizer can (optimizer) be addressed either via Qoption iropt in Fortran and C++; or via -W2 in C.) -Qoption iropt -Addint:sf= (optimizer) When considering whether to interchange loops, set memory store operation weight to n. A higher value of n indicates a greater performance cost for stores. -Qoption iropt -Ainline[:cp=][:cs=][:inc=][:irs=][:mi][:recursion=1] (optimizer) See -W2,[:cp=][:cs=][:inc=][:irs=][:mi][:recursion=1] -Qoption iropt -Apf:pdl=1 (optimizer) Do prefetching for one-level indirect memory references. -Qoption iropt -Atile:skewp[:b] (optimizer) Perform loop tiling which is enabled by loop skewing. Loop skewing is a transformation that transforms a non-fully interchangeable loop nest to a fully interchangeable loop nest. The optional b sets the tiling block size to n. -Qoption iropt -Aujam:inner=g (optimizer) Increase the probability that small-trip-count inner loops will be fully unrolled. -Qoption iropt -Mt See -W2,-Mt RM_SOURCES = lapak.f90 This option allows building the benchmark 178.galgel (SPEC tools) without its copy of the lapak sources; instead, the lapak entry points in the sunperf library are used. rm -rf ./feedback.profile ./SunWS_cache (Unix) Remove any profile feedback information from previous runs. -stackvar Allocate routine local variables on the stack. (Fortran) submit=echo 'pbind -b...' > dobmk; sh dobmk (SPEC tools, Unix) When running multiple copies of benchmarks, the SPEC config file feature submit is sometimes used to cause individual jobs to be bound to specific processors: submit= causes the SPEC tools to use this line when submitting jobs. echo ...> dobmk causes the generated commands to be written to a file, namely dobmk. pbind -b causes this copy's processes to be bound to the CPU specified by the expression that follows it. See the config file used in the submission for the exact syntax, which tends to be cumbersome because of the need to carefully quote parts of the expression. When all expressions are evaluated, each CPU ends up with exactly one copy of each benchmark. The pbind expression may include: $SPECUSERNUM: the SPEC tools-assigned number for this copy of the benchmark. psrinfo: find out what processors are available grep off-line: search the psrinfo output for information regarding off-line cpus awk...print \$1: Pick out the line corresponding to this copy of the benchmark and use the CPU number mentioned at the start of this line. sh dobmk actually runs the benchmark. -W, Pass flags along to compiler phase (2=optimizer, c=code genetator). -W2,-Abcopy Increase the probability that the compiler will (optimizer) perform memcpy/memset transformations. -W2,-Abopt Enable aggressive optimizations of all branches. (optimizer) -W2,-Aheap Allows the compiler to recognize malloc-like (optimizer) memory allocation functions. -W2,-Ainline[:cp=][:cs=][:inc=][:irs=][:mi][:recursion=1] (optimizer) Control the optimizer's loop inliner: cp= The minimum call site frequency counter in order to consider a routine for inlining. cs= Set inline callee size limit to n. The unit roughly corresponds to the number of instructions. inc= The inliner is allowed to increase the size of the program by up to n%. irs= Allow routines to increase by up to n. The unit roughly corresponds to the number of instructions. mi Perform maximum inlining (without considering code size increase). recursion=1 Allow routines that are called recursively to still be eligible for inlining. -W2,-crit Enable optimization of critical control paths. (optimizer) -W2,-Amemopt:arrayloc Reconstruct array subscripts during memory (optimizer) allocation merging and data layout program transformation. -W2,-Apf:llist=:noinnerllist (optimizer) Do speculative prefetching for link-list data structures: llist= perform prefetching n iterations ahead noinnerllist do not attempt for innermost loops. -W2,-Ashort_ldst Convert multiple short memory operations into (optimizer) single long memory operations. -W2,-Aunroll Enables outer-loop unrolling. (optimizer) -W2,-Mr Maximum code increase due to inlining is limited (optimizer) to n triples. -W2,-Ms Maximum level of recursive inlining. (optimizer) -W2,-Mt The maximum size of a routine body eligible for (optimizer) inlining is limited to n triples. -W2,-reroll=1 Turns on loop rerolling. (optimizer) -W2,-whole Do whole program optimizations. (optimizer) -Wc,-Qdepgraph-early_cross_call=1 (code generator) There are several scheduling passes in the compiler. This option allows early passes to move instructions across call instructions. -Wc,-Qeps:do_spec_load=1 (code generator) Allow generating speculative load during EPS. -Wc,-Qeps:enabled=1 Use enhanced pipeline scheduling(EPS) (code generator) and selective scheduling algorithms for instruction scheduling. -Wc,-Qeps:rp_filtering_margin=100 (code generator) Turn off register pressure heuristics in EPS. -Wc,-Qeps:ws= Set the EPS window size, that is, the number (code generator) of instructions it will consider across all paths when trying to find independent instructions to schedule a parallel group. Larger values may result in better run time, at the cost of increased compile time. -Wc,-Qgsched-T Sets the aggressiveness of the trace (code generator) formation, where n is 4, 5, or 6. The higher the value of n, the lower the branch probability needed to include