CPU2006 Result Flag Description

Operating systems: SUSE Linux Enterprise 10, SUSE Linux Enterprise 11, and Red Hat Enterprise Linux Advanced Platform 5 and 6

Base Optimization Flags

C benchmarks

- -ffast-math
- EXTRA_LDFLAGS, OPTIMIZE
- Sets the following flags:
  - -fno-math-errno
  - -funsafe-math-optimizations
  - -fno-trapping-math
  - -ffinite-math-only
  - -fno-signaling-nans
- Includes:
- -O3
- EXTRA_LDFLAGS, OPTIMIZE
- Optimize yet more. -O3 turns on all optimizations specified by -O2 and also turns on the -finline-functions, -funswitch-loops and -fgcse-after-reload options.
- Includes:
- -mcpu=power7
- EXTRA_LDFLAGS, OPTIMIZE
- Tune to cpu-type everything applicable about the generated code, except for the ABI and the set of available instructions.
  
  A sample list of supported values for this flag are
  - power
  - power2
  - power3
  - power4
  - power5
  - power5+
  - power6
  - power7
- -mtune=power7
- EXTRA_LDFLAGS, OPTIMIZE
- Sets the instruction scheduling parameters for a particular machine type, but does not set the architecture type, register usage, or choice of mnemonics, as -mcpu=cpu_type would. The same values for cpu_type are used for -mtune as for -mcpu. If both are specified, the code generated will use the architecture, registers, and mnemonics set by -mcpu, but the scheduling parameters set by -mtune.
- -mrecip=rsqrt
- EXTRA_LDFLAGS, OPTIMIZE
- This option will enable GCC to use RCPSS and RSQRTSS instructions (and their vectorized variants RCPPS and RSQRTPS) with an additional Newton-Raphson step to increase precision instead of DIVSS and SQRTSS (and their vectorized variants) for single precision floating point arguments. These instructions are generated only when -funsafe-math-optimizations is enabled together with -finite-math-only and -fno-trapping-math.
- -fpeel-loops
- EXTRA_LDFLAGS, OPTIMIZE
- Peels the loops for that there is enough information that they do not roll much (from profile feedback). It also turns on complete loop peeling (i.e. complete removal of loops with small constant number of iterations). Enabled with -fprofile-use
- -funroll-loops
- EXTRA_LDFLAGS, OPTIMIZE
- Unroll loops whose number of iterations can be determined at compile time or upon entry to the loop. -funroll-loops implies both -fstrength-reduce and -frerun-cse-after-loop. This option makes code larger, and may or may not make it run faster.
- Includes:
  - -fstrength-reduce
  - -frerun-cse-after-loop
- -mpopcntd
- EXTRA_LDFLAGS, OPTIMIZE
- Allows GCC to generate the popcount instruction implemented on the POWER7 processor and other processors that support the PowerPC V2.06 architecture.
- -m32
- EXTRA_LDFLAGS, OPTIMIZE
- Generate code for a 32-bit environment. The 32-bit environment sets int, long and pointer to 32 bits and generates code that runs on 32 bits system.
- -fvect-cost-model
- OPTIMIZE
- Enable cost model for vectorization.
- -mveclibabi=mass
- OPTIMIZE
- Specifies the Mathematical Acceleration Subsystem (MASS) ABI to use for vectorizing intrinsics using an external library.
- -Wl,-q
- EXTRA_LDFLAGS
- Pass the -q flag to the linker causing the final executable to have the relocation information.
- -Wl,-Map=link.map,--cref
- EXTRA_LDFLAGS
- Pass "-Map=link.map --cref" to the linker causing to print a linker map to link.map and output a cross reference table to the map file.
- -L /opt/at6.0/lib
- EXTRA_LDFLAGS
- Add directory /opt/at6.0/lib to the list of directories to be searched for -l.
- -L /opt/ibmcmp/xlmass/7.1/lib
- EXTRA_LDFLAGS
- Add directory /opt/ibmcmp/xlmass/7.1/lib to the list of directories to be searched for -l.
- -Wl,-rpath,/opt/at6.0/lib
- EXTRA_LDFLAGS
- Pass "-rpath /opt/at6.0/lib" to the linker causing the Advance Toolchain library directory (/opt/at6.0/lib) to be added to the runtime library search path.
- -Wl,-rpath,/opt/ibmcmp/xlmass/7.1/lib
- EXTRA_LDFLAGS
- Pass "-rpath /opt/ibmcmp/xlmass/7.1/lib" to the linker causing the IBM XL Mathematical Acceleration Subsystem (MASS) library directory (/opt/ibmcmp/xlmass/7.1/lib) to be added to the runtime library search path.
- -lhugetlbfs
- EXTRA_LIBS
- Link with libhugetlbfs.so. This enables heap to be backed by the 16 Megabyte pages.
- -lmassvp7
- EXTRA_LIBS
- Link the mathematical acceleration subsystem (MASS) vector libraries, which contain libraries of tuned mathematical intrinsic functions.
- -lmass_simdp7
- EXTRA_LIBS
- Link the mathematical acceleration subsystem (MASS) SIMD libraries, which contain libraries of tuned mathematical intrinsic functions.
- -lmass
- EXTRA_LIBS
- Link the mathematical acceleration subsystem (MASS) libraries, which contain libraries of tuned mathematical intrinsic functions.

C++ benchmarks

- -ffast-math
- EXTRA_LDFLAGS, OPTIMIZE
- Sets the following flags:
  - -fno-math-errno
  - -funsafe-math-optimizations
  - -fno-trapping-math
  - -ffinite-math-only
  - -fno-signaling-nans
- Includes:
- -O3
- EXTRA_LDFLAGS, OPTIMIZE
- Optimize yet more. -O3 turns on all optimizations specified by -O2 and also turns on the -finline-functions, -funswitch-loops and -fgcse-after-reload options.
- Includes:
- -mcpu=power7
- EXTRA_LDFLAGS, OPTIMIZE
- Tune to cpu-type everything applicable about the generated code, except for the ABI and the set of available instructions.
  
  A sample list of supported values for this flag are
  - power
  - power2
  - power3
  - power4
  - power5
  - power5+
  - power6
  - power7
- -mtune=power7
- EXTRA_LDFLAGS, OPTIMIZE
- Sets the instruction scheduling parameters for a particular machine type, but does not set the architecture type, register usage, or choice of mnemonics, as -mcpu=cpu_type would. The same values for cpu_type are used for -mtune as for -mcpu. If both are specified, the code generated will use the architecture, registers, and mnemonics set by -mcpu, but the scheduling parameters set by -mtune.
- -mrecip=rsqrt
- EXTRA_LDFLAGS, OPTIMIZE
- This option will enable GCC to use RCPSS and RSQRTSS instructions (and their vectorized variants RCPPS and RSQRTPS) with an additional Newton-Raphson step to increase precision instead of DIVSS and SQRTSS (and their vectorized variants) for single precision floating point arguments. These instructions are generated only when -funsafe-math-optimizations is enabled together with -finite-math-only and -fno-trapping-math.
- -fpeel-loops
- EXTRA_LDFLAGS, OPTIMIZE
- Peels the loops for that there is enough information that they do not roll much (from profile feedback). It also turns on complete loop peeling (i.e. complete removal of loops with small constant number of iterations). Enabled with -fprofile-use
- -funroll-loops
- EXTRA_LDFLAGS, OPTIMIZE
- Unroll loops whose number of iterations can be determined at compile time or upon entry to the loop. -funroll-loops implies both -fstrength-reduce and -frerun-cse-after-loop. This option makes code larger, and may or may not make it run faster.
- Includes:
  - -fstrength-reduce
  - -frerun-cse-after-loop
- -mpopcntd
- EXTRA_LDFLAGS, OPTIMIZE
- Allows GCC to generate the popcount instruction implemented on the POWER7 processor and other processors that support the PowerPC V2.06 architecture.
- -m32
- EXTRA_LDFLAGS, OPTIMIZE
- Generate code for a 32-bit environment. The 32-bit environment sets int, long and pointer to 32 bits and generates code that runs on 32 bits system.
- -fvect-cost-model
- OPTIMIZE
- Enable cost model for vectorization.
- -mveclibabi=mass
- OPTIMIZE
- Specifies the Mathematical Acceleration Subsystem (MASS) ABI to use for vectorizing intrinsics using an external library.
- --param max-inline-insns-auto=200
- CXXOPTIMIZE
- When you use -finline-functions (included in -O3), a lot of functions that would otherwise not be considered for inlining by the compiler will be investigated. To those functions, a different (more restrictive) limit compared to functions declared inline can be applied. Set the value to 200 via --param.
- -fno-associative-math
- CXXOPTIMIZE
- Do not allow the reciprocal of a value to be used instead of dividing by the value if this enables optimizations.
- -flto
- CXXOPTIMIZE
- This option runs the standard link-time optimizer. When invoked with source code, it generates GIMPLE (one of GCC's internal representations) and writes it to special ELF sections in the object file. When the object files are linked together, all the function bodies are read from these ELF sections and instantiated as if they had been part of the same translation unit.
- -fwhole-program
- CXXOPTIMIZE
- Assume that the current compilation unit represents whole program being compiled. All public functions and variables with the exception of "main" and those merged by attribute "externally_visible" become static functions and in a affect gets more aggressively optimized by interprocedural optimizers.
- -fuse-linker-plugin
- CXXOPTIMIZE
- Enables the use of linker plugin during link time optimization. This option relies on the linker plugin support in linker that is available in gold or in GNU ld 2.21 or newer. This option enables the extraction of object files with GIMPLE bytecode out of library archives. This improves the quality of optimization by exposing more code the the link time optimizer. This information specify what symbols can be accessed externally (by non-LTO object or during dynamic linking). Resulting code quality improvements on binaries (and shared libraries that do use hidden visibility) is similar to -fwhole-program. See -flto for a description on the effect of this flag and how to use it. Enabled by default when LTO support in GCC is enabled and GCC was compiled with a linker supporting plugins (GNU ld 2.21 or newer or gold).
- -Wl,-q
- EXTRA_LDFLAGS
- Pass the -q flag to the linker causing the final executable to have the relocation information.
- -Wl,-Map=link.map,--cref
- EXTRA_LDFLAGS
- Pass "-Map=link.map --cref" to the linker causing to print a linker map to link.map and output a cross reference table to the map file.
- -L /opt/at6.0/lib
- EXTRA_LDFLAGS
- Add directory /opt/at6.0/lib to the list of directories to be searched for -l.
- -L /opt/ibmcmp/xlmass/7.1/lib
- EXTRA_LDFLAGS
- Add directory /opt/ibmcmp/xlmass/7.1/lib to the list of directories to be searched for -l.
- -Wl,-rpath,/opt/at6.0/lib
- EXTRA_LDFLAGS
- Pass "-rpath /opt/at6.0/lib" to the linker causing the Advance Toolchain library directory (/opt/at6.0/lib) to be added to the runtime library search path.
- -Wl,-rpath,/opt/ibmcmp/xlmass/7.1/lib
- EXTRA_LDFLAGS
- Pass "-rpath /opt/ibmcmp/xlmass/7.1/lib" to the linker causing the IBM XL Mathematical Acceleration Subsystem (MASS) library directory (/opt/ibmcmp/xlmass/7.1/lib) to be added to the runtime library search path.
- -lhugetlbfs
- EXTRA_LIBS
- Link with libhugetlbfs.so. This enables heap to be backed by the 16 Megabyte pages.
- -lmassvp7
- EXTRA_LIBS
- Link the mathematical acceleration subsystem (MASS) vector libraries, which contain libraries of tuned mathematical intrinsic functions.
- -lmass_simdp7
- EXTRA_LIBS
- Link the mathematical acceleration subsystem (MASS) SIMD libraries, which contain libraries of tuned mathematical intrinsic functions.
- -lmass
- EXTRA_LIBS
- Link the mathematical acceleration subsystem (MASS) libraries, which contain libraries of tuned mathematical intrinsic functions.

Fortran benchmarks

- -ffast-math
- EXTRA_LDFLAGS, OPTIMIZE
- Sets the following flags:
  - -fno-math-errno
  - -funsafe-math-optimizations
  - -fno-trapping-math
  - -ffinite-math-only
  - -fno-signaling-nans
- Includes:
- -O3
- EXTRA_LDFLAGS, OPTIMIZE
- Optimize yet more. -O3 turns on all optimizations specified by -O2 and also turns on the -finline-functions, -funswitch-loops and -fgcse-after-reload options.
- Includes:
- -mcpu=power7
- EXTRA_LDFLAGS, OPTIMIZE
- Tune to cpu-type everything applicable about the generated code, except for the ABI and the set of available instructions.
  
