CPU2017 Flag Description
IBM Corporation IBM Power E1050 (2.95 - 3.90 GHz, 96 core, AIX)

Base Optimization Flags

C benchmarks

• • -m64
  • 
  • CC, COPTIMIZE, LD

  • Generates 64-bit ABI binaries. Used as a compiler option for C/C++ programs. If -m64 is not specified, the default is to generate 32-bit ABI binaries on AIX.

• • -Wl,-bbigtoc
  • 
  • LDCFLAGS

  • The -bbigtoc option instructs the linker to generate extra code if the size of TOC is greater than 64K. For C/C++ programs it is invoked as -Wl,-bplugin_opt:-bbigtoc and for Fortran programs it is invoked as -bbigtoc.

• • -Wl,-blpdata
  • 
  • LDCFLAGS

  • The -blpdata is a linker option and it sets the bit in the file's XCOFF header indicating that this executable will request the use of large pages when they are available on the system and when the user has an appropriate privilege. For C/C++ programs it is invoked as -Wl,-blpdata and for Fortran programs it is invoked as -blpdata.

• • -lc++
  • 
  • LDCFLAGS

  • Link the standard C++ library that supports C++11 features.

• • -lpthread
  • 
  • LDCFLAGS

  • Link the standard Posix threading library.

• • -Wl,-bplugin_opt:--enable-ppc-gen-scalar-mass
  • 
  • LDCFLAGS

  • Enable use of scalar MASS library while using link time optimization(LTO).

• • -Wl,-bplugin_opt:--data-layout-opt=1
  • 
  • LDCFLAGS

  • This option instructs the linker to analyze the whole program to determine if the layout of data can be transformed to improve the cache utilization and memory bandwidth of the program. This option has four levels [0, 1, 2, 3]. This is effective only under -flto and must be specified on both the compilation and linking steps with same option level.

    • data-layout-opt=0 disables the data layout transformation. No compiler analysis is performed for data layout optimization.
    • data-layout-opt=1 enables the data layout transformation with whole program analysis. This is performed with a strict safety analysis.
    • data-layout-opt=2 enables the data layout transformation with whole program more aggressive analysis. This is performed with a strict safety analysis.
    • data-layout-opt=3 enables the data layout transformation and data compression with whole program analysis. The compiler may implicitly change the default data type size to a non-default data type size.
• • -Wl,-bplugin_opt:-fold-complex-pointer-compare=false
  • 
  • EXTRA_LDFLAGS

  • <suboption> must be one of the following suboptions:

    • true : Enable simplification of compare instructions which rely on 'inbounds' semantics.
    • false : Disable simplification of compare instructions which rely on 'inbounds' semantics.

    Default: -fold-complex-pointer-compare=true

• • -mcpu=native
  • 
  • COPTIMIZE
  • Specifies the processor architecture for which the code(instructions) should be generated.

    Supported values for this flag are :

    • native - Automatically detects the specific architecture of the compiling machine. It assumes that the execution environment will be the same as the compilation environment.
    • pwr7 - Produces object code containing instructions that run on the Power7 hardware platform and on newer architectures.
    • pwr8 - Produces object code containing instructions that run on the Power8 hardware platform and on newer architectures.
    • pwr9 - Produces object code containing instructions that run on the Power9 hardware platform and on newer architectures.
• • -mabi=vec-extabi
  • 
  • COPTIMIZE

  • Enable the extended Altivec ABI on AIX, which uses volatile and nonvolatile vector registers.

• • -fapprox-func
  • 
  • COPTIMIZE

  • Allows certain math function calls (such as log, sqrt, pow, etc) to be replaced with an approximately equivalent set of instructions or alternative math function calls. For example, a pow(x, 0.25) may be replaced with sqrt(sqrt(x)), despite being an inexact result in cases where x is -0.0 or -inf. Defaults to -fno-approx-func.

• • -mllvm -enable-ppc-gen-scalar-mass
  • 
  • COPTIMIZE

  • This flag is to enable automatic use of the MASS scalar library and call scalar functions in your program.

• • -mllvm -vector-library=MASSV
  • 
  • COPTIMIZE

  • This flag is to enable automatic use of the MASS SIMD library in your program.

• • -flto
  • 
  • COPTIMIZE

  • Generate output files in LLVM format, suitable for link time optimization. When used with -S this generates LLVM intermediate language assembly files, otherwise this generates LLVM bitcode format object files. The LLVM bitcode generated is suitable for Link Time Optimization (LTO), when all modules are merged into a single combined module for optimization.

• • -mllvm -data-layout-opt=1
  • 
  • COPTIMIZE

  • This option instructs the compiler to analyze the whole program to determine if the layout of data can be transformed to improve the cache utilization and m emory bandwidth of the program. This option has four levels [0, 1, 2, 3]. This is effective only under -flto and must be specified on both the compilation an d linking steps with same option level.

    • data-layout-opt=0 disables the data layout transformation. No compiler analysis is performed for data layout optimization.
    • data-layout-opt=1 enables the data layout transformation with whole program analysis. This is performed with a strict safety analysis.
    • data-layout-opt=2 enables the data layout transformation with whole program more aggressive analysis. This is performed with a strict safety ana lysis.
    • data-layout-opt=3 enables the data layout transformation and data compression with whole program analysis. The compiler may implicitly change th e default data type size to a non-default data type size.
• • -O3
  • 
  • EXTRA_COPTIMIZE
  • -O3 Instruct the compiler to execute optimizations that might take longer to perform or that may generate larger code (in an attempt to make the program run faster).
• • -fcommon
  • 
  • EXTRA_COPTIMIZE

  • Place uninitialized global variables in a common block.

• • -mllvm -fold-complex-pointer-compare=false
  • 
  • EXTRA_COPTIMIZE

  • <suboption> must be one of the following suboptions:

    • true : Enable simplification of compare instructions which rely on 'inbounds' semantics.
    • false : Disable simplification of compare instructions which rely on 'inbounds' semantics.

    Default: -fold-complex-pointer-compare=true

• • -Wno-implicit-function-declaration
  • 
  • EXTRA_COPTIMIZE

  • This turns off diagnostics of benchmark code that is treated as errors. Specifically, it turns these error off: 1. error: call to undeclared function A; ISO C99 and later do not support implicit function declarations 2. error: call to undeclared library function 'A' with type B; ISO C99 and later do not support implicit function declarations

• • -L/opt/IBM/xlmass/10.1.x/lib
  • 
  • LIBS

  • Searches the directory path for library files specified by the "-l" option at link time.

• • -lmass
  • 
  • LIBS

  • Link the mathematical acceleration subsystem libraries (MASS), which consist of a set of mathematical functions for C,C++ and Fortran language applications that are tuned for specific Power processor architectures.

• • -L/opt/IBM/openxlf/17.1.x/lib
  • 
  • LIBS

  • Searches the directory path for library files specified by the "-l" option at link time.

• • -lxlopt
  • 
  • LIBS

  • Enables use of certain functions from the MASS library for scalars and vectors.

C++ benchmarks

• • -m64
  • 
  • CXX, CXXOPTIMIZE, LD

  • Generates 64-bit ABI binaries. Used as a compiler option for C/C++ programs. If -m64 is not specified, the default is to generate 32-bit ABI binaries on AIX.

• • -Wl,-bbigtoc
  • 
  • LDCXXFLAGS

  • The -bbigtoc option instructs the linker to generate extra code if the size of TOC is greater than 64K. For C/C++ programs it is invoked as -Wl,-bplugin_opt:-bbigtoc and for Fortran programs it is invoked as -bbigtoc.

• • -Wl,-blpdata
  • 
  • LDCXXFLAGS

  • The -blpdata is a linker option and it sets the bit in the file's XCOFF header indicating that this executable will request the use of large pages when they are available on the system and when the user has an appropriate privilege. For C/C++ programs it is invoked as -Wl,-blpdata and for Fortran programs it is invoked as -blpdata.

• • -lc++
  • 
  • LDCXXFLAGS

  • Link the standard C++ library that supports C++11 features.

• • -lpthread
  • 
  • LDCXXFLAGS

  • Link the standard Posix threading library.

• • -Wl,-bplugin_opt:--enable-ppc-gen-scalar-mass
  • 
  • LDCXXFLAGS

  • Enable use of scalar MASS library while using link time optimization(LTO).

• • -Wl,-bplugin_opt:-fold-complex-pointer-compare=false
  • 
  • EXTRA_LDFLAGS

  • <suboption> must be one of the following suboptions:

    • true : Enable simplification of compare instructions which rely on 'inbounds' semantics.
    • false : Disable simplification of compare instructions which rely on 'inbounds' semantics.

    Default: -fold-complex-pointer-compare=true

• • -mcpu=native
  • 
  • CXXOPTIMIZE
  • Specifies the processor architecture for which the code(instructions) should be generated.

