MPI2007 Flag Description
Linux Networx LS-1, Scali MPI Connect 5.6.1, Intel 9.1 compilers

Base Optimization Flags

C benchmarks

• • -O3
  • 
  • COPTIMIZE

  • Enables O2 optimizations plus more aggressive optimizations, such as prefetching, scalar replacement, and loop and memory access transformations. Enables optimizations for maximum speed, such as:
    - Loop unrolling, including instruction scheduling
    - Code replication to eliminate branches
    - Padding the size of certain power-of-two arrays to allow more efficient cache use.
    On Intel Itanium processors, the O3 option enables optimizations for technical computing applications (loop-intensive code):
    loop optimizations and data prefetch. The O3 optimizations may not cause higher performance unless loop and memory access transformations take place. The optimizations may slow down code in some cases compared to O2 optimizations.
    The O3 option is recommended for applications that have loops that heavily use floating-point calculations and process large data sets.

  • Includes:
    • -O2
      • -O1
        • -unrolln
        • -builtin
        • -mno-ieee-fp
        • -fomit-frame-pointer
        • -ffunction-sections
• • -no-prec-div
  • 
  • COPTIMIZE

  • Enables optimizations that give slightly less precise results than full IEEE division. With some optimizations, such as -xN and -xB, the compiler may change floating-point division computations into multiplication by the reciprocal of the denominator. For example, A/B is computed as A * (1/B) to improve the speed of the computation. The default is -prec-div, which provides fully precise IEEE division. It improves precision of floating-point divides by disabling floating-point division-to-multiplication optimizations, resulting in greater accuracy with some loss of performance.

• • -ftz
  • 
  • COPTIMIZE

  • Flushes denormal floating point results to zero when the application is in gradual underflow mode.

• • -fno-alias
  • 
  • COPTIMIZE

  • Tells the compiler not to assume aliasing in the program (DEFAULT = -falias).

• • -xT
  • 
  • COPTIMIZE

  • Generates specialized code to run exclusively on processors with the extensions T. This option can generate SSSE3, SSE3, SSE2, and SSE instructions for Intel processors, and it can optimize for the Intel (R) Core (TM) 2 Duo processor family.

C++ benchmarks

126.lammps

• • -O3
  • 
  • CXXOPTIMIZE

  • Enables O2 optimizations plus more aggressive optimizations, such as prefetching, scalar replacement, and loop and memory access transformations. Enables optimizations for maximum speed, such as:
    - Loop unrolling, including instruction scheduling
    - Code replication to eliminate branches
    - Padding the size of certain power-of-two arrays to allow more efficient cache use.
    On Intel Itanium processors, the O3 option enables optimizations for technical computing applications (loop-intensive code):
    loop optimizations and data prefetch. The O3 optimizations may not cause higher performance unless loop and memory access transformations take place. The optimizations may slow down code in some cases compared to O2 optimizations.
    The O3 option is recommended for applications that have loops that heavily use floating-point calculations and process large data sets.

  • Includes:
    • -O2
      • -O1
        • -unrolln
        • -builtin
        • -mno-ieee-fp
        • -fomit-frame-pointer
        • -ffunction-sections
• • -no-prec-div
  • 
  • CXXOPTIMIZE

  • Enables optimizations that give slightly less precise results than full IEEE division. With some optimizations, such as -xN and -xB, the compiler may change floating-point division computations into multiplication by the reciprocal of the denominator. For example, A/B is computed as A * (1/B) to improve the speed of the computation. The default is -prec-div, which provides fully precise IEEE division. It improves precision of floating-point divides by disabling floating-point division-to-multiplication optimizations, resulting in greater accuracy with some loss of performance.

• • -ftz
  • 
  • CXXOPTIMIZE

  • Flushes denormal floating point results to zero when the application is in gradual underflow mode.

• • -fno-alias
  • 
  • CXXOPTIMIZE

  • Tells the compiler not to assume aliasing in the program (DEFAULT = -falias).

• • -xT
  • 
  • CXXOPTIMIZE

  • Generates specialized code to run exclusively on processors with the extensions T. This option can generate SSSE3, SSE3, SSE2, and SSE instructions for Intel processors, and it can optimize for the Intel (R) Core (TM) 2 Duo processor family.