  A sample list of supported values for this flag are
  - power
  - power2
  - power3
  - power4
  - power5
  - power5+
  - power6
  - power7
- -mtune=power7
- EXTRA_LDFLAGS, OPTIMIZE
- Sets the instruction scheduling parameters for a particular machine type, but does not set the architecture type, register usage, or choice of mnemonics, as -mcpu=cpu_type would. The same values for cpu_type are used for -mtune as for -mcpu. If both are specified, the code generated will use the architecture, registers, and mnemonics set by -mcpu, but the scheduling parameters set by -mtune.
- -mrecip=rsqrt
- EXTRA_LDFLAGS, OPTIMIZE
- This option will enable GCC to use RCPSS and RSQRTSS instructions (and their vectorized variants RCPPS and RSQRTPS) with an additional Newton-Raphson step to increase precision instead of DIVSS and SQRTSS (and their vectorized variants) for single precision floating point arguments. These instructions are generated only when -funsafe-math-optimizations is enabled together with -finite-math-only and -fno-trapping-math.
- -fpeel-loops
- EXTRA_LDFLAGS, OPTIMIZE
- Peels the loops for that there is enough information that they do not roll much (from profile feedback). It also turns on complete loop peeling (i.e. complete removal of loops with small constant number of iterations). Enabled with -fprofile-use
- -funroll-loops
- EXTRA_LDFLAGS, OPTIMIZE
- Unroll loops whose number of iterations can be determined at compile time or upon entry to the loop. -funroll-loops implies both -fstrength-reduce and -frerun-cse-after-loop. This option makes code larger, and may or may not make it run faster.
- Includes:
  - -fstrength-reduce
  - -frerun-cse-after-loop
- -mpopcntd
- EXTRA_LDFLAGS, OPTIMIZE
- Allows GCC to generate the popcount instruction implemented on the POWER7 processor and other processors that support the PowerPC V2.06 architecture.
- -m32
- EXTRA_LDFLAGS, OPTIMIZE
- Generate code for a 32-bit environment. The 32-bit environment sets int, long and pointer to 32 bits and generates code that runs on 32 bits system.
- -fvect-cost-model
- OPTIMIZE
- Enable cost model for vectorization.
- -mveclibabi=mass
- OPTIMIZE
- Specifies the Mathematical Acceleration Subsystem (MASS) ABI to use for vectorizing intrinsics using an external library.
- -Wl,-q
- EXTRA_LDFLAGS
- Pass the -q flag to the linker causing the final executable to have the relocation information.
- -Wl,-Map=link.map,--cref
- EXTRA_LDFLAGS
- Pass "-Map=link.map --cref" to the linker causing to print a linker map to link.map and output a cross reference table to the map file.
- -L /opt/at6.0/lib
- EXTRA_LDFLAGS
- Add directory /opt/at6.0/lib to the list of directories to be searched for -l.
- -L /opt/ibmcmp/xlmass/7.1/lib
- EXTRA_LDFLAGS
- Add directory /opt/ibmcmp/xlmass/7.1/lib to the list of directories to be searched for -l.
- -Wl,-rpath,/opt/at6.0/lib
- EXTRA_LDFLAGS
- Pass "-rpath /opt/at6.0/lib" to the linker causing the Advance Toolchain library directory (/opt/at6.0/lib) to be added to the runtime library search path.
- -Wl,-rpath,/opt/ibmcmp/xlmass/7.1/lib
- EXTRA_LDFLAGS
- Pass "-rpath /opt/ibmcmp/xlmass/7.1/lib" to the linker causing the IBM XL Mathematical Acceleration Subsystem (MASS) library directory (/opt/ibmcmp/xlmass/7.1/lib) to be added to the runtime library search path.
- -lhugetlbfs
- EXTRA_LIBS
- Link with libhugetlbfs.so. This enables heap to be backed by the 16 Megabyte pages.
- -lmassvp7
- EXTRA_LIBS
- Link the mathematical acceleration subsystem (MASS) vector libraries, which contain libraries of tuned mathematical intrinsic functions.
- -lmass_simdp7
- EXTRA_LIBS
- Link the mathematical acceleration subsystem (MASS) SIMD libraries, which contain libraries of tuned mathematical intrinsic functions.
- -lmass
- EXTRA_LIBS
- Link the mathematical acceleration subsystem (MASS) libraries, which contain libraries of tuned mathematical intrinsic functions.

Benchmarks using both Fortran and C

- -ffast-math
- EXTRA_LDFLAGS, OPTIMIZE
- Sets the following flags:
  - -fno-math-errno
  - -funsafe-math-optimizations
  - -fno-trapping-math
  - -ffinite-math-only
  - -fno-signaling-nans
- Includes:
- -O3
- EXTRA_LDFLAGS, OPTIMIZE
- Optimize yet more. -O3 turns on all optimizations specified by -O2 and also turns on the -finline-functions, -funswitch-loops and -fgcse-after-reload options.
- Includes:
- -mcpu=power7
- EXTRA_LDFLAGS, OPTIMIZE
- Tune to cpu-type everything applicable about the generated code, except for the ABI and the set of available instructions.
  
  A sample list of supported values for this flag are
  - power
  - power2
  - power3
  - power4
  - power5
  - power5+
  - power6
  - power7
- -mtune=power7
- EXTRA_LDFLAGS, OPTIMIZE
- Sets the instruction scheduling parameters for a particular machine type, but does not set the architecture type, register usage, or choice of mnemonics, as -mcpu=cpu_type would. The same values for cpu_type are used for -mtune as for -mcpu. If both are specified, the code generated will use the architecture, registers, and mnemonics set by -mcpu, but the scheduling parameters set by -mtune.
- -mrecip=rsqrt
- EXTRA_LDFLAGS, OPTIMIZE
- This option will enable GCC to use RCPSS and RSQRTSS instructions (and their vectorized variants RCPPS and RSQRTPS) with an additional Newton-Raphson step to increase precision instead of DIVSS and SQRTSS (and their vectorized variants) for single precision floating point arguments. These instructions are generated only when -funsafe-math-optimizations is enabled together with -finite-math-only and -fno-trapping-math.
- -fpeel-loops
- EXTRA_LDFLAGS, OPTIMIZE
- Peels the loops for that there is enough information that they do not roll much (from profile feedback). It also turns on complete loop peeling (i.e. complete removal of loops with small constant number of iterations). Enabled with -fprofile-use
- -funroll-loops
- EXTRA_LDFLAGS, OPTIMIZE
- Unroll loops whose number of iterations can be determined at compile time or upon entry to the loop. -funroll-loops implies both -fstrength-reduce and -frerun-cse-after-loop. This option makes code larger, and may or may not make it run faster.
- Includes:
  - -fstrength-reduce
  - -frerun-cse-after-loop
- -mpopcntd
- EXTRA_LDFLAGS, OPTIMIZE
- Allows GCC to generate the popcount instruction implemented on the POWER7 processor and other processors that support the PowerPC V2.06 architecture.
- -m32
- EXTRA_LDFLAGS, OPTIMIZE
- Generate code for a 32-bit environment. The 32-bit environment sets int, long and pointer to 32 bits and generates code that runs on 32 bits system.
- -fvect-cost-model
- OPTIMIZE
- Enable cost model for vectorization.
- -mveclibabi=mass
- OPTIMIZE
- Specifies the Mathematical Acceleration Subsystem (MASS) ABI to use for vectorizing intrinsics using an external library.
- -Wl,-q
- EXTRA_LDFLAGS
- Pass the -q flag to the linker causing the final executable to have the relocation information.
- -Wl,-Map=link.map,--cref
- EXTRA_LDFLAGS
- Pass "-Map=link.map --cref" to the linker causing to print a linker map to link.map and output a cross reference table to the map file.
- -L /opt/at6.0/lib
- EXTRA_LDFLAGS
- Add directory /opt/at6.0/lib to the list of directories to be searched for -l.
- -L /opt/ibmcmp/xlmass/7.1/lib
- EXTRA_LDFLAGS
- Add directory /opt/ibmcmp/xlmass/7.1/lib to the list of directories to be searched for -l.
- -Wl,-rpath,/opt/at6.0/lib
- EXTRA_LDFLAGS
- Pass "-rpath /opt/at6.0/lib" to the linker causing the Advance Toolchain library directory (/opt/at6.0/lib) to be added to the runtime library search path.
- -Wl,-rpath,/opt/ibmcmp/xlmass/7.1/lib
- EXTRA_LDFLAGS
- Pass "-rpath /opt/ibmcmp/xlmass/7.1/lib" to the linker causing the IBM XL Mathematical Acceleration Subsystem (MASS) library directory (/opt/ibmcmp/xlmass/7.1/lib) to be added to the runtime library search path.
- -lhugetlbfs
- EXTRA_LIBS
- Link with libhugetlbfs.so. This enables heap to be backed by the 16 Megabyte pages.
- -lmassvp7
- EXTRA_LIBS
- Link the mathematical acceleration subsystem (MASS) vector libraries, which contain libraries of tuned mathematical intrinsic functions.
- -lmass_simdp7
- EXTRA_LIBS
- Link the mathematical acceleration subsystem (MASS) SIMD libraries, which contain libraries of tuned mathematical intrinsic functions.
- -lmass
- EXTRA_LIBS
- Link the mathematical acceleration subsystem (MASS) libraries, which contain libraries of tuned mathematical intrinsic functions.

Peak Optimization Flags

C benchmarks

433.milc

- -fprofile-generate
- PASS1_CFLAGS, PASS1_LDFLAGS
- Enable options usually used for instrumenting application to produce profile useful for later recompilation with profile feedback based optimization.
- -fprofile-use
- PASS2_CFLAGS, PASS2_LDFLAGS
- Enable profile feedback directed optimizations, and optimizations generally profitable only with profile feedback available.
- -ffast-math
- EXTRA_LDFLAGS, OPTIMIZE
- Sets the following flags:
  - -fno-math-errno
  - -funsafe-math-optimizations
  - -fno-trapping-math
  - -ffinite-math-only
  - -fno-signaling-nans
- Includes:
- -O3
- EXTRA_LDFLAGS, OPTIMIZE
- Optimize yet more. -O3 turns on all optimizations specified by -O2 and also turns on the -finline-functions, -funswitch-loops and -fgcse-after-reload options.
- Includes:
- -mcpu=power7
- EXTRA_LDFLAGS, OPTIMIZE
- Tune to cpu-type everything applicable about the generated code, except for the ABI and the set of available instructions.
  
  A sample list of supported values for this flag are
  - power
  - power2
  - power3
  - power4
  - power5
  - power5+
  - power6
  - power7
- -mtune=power7
- EXTRA_LDFLAGS, OPTIMIZE
- Sets the instruction scheduling parameters for a particular machine type, but does not set the architecture type, register usage, or choice of mnemonics, as -mcpu=cpu_type would. The same values for cpu_type are used for -mtune as for -mcpu. If both are specified, the code generated will use the architecture, registers, and mnemonics set by -mcpu, but the scheduling parameters set by -mtune.
- -mrecip=rsqrt
- EXTRA_LDFLAGS, OPTIMIZE
- This option will enable GCC to use RCPSS and RSQRTSS instructions (and their vectorized variants RCPPS and RSQRTPS) with an additional Newton-Raphson step to increase precision instead of DIVSS and SQRTSS (and their vectorized variants) for single precision floating point arguments. These instructions are generated only when -funsafe-math-optimizations is enabled together with -finite-math-only and -fno-trapping-math.
- -fpeel-loops
- EXTRA_LDFLAGS, OPTIMIZE
- Peels the loops for that there is enough information that they do not roll much (from profile feedback). It also turns on complete loop peeling (i.e. complete removal of loops with small constant number of iterations). Enabled with -fprofile-use
- -funroll-loops
- EXTRA_LDFLAGS, OPTIMIZE
- Unroll loops whose number of iterations can be determined at compile time or upon entry to the loop. -funroll-loops implies both -fstrength-reduce and -frerun-cse-after-loop. This option makes code larger, and may or may not make it run faster.
- Includes:
  - -fstrength-reduce
  - -frerun-cse-after-loop
- -mpopcntd
- EXTRA_LDFLAGS, OPTIMIZE
- Allows GCC to generate the popcount instruction implemented on the POWER7 processor and other processors that support the PowerPC V2.06 architecture.
- -m32
- EXTRA_LDFLAGS, OPTIMIZE
- Generate code for a 32-bit environment. The 32-bit environment sets int, long and pointer to 32 bits and generates code that runs on 32 bits system.
- -fvect-cost-model
- OPTIMIZE
- Enable cost model for vectorization.
- -mveclibabi=mass
- OPTIMIZE
- Specifies the Mathematical Acceleration Subsystem (MASS) ABI to use for vectorizing intrinsics using an external library.
- -flto
- OPTIMIZE
- This option runs the standard link-time optimizer. When invoked with source code, it generates GIMPLE (one of GCC's internal representations) and writes it to special ELF sections in the object file. When the object files are linked together, all the function bodies are read from these ELF sections and instantiated as if they had been part of the same translation unit.
- -fwhole-program
- OPTIMIZE
- Assume that the current compilation unit represents whole program being compiled. All public functions and variables with the exception of "main" and those merged by attribute "externally_visible" become static functions and in a affect gets more aggressively optimized by interprocedural optimizers.
- -fuse-linker-plugin
- OPTIMIZE
- Enables the use of linker plugin during link time optimization. This option relies on the linker plugin support in linker that is available in gold or in GNU ld 2.21 or newer. This option enables the extraction of object files with GIMPLE bytecode out of library archives. This improves the quality of optimization by exposing more code the the link time optimizer. This information specify what symbols can be accessed externally (by non-LTO object or during dynamic linking). Resulting code quality improvements on binaries (and shared libraries that do use hidden visibility) is similar to -fwhole-program. See -flto for a description on the effect of this flag and how to use it. Enabled by default when LTO support in GCC is enabled and GCC was compiled with a linker supporting plugins (GNU ld 2.21 or newer or gold).
- -Wl,-q
- EXTRA_LDFLAGS
- Pass the -q flag to the linker causing the final executable to have the relocation information.
- -Wl,-Map=link.map,--cref
- EXTRA_LDFLAGS
- Pass "-Map=link.map --cref" to the linker causing to print a linker map to link.map and output a cross reference table to the map file.
- -L /opt/at6.0/lib
- EXTRA_LDFLAGS
- Add directory /opt/at6.0/lib to the list of directories to be searched for -l.
- -L /opt/ibmcmp/xlmass/7.1/lib
- EXTRA_LDFLAGS
- Add directory /opt/ibmcmp/xlmass/7.1/lib to the list of directories to be searched for -l.
- -Wl,-rpath,/opt/at6.0/lib
- EXTRA_LDFLAGS
- Pass "-rpath /opt/at6.0/lib" to the linker causing the Advance Toolchain library directory (/opt/at6.0/lib) to be added to the runtime library search path.
- -Wl,-rpath,/opt/ibmcmp/xlmass/7.1/lib
- EXTRA_LDFLAGS
- Pass "-rpath /opt/ibmcmp/xlmass/7.1/lib" to the linker causing the IBM XL Mathematical Acceleration Subsystem (MASS) library directory (/opt/ibmcmp/xlmass/7.1/lib) to be added to the runtime library search path.
- -lhugetlbfs
- EXTRA_LIBS
- Link with libhugetlbfs.so. This enables heap to be backed by the 16 Megabyte pages.
- -lmassvp7
- EXTRA_LIBS
- Link the mathematical acceleration subsystem (MASS) vector libraries, which contain libraries of tuned mathematical intrinsic functions.
- -lmass_simdp7
- EXTRA_LIBS
- Link the mathematical acceleration subsystem (MASS) SIMD libraries, which contain libraries of tuned mathematical intrinsic functions.
- -lmass
- EXTRA_LIBS
- Link the mathematical acceleration subsystem (MASS) libraries, which contain libraries of tuned mathematical intrinsic functions.