    Supported values for this flag are :

    • native - Automatically detects the specific architecture of the compiling machine. It assumes that the execution environment will be the same as the compilation environment.
    • pwr7 - Produces object code containing instructions that run on the Power7 hardware platform and on newer architectures.
    • pwr8 - Produces object code containing instructions that run on the Power8 hardware platform and on newer architectures.
    • pwr9 - Produces object code containing instructions that run on the Power9 hardware platform and on newer architectures.
• • -mabi=vec-extabi
  • 
  • CXXOPTIMIZE

  • Enable the extended Altivec ABI on AIX, which uses volatile and nonvolatile vector registers.

• • -fapprox-func
  • 
  • CXXOPTIMIZE

  • Allows certain math function calls (such as log, sqrt, pow, etc) to be replaced with an approximately equivalent set of instructions or alternative math function calls. For example, a pow(x, 0.25) may be replaced with sqrt(sqrt(x)), despite being an inexact result in cases where x is -0.0 or -inf. Defaults to -fno-approx-func.

• • -mllvm -enable-ppc-gen-scalar-mass
  • 
  • CXXOPTIMIZE

  • This flag is to enable automatic use of the MASS scalar library and call scalar functions in your program.

• • -mllvm -vector-library=MASSV
  • 
  • CXXOPTIMIZE

  • This flag is to enable automatic use of the MASS SIMD library in your program.

• • -flto
  • 
  • CXXOPTIMIZE

  • Generate output files in LLVM format, suitable for link time optimization. When used with -S this generates LLVM intermediate language assembly files, otherwise this generates LLVM bitcode format object files. The LLVM bitcode generated is suitable for Link Time Optimization (LTO), when all modules are merged into a single combined module for optimization.

• • -O3
  • 
  • EXTRA_CXXOPTIMIZE
  • -O3 Instruct the compiler to execute optimizations that might take longer to perform or that may generate larger code (in an attempt to make the program run faster).
• • -L/opt/IBM/xlmass/10.1.x/lib
  • 
  • LIBS

  • Searches the directory path for library files specified by the "-l" option at link time.

• • -lmass
  • 
  • LIBS

  • Link the mathematical acceleration subsystem libraries (MASS), which consist of a set of mathematical functions for C,C++ and Fortran language applications that are tuned for specific Power processor architectures.

• • -L/opt/IBM/openxlf/17.1.x/lib
  • 
  • LIBS

  • Searches the directory path for library files specified by the "-l" option at link time.

• • -lxlopt
  • 
  • LIBS

  • Enables use of certain functions from the MASS library for scalars and vectors.

Fortran benchmarks

• • -q64
  • 
  • FC, FOPTIMIZE, LD

  • Generates 64-bit ABI binaries. Used as a compiler option for Fortran programs. If -q64 is not specified, the default is to generate 32-bit ABI binaries on AIX.

• • -bbigtoc
  • 
  • LDFFLAGS

  • The -bbigtoc option instructs the linker to generate extra code if the size of TOC is greater than 64K. For C/C++ programs it is invoked as -Wl,-bplugin_opt:-bbigtoc and for Fortran programs it is invoked as -bbigtoc.

• • -blpdata
  • 
  • LDFFLAGS

  • The -blpdata is a linker option and it sets the bit in the file's XCOFF header indicating that this executable will request the use of large pages when they are available on the system and when the user has an appropriate privilege. For C/C++ programs it is invoked as -Wl,-blpdata and for Fortran programs it is invoked as -blpdata.

• • -lpthread
  • 
  • LDFFLAGS

  • Link the standard Posix threading library.

• • -Wl,-bplugin_opt:-fold-complex-pointer-compare=false
  • 
  • EXTRA_LDFLAGS

  • <suboption> must be one of the following suboptions:

    • true : Enable simplification of compare instructions which rely on 'inbounds' semantics.
    • false : Disable simplification of compare instructions which rely on 'inbounds' semantics.

    Default: -fold-complex-pointer-compare=true

• • -O3
  • 
  • FOPTIMIZE
  • -O3 Instruct the compiler to execute optimizations that might take longer to perform or that may generate larger code (in an attempt to make the program run faster).
• • -qarch=auto
  • 
  • FOPTIMIZE
  • Specifies the processor architecture for which the code(instructions) should be generated.

    Supported values for this flag are :

    • auto - Automatically detects the specific architecture of the compiling machine. It assumes that the execution environment will be the same as the compilation environment.
    • pwr7 - Produces object code containing instructions that run on the Power7 hardware platform and on newer architectures.
    • pwr8 - Produces object code containing instructions that run on the Power8 hardware platform and on newer architectures.
    • pwr9 - Produces object code containing instructions that run on the Power9 hardware platform and on newer architectures.
• • -qvecnvol
  • 
  • FOPTIMIZE

  • Instructs the compiler to use both volatile and nonvolatile vector registers. Volatile vector registers are those whose value is not preserved across function calls or across save context, jump or switch context system library functions.

• • -mllvm -enable-ppc-gen-scalar-mass
  • 
  • FOPTIMIZE

  • This flag is to enable automatic use of the MASS scalar library and call scalar functions in your program.

• • -mllvm -vector-library=MASSV
  • 
  • FOPTIMIZE

  • This flag is to enable automatic use of the MASS SIMD library in your program.

• • -qlto
  • 
  • FOPTIMIZE

  • Generate output files in LLVM format, suitable for link time optimization. When used with -S this generates LLVM intermediate language assembly files, otherwise this generates LLVM bitcode format object files. The LLVM bitcode generated is suitable for Link Time Optimization (LTO), when all modules are merged into a single combined module for optimization.

• • -L/opt/IBM/xlmass/10.1.x/lib
  • 
  • LIBS

  • Searches the directory path for library files specified by the "-l" option at link time.

• • -lmass
  • 
  • LIBS

  • Link the mathematical acceleration subsystem libraries (MASS), which consist of a set of mathematical functions for C,C++ and Fortran language applications that are tuned for specific Power processor architectures.

• • -L/opt/IBM/openxlf/17.1.x/lib
  • 
  • LIBS

  • Searches the directory path for library files specified by the "-l" option at link time.

• • -lxlopt
  • 
  • LIBS

  • Enables use of certain functions from the MASS library for scalars and vectors.

Peak Optimization Flags

C benchmarks

500.perlbench_r

• • -m32
  • 
  • CC, LD

  • Generates 32-bit ABI binaries. Used as a compiler option for C/C++ programs.

• • -Wl,-bplugin_opt:--enable-ppc-gen-scalar-mass
  • 
  • LDCFLAGS

  • Enable use of scalar MASS library while using link time optimization(LTO).

• • -Wl,-bmaxdata:0xD0000000/dsa
  • 
  • LDCFLAGS

  • Causes the system loader to put the heap in its own segment of the size specified. This is only required for 32-bit applications, as their segments are 256M. A 32 bit application can be made to use a large or very large memory model. A large memory model allows up to 2GB to be used for data and a very large memory model allows up to 3.25GB to be used for data. Values above 2GB must use the dsa, or dynamic segment allocation parameter as in -bmaxdata:0xD0000000/dsa.

• • -Wl,-bplugin_opt:--ppc-set-dscr=1
  • 
  • LDCFLAGS

  • Pass on the DSCR (data stream control register) value to the linker. This value controls the aggressiveness of the hardware data prefetch engine.

• • -Wl,-bplugin_opt:-inline-hot-callsites-aggressively
  • 
  • PASS2_LDFLAGS

  • Enables aggressive heuristics to decide which frequently executed functions should be inlined.

• • -Wl,-blpdata
  • 
  • PASS1_LDCFLAGS, PASS2_LDCFLAGS

  • The -blpdata is a linker option and it sets the bit in the file's XCOFF header indicating that this executable will request the use of large pages when they are available on the system and when the user has an appropriate privilege. For C/C++ programs it is invoked as -Wl,-blpdata and for Fortran programs it is invoked as -blpdata.

• • -lpthread
  • 
  • PASS1_LDCFLAGS, PASS2_LDCFLAGS

  • Link the standard Posix threading library.

• • -O3
  • 
  • COPTIMIZE
  • -O3 Instruct the compiler to execute optimizations that might take longer to perform or that may generate larger code (in an attempt to make the program run faster).
• • -mcpu=native
  • 
  • COPTIMIZE
  • Specifies the processor architecture for which the code(instructions) should be generated.

    Supported values for this flag are :

    • native - Automatically detects the specific architecture of the compiling machine. It assumes that the execution environment will be the same as the compilation environment.
    • pwr7 - Produces object code containing instructions that run on the Power7 hardware platform and on newer architectures.
    • pwr8 - Produces object code containing instructions that run on the Power8 hardware platform and on newer architectures.
    • pwr9 - Produces object code containing instructions that run on the Power9 hardware platform and on newer architectures.
• • -mabi=vec-extabi
  • 
  • COPTIMIZE

  • Enable the extended Altivec ABI on AIX, which uses volatile and nonvolatile vector registers.

• • -flto
  • 
  • COPTIMIZE

  • Generate output files in LLVM format, suitable for link time optimization. When used with -S this generates LLVM intermediate language assembly files, otherwise this generates LLVM bitcode format object files. The LLVM bitcode generated is suitable for Link Time Optimization (LTO), when all modules are merged into a single combined module for optimization.