a basic block in a trace. -Wc,-Qgsched-trace_late=1 (code generator) Turns on the late trace scheduler. -Wc,-Qicache-chbab=1 Turn on optimization to reduce branch (code generator) after branch penalty: nops will be inserted to prevent one branch from occupying the delay slot of another branch. -Wc,-Qipa:valueprediction (code generator) Use profile feedback data to predict values and attempt to generate faster code along these control paths, even at the expense of possibly slower code along paths leading to different values. Correct code is generated for all paths. -Wc,-Qiselect-funcalign= (code generator) Do function entry alignment at n-byte boundaries. -Wc,-Qiselect-sw_pf_tbl_th= (code generator) Peels the most frequent test branches/cases off a switch until the branch probability reaches less than 1/n. This is effective only when profile feedback is used. -Wc,-Qlp=[-av=][-t=][-fa=][-fl=] (code generator) Control irregular loop prefetching: lp= Turns the module on (1) or off (0) (default is on for F90; off for C/C++) -av= Sets the prefetch look ahead distance, in bytes. Default is 256. -t= Sets the number of attempts at prefetching. If not specified, t=2 if -xprefetch_level=3 has been set; otherwise, defaults to t=1. -fa= 1=Force user settings to override internally computed values. -fl= 1=Force the optimization to be turned on for all languages. -Wc,-Qms_pipe+intdivusefp (code generator) In pipelined loops, use floating point divide instructions for signed integer division. -Wc,-Qms_pipe+prefolim= (code generator) Set number of outstanding prefetches in pipelined loops to -Wc,-Qms_pipe+unoovf Assert (to the pipeliner) that unsigned (code generator) int computations will not overflow. -Wc,-Qms_pipe-prefst Turn off prefetching for stores in the pipeliner. (code generator) -Wc,-Qms_pipe-pref Turn off prefetching within modulo scheduling. (code generator) -Wc,-Qpeep-Sh0 Reduce the probability that the compiler will hoist sethi (code generator) instructions out of loops. -xalias_level=[basic|std|strong] (C) Allows the compiler to perform type-based alias analysis at the specified alias level: basic Assume that memory references that involve different C basic types do not alias each other. std Assume aliasing rules described in the ISO 1999 C standard. strong In addition to the restrictions at the std level, assume that pointers of type char * are used only to access an object of type char; and assume that there are no interior pointers. -xalias_level=compatible (C++) Allows the compiler to assume that layout-incompatible types are not aliased. -xarch= Limit the set of instructions the compiler may use (C, C++, Fortran) to generic, generic64, native, native64, v7, v8a, v8, v8plus, v8plusa, v8plusb, v9, v9a, v9b. Typical settings include: UltraSPARC-II, 32-bit mode: v8plusa UltraSPARC-II, 64-bit mode: v9a UltraSPARC-III, 32-bit mode: v8plusb UltraSPARC-III, 64-bit mode: v9b For more information, see the Fortran User's Guide at docs.sun.com -xbuiltin=%all Substitute intrinsic functions or inline system (C, C++) functions where profitable for performance. -xchip= Specifies the target processor for use by the (C, C++, Fortran) optimizer. c must be one of: generic, generic64, native, native64, old, super, super2, micro, micro2, hyper, hyper2, powerup, ultra, ultra2, ultra2i, ultra3, ultra3cu, 386, 486, pentium, pentium_pro, 603, 604. -xcache= Defines the cache properties for use by the (C, C++, Fortran) optimizer. c must be one of the following: native (set parameters for the host environment) * s1/l1/a1 * s1/l1/a1:s2/l2/a2 * s1/l1/a1:s2/l2/a2:s3/l3/a3 The si/li/ai are defined as follows: si The size of the data cache at level i, in kilobytes. li The line size of the data cache at level i, in bytes. ai The associativety of the data cache at level i. -xdepend Analyze loops for inter-iteration data dependencies, (C, Fortran) and do loop restructuring. -xinline= Turn off inlining. (C, C++, Fortran) -xipo[=2] Perform optimizations across all object files in the (C, C++, Fortran) link step: 0=off 1=on 2=performs whole-program detection and analysis -xlibmil Use inline expansion for math library, libm. (C, C++, Fortran) -xlibmopt Select the optimized math library. (C++, Fortran) -xlic_lib=sunperf Link with Sun supplied licensed sunperf library. (C, C++, Fortran) -xlinkopt Perform link-time optimizations, such as branch (C, C++, Fortran) optimization and cache coloring. -xO Specify optimization level n: (C, C++, Fortran) -xO1 Does only basic local optimizations (peephole). -xO2 Do basic local and global optimizations, such as induction variable elimination, common subexpression elimination, constant propogation, register allocation, and basic block merging. -xO3 Add global optimizations at the function level, loop unrolling, and software pipelining. -xO4 Adds automatic inlining of functions in the same file. -xO5 Uses optmization algorithms that may take significantly more compilation time or that do not have as high a probability of improving execution time, such as speculative code motion. -xpad=common[:] If multiple same-sized arrays are placed in common, (Fortran) insert padding between them for better use of cache. n specifies the amount of padding to apply, in units that are the same size as the array elements. If no parameter is specified then the