470.lbm

- -fprofile-generate
- PASS1_CFLAGS, PASS1_LDFLAGS
- Enable options usually used for instrumenting application to produce profile useful for later recompilation with profile feedback based optimization.
- -fprofile-use
- PASS2_CFLAGS, PASS2_LDFLAGS
- Enable profile feedback directed optimizations, and optimizations generally profitable only with profile feedback available.
- -ffast-math
- EXTRA_LDFLAGS, OPTIMIZE
- Sets the following flags:
  - -fno-math-errno
  - -funsafe-math-optimizations
  - -fno-trapping-math
  - -ffinite-math-only
  - -fno-signaling-nans
- Includes:
- -O3
- EXTRA_LDFLAGS, OPTIMIZE
- Optimize yet more. -O3 turns on all optimizations specified by -O2 and also turns on the -finline-functions, -funswitch-loops and -fgcse-after-reload options.
- Includes:
- -mcpu=power7
- EXTRA_LDFLAGS, OPTIMIZE
- Tune to cpu-type everything applicable about the generated code, except for the ABI and the set of available instructions.
  
  A sample list of supported values for this flag are
  - power
  - power2
  - power3
  - power4
  - power5
  - power5+
  - power6
  - power7
- -mtune=power7
- EXTRA_LDFLAGS, OPTIMIZE
- Sets the instruction scheduling parameters for a particular machine type, but does not set the architecture type, register usage, or choice of mnemonics, as -mcpu=cpu_type would. The same values for cpu_type are used for -mtune as for -mcpu. If both are specified, the code generated will use the architecture, registers, and mnemonics set by -mcpu, but the scheduling parameters set by -mtune.
- -mrecip=rsqrt
- EXTRA_LDFLAGS, OPTIMIZE
- This option will enable GCC to use RCPSS and RSQRTSS instructions (and their vectorized variants RCPPS and RSQRTPS) with an additional Newton-Raphson step to increase precision instead of DIVSS and SQRTSS (and their vectorized variants) for single precision floating point arguments. These instructions are generated only when -funsafe-math-optimizations is enabled together with -finite-math-only and -fno-trapping-math.
- -fpeel-loops
- EXTRA_LDFLAGS, OPTIMIZE
- Peels the loops for that there is enough information that they do not roll much (from profile feedback). It also turns on complete loop peeling (i.e. complete removal of loops with small constant number of iterations). Enabled with -fprofile-use
- -funroll-loops
- EXTRA_LDFLAGS, OPTIMIZE
- Unroll loops whose number of iterations can be determined at compile time or upon entry to the loop. -funroll-loops implies both -fstrength-reduce and -frerun-cse-after-loop. This option makes code larger, and may or may not make it run faster.
- Includes:
  - -fstrength-reduce
  - -frerun-cse-after-loop
- -mpopcntd
- EXTRA_LDFLAGS, OPTIMIZE
- Allows GCC to generate the popcount instruction implemented on the POWER7 processor and other processors that support the PowerPC V2.06 architecture.
- -m64
- EXTRA_LDFLAGS, OPTIMIZE
- Generate code for a 64-bit environment. The 64-bit environment sets int, long and pointer to 64 bits and generates code that runs on 64 bits system.
- -mcmodel=medium
- EXTRA_LDFLAGS, OPTIMIZE
- Generate code for the medium model: The program is linked in the lower 2 GB of the address space. Small symbols are also placed there. Symbols with sizes larger than -mlarge-data-threshold are put into large data or bss sections and can be located above 2GB. Programs can be statically or dynamically linked.
- -fvect-cost-model
- OPTIMIZE
- Enable cost model for vectorization.
- -mveclibabi=mass
- OPTIMIZE
- Specifies the Mathematical Acceleration Subsystem (MASS) ABI to use for vectorizing intrinsics using an external library.
- -flto
- OPTIMIZE
- This option runs the standard link-time optimizer. When invoked with source code, it generates GIMPLE (one of GCC's internal representations) and writes it to special ELF sections in the object file. When the object files are linked together, all the function bodies are read from these ELF sections and instantiated as if they had been part of the same translation unit.
- -fwhole-program
- OPTIMIZE
- Assume that the current compilation unit represents whole program being compiled. All public functions and variables with the exception of "main" and those merged by attribute "externally_visible" become static functions and in a affect gets more aggressively optimized by interprocedural optimizers.
- -fuse-linker-plugin
- OPTIMIZE
- Enables the use of linker plugin during link time optimization. This option relies on the linker plugin support in linker that is available in gold or in GNU ld 2.21 or newer. This option enables the extraction of object files with GIMPLE bytecode out of library archives. This improves the quality of optimization by exposing more code the the link time optimizer. This information specify what symbols can be accessed externally (by non-LTO object or during dynamic linking). Resulting code quality improvements on binaries (and shared libraries that do use hidden visibility) is similar to -fwhole-program. See -flto for a description on the effect of this flag and how to use it. Enabled by default when LTO support in GCC is enabled and GCC was compiled with a linker supporting plugins (GNU ld 2.21 or newer or gold).
- -Wl,-q
- EXTRA_LDFLAGS
- Pass the -q flag to the linker causing the final executable to have the relocation information.
- -Wl,-Map=link.map,--cref
- EXTRA_LDFLAGS
- Pass "-Map=link.map --cref" to the linker causing to print a linker map to link.map and output a cross reference table to the map file.
- -L /opt/at6.0/lib64
- EXTRA_LDFLAGS
- Add directory /opt/at6.0/lib64 to the list of directories to be searched for -l.
- -L /opt/ibmcmp/xlmass/7.1/lib64
- EXTRA_LDFLAGS
- Add directory /opt/ibmcmp/xlmass/7.1/lib64 to the list of directories to be searched for -l.
- -Wl,-rpath,/opt/at6.0/lib64
- EXTRA_LDFLAGS
- Pass "-rpath /opt/at6.0/lib64" to the linker causing the Advance Toolchain for 64-bit library directory (/opt/at6.0/lib64) to be added to the runtime library search path.
- -Wl,-rpath,/opt/ibmcmp/xlmass/7.1/lib64
- EXTRA_LDFLAGS
- Pass "-rpath /opt/ibmcmp/xlmass/7.1/lib64" to the linker causing the IBM XL Mathematical Acceleration Subsystem (MASS) for 64-bit library directory (/opt/ibmcmp/xlmass/7.1/lib64) to be added to the runtime library search path.
- -Wl,-dynamic-linker,/opt/at6.0/lib64/ld64.so.1
- EXTRA_LDFLAGS
- Pass the "-dynamic-linker=/opt/at6.0/lib64/ld64.so.1" flag to the linker causing to set the dynamic linker of Advance Toolchain 64-bit (/opt/at6.0/lib64/ld64.so.1).
- -lhugetlbfs
- EXTRA_LIBS
- Link with libhugetlbfs.so. This enables heap to be backed by the 16 Megabyte pages.
- -lmassvp7_64
- EXTRA_LIBS
- Link the mathematical acceleration subsystem (MASS) vector libraries for 64-bit, which contain libraries of tuned mathematical intrinsic functions.
- -lmass_simdp7_64
- EXTRA_LIBS
- Link the mathematical acceleration subsystem (MASS) SIMD libraries for 64-bit, which contain libraries of tuned mathematical intrinsic functions.
- -lmass_64
- EXTRA_LIBS
- Link the mathematical acceleration subsystem (MASS) libraries for 64-bit, which contain libraries of tuned mathematical intrinsic functions.

482.sphinx3

- -fprofile-generate
- PASS1_CFLAGS, PASS1_LDFLAGS
- Enable options usually used for instrumenting application to produce profile useful for later recompilation with profile feedback based optimization.
- -fprofile-use
- PASS2_CFLAGS, PASS2_LDFLAGS
- Enable profile feedback directed optimizations, and optimizations generally profitable only with profile feedback available.
- -ffast-math
- EXTRA_LDFLAGS, OPTIMIZE
- Sets the following flags:
  - -fno-math-errno
  - -funsafe-math-optimizations
  - -fno-trapping-math
  - -ffinite-math-only
  - -fno-signaling-nans
- Includes:
- -O3
- EXTRA_LDFLAGS, OPTIMIZE
- Optimize yet more. -O3 turns on all optimizations specified by -O2 and also turns on the -finline-functions, -funswitch-loops and -fgcse-after-reload options.
- Includes:
- -mcpu=power7
- EXTRA_LDFLAGS, OPTIMIZE
- Tune to cpu-type everything applicable about the generated code, except for the ABI and the set of available instructions.
  
  A sample list of supported values for this flag are
  - power
  - power2
  - power3
  - power4
  - power5
  - power5+
  - power6
  - power7
- -mtune=power7
- EXTRA_LDFLAGS, OPTIMIZE
- Sets the instruction scheduling parameters for a particular machine type, but does not set the architecture type, register usage, or choice of mnemonics, as -mcpu=cpu_type would. The same values for cpu_type are used for -mtune as for -mcpu. If both are specified, the code generated will use the architecture, registers, and mnemonics set by -mcpu, but the scheduling parameters set by -mtune.
- -mrecip=rsqrt
- EXTRA_LDFLAGS, OPTIMIZE
- This option will enable GCC to use RCPSS and RSQRTSS instructions (and their vectorized variants RCPPS and RSQRTPS) with an additional Newton-Raphson step to increase precision instead of DIVSS and SQRTSS (and their vectorized variants) for single precision floating point arguments. These instructions are generated only when -funsafe-math-optimizations is enabled together with -finite-math-only and -fno-trapping-math.
- -fpeel-loops
- EXTRA_LDFLAGS, OPTIMIZE
- Peels the loops for that there is enough information that they do not roll much (from profile feedback). It also turns on complete loop peeling (i.e. complete removal of loops with small constant number of iterations). Enabled with -fprofile-use
- -funroll-loops
- EXTRA_LDFLAGS, OPTIMIZE
- Unroll loops whose number of iterations can be determined at compile time or upon entry to the loop. -funroll-loops implies both -fstrength-reduce and -frerun-cse-after-loop. This option makes code larger, and may or may not make it run faster.
- Includes:
  - -fstrength-reduce
  - -frerun-cse-after-loop
- -mpopcntd
- EXTRA_LDFLAGS, OPTIMIZE
- Allows GCC to generate the popcount instruction implemented on the POWER7 processor and other processors that support the PowerPC V2.06 architecture.
- -m32
- EXTRA_LDFLAGS, OPTIMIZE
- Generate code for a 32-bit environment. The 32-bit environment sets int, long and pointer to 32 bits and generates code that runs on 32 bits system.
- -fvect-cost-model
- OPTIMIZE
- Enable cost model for vectorization.
- -mveclibabi=mass
- OPTIMIZE
- Specifies the Mathematical Acceleration Subsystem (MASS) ABI to use for vectorizing intrinsics using an external library.
- -Wl,-q
- EXTRA_LDFLAGS
- Pass the -q flag to the linker causing the final executable to have the relocation information.
- -Wl,-Map=link.map,--cref
- EXTRA_LDFLAGS
- Pass "-Map=link.map --cref" to the linker causing to print a linker map to link.map and output a cross reference table to the map file.
- -L /opt/at6.0/lib
- EXTRA_LDFLAGS
- Add directory /opt/at6.0/lib to the list of directories to be searched for -l.
- -L /opt/ibmcmp/xlmass/7.1/lib
- EXTRA_LDFLAGS
- Add directory /opt/ibmcmp/xlmass/7.1/lib to the list of directories to be searched for -l.
- -Wl,-rpath,/opt/at6.0/lib
- EXTRA_LDFLAGS
- Pass "-rpath /opt/at6.0/lib" to the linker causing the Advance Toolchain library directory (/opt/at6.0/lib) to be added to the runtime library search path.
- -Wl,-rpath,/opt/ibmcmp/xlmass/7.1/lib
- EXTRA_LDFLAGS
- Pass "-rpath /opt/ibmcmp/xlmass/7.1/lib" to the linker causing the IBM XL Mathematical Acceleration Subsystem (MASS) library directory (/opt/ibmcmp/xlmass/7.1/lib) to be added to the runtime library search path.
- -lhugetlbfs
- EXTRA_LIBS
- Link with libhugetlbfs.so. This enables heap to be backed by the 16 Megabyte pages.
- -lmassvp7
- EXTRA_LIBS
- Link the mathematical acceleration subsystem (MASS) vector libraries, which contain libraries of tuned mathematical intrinsic functions.
- -lmass_simdp7
- EXTRA_LIBS
- Link the mathematical acceleration subsystem (MASS) SIMD libraries, which contain libraries of tuned mathematical intrinsic functions.
- -lmass
- EXTRA_LIBS
- Link the mathematical acceleration subsystem (MASS) libraries, which contain libraries of tuned mathematical intrinsic functions.

C++ benchmarks

447.dealII

- -ffast-math
- EXTRA_LDFLAGS, OPTIMIZE
- Sets the following flags:
  - -fno-math-errno
  - -funsafe-math-optimizations
  - -fno-trapping-math
  - -ffinite-math-only
  - -fno-signaling-nans
- Includes:
- -O3
- EXTRA_LDFLAGS, OPTIMIZE
- Optimize yet more. -O3 turns on all optimizations specified by -O2 and also turns on the -finline-functions, -funswitch-loops and -fgcse-after-reload options.
- Includes:
- -mcpu=power7
- EXTRA_LDFLAGS, OPTIMIZE
- Tune to cpu-type everything applicable about the generated code, except for the ABI and the set of available instructions.
  