• • -fapprox-func
  • 
  • COPTIMIZE

  • Allows certain math function calls (such as log, sqrt, pow, etc) to be replaced with an approximately equivalent set of instructions or alternative math function calls. For example, a pow(x, 0.25) may be replaced with sqrt(sqrt(x)), despite being an inexact result in cases where x is -0.0 or -inf. Defaults to -fno-approx-func.

• • -mllvm -enable-ppc-gen-scalar-mass
  • 
  • COPTIMIZE

  • This flag is to enable automatic use of the MASS scalar library and call scalar functions in your program.

• • -mllvm -vector-library=MASSV
  • 
  • COPTIMIZE

  • This flag is to enable automatic use of the MASS SIMD library in your program.

• • -fprofile-generate
  • 
  • PASS1_COPTIMIZE

  • This option is used in the first pass of a profile guided optimization(PGO) compile and causes PGO information to be generated. The profile guided optimization gathers data on both execution path and data values. It does not use hardware counters, nor gather any data other than path and data values for PGO specific optimizations. Instructs the compiler to generate profiling instrumentation in the generated binary. This compiler option is used for C/C++ programs.

• • -fprofile-use=default.profdata
  • 
  • PASS2_COPTIMIZE

  • This option is used in the second pass of a profile guided optimization(PGO) compile. This flag instructs the compiler to use the profiling information in the file specified by the argument to -fprofile-use. This file summarizes profile information that was emitted into default*profraw files by the -fprofile-generate option. The profiling information is utilized to guide the compiler optimizer in making better optimization decisions. This compiler option is used for C/C++ programs.

• • -mcmodel=large
  • 
  • PASS1_COPTIMIZE

  • Allows the Table of Contents to be larger than 64 Kb in size, allowing more addresses to be stored in the table.

• • -fno-strict-aliasing
  • 
  • EXTRA_COPTIMIZE

  • Instructs the compiler that it must not assume that the aliasing requirements from the language standard specification are honored by the program.

• • -Wno-implicit-function-declaration
  • 
  • EXTRA_COPTIMIZE

  • This turns off diagnostics of benchmark code that is treated as errors. Specifically, it turns these error off: 1. error: call to undeclared function A; ISO C99 and later do not support implicit function declarations 2. error: call to undeclared library function 'A' with type B; ISO C99 and later do not support implicit function declarations

• • -L/opt/IBM/xlmass/10.1.x/lib
  • 
  • LIBS

  • Searches the directory path for library files specified by the "-l" option at link time.

• • -lmass
  • 
  • LIBS

  • Link the mathematical acceleration subsystem libraries (MASS), which consist of a set of mathematical functions for C,C++ and Fortran language applications that are tuned for specific Power processor architectures.

• • -L/opt/IBM/openxlf/17.1.x/lib
  • 
  • LIBS

  • Searches the directory path for library files specified by the "-l" option at link time.

• • -lxlopt
  • 
  • LIBS

  • Enables use of certain functions from the MASS library for scalars and vectors.

502.gcc_r

• • -m32
  • 
  • CC, LD

  • Generates 32-bit ABI binaries. Used as a compiler option for C/C++ programs.

• • -Wl,-bplugin_opt:--enable-ppc-gen-scalar-mass
  • 
  • LDCFLAGS

  • Enable use of scalar MASS library while using link time optimization(LTO).

• • -Wl,-bmaxdata:0x50000000
  • 
  • LDCFLAGS

  • Causes the system loader to put the heap in its own segment of the size specified. This is only required for 32-bit applications, as their segments are 256M. A 32 bit application can be made to use a large or very large memory model. A large memory model allows up to 2GB to be used for data and a very large memory model allows up to 3.25GB to be used for data. Values above 2GB must use the dsa, or dynamic segment allocation parameter as in -bmaxdata:0xD0000000/dsa.

• • -Wl,-bplugin_opt:--ppc-set-dscr=0x184
  • 
  • LDCFLAGS

  • Pass on the DSCR (data stream control register) value to the linker. This value controls the aggressiveness of the hardware data prefetch engine.

• • -Wl,-bplugin_opt:-fold-complex-pointer-compare=false
  • 
  • EXTRA_LDFLAGS

  • <suboption> must be one of the following suboptions:

    • true : Enable simplification of compare instructions which rely on 'inbounds' semantics.
    • false : Disable simplification of compare instructions which rely on 'inbounds' semantics.

    Default: -fold-complex-pointer-compare=true

• • -Wl,-blpdata
  • 
  • PASS1_LDCFLAGS, PASS2_LDCFLAGS

  • The -blpdata is a linker option and it sets the bit in the file's XCOFF header indicating that this executable will request the use of large pages when they are available on the system and when the user has an appropriate privilege. For C/C++ programs it is invoked as -Wl,-blpdata and for Fortran programs it is invoked as -blpdata.

• • -lpthread
  • 
  • PASS1_LDCFLAGS, PASS2_LDCFLAGS

  • Link the standard Posix threading library.

• • -O3
  • 
  • COPTIMIZE
  • -O3 Instruct the compiler to execute optimizations that might take longer to perform or that may generate larger code (in an attempt to make the program run faster).
• • -mcpu=native
  • 
  • COPTIMIZE
  • Specifies the processor architecture for which the code(instructions) should be generated.

    Supported values for this flag are :

    • native - Automatically detects the specific architecture of the compiling machine. It assumes that the execution environment will be the same as the compilation environment.
    • pwr7 - Produces object code containing instructions that run on the Power7 hardware platform and on newer architectures.
    • pwr8 - Produces object code containing instructions that run on the Power8 hardware platform and on newer architectures.
    • pwr9 - Produces object code containing instructions that run on the Power9 hardware platform and on newer architectures.
• • -mabi=vec-extabi
  • 
  • COPTIMIZE

  • Enable the extended Altivec ABI on AIX, which uses volatile and nonvolatile vector registers.

• • -flto
  • 
  • COPTIMIZE

  • Generate output files in LLVM format, suitable for link time optimization. When used with -S this generates LLVM intermediate language assembly files, otherwise this generates LLVM bitcode format object files. The LLVM bitcode generated is suitable for Link Time Optimization (LTO), when all modules are merged into a single combined module for optimization.

• • -fapprox-func
  • 
  • COPTIMIZE

  • Allows certain math function calls (such as log, sqrt, pow, etc) to be replaced with an approximately equivalent set of instructions or alternative math function calls. For example, a pow(x, 0.25) may be replaced with sqrt(sqrt(x)), despite being an inexact result in cases where x is -0.0 or -inf. Defaults to -fno-approx-func.

• • -mllvm -enable-ppc-gen-scalar-mass
  • 
  • COPTIMIZE

  • This flag is to enable automatic use of the MASS scalar library and call scalar functions in your program.

• • -mllvm -vector-library=MASSV
  • 
  • COPTIMIZE

  • This flag is to enable automatic use of the MASS SIMD library in your program.

• • -fprofile-generate
  • 
  • PASS1_COPTIMIZE

  • This option is used in the first pass of a profile guided optimization(PGO) compile and causes PGO information to be generated. The profile guided optimization gathers data on both execution path and data values. It does not use hardware counters, nor gather any data other than path and data values for PGO specific optimizations. Instructs the compiler to generate profiling instrumentation in the generated binary. This compiler option is used for C/C++ programs.

• • -fprofile-use=default.profdata
  • 
  • PASS2_COPTIMIZE

  • This option is used in the second pass of a profile guided optimization(PGO) compile. This flag instructs the compiler to use the profiling information in the file specified by the argument to -fprofile-use. This file summarizes profile information that was emitted into default*profraw files by the -fprofile-generate option. The profiling information is utilized to guide the compiler optimizer in making better optimization decisions. This compiler option is used for C/C++ programs.

• • -mcmodel=large
  • 
  • PASS1_COPTIMIZE

  • Allows the Table of Contents to be larger than 64 Kb in size, allowing more addresses to be stored in the table.

• • -mllvm -fold-complex-pointer-compare=false
  • 
  • EXTRA_COPTIMIZE

  • <suboption> must be one of the following suboptions:

    • true : Enable simplification of compare instructions which rely on 'inbounds' semantics.
    • false : Disable simplification of compare instructions which rely on 'inbounds' semantics.

    Default: -fold-complex-pointer-compare=true

• • -L/opt/IBM/xlmass/10.1.x/lib
  • 
  • LIBS

  • Searches the directory path for library files specified by the "-l" option at link time.

• • -lmass
  • 
  • LIBS

  • Link the mathematical acceleration subsystem libraries (MASS), which consist of a set of mathematical functions for C,C++ and Fortran language applications that are tuned for specific Power processor architectures.

• • -L/opt/IBM/openxlf/17.1.x/lib
  • 
  • LIBS

  • Searches the directory path for library files specified by the "-l" option at link time.