compiler selects one automatically. -xpad=local Pad local variables, for better use of cache. (Fortran) -xprefetch=auto,explicit (C, C++, Fortran) Allow generation of prefetch instructions. -xprefetch and -xprefetch=yes is a synonym for -xprefetch=auto,explicit. -xprefetch=latx: Adjust the compiler's assumptions about prefetch latency (C, C++, Fortran) by the specified factor. Typically values in the range of 0.5 to 2.0 will be useful. A lower number might indicate that data will usually be cache resident; a higher number might indicate a relatively larger gap between the processor speed and the memory speed (compared to the assumptions built into the compiler). -xprefetch=no%auto Turn off prefetch instruction generation. (C, C++, Fortran) -xprefetch_level= Control the level of searching that the compiler does (C, C++, Fortran) for prefetch opportunities by setting n to 1, 2, or 3, where higher numbers mean to do more searching. The default is 2. -xprofile=collect:./feedback (C, C++, Fortran) Collect profile data for feedback-directed optimization, and store it in a sub directory of the current directory, named ./feedback. -xprofile=use:./feedback (C, C++, Fortran) Use data collected for profile feedback. Look for it in a subdirectory of the current directory, named ./feedback. -xregs=syst Allows use of the system reserved registers %g6 and (C, C++, Fortran) %g7, and %g5 if not already allowed by -xarch value. -xrestrict Treat pointer-valued function parameters as (C) restricted pointers. -xsafe=mem Enables the use of non-faulting loads when used in (C, C++, Fortran) conjunction with -xarch=v8plus. Assumes that no memory based traps will occur. -xsfpconst Represents unsuffixed floating-point constants (C, C++, Fortran) as single precision. -xtarget=[system_name] Selects options appropriate for the system where (C, C++, Fortran) the compile is taking place, including architecture, chip, and cache sizes. (These can also be controlled separately, via -xarch, -xchip, and -xcache, respectively.) -xunroll=n Specifies whether or not the compiler optimizes (C, C++, Fortran) (unrolls) loops. n is a positive integer. When n is 1, it is a command and the compiler unrolls no loops. When n is greater than 1, -xunroll=n merely suggests to the compiler that it unroll loops n times. -xvector Allow the compiler to transform math library calls within (C, Fortran) loops into calls to the vector math library. ----------------------------------------------------------------------------------- [4] Environment Variables Flag Remark ----------------------------------------------------------------------------------- LD_LIBRARY_PATH=

Specify the locations to resolve dynamic link dependencies. PRISM_HEAP= Set the heap size limit for large pages. PRISM_MODE=2 Large page mode: Attempt to put text, data and heap all into large pages. ulimit -s unlimited Allow stack size to grow without limit. ----------------------------------------------------------------------------------- [5] Kernel Parameters (/etc/system) System Tunable Remark ----------------------------------------------------------------------------------- autoup The frequency of file system sync operations. consistent_coloring Controls the page coloring policy. It can be set to one of the following: 0 (default) dynamic (uses various vaddr bits) 1 static (virtual=paddr) shmsys:shminfo_shmmax Maximum size of system V shared memory segment that can be created. shmsys:shminfo_shmmin Minimum size of system V shared memory segment that can be created. shmsys:shminfo_shmmni System wide limit on number of shared memory segments that can be created. shmsys:shminfo_shmseg Limit on the number of shared memory segments that any one process can create. tune_t_fsflushr The number of seconds between fsflush invocations for checking dirty memory. ----------------------------------------------------------------------------------- [6] Configuration file for large page manager (/etc/opt/FJSVpnrm/lpg.conf) Tunable Remark ----------------------------------------------------------------------------------- JOB=size [unit] Specify the large page memory resource size for in-job processes. "unit" can be T (terra byte), G (giga byte), or M (mega byte) after size. LIMITPOLICY=[job|proc] Define the memory allocation/limitation type. Default is job. job Limits for each job in Node. proc Limits for each process resource set. SHMSEGSIZE=size[unit] Size of large page segment. "unit" can be M for mega-byte and G for giga-byte. TSS=size[unit] Size of total memory, to be used for large page segments. At start of the system, this amount of memory is reserved and initialized. "unit" can be M for mega-byte and G for giga-byte. -------------------------------------------------------------------------------- [7] Commands for feedback control Command Remark -------------------------------------------------------------------------------- Paralllenavi compiler: fdo_pre0 = rm -rf `pwd`*.f.d fdo_pre0 = rm -rf `pwd`*.fbk remove the profile data generated at the last feedback-optimized compilation. Sun Studio 8 compiler: fdo_pre0 = rm -rf `pwd`/..feedback.profile fdo_pre0 = rm -rf `pwd`/SunWS_cache remove the profile data generated at the last feedback-optimized compilation. --------------------------------------------------------------------------------