  A sample list of supported values for this flag are
  - power
  - power2
  - power3
  - power4
  - power5
  - power5+
  - power6
  - power7
- -mtune=power7
- EXTRA_LDFLAGS, OPTIMIZE
- Sets the instruction scheduling parameters for a particular machine type, but does not set the architecture type, register usage, or choice of mnemonics, as -mcpu=cpu_type would. The same values for cpu_type are used for -mtune as for -mcpu. If both are specified, the code generated will use the architecture, registers, and mnemonics set by -mcpu, but the scheduling parameters set by -mtune.
- -mrecip=rsqrt
- EXTRA_LDFLAGS, OPTIMIZE
- This option will enable GCC to use RCPSS and RSQRTSS instructions (and their vectorized variants RCPPS and RSQRTPS) with an additional Newton-Raphson step to increase precision instead of DIVSS and SQRTSS (and their vectorized variants) for single precision floating point arguments. These instructions are generated only when -funsafe-math-optimizations is enabled together with -finite-math-only and -fno-trapping-math.
- -fpeel-loops
- EXTRA_LDFLAGS, OPTIMIZE
- Peels the loops for that there is enough information that they do not roll much (from profile feedback). It also turns on complete loop peeling (i.e. complete removal of loops with small constant number of iterations). Enabled with -fprofile-use
- -funroll-loops
- EXTRA_LDFLAGS, OPTIMIZE
- Unroll loops whose number of iterations can be determined at compile time or upon entry to the loop. -funroll-loops implies both -fstrength-reduce and -frerun-cse-after-loop. This option makes code larger, and may or may not make it run faster.
- Includes:
  - -fstrength-reduce
  - -frerun-cse-after-loop
- -mpopcntd
- EXTRA_LDFLAGS, OPTIMIZE
- Allows GCC to generate the popcount instruction implemented on the POWER7 processor and other processors that support the PowerPC V2.06 architecture.
- -m32
- EXTRA_LDFLAGS, OPTIMIZE
- Generate code for a 32-bit environment. The 32-bit environment sets int, long and pointer to 32 bits and generates code that runs on 32 bits system.
- -fvect-cost-model
- OPTIMIZE
- Enable cost model for vectorization.
- -mveclibabi=mass
- OPTIMIZE
- Specifies the Mathematical Acceleration Subsystem (MASS) ABI to use for vectorizing intrinsics using an external library.
- -flto
- OPTIMIZE
- This option runs the standard link-time optimizer. When invoked with source code, it generates GIMPLE (one of GCC's internal representations) and writes it to special ELF sections in the object file. When the object files are linked together, all the function bodies are read from these ELF sections and instantiated as if they had been part of the same translation unit.
- -fwhole-program
- OPTIMIZE
- Assume that the current compilation unit represents whole program being compiled. All public functions and variables with the exception of "main" and those merged by attribute "externally_visible" become static functions and in a affect gets more aggressively optimized by interprocedural optimizers.
- -fuse-linker-plugin
- OPTIMIZE
- Enables the use of linker plugin during link time optimization. This option relies on the linker plugin support in linker that is available in gold or in GNU ld 2.21 or newer. This option enables the extraction of object files with GIMPLE bytecode out of library archives. This improves the quality of optimization by exposing more code the the link time optimizer. This information specify what symbols can be accessed externally (by non-LTO object or during dynamic linking). Resulting code quality improvements on binaries (and shared libraries that do use hidden visibility) is similar to -fwhole-program. See -flto for a description on the effect of this flag and how to use it. Enabled by default when LTO support in GCC is enabled and GCC was compiled with a linker supporting plugins (GNU ld 2.21 or newer or gold).
- --param max-inline-insns-auto=200
- CXXOPTIMIZE
- When you use -finline-functions (included in -O3), a lot of functions that would otherwise not be considered for inlining by the compiler will be investigated. To those functions, a different (more restrictive) limit compared to functions declared inline can be applied. Set the value to 200 via --param.
- -fno-associative-math
- CXXOPTIMIZE
- Do not allow the reciprocal of a value to be used instead of dividing by the value if this enables optimizations.
- -Wl,-q
- EXTRA_LDFLAGS
- Pass the -q flag to the linker causing the final executable to have the relocation information.
- -Wl,-Map=link.map,--cref
- EXTRA_LDFLAGS
- Pass "-Map=link.map --cref" to the linker causing to print a linker map to link.map and output a cross reference table to the map file.
- -L /opt/at6.0/lib
- EXTRA_LDFLAGS
- Add directory /opt/at6.0/lib to the list of directories to be searched for -l.
- -L /opt/ibmcmp/xlmass/7.1/lib
- EXTRA_LDFLAGS
- Add directory /opt/ibmcmp/xlmass/7.1/lib to the list of directories to be searched for -l.
- -Wl,-rpath,/opt/at6.0/lib
- EXTRA_LDFLAGS
- Pass "-rpath /opt/at6.0/lib" to the linker causing the Advance Toolchain library directory (/opt/at6.0/lib) to be added to the runtime library search path.
- -Wl,-rpath,/opt/ibmcmp/xlmass/7.1/lib
- EXTRA_LDFLAGS
- Pass "-rpath /opt/ibmcmp/xlmass/7.1/lib" to the linker causing the IBM XL Mathematical Acceleration Subsystem (MASS) library directory (/opt/ibmcmp/xlmass/7.1/lib) to be added to the runtime library search path.
- -lhugetlbfs
- EXTRA_LIBS
- Link with libhugetlbfs.so. This enables heap to be backed by the 16 Megabyte pages.
- -lmassvp7
- EXTRA_LIBS
- Link the mathematical acceleration subsystem (MASS) vector libraries, which contain libraries of tuned mathematical intrinsic functions.
- -lmass_simdp7
- EXTRA_LIBS
- Link the mathematical acceleration subsystem (MASS) SIMD libraries, which contain libraries of tuned mathematical intrinsic functions.
- -lmass
- EXTRA_LIBS
- Link the mathematical acceleration subsystem (MASS) libraries, which contain libraries of tuned mathematical intrinsic functions.
- -ltcmalloc
- EXTRA_LIBS
- Link with tcmalloc's library for Linux on POWER. This is a library that optimizes calls to new, delete, malloc and free.
- -lstdc++ -lpthread
- EXTRA_LIBS
- Link the C++ standard library.

453.povray

- -fprofile-generate
- PASS1_CXXFLAGS, PASS1_LDFLAGS
- Enable options usually used for instrumenting application to produce profile useful for later recompilation with profile feedback based optimization.
- -fprofile-use
- PASS2_CXXFLAGS, PASS2_LDFLAGS
- Enable profile feedback directed optimizations, and optimizations generally profitable only with profile feedback available.
- -ffast-math
- EXTRA_LDFLAGS, OPTIMIZE
- Sets the following flags:
  - -fno-math-errno
  - -funsafe-math-optimizations
  - -fno-trapping-math
  - -ffinite-math-only
  - -fno-signaling-nans
- Includes:
- -O3
- EXTRA_LDFLAGS, OPTIMIZE
- Optimize yet more. -O3 turns on all optimizations specified by -O2 and also turns on the -finline-functions, -funswitch-loops and -fgcse-after-reload options.
- Includes:
- -mcpu=power7
- EXTRA_LDFLAGS, OPTIMIZE
- Tune to cpu-type everything applicable about the generated code, except for the ABI and the set of available instructions.
  
  A sample list of supported values for this flag are
  - power
  - power2
  - power3
  - power4
  - power5
  - power5+
  - power6
  - power7
- -mtune=power7
- EXTRA_LDFLAGS, OPTIMIZE
- Sets the instruction scheduling parameters for a particular machine type, but does not set the architecture type, register usage, or choice of mnemonics, as -mcpu=cpu_type would. The same values for cpu_type are used for -mtune as for -mcpu. If both are specified, the code generated will use the architecture, registers, and mnemonics set by -mcpu, but the scheduling parameters set by -mtune.
- -mrecip=rsqrt
- EXTRA_LDFLAGS, OPTIMIZE
- This option will enable GCC to use RCPSS and RSQRTSS instructions (and their vectorized variants RCPPS and RSQRTPS) with an additional Newton-Raphson step to increase precision instead of DIVSS and SQRTSS (and their vectorized variants) for single precision floating point arguments. These instructions are generated only when -funsafe-math-optimizations is enabled together with -finite-math-only and -fno-trapping-math.
- -fpeel-loops
- EXTRA_LDFLAGS, OPTIMIZE
- Peels the loops for that there is enough information that they do not roll much (from profile feedback). It also turns on complete loop peeling (i.e. complete removal of loops with small constant number of iterations). Enabled with -fprofile-use
- -funroll-loops
- EXTRA_LDFLAGS, OPTIMIZE
- Unroll loops whose number of iterations can be determined at compile time or upon entry to the loop. -funroll-loops implies both -fstrength-reduce and -frerun-cse-after-loop. This option makes code larger, and may or may not make it run faster.
- Includes:
  - -fstrength-reduce
  - -frerun-cse-after-loop
- -mpopcntd
- EXTRA_LDFLAGS, OPTIMIZE
- Allows GCC to generate the popcount instruction implemented on the POWER7 processor and other processors that support the PowerPC V2.06 architecture.
- -m32
- EXTRA_LDFLAGS, OPTIMIZE
- Generate code for a 32-bit environment. The 32-bit environment sets int, long and pointer to 32 bits and generates code that runs on 32 bits system.
- -fvect-cost-model
- OPTIMIZE
- Enable cost model for vectorization.
- -mveclibabi=mass
- OPTIMIZE
- Specifies the Mathematical Acceleration Subsystem (MASS) ABI to use for vectorizing intrinsics using an external library.
- -flto
- OPTIMIZE
- This option runs the standard link-time optimizer. When invoked with source code, it generates GIMPLE (one of GCC's internal representations) and writes it to special ELF sections in the object file. When the object files are linked together, all the function bodies are read from these ELF sections and instantiated as if they had been part of the same translation unit.
- -fwhole-program
- OPTIMIZE
- Assume that the current compilation unit represents whole program being compiled. All public functions and variables with the exception of "main" and those merged by attribute "externally_visible" become static functions and in a affect gets more aggressively optimized by interprocedural optimizers.
- -fuse-linker-plugin
- OPTIMIZE
- Enables the use of linker plugin during link time optimization. This option relies on the linker plugin support in linker that is available in gold or in GNU ld 2.21 or newer. This option enables the extraction of object files with GIMPLE bytecode out of library archives. This improves the quality of optimization by exposing more code the the link time optimizer. This information specify what symbols can be accessed externally (by non-LTO object or during dynamic linking). Resulting code quality improvements on binaries (and shared libraries that do use hidden visibility) is similar to -fwhole-program. See -flto for a description on the effect of this flag and how to use it. Enabled by default when LTO support in GCC is enabled and GCC was compiled with a linker supporting plugins (GNU ld 2.21 or newer or gold).
- --param max-inline-insns-auto=200
- CXXOPTIMIZE
- When you use -finline-functions (included in -O3), a lot of functions that would otherwise not be considered for inlining by the compiler will be investigated. To those functions, a different (more restrictive) limit compared to functions declared inline can be applied. Set the value to 200 via --param.
- -fno-associative-math
- CXXOPTIMIZE
- Do not allow the reciprocal of a value to be used instead of dividing by the value if this enables optimizations.
- -Wl,-q
- EXTRA_LDFLAGS
- Pass the -q flag to the linker causing the final executable to have the relocation information.
- -Wl,-Map=link.map,--cref
- EXTRA_LDFLAGS
- Pass "-Map=link.map --cref" to the linker causing to print a linker map to link.map and output a cross reference table to the map file.
- -L /opt/at6.0/lib
- EXTRA_LDFLAGS
- Add directory /opt/at6.0/lib to the list of directories to be searched for -l.
- -L /opt/ibmcmp/xlmass/7.1/lib
- EXTRA_LDFLAGS
- Add directory /opt/ibmcmp/xlmass/7.1/lib to the list of directories to be searched for -l.
- -Wl,-rpath,/opt/at6.0/lib
- EXTRA_LDFLAGS
- Pass "-rpath /opt/at6.0/lib" to the linker causing the Advance Toolchain library directory (/opt/at6.0/lib) to be added to the runtime library search path.
- -Wl,-rpath,/opt/ibmcmp/xlmass/7.1/lib
- EXTRA_LDFLAGS
- Pass "-rpath /opt/ibmcmp/xlmass/7.1/lib" to the linker causing the IBM XL Mathematical Acceleration Subsystem (MASS) library directory (/opt/ibmcmp/xlmass/7.1/lib) to be added to the runtime library search path.
- -lhugetlbfs
- EXTRA_LIBS
- Link with libhugetlbfs.so. This enables heap to be backed by the 16 Megabyte pages.
- -lmassvp7
- EXTRA_LIBS
- Link the mathematical acceleration subsystem (MASS) vector libraries, which contain libraries of tuned mathematical intrinsic functions.
- -lmass_simdp7
- EXTRA_LIBS
- Link the mathematical acceleration subsystem (MASS) SIMD libraries, which contain libraries of tuned mathematical intrinsic functions.
- -lmass
- EXTRA_LIBS
- Link the mathematical acceleration subsystem (MASS) libraries, which contain libraries of tuned mathematical intrinsic functions.
- -ltcmalloc
- EXTRA_LIBS
- Link with tcmalloc's library for Linux on POWER. This is a library that optimizes calls to new, delete, malloc and free.
- -lstdc++ -lpthread
- EXTRA_LIBS
- Link the C++ standard library.

Fortran benchmarks

410.bwaves

- -fprofile-generate
- PASS1_FFLAGS, PASS1_LDFLAGS
- Enable options usually used for instrumenting application to produce profile useful for later recompilation with profile feedback based optimization.
- -fprofile-use
- PASS2_FFLAGS, PASS2_LDFLAGS
- Enable profile feedback directed optimizations, and optimizations generally profitable only with profile feedback available.
- -ffast-math
- EXTRA_LDFLAGS, OPTIMIZE
- Sets the following flags:
  - -fno-math-errno
  - -funsafe-math-optimizations
  - -fno-trapping-math
  - -ffinite-math-only
  - -fno-signaling-nans
- Includes:
- -O3
- EXTRA_LDFLAGS, OPTIMIZE
- Optimize yet more. -O3 turns on all optimizations specified by -O2 and also turns on the -finline-functions, -funswitch-loops and -fgcse-after-reload options.
- Includes:
- -mcpu=power7
- EXTRA_LDFLAGS, OPTIMIZE
- Tune to cpu-type everything applicable about the generated code, except for the ABI and the set of available instructions.
  