• • -lxlopt
  • 
  • LIBS

  • Enables use of certain functions from the MASS library for scalars and vectors.

505.mcf_r

• • -m64
  • 
  • CC, LD

  • Generates 64-bit ABI binaries. Used as a compiler option for C/C++ programs. If -m64 is not specified, the default is to generate 32-bit ABI binaries on AIX.

• • -Wl,-bplugin_opt:--enable-ppc-gen-scalar-mass
  • 
  • LDCFLAGS

  • Enable use of scalar MASS library while using link time optimization(LTO).

• • -Wl,-bplugin_opt:--ppc-set-dscr=0x1c7
  • 
  • LDCFLAGS

  • Pass on the DSCR (data stream control register) value to the linker. This value controls the aggressiveness of the hardware data prefetch engine.

• • -Wl,-bplugin_opt:--data-layout-opt=3
  • 
  • PASS2_LDFLAGS

  • This option instructs the linker to analyze the whole program to determine if the layout of data can be transformed to improve the cache utilization and memory bandwidth of the program. This option has four levels [0, 1, 2, 3]. This is effective only under -flto and must be specified on both the compilation and linking steps with same option level.

    • data-layout-opt=0 disables the data layout transformation. No compiler analysis is performed for data layout optimization.
    • data-layout-opt=1 enables the data layout transformation with whole program analysis. This is performed with a strict safety analysis.
    • data-layout-opt=2 enables the data layout transformation with whole program more aggressive analysis. This is performed with a strict safety analysis.
    • data-layout-opt=3 enables the data layout transformation and data compression with whole program analysis. The compiler may implicitly change the default data type size to a non-default data type size.
• • -Wl,-blpdata
  • 
  • PASS1_LDCFLAGS, PASS2_LDCFLAGS

  • The -blpdata is a linker option and it sets the bit in the file's XCOFF header indicating that this executable will request the use of large pages when they are available on the system and when the user has an appropriate privilege. For C/C++ programs it is invoked as -Wl,-blpdata and for Fortran programs it is invoked as -blpdata.

• • -lpthread
  • 
  • PASS1_LDCFLAGS, PASS2_LDCFLAGS

  • Link the standard Posix threading library.

• • -O3
  • 
  • COPTIMIZE
  • -O3 Instruct the compiler to execute optimizations that might take longer to perform or that may generate larger code (in an attempt to make the program run faster).
• • -mcpu=native
  • 
  • COPTIMIZE
  • Specifies the processor architecture for which the code(instructions) should be generated.

    Supported values for this flag are :

    • native - Automatically detects the specific architecture of the compiling machine. It assumes that the execution environment will be the same as the compilation environment.
    • pwr7 - Produces object code containing instructions that run on the Power7 hardware platform and on newer architectures.
    • pwr8 - Produces object code containing instructions that run on the Power8 hardware platform and on newer architectures.
    • pwr9 - Produces object code containing instructions that run on the Power9 hardware platform and on newer architectures.
• • -mabi=vec-extabi
  • 
  • COPTIMIZE

  • Enable the extended Altivec ABI on AIX, which uses volatile and nonvolatile vector registers.

• • -flto
  • 
  • COPTIMIZE

  • Generate output files in LLVM format, suitable for link time optimization. When used with -S this generates LLVM intermediate language assembly files, otherwise this generates LLVM bitcode format object files. The LLVM bitcode generated is suitable for Link Time Optimization (LTO), when all modules are merged into a single combined module for optimization.

• • -fapprox-func
  • 
  • COPTIMIZE

  • Allows certain math function calls (such as log, sqrt, pow, etc) to be replaced with an approximately equivalent set of instructions or alternative math function calls. For example, a pow(x, 0.25) may be replaced with sqrt(sqrt(x)), despite being an inexact result in cases where x is -0.0 or -inf. Defaults to -fno-approx-func.

• • -mllvm -enable-ppc-gen-scalar-mass
  • 
  • COPTIMIZE

  • This flag is to enable automatic use of the MASS scalar library and call scalar functions in your program.

• • -mllvm -vector-library=MASSV
  • 
  • COPTIMIZE

  • This flag is to enable automatic use of the MASS SIMD library in your program.

• • -fprofile-generate
  • 
  • PASS1_COPTIMIZE

  • This option is used in the first pass of a profile guided optimization(PGO) compile and causes PGO information to be generated. The profile guided optimization gathers data on both execution path and data values. It does not use hardware counters, nor gather any data other than path and data values for PGO specific optimizations. Instructs the compiler to generate profiling instrumentation in the generated binary. This compiler option is used for C/C++ programs.

• • -fprofile-use=default.profdata
  • 
  • PASS2_COPTIMIZE

  • This option is used in the second pass of a profile guided optimization(PGO) compile. This flag instructs the compiler to use the profiling information in the file specified by the argument to -fprofile-use. This file summarizes profile information that was emitted into default*profraw files by the -fprofile-generate option. The profiling information is utilized to guide the compiler optimizer in making better optimization decisions. This compiler option is used for C/C++ programs.

• • -mcmodel=large
  • 
  • PASS1_COPTIMIZE

  • Allows the Table of Contents to be larger than 64 Kb in size, allowing more addresses to be stored in the table.

• • -mllvm -data-layout-opt=3
  • 
  • EXTRA_COPTIMIZE

  • This option instructs the compiler to analyze the whole program to determine if the layout of data can be transformed to improve the cache utilization and m emory bandwidth of the program. This option has four levels [0, 1, 2, 3]. This is effective only under -flto and must be specified on both the compilation an d linking steps with same option level.

    • data-layout-opt=0 disables the data layout transformation. No compiler analysis is performed for data layout optimization.
    • data-layout-opt=1 enables the data layout transformation with whole program analysis. This is performed with a strict safety analysis.
    • data-layout-opt=2 enables the data layout transformation with whole program more aggressive analysis. This is performed with a strict safety ana lysis.
    • data-layout-opt=3 enables the data layout transformation and data compression with whole program analysis. The compiler may implicitly change th e default data type size to a non-default data type size.
• • -L/opt/IBM/xlmass/10.1.x/lib
  • 
  • LIBS

  • Searches the directory path for library files specified by the "-l" option at link time.

• • -lmass
  • 
  • LIBS

  • Link the mathematical acceleration subsystem libraries (MASS), which consist of a set of mathematical functions for C,C++ and Fortran language applications that are tuned for specific Power processor architectures.

• • -L/opt/IBM/openxlf/17.1.x/lib
  • 
  • LIBS

  • Searches the directory path for library files specified by the "-l" option at link time.

• • -lxlopt
  • 
  • LIBS

  • Enables use of certain functions from the MASS library for scalars and vectors.

525.x264_r

• • -m64
  • 
  • CC, LD

  • Generates 64-bit ABI binaries. Used as a compiler option for C/C++ programs. If -m64 is not specified, the default is to generate 32-bit ABI binaries on AIX.

• • -Wl,-bplugin_opt:--ppc-set-dscr=0x80
  • 
  • LDFLAGS

  • Pass on the DSCR (data stream control register) value to the linker. This value controls the aggressiveness of the hardware data prefetch engine.

• • -Wl,-bplugin_opt:--enable-ppc-gen-scalar-mass
  • 
  • LDCFLAGS

  • Enable use of scalar MASS library while using link time optimization(LTO).

• • -Wl,-blpdata
  • 
  • PASS1_LDCFLAGS, PASS2_LDCFLAGS

  • The -blpdata is a linker option and it sets the bit in the file's XCOFF header indicating that this executable will request the use of large pages when they are available on the system and when the user has an appropriate privilege. For C/C++ programs it is invoked as -Wl,-blpdata and for Fortran programs it is invoked as -blpdata.

• • -lpthread
  • 
  • PASS1_LDCFLAGS, PASS2_LDCFLAGS

  • Link the standard Posix threading library.

• • -Wl,-bplugin_opt:-enable-aggressive-vectorization
  • 
  • PASS2_LDCFLAGS

  • Instructs the linker to enable aggressive heuristics for loop vectorization.

• • -O3
  • 
  • COPTIMIZE
  • -O3 Instruct the compiler to execute optimizations that might take longer to perform or that may generate larger code (in an attempt to make the program run faster).
• • -mcpu=native
  • 
  • COPTIMIZE
  • Specifies the processor architecture for which the code(instructions) should be generated.

    Supported values for this flag are :

    • native - Automatically detects the specific architecture of the compiling machine. It assumes that the execution environment will be the same as the compilation environment.
    • pwr7 - Produces object code containing instructions that run on the Power7 hardware platform and on newer architectures.
    • pwr8 - Produces object code containing instructions that run on the Power8 hardware platform and on newer architectures.
    • pwr9 - Produces object code containing instructions that run on the Power9 hardware platform and on newer architectures.
• • -mabi=vec-extabi
  • 
  • COPTIMIZE

  • Enable the extended Altivec ABI on AIX, which uses volatile and nonvolatile vector registers.

