The following performance options are listed in alphabetical order.
Action |
Option |
Details |
---|---|---|
Selects a combination of compilation options for optimum execution speed. |
-fast | |
Strips the symbol table out of the executable. |
-s | |
Specifies target architecture instruction set. |
-xarch=a | |
SPARC: Defines target cache properties for the optimizer. |
-xcache=c | |
Compiles for generic SPARC architecture. |
-xcg89 | |
Compiles for SPARC V8 architecture. |
-xcg92 | |
Specifies target processor chip. |
-xchip=c | |
Enables linker reordering of functions. |
-xF | |
Inlines the specified routines to optimize for speed. |
-xinline=rlst | |
Inlines selected libm library routines for optimization. |
-xlibmil | |
SPARC: Uses a library of optimized math routines. |
-xlibmopt | |
Cancels -xlibmil on the command line. |
-xnolibmil | |
Does not use math library routines. |
-xnolibmopt | |
Specifies optimization level to level. |
-xOlevel | |
SPARC: Controls scratch register use. |
-xregs=rlst | |
SPARC: Allows no memory-based traps. |
-xsafe=mem | |
SPARC: Does not allow optimizations that increase code size. |
-xspace | |
Specifies a target instruction set and optimization system. |
-xtarget=t | |
Enables unrolling of loops where possible. |
-xunroll=n |