  A sample list of supported values for this flag are
  - power
  - power2
  - power3
  - power4
  - power5
  - power5+
  - power6
  - power7
- -mtune=power7
- EXTRA_LDFLAGS, OPTIMIZE
- Sets the instruction scheduling parameters for a particular machine type, but does not set the architecture type, register usage, or choice of mnemonics, as -mcpu=cpu_type would. The same values for cpu_type are used for -mtune as for -mcpu. If both are specified, the code generated will use the architecture, registers, and mnemonics set by -mcpu, but the scheduling parameters set by -mtune.
- -mrecip=rsqrt
- EXTRA_LDFLAGS, OPTIMIZE
- This option will enable GCC to use RCPSS and RSQRTSS instructions (and their vectorized variants RCPPS and RSQRTPS) with an additional Newton-Raphson step to increase precision instead of DIVSS and SQRTSS (and their vectorized variants) for single precision floating point arguments. These instructions are generated only when -funsafe-math-optimizations is enabled together with -finite-math-only and -fno-trapping-math.
- -fpeel-loops
- EXTRA_LDFLAGS, OPTIMIZE
- Peels the loops for that there is enough information that they do not roll much (from profile feedback). It also turns on complete loop peeling (i.e. complete removal of loops with small constant number of iterations). Enabled with -fprofile-use
- -funroll-loops
- EXTRA_LDFLAGS, OPTIMIZE
- Unroll loops whose number of iterations can be determined at compile time or upon entry to the loop. -funroll-loops implies both -fstrength-reduce and -frerun-cse-after-loop. This option makes code larger, and may or may not make it run faster.
- Includes:
  - -fstrength-reduce
  - -frerun-cse-after-loop
- -mpopcntd
- EXTRA_LDFLAGS, OPTIMIZE
- Allows GCC to generate the popcount instruction implemented on the POWER7 processor and other processors that support the PowerPC V2.06 architecture.
- -m64
- EXTRA_LDFLAGS, OPTIMIZE
- Generate code for a 64-bit environment. The 64-bit environment sets int, long and pointer to 64 bits and generates code that runs on 64 bits system.
- -mcmodel=medium
- EXTRA_LDFLAGS, OPTIMIZE
- Generate code for the medium model: The program is linked in the lower 2 GB of the address space. Small symbols are also placed there. Symbols with sizes larger than -mlarge-data-threshold are put into large data or bss sections and can be located above 2GB. Programs can be statically or dynamically linked.
- -fvect-cost-model
- OPTIMIZE
- Enable cost model for vectorization.
- -mveclibabi=mass
- OPTIMIZE
- Specifies the Mathematical Acceleration Subsystem (MASS) ABI to use for vectorizing intrinsics using an external library.
- -flto
- OPTIMIZE
- This option runs the standard link-time optimizer. When invoked with source code, it generates GIMPLE (one of GCC's internal representations) and writes it to special ELF sections in the object file. When the object files are linked together, all the function bodies are read from these ELF sections and instantiated as if they had been part of the same translation unit.
- -fwhole-program
- OPTIMIZE
- Assume that the current compilation unit represents whole program being compiled. All public functions and variables with the exception of "main" and those merged by attribute "externally_visible" become static functions and in a affect gets more aggressively optimized by interprocedural optimizers.
- -fuse-linker-plugin
- OPTIMIZE
- Enables the use of linker plugin during link time optimization. This option relies on the linker plugin support in linker that is available in gold or in GNU ld 2.21 or newer. This option enables the extraction of object files with GIMPLE bytecode out of library archives. This improves the quality of optimization by exposing more code the the link time optimizer. This information specify what symbols can be accessed externally (by non-LTO object or during dynamic linking). Resulting code quality improvements on binaries (and shared libraries that do use hidden visibility) is similar to -fwhole-program. See -flto for a description on the effect of this flag and how to use it. Enabled by default when LTO support in GCC is enabled and GCC was compiled with a linker supporting plugins (GNU ld 2.21 or newer or gold).
- -Wl,-q
- EXTRA_LDFLAGS
- Pass the -q flag to the linker causing the final executable to have the relocation information.
- -Wl,-Map=link.map,--cref
- EXTRA_LDFLAGS
- Pass "-Map=link.map --cref" to the linker causing to print a linker map to link.map and output a cross reference table to the map file.
- -L /opt/at6.0/lib64
- EXTRA_LDFLAGS
- Add directory /opt/at6.0/lib64 to the list of directories to be searched for -l.
- -L /opt/ibmcmp/xlmass/7.1/lib64
- EXTRA_LDFLAGS
- Add directory /opt/ibmcmp/xlmass/7.1/lib64 to the list of directories to be searched for -l.
- -Wl,-rpath,/opt/at6.0/lib64
- EXTRA_LDFLAGS
- Pass "-rpath /opt/at6.0/lib64" to the linker causing the Advance Toolchain for 64-bit library directory (/opt/at6.0/lib64) to be added to the runtime library search path.
- -Wl,-rpath,/opt/ibmcmp/xlmass/7.1/lib64
- EXTRA_LDFLAGS
- Pass "-rpath /opt/ibmcmp/xlmass/7.1/lib64" to the linker causing the IBM XL Mathematical Acceleration Subsystem (MASS) for 64-bit library directory (/opt/ibmcmp/xlmass/7.1/lib64) to be added to the runtime library search path.
- -Wl,-dynamic-linker,/opt/at6.0/lib64/ld64.so.1
- EXTRA_LDFLAGS
- Pass the "-dynamic-linker=/opt/at6.0/lib64/ld64.so.1" flag to the linker causing to set the dynamic linker of Advance Toolchain 64-bit (/opt/at6.0/lib64/ld64.so.1).
- -lhugetlbfs
- EXTRA_LIBS
- Link with libhugetlbfs.so. This enables heap to be backed by the 16 Megabyte pages.
- -lmassvp7_64
- EXTRA_LIBS
- Link the mathematical acceleration subsystem (MASS) vector libraries for 64-bit, which contain libraries of tuned mathematical intrinsic functions.
- -lmass_simdp7_64
- EXTRA_LIBS
- Link the mathematical acceleration subsystem (MASS) SIMD libraries for 64-bit, which contain libraries of tuned mathematical intrinsic functions.
- -lmass_64
- EXTRA_LIBS
- Link the mathematical acceleration subsystem (MASS) libraries for 64-bit, which contain libraries of tuned mathematical intrinsic functions.

416.gamess

- -fprofile-generate
- PASS1_FFLAGS, PASS1_LDFLAGS
- Enable options usually used for instrumenting application to produce profile useful for later recompilation with profile feedback based optimization.
- -fprofile-use
- PASS2_FFLAGS, PASS2_LDFLAGS
- Enable profile feedback directed optimizations, and optimizations generally profitable only with profile feedback available.
- -ffast-math
- EXTRA_LDFLAGS, OPTIMIZE
- Sets the following flags:
  - -fno-math-errno
  - -funsafe-math-optimizations
  - -fno-trapping-math
  - -ffinite-math-only
  - -fno-signaling-nans
- Includes:
- -O3
- EXTRA_LDFLAGS, OPTIMIZE
- Optimize yet more. -O3 turns on all optimizations specified by -O2 and also turns on the -finline-functions, -funswitch-loops and -fgcse-after-reload options.
- Includes:
- -mcpu=power7
- EXTRA_LDFLAGS, OPTIMIZE
- Tune to cpu-type everything applicable about the generated code, except for the ABI and the set of available instructions.
  
  A sample list of supported values for this flag are
  - power
  - power2
  - power3
  - power4
  - power5
  - power5+
  - power6
  - power7
- -mtune=power7
- EXTRA_LDFLAGS, OPTIMIZE
- Sets the instruction scheduling parameters for a particular machine type, but does not set the architecture type, register usage, or choice of mnemonics, as -mcpu=cpu_type would. The same values for cpu_type are used for -mtune as for -mcpu. If both are specified, the code generated will use the architecture, registers, and mnemonics set by -mcpu, but the scheduling parameters set by -mtune.
- -mrecip=rsqrt
- EXTRA_LDFLAGS, OPTIMIZE
- This option will enable GCC to use RCPSS and RSQRTSS instructions (and their vectorized variants RCPPS and RSQRTPS) with an additional Newton-Raphson step to increase precision instead of DIVSS and SQRTSS (and their vectorized variants) for single precision floating point arguments. These instructions are generated only when -funsafe-math-optimizations is enabled together with -finite-math-only and -fno-trapping-math.
- -fpeel-loops
- EXTRA_LDFLAGS, OPTIMIZE
- Peels the loops for that there is enough information that they do not roll much (from profile feedback). It also turns on complete loop peeling (i.e. complete removal of loops with small constant number of iterations). Enabled with -fprofile-use
- -funroll-loops
- EXTRA_LDFLAGS, OPTIMIZE
- Unroll loops whose number of iterations can be determined at compile time or upon entry to the loop. -funroll-loops implies both -fstrength-reduce and -frerun-cse-after-loop. This option makes code larger, and may or may not make it run faster.
- Includes:
  - -fstrength-reduce
  - -frerun-cse-after-loop
- -mpopcntd
- EXTRA_LDFLAGS, OPTIMIZE
- Allows GCC to generate the popcount instruction implemented on the POWER7 processor and other processors that support the PowerPC V2.06 architecture.
- -m32
- EXTRA_LDFLAGS, OPTIMIZE
- Generate code for a 32-bit environment. The 32-bit environment sets int, long and pointer to 32 bits and generates code that runs on 32 bits system.
- -fvect-cost-model
- OPTIMIZE
- Enable cost model for vectorization.
- -mveclibabi=mass
- OPTIMIZE
- Specifies the Mathematical Acceleration Subsystem (MASS) ABI to use for vectorizing intrinsics using an external library.
- -Wl,-q
- EXTRA_LDFLAGS
- Pass the -q flag to the linker causing the final executable to have the relocation information.
- -Wl,-Map=link.map,--cref
- EXTRA_LDFLAGS
- Pass "-Map=link.map --cref" to the linker causing to print a linker map to link.map and output a cross reference table to the map file.
- -L /opt/at6.0/lib
- EXTRA_LDFLAGS
- Add directory /opt/at6.0/lib to the list of directories to be searched for -l.
- -L /opt/ibmcmp/xlmass/7.1/lib
- EXTRA_LDFLAGS
- Add directory /opt/ibmcmp/xlmass/7.1/lib to the list of directories to be searched for -l.
- -Wl,-rpath,/opt/at6.0/lib
- EXTRA_LDFLAGS
- Pass "-rpath /opt/at6.0/lib" to the linker causing the Advance Toolchain library directory (/opt/at6.0/lib) to be added to the runtime library search path.
- -Wl,-rpath,/opt/ibmcmp/xlmass/7.1/lib
- EXTRA_LDFLAGS
- Pass "-rpath /opt/ibmcmp/xlmass/7.1/lib" to the linker causing the IBM XL Mathematical Acceleration Subsystem (MASS) library directory (/opt/ibmcmp/xlmass/7.1/lib) to be added to the runtime library search path.
- -lhugetlbfs
- EXTRA_LIBS
- Link with libhugetlbfs.so. This enables heap to be backed by the 16 Megabyte pages.
- -lmassvp7
- EXTRA_LIBS
- Link the mathematical acceleration subsystem (MASS) vector libraries, which contain libraries of tuned mathematical intrinsic functions.
- -lmass_simdp7
- EXTRA_LIBS
- Link the mathematical acceleration subsystem (MASS) SIMD libraries, which contain libraries of tuned mathematical intrinsic functions.
- -lmass
- EXTRA_LIBS
- Link the mathematical acceleration subsystem (MASS) libraries, which contain libraries of tuned mathematical intrinsic functions.
- -ltcmalloc
- EXTRA_LIBS
- Link with tcmalloc's library for Linux on POWER. This is a library that optimizes calls to new, delete, malloc and free.
- -lstdc++ -lpthread
- EXTRA_LIBS
- Link the C++ standard library.

434.zeusmp

- -fprofile-generate
- PASS1_FFLAGS, PASS1_LDFLAGS
- Enable options usually used for instrumenting application to produce profile useful for later recompilation with profile feedback based optimization.
- -fprofile-use
- PASS2_FFLAGS, PASS2_LDFLAGS
- Enable profile feedback directed optimizations, and optimizations generally profitable only with profile feedback available.
- -ffast-math
- EXTRA_LDFLAGS, OPTIMIZE
- Sets the following flags:
  - -fno-math-errno
  - -funsafe-math-optimizations
  - -fno-trapping-math
  - -ffinite-math-only
  - -fno-signaling-nans
- Includes:
- -O3
- EXTRA_LDFLAGS, OPTIMIZE
- Optimize yet more. -O3 turns on all optimizations specified by -O2 and also turns on the -finline-functions, -funswitch-loops and -fgcse-after-reload options.
- Includes:
- -mcpu=power7
- EXTRA_LDFLAGS, OPTIMIZE
- Tune to cpu-type everything applicable about the generated code, except for the ABI and the set of available instructions.
  