• • -flto
  • 
  • COPTIMIZE

  • Generate output files in LLVM format, suitable for link time optimization. When used with -S this generates LLVM intermediate language assembly files, otherwise this generates LLVM bitcode format object files. The LLVM bitcode generated is suitable for Link Time Optimization (LTO), when all modules are merged into a single combined module for optimization.

• • -fapprox-func
  • 
  • COPTIMIZE

  • Allows certain math function calls (such as log, sqrt, pow, etc) to be replaced with an approximately equivalent set of instructions or alternative math function calls. For example, a pow(x, 0.25) may be replaced with sqrt(sqrt(x)), despite being an inexact result in cases where x is -0.0 or -inf. Defaults to -fno-approx-func.

• • -mllvm -enable-ppc-gen-scalar-mass
  • 
  • COPTIMIZE

  • This flag is to enable automatic use of the MASS scalar library and call scalar functions in your program.

• • -mllvm -vector-library=MASSV
  • 
  • COPTIMIZE

  • This flag is to enable automatic use of the MASS SIMD library in your program.

• • -fprofile-generate
  • 
  • PASS1_COPTIMIZE

  • This option is used in the first pass of a profile guided optimization(PGO) compile and causes PGO information to be generated. The profile guided optimization gathers data on both execution path and data values. It does not use hardware counters, nor gather any data other than path and data values for PGO specific optimizations. Instructs the compiler to generate profiling instrumentation in the generated binary. This compiler option is used for C/C++ programs.

• • -fprofile-use=default.profdata
  • 
  • PASS2_COPTIMIZE

  • This option is used in the second pass of a profile guided optimization(PGO) compile. This flag instructs the compiler to use the profiling information in the file specified by the argument to -fprofile-use. This file summarizes profile information that was emitted into default*profraw files by the -fprofile-generate option. The profiling information is utilized to guide the compiler optimizer in making better optimization decisions. This compiler option is used for C/C++ programs.

• • -mcmodel=large
  • 
  • PASS1_COPTIMIZE

  • Allows the Table of Contents to be larger than 64 Kb in size, allowing more addresses to be stored in the table.

• • -fcommon
  • 
  • EXTRA_COPTIMIZE

  • Place uninitialized global variables in a common block.

• • -frestrict-args
  • 
  • EXTRA_COPTIMIZE

  • Allows the compiler to assume that function parameters of pointer type are restrict qualified.

• • -mllvm -enable-aggressive-vectorization
  • 
  • EXTRA_COPTIMIZE

  • Instructs the compiler to enable aggressive heuristics for loop vectorization.

• • -mllvm -ppc-enable-redxnintr
  • 
  • EXTRA_COPTIMIZE

  • Allow the compiler to generate PPC hardware reduction instructions for reductions without proving signed overflow safety.

• • -L/opt/IBM/xlmass/10.1.x/lib
  • 
  • LIBS

  • Searches the directory path for library files specified by the "-l" option at link time.

• • -lmass
  • 
  • LIBS

  • Link the mathematical acceleration subsystem libraries (MASS), which consist of a set of mathematical functions for C,C++ and Fortran language applications that are tuned for specific Power processor architectures.

• • -L/opt/IBM/openxlf/17.1.x/lib
  • 
  • LIBS

  • Searches the directory path for library files specified by the "-l" option at link time.

• • -lxlopt
  • 
  • LIBS

  • Enables use of certain functions from the MASS library for scalars and vectors.

557.xz_r

• • -m64
  • 
  • CC, LD

  • Generates 64-bit ABI binaries. Used as a compiler option for C/C++ programs. If -m64 is not specified, the default is to generate 32-bit ABI binaries on AIX.

• • -Wl,-bplugin_opt:--enable-ppc-gen-scalar-mass
  • 
  • LDCFLAGS

  • Enable use of scalar MASS library while using link time optimization(LTO).

• • -Wl,-bplugin_opt:-aggressive-late-full-unroll
  • 
  • PASS2_LDFLAGS

  • Instructs the linker to use more aggressive heuristics for the loop unroll optimization when considering whether it is beneficial to fully unroll a loop.

• • -Wl,-blpdata
  • 
  • PASS1_LDCFLAGS, PASS2_LDCFLAGS

  • The -blpdata is a linker option and it sets the bit in the file's XCOFF header indicating that this executable will request the use of large pages when they are available on the system and when the user has an appropriate privilege. For C/C++ programs it is invoked as -Wl,-blpdata and for Fortran programs it is invoked as -blpdata.

• • -lpthread
  • 
  • PASS1_LDCFLAGS, PASS2_LDCFLAGS

  • Link the standard Posix threading library.

• • -mllvm -aggressive-late-full-unroll
  • 
  • OPTIMIZE

  • Instructs the compiler optimizer to use more aggressive heuristics for the loop unroll optimization when considering whether it is beneficial to fully unroll a loop.

• • -mllvm -unroll-threshold=500
  • 
  • OPTIMIZE

  • Control how many loops get completely unrolled. For example, if unroll threshold is set to 500 then only loops with 500 or fewer instructions will be unrolled.

• • -mllvm -inline-threshold=2000
  • 
  • OPTIMIZE

  • Control the amount of inlining to the value passed as the argument. The inline threshold is used in the inliner heuristics to decide which functions should be inlined.

• • -O3
  • 
  • COPTIMIZE
  • -O3 Instruct the compiler to execute optimizations that might take longer to perform or that may generate larger code (in an attempt to make the program run faster).
• • -mcpu=native
  • 
  • COPTIMIZE
  • Specifies the processor architecture for which the code(instructions) should be generated.

    Supported values for this flag are :

    • native - Automatically detects the specific architecture of the compiling machine. It assumes that the execution environment will be the same as the compilation environment.
    • pwr7 - Produces object code containing instructions that run on the Power7 hardware platform and on newer architectures.
    • pwr8 - Produces object code containing instructions that run on the Power8 hardware platform and on newer architectures.
    • pwr9 - Produces object code containing instructions that run on the Power9 hardware platform and on newer architectures.
• • -mabi=vec-extabi
  • 
  • COPTIMIZE

  • Enable the extended Altivec ABI on AIX, which uses volatile and nonvolatile vector registers.

• • -flto
  • 
  • COPTIMIZE

  • Generate output files in LLVM format, suitable for link time optimization. When used with -S this generates LLVM intermediate language assembly files, otherwise this generates LLVM bitcode format object files. The LLVM bitcode generated is suitable for Link Time Optimization (LTO), when all modules are merged into a single combined module for optimization.

• • -fapprox-func
  • 
  • COPTIMIZE

  • Allows certain math function calls (such as log, sqrt, pow, etc) to be replaced with an approximately equivalent set of instructions or alternative math function calls. For example, a pow(x, 0.25) may be replaced with sqrt(sqrt(x)), despite being an inexact result in cases where x is -0.0 or -inf. Defaults to -fno-approx-func.

• • -mllvm -enable-ppc-gen-scalar-mass
  • 
  • COPTIMIZE

  • This flag is to enable automatic use of the MASS scalar library and call scalar functions in your program.

• • -mllvm -vector-library=MASSV
  • 
  • COPTIMIZE

  • This flag is to enable automatic use of the MASS SIMD library in your program.

• • -fprofile-generate
  • 
  • PASS1_COPTIMIZE

  • This option is used in the first pass of a profile guided optimization(PGO) compile and causes PGO information to be generated. The profile guided optimization gathers data on both execution path and data values. It does not use hardware counters, nor gather any data other than path and data values for PGO specific optimizations. Instructs the compiler to generate profiling instrumentation in the generated binary. This compiler option is used for C/C++ programs.

• • -fprofile-use=default.profdata
  • 
  • PASS2_COPTIMIZE

  • This option is used in the second pass of a profile guided optimization(PGO) compile. This flag instructs the compiler to use the profiling information in the file specified by the argument to -fprofile-use. This file summarizes profile information that was emitted into default*profraw files by the -fprofile-generate option. The profiling information is utilized to guide the compiler optimizer in making better optimization decisions. This compiler option is used for C/C++ programs.

• • -mcmodel=large
  • 
  • PASS1_COPTIMIZE

  • Allows the Table of Contents to be larger than 64 Kb in size, allowing more addresses to be stored in the table.

• • -L/opt/IBM/xlmass/10.1.x/lib
  • 
  • LIBS

  • Searches the directory path for library files specified by the "-l" option at link time.

• • -lmass
  • 
  • LIBS

  • Link the mathematical acceleration subsystem libraries (MASS), which consist of a set of mathematical functions for C,C++ and Fortran language applications that are tuned for specific Power processor architectures.

• • -L/opt/IBM/openxlf/17.1.x/lib
  • 
  • LIBS

  • Searches the directory path for library files specified by the "-l" option at link time.

• • -lxlopt
  • 
  • LIBS

  • Enables use of certain functions from the MASS library for scalars and vectors.

C++ benchmarks

520.omnetpp_r

• • -m32
  • 
  • CXX, LD

  • Generates 32-bit ABI binaries. Used as a compiler option for C/C++ programs.

