--------------------------------------------------------- Descriptions of options used by HP for SPEC95 Newsletter: --------------------------------------------------------- +Ae Enable ANSI compatibility mode with UNIX extensions. -D Set definition for preprocessor. +ESfic Replace millicode calls with in-line code for for indirect calls. +ESlit Put literals in read-only text space. +ESsfc Replace millicode calls with in-line code when performing function pointer comparison. -Wl,-a,archive Link with archive format libraries rather than shared. fastmem.o Replacement for malloc/free that assumes ANSI compliant code. +O1 Perform optimizations within basic blocks only. +O2 Perform +O1 optimizations. In addition perform global optimizations at the procedure level. (Same as -O.) +O3 Perform +O2 optimizations as well as interprocedural global optimizations within a file (including inlining). +O4 Perform +O3 optimizations. Also apply interprocedural optimizations across all files of the program. +Ocachepadcommon Pad common blocks to avoid cache collisions. +Odataprefetch Generate data prefetch instructions for data structures referenced within innermost loops. +Oentrysched Perform instruction scheduling on a subprogram's entry and exit sequences. +O[no]fastaccess Enable [disable] fast access to global data items. +O[no]fltacc Disable [enable] optimizations that cause imprecise floating-point results. +Oinline_budget Perform more aggressive inlining. +O[no]libcalls Invoke [do not invoke] low call-overhead versions of select library routines. +O[no]limit Limit [do not limit] optimization by compile-time considerations. +O[no]loop_unroll Enable [disable] loop unrolling. +O[no]parallel Transform [do not transform] eligible loops for parallel execution on multiprocessor machines. +O[no]parmsoverlap Assume that multiple reference parameters do [do not] refer to overlapping memory items. +Optrs_strongly_typed Assume that all pointers are type safe. +O[no]ptrs_to_globals Optimize with assumption that global variables may [may not] be modified through pointers. +Oregionsched Perform scheduling across branches. +Ostaticprediction Enable static branch prediction. +Ovectorize Replace eligible loops with calls to the vector library. +Owhole_program_mode Hold global variables in registers longer and delete inlined or cloned global procedures. +Oaggressive Apply aggressive optimizations which include scheduling on procedure entry and exit code, replacing standard library calls with inlined code sequences, no auto-initialization of uninitialized local scalar non-static variables, and scheduling beyond basic block boundaries. +Oall Apply all optimizations that would be enabled by specifying +O4, +Oaggressive, and +Onolimit on the same command line. Detailed options countermanding components of +Oall override that component. +DA Generate code for the architecture specified. +P/+I Optimize/Instrument for profile based optimization.