  A sample list of supported values for this flag are
  - power
  - power2
  - power3
  - power4
  - power5
  - power5+
  - power6
  - power7
- -mtune=power7
- EXTRA_LDFLAGS, OPTIMIZE
- Sets the instruction scheduling parameters for a particular machine type, but does not set the architecture type, register usage, or choice of mnemonics, as -mcpu=cpu_type would. The same values for cpu_type are used for -mtune as for -mcpu. If both are specified, the code generated will use the architecture, registers, and mnemonics set by -mcpu, but the scheduling parameters set by -mtune.
- -mrecip=rsqrt
- EXTRA_LDFLAGS, OPTIMIZE
- This option will enable GCC to use RCPSS and RSQRTSS instructions (and their vectorized variants RCPPS and RSQRTPS) with an additional Newton-Raphson step to increase precision instead of DIVSS and SQRTSS (and their vectorized variants) for single precision floating point arguments. These instructions are generated only when -funsafe-math-optimizations is enabled together with -finite-math-only and -fno-trapping-math.
- -fpeel-loops
- EXTRA_LDFLAGS, OPTIMIZE
- Peels the loops for that there is enough information that they do not roll much (from profile feedback). It also turns on complete loop peeling (i.e. complete removal of loops with small constant number of iterations). Enabled with -fprofile-use
- -funroll-loops
- EXTRA_LDFLAGS, OPTIMIZE
- Unroll loops whose number of iterations can be determined at compile time or upon entry to the loop. -funroll-loops implies both -fstrength-reduce and -frerun-cse-after-loop. This option makes code larger, and may or may not make it run faster.
- Includes:
  - -fstrength-reduce
  - -frerun-cse-after-loop
- -mpopcntd
- EXTRA_LDFLAGS, OPTIMIZE
- Allows GCC to generate the popcount instruction implemented on the POWER7 processor and other processors that support the PowerPC V2.06 architecture.
- -m32
- EXTRA_LDFLAGS, OPTIMIZE
- Generate code for a 32-bit environment. The 32-bit environment sets int, long and pointer to 32 bits and generates code that runs on 32 bits system.
- -fvect-cost-model
- OPTIMIZE
- Enable cost model for vectorization.
- -mveclibabi=mass
- OPTIMIZE
- Specifies the Mathematical Acceleration Subsystem (MASS) ABI to use for vectorizing intrinsics using an external library.
- -flto
- OPTIMIZE
- This option runs the standard link-time optimizer. When invoked with source code, it generates GIMPLE (one of GCC's internal representations) and writes it to special ELF sections in the object file. When the object files are linked together, all the function bodies are read from these ELF sections and instantiated as if they had been part of the same translation unit.
- -fwhole-program
- OPTIMIZE
- Assume that the current compilation unit represents whole program being compiled. All public functions and variables with the exception of "main" and those merged by attribute "externally_visible" become static functions and in a affect gets more aggressively optimized by interprocedural optimizers.
- -fuse-linker-plugin
- OPTIMIZE
- Enables the use of linker plugin during link time optimization. This option relies on the linker plugin support in linker that is available in gold or in GNU ld 2.21 or newer. This option enables the extraction of object files with GIMPLE bytecode out of library archives. This improves the quality of optimization by exposing more code the the link time optimizer. This information specify what symbols can be accessed externally (by non-LTO object or during dynamic linking). Resulting code quality improvements on binaries (and shared libraries that do use hidden visibility) is similar to -fwhole-program. See -flto for a description on the effect of this flag and how to use it. Enabled by default when LTO support in GCC is enabled and GCC was compiled with a linker supporting plugins (GNU ld 2.21 or newer or gold).
- -Wl,-q
- EXTRA_LDFLAGS
- Pass the -q flag to the linker causing the final executable to have the relocation information.
- -Wl,-Map=link.map,--cref
- EXTRA_LDFLAGS
- Pass "-Map=link.map --cref" to the linker causing to print a linker map to link.map and output a cross reference table to the map file.
- -L /opt/at6.0/lib
- EXTRA_LDFLAGS
- Add directory /opt/at6.0/lib to the list of directories to be searched for -l.
- -L /opt/ibmcmp/xlmass/7.1/lib
- EXTRA_LDFLAGS
- Add directory /opt/ibmcmp/xlmass/7.1/lib to the list of directories to be searched for -l.
- -Wl,-rpath,/opt/at6.0/lib
- EXTRA_LDFLAGS
- Pass "-rpath /opt/at6.0/lib" to the linker causing the Advance Toolchain library directory (/opt/at6.0/lib) to be added to the runtime library search path.
- -Wl,-rpath,/opt/ibmcmp/xlmass/7.1/lib
- EXTRA_LDFLAGS
- Pass "-rpath /opt/ibmcmp/xlmass/7.1/lib" to the linker causing the IBM XL Mathematical Acceleration Subsystem (MASS) library directory (/opt/ibmcmp/xlmass/7.1/lib) to be added to the runtime library search path.
- -lhugetlbfs
- EXTRA_LIBS
- Link with libhugetlbfs.so. This enables heap to be backed by the 16 Megabyte pages.
- -lmassvp7
- EXTRA_LIBS
- Link the mathematical acceleration subsystem (MASS) vector libraries, which contain libraries of tuned mathematical intrinsic functions.
- -lmass_simdp7
- EXTRA_LIBS
- Link the mathematical acceleration subsystem (MASS) SIMD libraries, which contain libraries of tuned mathematical intrinsic functions.
- -lmass
- EXTRA_LIBS
- Link the mathematical acceleration subsystem (MASS) libraries, which contain libraries of tuned mathematical intrinsic functions.
- -ltcmalloc
- EXTRA_LIBS
- Link with tcmalloc's library for Linux on POWER. This is a library that optimizes calls to new, delete, malloc and free.
- -lstdc++ -lpthread
- EXTRA_LIBS
- Link the C++ standard library.

437.leslie3d

- -fprofile-generate
- PASS1_FFLAGS, PASS1_LDFLAGS
- Enable options usually used for instrumenting application to produce profile useful for later recompilation with profile feedback based optimization.
- -fprofile-use
- PASS2_FFLAGS, PASS2_LDFLAGS
- Enable profile feedback directed optimizations, and optimizations generally profitable only with profile feedback available.
- -ffast-math
- EXTRA_LDFLAGS, OPTIMIZE
- Sets the following flags:
  - -fno-math-errno
  - -funsafe-math-optimizations
  - -fno-trapping-math
  - -ffinite-math-only
  - -fno-signaling-nans
- Includes:
- -O3
- EXTRA_LDFLAGS, OPTIMIZE
- Optimize yet more. -O3 turns on all optimizations specified by -O2 and also turns on the -finline-functions, -funswitch-loops and -fgcse-after-reload options.
- Includes:
- -mcpu=power7
- EXTRA_LDFLAGS, OPTIMIZE
- Tune to cpu-type everything applicable about the generated code, except for the ABI and the set of available instructions.
  
  A sample list of supported values for this flag are
  - power
  - power2
  - power3
  - power4
  - power5
  - power5+
  - power6
  - power7
- -mtune=power7
- EXTRA_LDFLAGS, OPTIMIZE
- Sets the instruction scheduling parameters for a particular machine type, but does not set the architecture type, register usage, or choice of mnemonics, as -mcpu=cpu_type would. The same values for cpu_type are used for -mtune as for -mcpu. If both are specified, the code generated will use the architecture, registers, and mnemonics set by -mcpu, but the scheduling parameters set by -mtune.
- -mrecip=rsqrt
- EXTRA_LDFLAGS, OPTIMIZE
- This option will enable GCC to use RCPSS and RSQRTSS instructions (and their vectorized variants RCPPS and RSQRTPS) with an additional Newton-Raphson step to increase precision instead of DIVSS and SQRTSS (and their vectorized variants) for single precision floating point arguments. These instructions are generated only when -funsafe-math-optimizations is enabled together with -finite-math-only and -fno-trapping-math.
- -fpeel-loops
- EXTRA_LDFLAGS, OPTIMIZE
- Peels the loops for that there is enough information that they do not roll much (from profile feedback). It also turns on complete loop peeling (i.e. complete removal of loops with small constant number of iterations). Enabled with -fprofile-use
- -funroll-loops
- EXTRA_LDFLAGS, OPTIMIZE
- Unroll loops whose number of iterations can be determined at compile time or upon entry to the loop. -funroll-loops implies both -fstrength-reduce and -frerun-cse-after-loop. This option makes code larger, and may or may not make it run faster.
- Includes:
  - -fstrength-reduce
  - -frerun-cse-after-loop
- -mpopcntd
- EXTRA_LDFLAGS, OPTIMIZE
- Allows GCC to generate the popcount instruction implemented on the POWER7 processor and other processors that support the PowerPC V2.06 architecture.
- -m32
- EXTRA_LDFLAGS, OPTIMIZE
- Generate code for a 32-bit environment. The 32-bit environment sets int, long and pointer to 32 bits and generates code that runs on 32 bits system.
- -fvect-cost-model
- OPTIMIZE
- Enable cost model for vectorization.
- -mveclibabi=mass
- OPTIMIZE
- Specifies the Mathematical Acceleration Subsystem (MASS) ABI to use for vectorizing intrinsics using an external library.
- -flto
- OPTIMIZE
- This option runs the standard link-time optimizer. When invoked with source code, it generates GIMPLE (one of GCC's internal representations) and writes it to special ELF sections in the object file. When the object files are linked together, all the function bodies are read from these ELF sections and instantiated as if they had been part of the same translation unit.
- -fwhole-program
- OPTIMIZE
- Assume that the current compilation unit represents whole program being compiled. All public functions and variables with the exception of "main" and those merged by attribute "externally_visible" become static functions and in a affect gets more aggressively optimized by interprocedural optimizers.
- -fuse-linker-plugin
- OPTIMIZE
- Enables the use of linker plugin during link time optimization. This option relies on the linker plugin support in linker that is available in gold or in GNU ld 2.21 or newer. This option enables the extraction of object files with GIMPLE bytecode out of library archives. This improves the quality of optimization by exposing more code the the link time optimizer. This information specify what symbols can be accessed externally (by non-LTO object or during dynamic linking). Resulting code quality improvements on binaries (and shared libraries that do use hidden visibility) is similar to -fwhole-program. See -flto for a description on the effect of this flag and how to use it. Enabled by default when LTO support in GCC is enabled and GCC was compiled with a linker supporting plugins (GNU ld 2.21 or newer or gold).
- -Wl,-q
- EXTRA_LDFLAGS
- Pass the -q flag to the linker causing the final executable to have the relocation information.
- -Wl,-Map=link.map,--cref
- EXTRA_LDFLAGS
- Pass "-Map=link.map --cref" to the linker causing to print a linker map to link.map and output a cross reference table to the map file.
- -L /opt/at6.0/lib
- EXTRA_LDFLAGS
- Add directory /opt/at6.0/lib to the list of directories to be searched for -l.
- -L /opt/ibmcmp/xlmass/7.1/lib
- EXTRA_LDFLAGS
- Add directory /opt/ibmcmp/xlmass/7.1/lib to the list of directories to be searched for -l.
- -Wl,-rpath,/opt/at6.0/lib
- EXTRA_LDFLAGS
- Pass "-rpath /opt/at6.0/lib" to the linker causing the Advance Toolchain library directory (/opt/at6.0/lib) to be added to the runtime library search path.
- -Wl,-rpath,/opt/ibmcmp/xlmass/7.1/lib
- EXTRA_LDFLAGS
- Pass "-rpath /opt/ibmcmp/xlmass/7.1/lib" to the linker causing the IBM XL Mathematical Acceleration Subsystem (MASS) library directory (/opt/ibmcmp/xlmass/7.1/lib) to be added to the runtime library search path.
- -lhugetlbfs
- EXTRA_LIBS
- Link with libhugetlbfs.so. This enables heap to be backed by the 16 Megabyte pages.
- -lmassvp7
- EXTRA_LIBS
- Link the mathematical acceleration subsystem (MASS) vector libraries, which contain libraries of tuned mathematical intrinsic functions.
- -lmass_simdp7
- EXTRA_LIBS
- Link the mathematical acceleration subsystem (MASS) SIMD libraries, which contain libraries of tuned mathematical intrinsic functions.
- -lmass
- EXTRA_LIBS
- Link the mathematical acceleration subsystem (MASS) libraries, which contain libraries of tuned mathematical intrinsic functions.

459.GemsFDTD

- -fprofile-generate
- PASS1_FFLAGS, PASS1_LDFLAGS
- Enable options usually used for instrumenting application to produce profile useful for later recompilation with profile feedback based optimization.
- -fprofile-use
- PASS2_FFLAGS, PASS2_LDFLAGS
- Enable profile feedback directed optimizations, and optimizations generally profitable only with profile feedback available.
- -ffast-math
- EXTRA_LDFLAGS, OPTIMIZE
- Sets the following flags:
  - -fno-math-errno
  - -funsafe-math-optimizations
  - -fno-trapping-math
  - -ffinite-math-only
  - -fno-signaling-nans
- Includes:
- -O3
- EXTRA_LDFLAGS, OPTIMIZE
- Optimize yet more. -O3 turns on all optimizations specified by -O2 and also turns on the -finline-functions, -funswitch-loops and -fgcse-after-reload options.
- Includes:
- -mcpu=power7
- EXTRA_LDFLAGS, OPTIMIZE
- Tune to cpu-type everything applicable about the generated code, except for the ABI and the set of available instructions.
  