• • -Wl,-bplugin_opt:--enable-ppc-gen-scalar-mass
  • 
  • LDCXXFLAGS

  • Enable use of scalar MASS library while using link time optimization(LTO).

• • -Wl,-bmaxdata:0x30000000
  • 
  • LDCXXFLAGS

  • Causes the system loader to put the heap in its own segment of the size specified. This is only required for 32-bit applications, as their segments are 256M. A 32 bit application can be made to use a large or very large memory model. A large memory model allows up to 2GB to be used for data and a very large memory model allows up to 3.25GB to be used for data. Values above 2GB must use the dsa, or dynamic segment allocation parameter as in -bmaxdata:0xD0000000/dsa.

• • -Wl,-bplugin_opt:--ppc-set-dscr=1
  • 
  • LDCXXFLAGS

  • Pass on the DSCR (data stream control register) value to the linker. This value controls the aggressiveness of the hardware data prefetch engine.

• • -Wl,-blpdata
  • 
  • PASS1_LDCXXFLAGS, PASS2_LDCXXFLAGS

  • The -blpdata is a linker option and it sets the bit in the file's XCOFF header indicating that this executable will request the use of large pages when they are available on the system and when the user has an appropriate privilege. For C/C++ programs it is invoked as -Wl,-blpdata and for Fortran programs it is invoked as -blpdata.

• • -lpthread
  • 
  • PASS1_LDCXXFLAGS, PASS2_LDCXXFLAGS

  • Link the standard Posix threading library.

• • -O3
  • 
  • CXXOPTIMIZE
  • -O3 Instruct the compiler to execute optimizations that might take longer to perform or that may generate larger code (in an attempt to make the program run faster).
• • -mcpu=native
  • 
  • CXXOPTIMIZE
  • Specifies the processor architecture for which the code(instructions) should be generated.

    Supported values for this flag are :

    • native - Automatically detects the specific architecture of the compiling machine. It assumes that the execution environment will be the same as the compilation environment.
    • pwr7 - Produces object code containing instructions that run on the Power7 hardware platform and on newer architectures.
    • pwr8 - Produces object code containing instructions that run on the Power8 hardware platform and on newer architectures.
    • pwr9 - Produces object code containing instructions that run on the Power9 hardware platform and on newer architectures.
• • -mabi=vec-extabi
  • 
  • CXXOPTIMIZE

  • Enable the extended Altivec ABI on AIX, which uses volatile and nonvolatile vector registers.

• • -flto
  • 
  • CXXOPTIMIZE

  • Generate output files in LLVM format, suitable for link time optimization. When used with -S this generates LLVM intermediate language assembly files, otherwise this generates LLVM bitcode format object files. The LLVM bitcode generated is suitable for Link Time Optimization (LTO), when all modules are merged into a single combined module for optimization.

• • -fapprox-func
  • 
  • CXXOPTIMIZE

  • Allows certain math function calls (such as log, sqrt, pow, etc) to be replaced with an approximately equivalent set of instructions or alternative math function calls. For example, a pow(x, 0.25) may be replaced with sqrt(sqrt(x)), despite being an inexact result in cases where x is -0.0 or -inf. Defaults to -fno-approx-func.

• • -mllvm -enable-ppc-gen-scalar-mass
  • 
  • CXXOPTIMIZE

  • This flag is to enable automatic use of the MASS scalar library and call scalar functions in your program.

• • -mllvm -vector-library=MASSV
  • 
  • CXXOPTIMIZE

  • This flag is to enable automatic use of the MASS SIMD library in your program.

• • -fprofile-generate
  • 
  • PASS1_CXXOPTIMIZE

  • This option is used in the first pass of a profile guided optimization(PGO) compile and causes PGO information to be generated. The profile guided optimization gathers data on both execution path and data values. It does not use hardware counters, nor gather any data other than path and data values for PGO specific optimizations. Instructs the compiler to generate profiling instrumentation in the generated binary. This compiler option is used for C/C++ programs.

• • -fprofile-use=default.profdata
  • 
  • PASS2_CXXOPTIMIZE

  • This option is used in the second pass of a profile guided optimization(PGO) compile. This flag instructs the compiler to use the profiling information in the file specified by the argument to -fprofile-use. This file summarizes profile information that was emitted into default*profraw files by the -fprofile-generate option. The profiling information is utilized to guide the compiler optimizer in making better optimization decisions. This compiler option is used for C/C++ programs.

• • -mcmodel=large
  • 
  • PASS1_CXXOPTIMIZE

  • Allows the Table of Contents to be larger than 64 Kb in size, allowing more addresses to be stored in the table.

• • -Wl,-bplugin_opt:-dynamic-cast-opt=on
  • 
  • PASS2_CXXOPTIMIZE

  • Tries to convert __dynamic_cast function call into an address comparision when possible. The default value is off.

• • -L/opt/IBM/xlmass/10.1.x/lib
  • 
  • LIBS

  • Searches the directory path for library files specified by the "-l" option at link time.

• • -lmass
  • 
  • LIBS

  • Link the mathematical acceleration subsystem libraries (MASS), which consist of a set of mathematical functions for C,C++ and Fortran language applications that are tuned for specific Power processor architectures.

• • -L/opt/IBM/openxlf/17.1.x/lib
  • 
  • LIBS

  • Searches the directory path for library files specified by the "-l" option at link time.

• • -lxlopt
  • 
  • LIBS

  • Enables use of certain functions from the MASS library for scalars and vectors.

523.xalancbmk_r

• • -m32
  • 
  • CXX, LD

  • Generates 32-bit ABI binaries. Used as a compiler option for C/C++ programs.

• • -Wl,-bplugin_opt:--ppc-set-dscr=0x80
  • 
  • LDFLAGS

  • Pass on the DSCR (data stream control register) value to the linker. This value controls the aggressiveness of the hardware data prefetch engine.

• • -Wl,-bplugin_opt:--enable-ppc-gen-scalar-mass
  • 
  • LDCXXFLAGS

  • Enable use of scalar MASS library while using link time optimization(LTO).

• • -Wl,-bmaxdata:0x30000000
  • 
  • LDCXXFLAGS

  • Causes the system loader to put the heap in its own segment of the size specified. This is only required for 32-bit applications, as their segments are 256M. A 32 bit application can be made to use a large or very large memory model. A large memory model allows up to 2GB to be used for data and a very large memory model allows up to 3.25GB to be used for data. Values above 2GB must use the dsa, or dynamic segment allocation parameter as in -bmaxdata:0xD0000000/dsa.

• • -Wl,-bplugin_opt:-inline-hot-callsites-aggressively
  • 
  • PASS2_LDFLAGS

  • Enables aggressive heuristics to decide which frequently executed functions should be inlined.

• • -Wl,-bplugin_opt:-enable-partial-inlining
  • 
  • PASS2_LDFLAGS

  • Instructs the linker to enable the partial inlining optimization.

• • -Wl,-blpdata
  • 
  • PASS1_LDCXXFLAGS, PASS2_LDCXXFLAGS

  • The -blpdata is a linker option and it sets the bit in the file's XCOFF header indicating that this executable will request the use of large pages when they are available on the system and when the user has an appropriate privilege. For C/C++ programs it is invoked as -Wl,-blpdata and for Fortran programs it is invoked as -blpdata.

• • -lpthread
  • 
  • PASS1_LDCXXFLAGS, PASS2_LDCXXFLAGS

  • Link the standard Posix threading library.

• • -mllvm -enable-partial-inlining
  • 
  • OPTIMIZE

  • Instructs the compiler to enable the partial inlining optimization.

• • -mllvm -enable-vec-find=true
  • 
  • OPTIMIZE

  • Allow the compiler to vectorize simple search loops if they operate on contiguous arrays.

• • -O3
  • 
  • CXXOPTIMIZE
  • -O3 Instruct the compiler to execute optimizations that might take longer to perform or that may generate larger code (in an attempt to make the program run faster).
• • -mcpu=native
  • 
  • CXXOPTIMIZE
  • Specifies the processor architecture for which the code(instructions) should be generated.

    Supported values for this flag are :

    • native - Automatically detects the specific architecture of the compiling machine. It assumes that the execution environment will be the same as the compilation environment.
    • pwr7 - Produces object code containing instructions that run on the Power7 hardware platform and on newer architectures.
    • pwr8 - Produces object code containing instructions that run on the Power8 hardware platform and on newer architectures.
    • pwr9 - Produces object code containing instructions that run on the Power9 hardware platform and on newer architectures.
• • -mabi=vec-extabi
  • 
  • CXXOPTIMIZE

  • Enable the extended Altivec ABI on AIX, which uses volatile and nonvolatile vector registers.

• • -flto
  • 
  • CXXOPTIMIZE

  • Generate output files in LLVM format, suitable for link time optimization. When used with -S this generates LLVM intermediate language assembly files, otherwise this generates LLVM bitcode format object files. The LLVM bitcode generated is suitable for Link Time Optimization (LTO), when all modules are merged into a single combined module for optimization.

• • -fapprox-func
  • 
  • CXXOPTIMIZE

  • Allows certain math function calls (such as log, sqrt, pow, etc) to be replaced with an approximately equivalent set of instructions or alternative math function calls. For example, a pow(x, 0.25) may be replaced with sqrt(sqrt(x)), despite being an inexact result in cases where x is -0.0 or -inf. Defaults to -fno-approx-func.

• • -mllvm -enable-ppc-gen-scalar-mass
  • 
  • CXXOPTIMIZE

  • This flag is to enable automatic use of the MASS scalar library and call scalar functions in your program.

• • -mllvm -vector-library=MASSV
  • 
  • CXXOPTIMIZE

  • This flag is to enable automatic use of the MASS SIMD library in your program.

• • -fprofile-generate
  • 
  • PASS1_CXXOPTIMIZE

  • This option is used in the first pass of a profile guided optimization(PGO) compile and causes PGO information to be generated. The profile guided optimization gathers data on both execution path and data values. It does not use hardware counters, nor gather any data other than path and data values for PGO specific optimizations. Instructs the compiler to generate profiling instrumentation in the generated binary. This compiler option is used for C/C++ programs.

• • -fprofile-use=default.profdata
  • 
  • PASS2_CXXOPTIMIZE

  • This option is used in the second pass of a profile guided optimization(PGO) compile. This flag instructs the compiler to use the profiling information in the file specified by the argument to -fprofile-use. This file summarizes profile information that was emitted into default*profraw files by the -fprofile-generate option. The profiling information is utilized to guide the compiler optimizer in making better optimization decisions. This compiler option is used for C/C++ programs.

• • -mcmodel=large
  • 
  • PASS1_CXXOPTIMIZE

  • Allows the Table of Contents to be larger than 64 Kb in size, allowing more addresses to be stored in the table.

• • -L/opt/IBM/xlmass/10.1.x/lib
  • 
  • LIBS

  • Searches the directory path for library files specified by the "-l" option at link time.

• • -lmass
  • 
  • LIBS

  • Link the mathematical acceleration subsystem libraries (MASS), which consist of a set of mathematical functions for C,C++ and Fortran language applications that are tuned for specific Power processor architectures.

• • -L/opt/IBM/openxlf/17.1.x/lib
  • 
  • LIBS

  • Searches the directory path for library files specified by the "-l" option at link time.

• • -lxlopt
  • 
  • LIBS

  • Enables use of certain functions from the MASS library for scalars and vectors.

531.deepsjeng_r

• • -m64
  • 
  • CXX, LD

  • Generates 64-bit ABI binaries. Used as a compiler option for C/C++ programs. If -m64 is not specified, the default is to generate 32-bit ABI binaries on AIX.

• • -Wl,-bplugin_opt:--ppc-set-dscr=1
  • 
  • LDFLAGS

  • Pass on the DSCR (data stream control register) value to the linker. This value controls the aggressiveness of the hardware data prefetch engine.

• • -Wl,-bplugin_opt:--enable-ppc-gen-scalar-mass
  • 
  • LDCXXFLAGS

  • Enable use of scalar MASS library while using link time optimization(LTO).

• • -Wl,-bplugin_opt:-inline-hot-callsites-aggressively
  • 
  • PASS2_LDFLAGS

  • Enables aggressive heuristics to decide which frequently executed functions should be inlined.

• • -Wl,-blpdata
  • 
  • PASS1_LDCXXFLAGS, PASS2_LDCXXFLAGS

  • The -blpdata is a linker option and it sets the bit in the file's XCOFF header indicating that this executable will request the use of large pages when they are available on the system and when the user has an appropriate privilege. For C/C++ programs it is invoked as -Wl,-blpdata and for Fortran programs it is invoked as -blpdata.

• • -lpthread
  • 
  • PASS1_LDCXXFLAGS, PASS2_LDCXXFLAGS

  • Link the standard Posix threading library.

• • -O3
  • 
  • CXXOPTIMIZE
  • -O3 Instruct the compiler to execute optimizations that might take longer to perform or that may generate larger code (in an attempt to make the program run faster).
• • -mcpu=native
  • 
  • CXXOPTIMIZE
  • Specifies the processor architecture for which the code(instructions) should be generated.

    Supported values for this flag are :

    • native - Automatically detects the specific architecture of the compiling machine. It assumes that the execution environment will be the same as the compilation environment.
    • pwr7 - Produces object code containing instructions that run on the Power7 hardware platform and on newer architectures.
    • pwr8 - Produces object code containing instructions that run on the Power8 hardware platform and on newer architectures.
    • pwr9 - Produces object code containing instructions that run on the Power9 hardware platform and on newer architectures.
• • -mabi=vec-extabi
  • 
  • CXXOPTIMIZE

  • Enable the extended Altivec ABI on AIX, which uses volatile and nonvolatile vector registers.

• • -flto
  • 
  • CXXOPTIMIZE

  • Generate output files in LLVM format, suitable for link time optimization. When used with -S this generates LLVM intermediate language assembly files, otherwise this generates LLVM bitcode format object files. The LLVM bitcode generated is suitable for Link Time Optimization (LTO), when all modules are merged into a single combined module for optimization.

• • -fapprox-func
  • 
  • CXXOPTIMIZE

  • Allows certain math function calls (such as log, sqrt, pow, etc) to be replaced with an approximately equivalent set of instructions or alternative math function calls. For example, a pow(x, 0.25) may be replaced with sqrt(sqrt(x)), despite being an inexact result in cases where x is -0.0 or -inf. Defaults to -fno-approx-func.

• • -mllvm -enable-ppc-gen-scalar-mass
  • 
  • CXXOPTIMIZE

  • This flag is to enable automatic use of the MASS scalar library and call scalar functions in your program.

• • -mllvm -vector-library=MASSV
  • 
  • CXXOPTIMIZE

  • This flag is to enable automatic use of the MASS SIMD library in your program.

• • -fprofile-generate
  • 
  • PASS1_CXXOPTIMIZE

  • This option is used in the first pass of a profile guided optimization(PGO) compile and causes PGO information to be generated. The profile guided optimization gathers data on both execution path and data values. It does not use hardware counters, nor gather any data other than path and data values for PGO specific optimizations. Instructs the compiler to generate profiling instrumentation in the generated binary. This compiler option is used for C/C++ programs.

• • -fprofile-use=default.profdata
  • 
  • PASS2_CXXOPTIMIZE

  • This option is used in the second pass of a profile guided optimization(PGO) compile. This flag instructs the compiler to use the profiling information in the file specified by the argument to -fprofile-use. This file summarizes profile information that was emitted into default*profraw files by the -fprofile-generate option. The profiling information is utilized to guide the compiler optimizer in making better optimization decisions. This compiler option is used for C/C++ programs.

• • -mcmodel=large
  • 
  • PASS1_CXXOPTIMIZE

  • Allows the Table of Contents to be larger than 64 Kb in size, allowing more addresses to be stored in the table.

• • -L/opt/IBM/xlmass/10.1.x/lib
  • 
  • LIBS

  • Searches the directory path for library files specified by the "-l" option at link time.

• • -lmass
  • 
  • LIBS

  • Link the mathematical acceleration subsystem libraries (MASS), which consist of a set of mathematical functions for C,C++ and Fortran language applications that are tuned for specific Power processor architectures.

• • -L/opt/IBM/openxlf/17.1.x/lib
  • 
  • LIBS

  • Searches the directory path for library files specified by the "-l" option at link time.

• • -lxlopt
  • 
  • LIBS

  • Enables use of certain functions from the MASS library for scalars and vectors.

541.leela_r

• • -m64
  • 
  • CXX, LD

  • Generates 64-bit ABI binaries. Used as a compiler option for C/C++ programs. If -m64 is not specified, the default is to generate 32-bit ABI binaries on AIX.

• • -Wl,-bplugin_opt:--enable-ppc-gen-scalar-mass
  • 
  • LDCXXFLAGS

  • Enable use of scalar MASS library while using link time optimization(LTO).

• • -Wl,-bplugin_opt:-inline-hot-callsites-aggressively
  • 
  • PASS2_LDFLAGS

  • Enables aggressive heuristics to decide which frequently executed functions should be inlined.

• • -Wl,-bplugin_opt:-unroll-runtime=false
  • 
  • PASS2_LDFLAGS

  • Instructs the linker to unroll loops with run-time trip counts. When set to false only unroll the loop if we are able to calculate the trip count at compile time.

• • -Wl,-blpdata
  • 
  • PASS1_LDCXXFLAGS, PASS2_LDCXXFLAGS

  • The -blpdata is a linker option and it sets the bit in the file's XCOFF header indicating that this executable will request the use of large pages when they are available on the system and when the user has an appropriate privilege. For C/C++ programs it is invoked as -Wl,-blpdata and for Fortran programs it is invoked as -blpdata.

• • -lpthread
  • 
  • PASS1_LDCXXFLAGS, PASS2_LDCXXFLAGS

  • Link the standard Posix threading library.