  A sample list of supported values for this flag are
  - power
  - power2
  - power3
  - power4
  - power5
  - power5+
  - power6
  - power7
- -mtune=power7
- EXTRA_LDFLAGS, OPTIMIZE
- Sets the instruction scheduling parameters for a particular machine type, but does not set the architecture type, register usage, or choice of mnemonics, as -mcpu=cpu_type would. The same values for cpu_type are used for -mtune as for -mcpu. If both are specified, the code generated will use the architecture, registers, and mnemonics set by -mcpu, but the scheduling parameters set by -mtune.
- -mrecip=rsqrt
- EXTRA_LDFLAGS, OPTIMIZE
- This option will enable GCC to use RCPSS and RSQRTSS instructions (and their vectorized variants RCPPS and RSQRTPS) with an additional Newton-Raphson step to increase precision instead of DIVSS and SQRTSS (and their vectorized variants) for single precision floating point arguments. These instructions are generated only when -funsafe-math-optimizations is enabled together with -finite-math-only and -fno-trapping-math.
- -fpeel-loops
- EXTRA_LDFLAGS, OPTIMIZE
- Peels the loops for that there is enough information that they do not roll much (from profile feedback). It also turns on complete loop peeling (i.e. complete removal of loops with small constant number of iterations). Enabled with -fprofile-use
- -funroll-loops
- EXTRA_LDFLAGS, OPTIMIZE
- Unroll loops whose number of iterations can be determined at compile time or upon entry to the loop. -funroll-loops implies both -fstrength-reduce and -frerun-cse-after-loop. This option makes code larger, and may or may not make it run faster.
- Includes:
  - -fstrength-reduce
  - -frerun-cse-after-loop
- -mpopcntd
- EXTRA_LDFLAGS, OPTIMIZE
- Allows GCC to generate the popcount instruction implemented on the POWER7 processor and other processors that support the PowerPC V2.06 architecture.
- -m64
- EXTRA_LDFLAGS, OPTIMIZE
- Generate code for a 64-bit environment. The 64-bit environment sets int, long and pointer to 64 bits and generates code that runs on 64 bits system.
- -mcmodel=medium
- EXTRA_LDFLAGS, OPTIMIZE
- Generate code for the medium model: The program is linked in the lower 2 GB of the address space. Small symbols are also placed there. Symbols with sizes larger than -mlarge-data-threshold are put into large data or bss sections and can be located above 2GB. Programs can be statically or dynamically linked.
- -fvect-cost-model
- OPTIMIZE
- Enable cost model for vectorization.
- -mveclibabi=mass
- OPTIMIZE
- Specifies the Mathematical Acceleration Subsystem (MASS) ABI to use for vectorizing intrinsics using an external library.
- -Wl,-q
- EXTRA_LDFLAGS
- Pass the -q flag to the linker causing the final executable to have the relocation information.
- -Wl,-Map=link.map,--cref
- EXTRA_LDFLAGS
- Pass "-Map=link.map --cref" to the linker causing to print a linker map to link.map and output a cross reference table to the map file.
- -L /opt/at6.0/lib64
- EXTRA_LDFLAGS
- Add directory /opt/at6.0/lib64 to the list of directories to be searched for -l.
- -L /opt/ibmcmp/xlmass/7.1/lib64
- EXTRA_LDFLAGS
- Add directory /opt/ibmcmp/xlmass/7.1/lib64 to the list of directories to be searched for -l.
- -Wl,-rpath,/opt/at6.0/lib64
- EXTRA_LDFLAGS
- Pass "-rpath /opt/at6.0/lib64" to the linker causing the Advance Toolchain for 64-bit library directory (/opt/at6.0/lib64) to be added to the runtime library search path.
- -Wl,-rpath,/opt/ibmcmp/xlmass/7.1/lib64
- EXTRA_LDFLAGS
- Pass "-rpath /opt/ibmcmp/xlmass/7.1/lib64" to the linker causing the IBM XL Mathematical Acceleration Subsystem (MASS) for 64-bit library directory (/opt/ibmcmp/xlmass/7.1/lib64) to be added to the runtime library search path.
- -Wl,-dynamic-linker,/opt/at6.0/lib64/ld64.so.1
- EXTRA_LDFLAGS
- Pass the "-dynamic-linker=/opt/at6.0/lib64/ld64.so.1" flag to the linker causing to set the dynamic linker of Advance Toolchain 64-bit (/opt/at6.0/lib64/ld64.so.1).
- -lhugetlbfs
- EXTRA_LIBS
- Link with libhugetlbfs.so. This enables heap to be backed by the 16 Megabyte pages.
- -lmassvp7_64
- EXTRA_LIBS
- Link the mathematical acceleration subsystem (MASS) vector libraries for 64-bit, which contain libraries of tuned mathematical intrinsic functions.
- -lmass_simdp7_64
- EXTRA_LIBS
- Link the mathematical acceleration subsystem (MASS) SIMD libraries for 64-bit, which contain libraries of tuned mathematical intrinsic functions.
- -lmass_64
- EXTRA_LIBS
- Link the mathematical acceleration subsystem (MASS) libraries for 64-bit, which contain libraries of tuned mathematical intrinsic functions.

Benchmarks using both Fortran and C

435.gromacs

- -fprofile-generate
- PASS1_CFLAGS, PASS1_FFLAGS, PASS1_LDFLAGS
- Enable options usually used for instrumenting application to produce profile useful for later recompilation with profile feedback based optimization.
- -fprofile-use
- PASS2_CFLAGS, PASS2_FFLAGS, PASS2_LDFLAGS
- Enable profile feedback directed optimizations, and optimizations generally profitable only with profile feedback available.
- -ffast-math
- EXTRA_LDFLAGS, OPTIMIZE
- Sets the following flags:
  - -fno-math-errno
  - -funsafe-math-optimizations
  - -fno-trapping-math
  - -ffinite-math-only
  - -fno-signaling-nans
- Includes:
- -O3
- EXTRA_LDFLAGS, OPTIMIZE
- Optimize yet more. -O3 turns on all optimizations specified by -O2 and also turns on the -finline-functions, -funswitch-loops and -fgcse-after-reload options.
- Includes:
- -mcpu=power7
- EXTRA_LDFLAGS, OPTIMIZE
- Tune to cpu-type everything applicable about the generated code, except for the ABI and the set of available instructions.
  
  A sample list of supported values for this flag are
  - power
  - power2
  - power3
  - power4
  - power5
  - power5+
  - power6
  - power7
- -mtune=power7
- EXTRA_LDFLAGS, OPTIMIZE
- Sets the instruction scheduling parameters for a particular machine type, but does not set the architecture type, register usage, or choice of mnemonics, as -mcpu=cpu_type would. The same values for cpu_type are used for -mtune as for -mcpu. If both are specified, the code generated will use the architecture, registers, and mnemonics set by -mcpu, but the scheduling parameters set by -mtune.
- -mrecip=rsqrt
- EXTRA_LDFLAGS, OPTIMIZE
- This option will enable GCC to use RCPSS and RSQRTSS instructions (and their vectorized variants RCPPS and RSQRTPS) with an additional Newton-Raphson step to increase precision instead of DIVSS and SQRTSS (and their vectorized variants) for single precision floating point arguments. These instructions are generated only when -funsafe-math-optimizations is enabled together with -finite-math-only and -fno-trapping-math.
- -fpeel-loops
- EXTRA_LDFLAGS, OPTIMIZE
- Peels the loops for that there is enough information that they do not roll much (from profile feedback). It also turns on complete loop peeling (i.e. complete removal of loops with small constant number of iterations). Enabled with -fprofile-use
- -funroll-loops
- EXTRA_LDFLAGS, OPTIMIZE
- Unroll loops whose number of iterations can be determined at compile time or upon entry to the loop. -funroll-loops implies both -fstrength-reduce and -frerun-cse-after-loop. This option makes code larger, and may or may not make it run faster.
- Includes:
  - -fstrength-reduce
  - -frerun-cse-after-loop
- -mpopcntd
- EXTRA_LDFLAGS, OPTIMIZE
- Allows GCC to generate the popcount instruction implemented on the POWER7 processor and other processors that support the PowerPC V2.06 architecture.
- -m32
- EXTRA_LDFLAGS, OPTIMIZE
- Generate code for a 32-bit environment. The 32-bit environment sets int, long and pointer to 32 bits and generates code that runs on 32 bits system.
- -fvect-cost-model
- OPTIMIZE
- Enable cost model for vectorization.
- -mveclibabi=mass
- OPTIMIZE
- Specifies the Mathematical Acceleration Subsystem (MASS) ABI to use for vectorizing intrinsics using an external library.
- -Wl,-q
- EXTRA_LDFLAGS
- Pass the -q flag to the linker causing the final executable to have the relocation information.
- -Wl,-Map=link.map,--cref
- EXTRA_LDFLAGS
- Pass "-Map=link.map --cref" to the linker causing to print a linker map to link.map and output a cross reference table to the map file.
- -L /opt/at6.0/lib
- EXTRA_LDFLAGS
- Add directory /opt/at6.0/lib to the list of directories to be searched for -l.
- -L /opt/ibmcmp/xlmass/7.1/lib
- EXTRA_LDFLAGS
- Add directory /opt/ibmcmp/xlmass/7.1/lib to the list of directories to be searched for -l.
- -Wl,-rpath,/opt/at6.0/lib
- EXTRA_LDFLAGS
- Pass "-rpath /opt/at6.0/lib" to the linker causing the Advance Toolchain library directory (/opt/at6.0/lib) to be added to the runtime library search path.
- -Wl,-rpath,/opt/ibmcmp/xlmass/7.1/lib
- EXTRA_LDFLAGS
- Pass "-rpath /opt/ibmcmp/xlmass/7.1/lib" to the linker causing the IBM XL Mathematical Acceleration Subsystem (MASS) library directory (/opt/ibmcmp/xlmass/7.1/lib) to be added to the runtime library search path.
- -lhugetlbfs
- EXTRA_LIBS
- Link with libhugetlbfs.so. This enables heap to be backed by the 16 Megabyte pages.
- -lmassvp7
- EXTRA_LIBS
- Link the mathematical acceleration subsystem (MASS) vector libraries, which contain libraries of tuned mathematical intrinsic functions.
- -lmass_simdp7
- EXTRA_LIBS
- Link the mathematical acceleration subsystem (MASS) SIMD libraries, which contain libraries of tuned mathematical intrinsic functions.
- -lmass
- EXTRA_LIBS
- Link the mathematical acceleration subsystem (MASS) libraries, which contain libraries of tuned mathematical intrinsic functions.

436.cactusADM

- -fprofile-generate
- PASS1_CFLAGS, PASS1_FFLAGS, PASS1_LDFLAGS
- Enable options usually used for instrumenting application to produce profile useful for later recompilation with profile feedback based optimization.
- -fprofile-use
- PASS2_CFLAGS, PASS2_FFLAGS, PASS2_LDFLAGS
- Enable profile feedback directed optimizations, and optimizations generally profitable only with profile feedback available.
- -ffast-math
- EXTRA_LDFLAGS, OPTIMIZE
- Sets the following flags:
  - -fno-math-errno
  - -funsafe-math-optimizations
  - -fno-trapping-math
  - -ffinite-math-only
  - -fno-signaling-nans
- Includes:
- -O3
- EXTRA_LDFLAGS, OPTIMIZE
- Optimize yet more. -O3 turns on all optimizations specified by -O2 and also turns on the -finline-functions, -funswitch-loops and -fgcse-after-reload options.
- Includes:
- -mcpu=power7
- EXTRA_LDFLAGS, OPTIMIZE
- Tune to cpu-type everything applicable about the generated code, except for the ABI and the set of available instructions.
  
  A sample list of supported values for this flag are
  - power
  - power2
  - power3
  - power4
  - power5
  - power5+
  - power6
  - power7
- -mtune=power7
- EXTRA_LDFLAGS, OPTIMIZE
- Sets the instruction scheduling parameters for a particular machine type, but does not set the architecture type, register usage, or choice of mnemonics, as -mcpu=cpu_type would. The same values for cpu_type are used for -mtune as for -mcpu. If both are specified, the code generated will use the architecture, registers, and mnemonics set by -mcpu, but the scheduling parameters set by -mtune.
- -mrecip=rsqrt
- EXTRA_LDFLAGS, OPTIMIZE
- This option will enable GCC to use RCPSS and RSQRTSS instructions (and their vectorized variants RCPPS and RSQRTPS) with an additional Newton-Raphson step to increase precision instead of DIVSS and SQRTSS (and their vectorized variants) for single precision floating point arguments. These instructions are generated only when -funsafe-math-optimizations is enabled together with -finite-math-only and -fno-trapping-math.
- -fpeel-loops
- EXTRA_LDFLAGS, OPTIMIZE
- Peels the loops for that there is enough information that they do not roll much (from profile feedback). It also turns on complete loop peeling (i.e. complete removal of loops with small constant number of iterations). Enabled with -fprofile-use
- -funroll-loops
- EXTRA_LDFLAGS, OPTIMIZE
- Unroll loops whose number of iterations can be determined at compile time or upon entry to the loop. -funroll-loops implies both -fstrength-reduce and -frerun-cse-after-loop. This option makes code larger, and may or may not make it run faster.
- Includes:
  - -fstrength-reduce
  - -frerun-cse-after-loop
- -mpopcntd
- EXTRA_LDFLAGS, OPTIMIZE
- Allows GCC to generate the popcount instruction implemented on the POWER7 processor and other processors that support the PowerPC V2.06 architecture.
- -m64
- EXTRA_LDFLAGS, OPTIMIZE
- Generate code for a 64-bit environment. The 64-bit environment sets int, long and pointer to 64 bits and generates code that runs on 64 bits system.
- -mcmodel=medium
- EXTRA_LDFLAGS, OPTIMIZE
- Generate code for the medium model: The program is linked in the lower 2 GB of the address space. Small symbols are also placed there. Symbols with sizes larger than -mlarge-data-threshold are put into large data or bss sections and can be located above 2GB. Programs can be statically or dynamically linked.
- -fvect-cost-model
- OPTIMIZE
- Enable cost model for vectorization.
- -mveclibabi=mass
- OPTIMIZE
- Specifies the Mathematical Acceleration Subsystem (MASS) ABI to use for vectorizing intrinsics using an external library.
- -Wl,-q
- EXTRA_LDFLAGS
- Pass the -q flag to the linker causing the final executable to have the relocation information.
- -Wl,-Map=link.map,--cref
- EXTRA_LDFLAGS
- Pass "-Map=link.map --cref" to the linker causing to print a linker map to link.map and output a cross reference table to the map file.
- -L /opt/at6.0/lib64
- EXTRA_LDFLAGS
- Add directory /opt/at6.0/lib64 to the list of directories to be searched for -l.
- -L /opt/ibmcmp/xlmass/7.1/lib64
- EXTRA_LDFLAGS
- Add directory /opt/ibmcmp/xlmass/7.1/lib64 to the list of directories to be searched for -l.
- -Wl,-rpath,/opt/at6.0/lib64
- EXTRA_LDFLAGS
- Pass "-rpath /opt/at6.0/lib64" to the linker causing the Advance Toolchain for 64-bit library directory (/opt/at6.0/lib64) to be added to the runtime library search path.
- -Wl,-rpath,/opt/ibmcmp/xlmass/7.1/lib64
- EXTRA_LDFLAGS
- Pass "-rpath /opt/ibmcmp/xlmass/7.1/lib64" to the linker causing the IBM XL Mathematical Acceleration Subsystem (MASS) for 64-bit library directory (/opt/ibmcmp/xlmass/7.1/lib64) to be added to the runtime library search path.
- -Wl,-dynamic-linker,/opt/at6.0/lib64/ld64.so.1
- EXTRA_LDFLAGS
- Pass the "-dynamic-linker=/opt/at6.0/lib64/ld64.so.1" flag to the linker causing to set the dynamic linker of Advance Toolchain 64-bit (/opt/at6.0/lib64/ld64.so.1).
- -lhugetlbfs
- EXTRA_LIBS
- Link with libhugetlbfs.so. This enables heap to be backed by the 16 Megabyte pages.
- -lmassvp7_64
- EXTRA_LIBS
- Link the mathematical acceleration subsystem (MASS) vector libraries for 64-bit, which contain libraries of tuned mathematical intrinsic functions.
- -lmass_simdp7_64
- EXTRA_LIBS
- Link the mathematical acceleration subsystem (MASS) SIMD libraries for 64-bit, which contain libraries of tuned mathematical intrinsic functions.
- -lmass_64
- EXTRA_LIBS
- Link the mathematical acceleration subsystem (MASS) libraries for 64-bit, which contain libraries of tuned mathematical intrinsic functions.

454.calculix

- -ffast-math
- EXTRA_LDFLAGS, OPTIMIZE
- Sets the following flags:
  - -fno-math-errno
  - -funsafe-math-optimizations
  - -fno-trapping-math
  - -ffinite-math-only
  - -fno-signaling-nans
- Includes:
- -O3
- EXTRA_LDFLAGS, OPTIMIZE
- Optimize yet more. -O3 turns on all optimizations specified by -O2 and also turns on the -finline-functions, -funswitch-loops and -fgcse-after-reload options.
- Includes:
- -mcpu=power7
- EXTRA_LDFLAGS, OPTIMIZE
- Tune to cpu-type everything applicable about the generated code, except for the ABI and the set of available instructions.
  
  A sample list of supported values for this flag are
  - power
  - power2
  - power3
  - power4
  - power5
  - power5+
  - power6
  - power7
- -mtune=power7
- EXTRA_LDFLAGS, OPTIMIZE
- Sets the instruction scheduling parameters for a particular machine type, but does not set the architecture type, register usage, or choice of mnemonics, as -mcpu=cpu_type would. The same values for cpu_type are used for -mtune as for -mcpu. If both are specified, the code generated will use the architecture, registers, and mnemonics set by -mcpu, but the scheduling parameters set by -mtune.
- -mrecip=rsqrt
- EXTRA_LDFLAGS, OPTIMIZE
- This option will enable GCC to use RCPSS and RSQRTSS instructions (and their vectorized variants RCPPS and RSQRTPS) with an additional Newton-Raphson step to increase precision instead of DIVSS and SQRTSS (and their vectorized variants) for single precision floating point arguments. These instructions are generated only when -funsafe-math-optimizations is enabled together with -finite-math-only and -fno-trapping-math.
- -fpeel-loops
- EXTRA_LDFLAGS, OPTIMIZE
- Peels the loops for that there is enough information that they do not roll much (from profile feedback). It also turns on complete loop peeling (i.e. complete removal of loops with small constant number of iterations). Enabled with -fprofile-use
- -funroll-loops
- EXTRA_LDFLAGS, OPTIMIZE
- Unroll loops whose number of iterations can be determined at compile time or upon entry to the loop. -funroll-loops implies both -fstrength-reduce and -frerun-cse-after-loop. This option makes code larger, and may or may not make it run faster.
- Includes:
  - -fstrength-reduce
  - -frerun-cse-after-loop
- -mpopcntd
- EXTRA_LDFLAGS, OPTIMIZE
- Allows GCC to generate the popcount instruction implemented on the POWER7 processor and other processors that support the PowerPC V2.06 architecture.
- -m32
- EXTRA_LDFLAGS, OPTIMIZE
- Generate code for a 32-bit environment. The 32-bit environment sets int, long and pointer to 32 bits and generates code that runs on 32 bits system.
- -fvect-cost-model
- OPTIMIZE
- Enable cost model for vectorization.
- -mveclibabi=mass
- OPTIMIZE
- Specifies the Mathematical Acceleration Subsystem (MASS) ABI to use for vectorizing intrinsics using an external library.
- -flto
- COPTIMIZE
- This option runs the standard link-time optimizer. When invoked with source code, it generates GIMPLE (one of GCC's internal representations) and writes it to special ELF sections in the object file. When the object files are linked together, all the function bodies are read from these ELF sections and instantiated as if they had been part of the same translation unit.
- -fwhole-program
- COPTIMIZE
- Assume that the current compilation unit represents whole program being compiled. All public functions and variables with the exception of "main" and those merged by attribute "externally_visible" become static functions and in a affect gets more aggressively optimized by interprocedural optimizers.
- -fuse-linker-plugin
- COPTIMIZE
- Enables the use of linker plugin during link time optimization. This option relies on the linker plugin support in linker that is available in gold or in GNU ld 2.21 or newer. This option enables the extraction of object files with GIMPLE bytecode out of library archives. This improves the quality of optimization by exposing more code the the link time optimizer. This information specify what symbols can be accessed externally (by non-LTO object or during dynamic linking). Resulting code quality improvements on binaries (and shared libraries that do use hidden visibility) is similar to -fwhole-program. See -flto for a description on the effect of this flag and how to use it. Enabled by default when LTO support in GCC is enabled and GCC was compiled with a linker supporting plugins (GNU ld 2.21 or newer or gold).
- -Wl,-q
- EXTRA_LDFLAGS
- Pass the -q flag to the linker causing the final executable to have the relocation information.
- -Wl,-Map=link.map,--cref
- EXTRA_LDFLAGS
- Pass "-Map=link.map --cref" to the linker causing to print a linker map to link.map and output a cross reference table to the map file.
- -L /opt/at6.0/lib
- EXTRA_LDFLAGS
- Add directory /opt/at6.0/lib to the list of directories to be searched for -l.
- -L /opt/ibmcmp/xlmass/7.1/lib
- EXTRA_LDFLAGS
- Add directory /opt/ibmcmp/xlmass/7.1/lib to the list of directories to be searched for -l.
- -Wl,-rpath,/opt/at6.0/lib
- EXTRA_LDFLAGS
- Pass "-rpath /opt/at6.0/lib" to the linker causing the Advance Toolchain library directory (/opt/at6.0/lib) to be added to the runtime library search path.
- -Wl,-rpath,/opt/ibmcmp/xlmass/7.1/lib
- EXTRA_LDFLAGS
- Pass "-rpath /opt/ibmcmp/xlmass/7.1/lib" to the linker causing the IBM XL Mathematical Acceleration Subsystem (MASS) library directory (/opt/ibmcmp/xlmass/7.1/lib) to be added to the runtime library search path.
- -lhugetlbfs
- EXTRA_LIBS
- Link with libhugetlbfs.so. This enables heap to be backed by the 16 Megabyte pages.
- -lmassvp7
- EXTRA_LIBS
- Link the mathematical acceleration subsystem (MASS) vector libraries, which contain libraries of tuned mathematical intrinsic functions.
- -lmass_simdp7
- EXTRA_LIBS
- Link the mathematical acceleration subsystem (MASS) SIMD libraries, which contain libraries of tuned mathematical intrinsic functions.
- -lmass
- EXTRA_LIBS
- Link the mathematical acceleration subsystem (MASS) libraries, which contain libraries of tuned mathematical intrinsic functions.
- -ltcmalloc
- EXTRA_LIBS
- Link with tcmalloc's library for Linux on POWER. This is a library that optimizes calls to new, delete, malloc and free.
- -lstdc++ -lpthread
- EXTRA_LIBS
- Link the C++ standard library.

481.wrf

- -ffast-math
- EXTRA_LDFLAGS, OPTIMIZE
- Sets the following flags:
  - -fno-math-errno
  - -funsafe-math-optimizations
  - -fno-trapping-math
  - -ffinite-math-only
  - -fno-signaling-nans
- Includes:
- -O3
- EXTRA_LDFLAGS, OPTIMIZE
- Optimize yet more. -O3 turns on all optimizations specified by -O2 and also turns on the -finline-functions, -funswitch-loops and -fgcse-after-reload options.
- Includes:
- -mcpu=power7
- EXTRA_LDFLAGS, OPTIMIZE
- Tune to cpu-type everything applicable about the generated code, except for the ABI and the set of available instructions.
  
  A sample list of supported values for this flag are
  - power
  - power2
  - power3
  - power4
  - power5
  - power5+
  - power6
  - power7
- -mtune=power7
- EXTRA_LDFLAGS, OPTIMIZE
- Sets the instruction scheduling parameters for a particular machine type, but does not set the architecture type, register usage, or choice of mnemonics, as -mcpu=cpu_type would. The same values for cpu_type are used for -mtune as for -mcpu. If both are specified, the code generated will use the architecture, registers, and mnemonics set by -mcpu, but the scheduling parameters set by -mtune.
- -mrecip=rsqrt
- EXTRA_LDFLAGS, OPTIMIZE
- This option will enable GCC to use RCPSS and RSQRTSS instructions (and their vectorized variants RCPPS and RSQRTPS) with an additional Newton-Raphson step to increase precision instead of DIVSS and SQRTSS (and their vectorized variants) for single precision floating point arguments. These instructions are generated only when -funsafe-math-optimizations is enabled together with -finite-math-only and -fno-trapping-math.
- -fpeel-loops
- EXTRA_LDFLAGS, OPTIMIZE
- Peels the loops for that there is enough information that they do not roll much (from profile feedback). It also turns on complete loop peeling (i.e. complete removal of loops with small constant number of iterations). Enabled with -fprofile-use
- -funroll-loops
- EXTRA_LDFLAGS, OPTIMIZE
- Unroll loops whose number of iterations can be determined at compile time or upon entry to the loop. -funroll-loops implies both -fstrength-reduce and -frerun-cse-after-loop. This option makes code larger, and may or may not make it run faster.
- Includes:
  - -fstrength-reduce
  - -frerun-cse-after-loop
- -mpopcntd
- EXTRA_LDFLAGS, OPTIMIZE
- Allows GCC to generate the popcount instruction implemented on the POWER7 processor and other processors that support the PowerPC V2.06 architecture.
- -m64
- EXTRA_LDFLAGS, OPTIMIZE
- Generate code for a 64-bit environment. The 64-bit environment sets int, long and pointer to 64 bits and generates code that runs on 64 bits system.
- -mcmodel=medium
- EXTRA_LDFLAGS, OPTIMIZE
- Generate code for the medium model: The program is linked in the lower 2 GB of the address space. Small symbols are also placed there. Symbols with sizes larger than -mlarge-data-threshold are put into large data or bss sections and can be located above 2GB. Programs can be statically or dynamically linked.
- -fvect-cost-model
- OPTIMIZE
- Enable cost model for vectorization.
- -mveclibabi=mass
- OPTIMIZE
- Specifies the Mathematical Acceleration Subsystem (MASS) ABI to use for vectorizing intrinsics using an external library.
- -Wl,-q
- EXTRA_LDFLAGS
- Pass the -q flag to the linker causing the final executable to have the relocation information.
- -Wl,-Map=link.map,--cref
- EXTRA_LDFLAGS
- Pass "-Map=link.map --cref" to the linker causing to print a linker map to link.map and output a cross reference table to the map file.
- -L /opt/at6.0/lib64
- EXTRA_LDFLAGS
- Add directory /opt/at6.0/lib64 to the list of directories to be searched for -l.
- -L /opt/ibmcmp/xlmass/7.1/lib64
- EXTRA_LDFLAGS
- Add directory /opt/ibmcmp/xlmass/7.1/lib64 to the list of directories to be searched for -l.
- -Wl,-rpath,/opt/at6.0/lib64
- EXTRA_LDFLAGS
- Pass "-rpath /opt/at6.0/lib64" to the linker causing the Advance Toolchain for 64-bit library directory (/opt/at6.0/lib64) to be added to the runtime library search path.
- -Wl,-rpath,/opt/ibmcmp/xlmass/7.1/lib64
- EXTRA_LDFLAGS
- Pass "-rpath /opt/ibmcmp/xlmass/7.1/lib64" to the linker causing the IBM XL Mathematical Acceleration Subsystem (MASS) for 64-bit library directory (/opt/ibmcmp/xlmass/7.1/lib64) to be added to the runtime library search path.
- -Wl,-dynamic-linker,/opt/at6.0/lib64/ld64.so.1
- EXTRA_LDFLAGS
- Pass the "-dynamic-linker=/opt/at6.0/lib64/ld64.so.1" flag to the linker causing to set the dynamic linker of Advance Toolchain 64-bit (/opt/at6.0/lib64/ld64.so.1).
- -lhugetlbfs
- EXTRA_LIBS
- Link with libhugetlbfs.so. This enables heap to be backed by the 16 Megabyte pages.
- -lmassvp7_64
- EXTRA_LIBS
- Link the mathematical acceleration subsystem (MASS) vector libraries for 64-bit, which contain libraries of tuned mathematical intrinsic functions.
- -lmass_simdp7_64
- EXTRA_LIBS
- Link the mathematical acceleration subsystem (MASS) SIMD libraries for 64-bit, which contain libraries of tuned mathematical intrinsic functions.
- -lmass_64
- EXTRA_LIBS
- Link the mathematical acceleration subsystem (MASS) libraries for 64-bit, which contain libraries of tuned mathematical intrinsic functions.

Implicitly Included Flags

This section contains descriptions of flags that were included implicitly by other flags, but which do not have a permanent home at SPEC.

For questions about the meanings of these flags, please contact the tester.
For other inquiries, please contact webmaster@spec.org
Copyright 2006-2014 Standard Performance Evaluation Corporation
Tested with SPEC CPU2006 v1.2.
Report generated on Thu Jul 24 16:16:33 2014 by SPEC CPU2006 flags formatter v6906.

	Indicates that the flag description came from the user flags file.
	Indicates that the flag description came from the suite-wide flags file.
	Indicates that the flag description came from a per-benchmark flags file.

CPU2006 Flag DescriptionIBM Corporation IBM PowerLinux 7R4 (4.0 GHz, 32 core, RHEL, GCC)

Base Compiler Invocation

Peak Compiler Invocation

Base Portability Flags

Peak Portability Flags

Base Optimization Flags

Peak Optimization Flags

Implicitly Included Flags

CPU2006 Flag Description
IBM Corporation IBM PowerLinux 7R4 (4.0 GHz, 32 core, RHEL, GCC)