• • -mllvm -aggressive-late-full-unroll
  • 
  • OPTIMIZE

  • Instructs the compiler optimizer to use more aggressive heuristics for the loop unroll optimization when considering whether it is beneficial to fully unroll a loop.

• • -mllvm -array-compress=true
  • 
  • OPTIMIZE

  • Instructs the compiler optimizer to compress a local array of integers into a smaller integer type when it is legal to do so (i.e. when all possible values stored by the program into the array elements are representable by a smaller integer type).

• • -mllvm -enable-lvi-memoryssa=true
  • 
  • OPTIMIZE

  • Instructs the compiler optimizer to use an advanced value tracking analysis that leverages LLVM's "memory ssa" representation to track values stored into memory locations.

• • -mllvm -unroll-runtime=false
  • 
  • OPTIMIZE

  • Unroll loops with run-time trip counts. When set to false only unroll the loop if we are able to calculate the trip count at compile time.

• • -O3
  • 
  • CXXOPTIMIZE
  • -O3 Instruct the compiler to execute optimizations that might take longer to perform or that may generate larger code (in an attempt to make the program run faster).
• • -mcpu=native
  • 
  • CXXOPTIMIZE
  • Specifies the processor architecture for which the code(instructions) should be generated.

    Supported values for this flag are :

    • native - Automatically detects the specific architecture of the compiling machine. It assumes that the execution environment will be the same as the compilation environment.
    • pwr7 - Produces object code containing instructions that run on the Power7 hardware platform and on newer architectures.
    • pwr8 - Produces object code containing instructions that run on the Power8 hardware platform and on newer architectures.
    • pwr9 - Produces object code containing instructions that run on the Power9 hardware platform and on newer architectures.
• • -mabi=vec-extabi
  • 
  • CXXOPTIMIZE

  • Enable the extended Altivec ABI on AIX, which uses volatile and nonvolatile vector registers.

• • -flto
  • 
  • CXXOPTIMIZE

  • Generate output files in LLVM format, suitable for link time optimization. When used with -S this generates LLVM intermediate language assembly files, otherwise this generates LLVM bitcode format object files. The LLVM bitcode generated is suitable for Link Time Optimization (LTO), when all modules are merged into a single combined module for optimization.

• • -fapprox-func
  • 
  • CXXOPTIMIZE

  • Allows certain math function calls (such as log, sqrt, pow, etc) to be replaced with an approximately equivalent set of instructions or alternative math function calls. For example, a pow(x, 0.25) may be replaced with sqrt(sqrt(x)), despite being an inexact result in cases where x is -0.0 or -inf. Defaults to -fno-approx-func.

• • -mllvm -enable-ppc-gen-scalar-mass
  • 
  • CXXOPTIMIZE

  • This flag is to enable automatic use of the MASS scalar library and call scalar functions in your program.

• • -mllvm -vector-library=MASSV
  • 
  • CXXOPTIMIZE

  • This flag is to enable automatic use of the MASS SIMD library in your program.

• • -fprofile-generate
  • 
  • PASS1_CXXOPTIMIZE

  • This option is used in the first pass of a profile guided optimization(PGO) compile and causes PGO information to be generated. The profile guided optimization gathers data on both execution path and data values. It does not use hardware counters, nor gather any data other than path and data values for PGO specific optimizations. Instructs the compiler to generate profiling instrumentation in the generated binary. This compiler option is used for C/C++ programs.

• • -fprofile-use=default.profdata
  • 
  • PASS2_CXXOPTIMIZE

  • This option is used in the second pass of a profile guided optimization(PGO) compile. This flag instructs the compiler to use the profiling information in the file specified by the argument to -fprofile-use. This file summarizes profile information that was emitted into default*profraw files by the -fprofile-generate option. The profiling information is utilized to guide the compiler optimizer in making better optimization decisions. This compiler option is used for C/C++ programs.

• • -mcmodel=large
  • 
  • PASS1_CXXOPTIMIZE

  • Allows the Table of Contents to be larger than 64 Kb in size, allowing more addresses to be stored in the table.

• • -L/opt/IBM/xlmass/10.1.x/lib
  • 
  • LIBS

  • Searches the directory path for library files specified by the "-l" option at link time.

• • -lmass
  • 
  • LIBS

  • Link the mathematical acceleration subsystem libraries (MASS), which consist of a set of mathematical functions for C,C++ and Fortran language applications that are tuned for specific Power processor architectures.

• • -L/opt/IBM/openxlf/17.1.x/lib
  • 
  • LIBS

  • Searches the directory path for library files specified by the "-l" option at link time.

• • -lxlopt
  • 
  • LIBS

  • Enables use of certain functions from the MASS library for scalars and vectors.

Fortran benchmarks

• • -q64
  • 
  • FC, LD

  • Generates 64-bit ABI binaries. Used as a compiler option for Fortran programs. If -q64 is not specified, the default is to generate 32-bit ABI binaries on AIX.

• • -Wl,-bplugin_opt:--ppc-set-dscr=1
  • 
  • LDFFLAGS

  • Pass on the DSCR (data stream control register) value to the linker. This value controls the aggressiveness of the hardware data prefetch engine.

• • -blpdata
  • 
  • PASS1_LDFFLAGS, PASS2_LDFFLAGS

  • The -blpdata is a linker option and it sets the bit in the file's XCOFF header indicating that this executable will request the use of large pages when they are available on the system and when the user has an appropriate privilege. For C/C++ programs it is invoked as -Wl,-blpdata and for Fortran programs it is invoked as -blpdata.

• • -qlto
  • 
  • FOPTIMIZE, PASS1_LDFFLAGS, PASS2_LDFFLAGS

  • Generate output files in LLVM format, suitable for link time optimization. When used with -S this generates LLVM intermediate language assembly files, otherwise this generates LLVM bitcode format object files. The LLVM bitcode generated is suitable for Link Time Optimization (LTO), when all modules are merged into a single combined module for optimization.

• • -O3
  • 
  • FOPTIMIZE
  • -O3 Instruct the compiler to execute optimizations that might take longer to perform or that may generate larger code (in an attempt to make the program run faster).
• • -qarch=auto
  • 
  • FOPTIMIZE
  • Specifies the processor architecture for which the code(instructions) should be generated.

    Supported values for this flag are :

    • auto - Automatically detects the specific architecture of the compiling machine. It assumes that the execution environment will be the same as the compilation environment.
    • pwr7 - Produces object code containing instructions that run on the Power7 hardware platform and on newer architectures.
    • pwr8 - Produces object code containing instructions that run on the Power8 hardware platform and on newer architectures.
    • pwr9 - Produces object code containing instructions that run on the Power9 hardware platform and on newer architectures.
• • -qvecnvol
  • 
  • FOPTIMIZE

  • Instructs the compiler to use both volatile and nonvolatile vector registers. Volatile vector registers are those whose value is not preserved across function calls or across save context, jump or switch context system library functions.

• • -mllvm -enable-ppc-gen-scalar-mass
  • 
  • FOPTIMIZE

  • This flag is to enable automatic use of the MASS scalar library and call scalar functions in your program.

• • -mllvm -vector-library=MASSV
  • 
  • FOPTIMIZE

  • This flag is to enable automatic use of the MASS SIMD library in your program.

• • -qprofile-generate
  • 
  • PASS1_FOPTIMIZE

  • This option is used in the first pass of a profile guided optimization(PGO) compile and causes PGO information to be generated. The profile guided optimization gathers data on both execution path and data values. It does not use hardware counters, nor gather any data other than path and data values for PGO specific optimizations. Instructs the compiler to generate profiling instrumentation in the generated binary. This compiler option is used for Fortran programs.

• • -qprofile-use=default.profdata
  • 
  • PASS2_FOPTIMIZE

  • This option is used in the second pass of a profile guided optimization(PGO) compile. This flag instructs the compiler to use the profiling information in the file specified by the argument to -qprofile-use. This file summarizes profile information that was emitted into default*profraw files by the -fprofile-generate option. The profiling information is utilized to guide the compiler optimizer in making better optimization decisions. This compiler option is used for Fortran programs.

• • -qpic=large
  • 
  • PASS1_FOPTIMIZE

  • Instructs the compiler to assume that the size of the TOC is larger than 64 Kb. When -qpic=large is in effect, the compiler generates two instructions for each TOC access to enlarge the accessing range. This helps avoid TOC overflow conditions when the Table of Contents is larger than 64 Kb.

• • -L/opt/IBM/xlmass/10.1.x/lib
  • 
  • LIBS

  • Searches the directory path for library files specified by the "-l" option at link time.

• • -lmass
  • 
  • LIBS

  • Link the mathematical acceleration subsystem libraries (MASS), which consist of a set of mathematical functions for C,C++ and Fortran language applications that are tuned for specific Power processor architectures.

• • -L/opt/IBM/openxlf/17.1.x/lib
  • 
  • LIBS

  • Searches the directory path for library files specified by the "-l" option at link time.

• • -lxlopt
  • 
  • LIBS

  • Enables use of certain functions from the MASS library for scalars and vectors.