Fortran benchmarks

• • -O3
  • 
  • FOPTIMIZE

  • Enables O2 optimizations plus more aggressive optimizations, such as prefetching, scalar replacement, and loop and memory access transformations. Enables optimizations for maximum speed, such as:
    - Loop unrolling, including instruction scheduling
    - Code replication to eliminate branches
    - Padding the size of certain power-of-two arrays to allow more efficient cache use.
    On Intel Itanium processors, the O3 option enables optimizations for technical computing applications (loop-intensive code):
    loop optimizations and data prefetch. The O3 optimizations may not cause higher performance unless loop and memory access transformations take place. The optimizations may slow down code in some cases compared to O2 optimizations.
    The O3 option is recommended for applications that have loops that heavily use floating-point calculations and process large data sets.

  • Includes:
    • -O2
      • -O1
        • -unrolln
        • -builtin
        • -mno-ieee-fp
        • -fomit-frame-pointer
        • -ffunction-sections
• • -no-prec-div
  • 
  • FOPTIMIZE

  • Enables optimizations that give slightly less precise results than full IEEE division. With some optimizations, such as -xN and -xB, the compiler may change floating-point division computations into multiplication by the reciprocal of the denominator. For example, A/B is computed as A * (1/B) to improve the speed of the computation. The default is -prec-div, which provides fully precise IEEE division. It improves precision of floating-point divides by disabling floating-point division-to-multiplication optimizations, resulting in greater accuracy with some loss of performance.

• • -ftz
  • 
  • FOPTIMIZE

  • Flushes denormal floating point results to zero when the application is in gradual underflow mode.

• • -fno-alias
  • 
  • FOPTIMIZE

  • Tells the compiler not to assume aliasing in the program (DEFAULT = -falias).

• • -xT
  • 
  • FOPTIMIZE

  • Generates specialized code to run exclusively on processors with the extensions T. This option can generate SSSE3, SSE3, SSE2, and SSE instructions for Intel processors, and it can optimize for the Intel (R) Core (TM) 2 Duo processor family.

Benchmarks using both Fortran and C

• • -O3
  • 
  • COPTIMIZE, FOPTIMIZE

  • Enables O2 optimizations plus more aggressive optimizations, such as prefetching, scalar replacement, and loop and memory access transformations. Enables optimizations for maximum speed, such as:
    - Loop unrolling, including instruction scheduling
    - Code replication to eliminate branches
    - Padding the size of certain power-of-two arrays to allow more efficient cache use.
    On Intel Itanium processors, the O3 option enables optimizations for technical computing applications (loop-intensive code):
    loop optimizations and data prefetch. The O3 optimizations may not cause higher performance unless loop and memory access transformations take place. The optimizations may slow down code in some cases compared to O2 optimizations.
    The O3 option is recommended for applications that have loops that heavily use floating-point calculations and process large data sets.

  • Includes:
    • -O2
      • -O1
        • -unrolln
        • -builtin
        • -mno-ieee-fp
        • -fomit-frame-pointer
        • -ffunction-sections
• • -no-prec-div
  • 
  • COPTIMIZE, FOPTIMIZE

  • Enables optimizations that give slightly less precise results than full IEEE division. With some optimizations, such as -xN and -xB, the compiler may change floating-point division computations into multiplication by the reciprocal of the denominator. For example, A/B is computed as A * (1/B) to improve the speed of the computation. The default is -prec-div, which provides fully precise IEEE division. It improves precision of floating-point divides by disabling floating-point division-to-multiplication optimizations, resulting in greater accuracy with some loss of performance.

• • -ftz
  • 
  • COPTIMIZE, FOPTIMIZE

  • Flushes denormal floating point results to zero when the application is in gradual underflow mode.

• • -fno-alias
  • 
  • COPTIMIZE, FOPTIMIZE

  • Tells the compiler not to assume aliasing in the program (DEFAULT = -falias).

• • -xT
  • 
  • COPTIMIZE, FOPTIMIZE

  • Generates specialized code to run exclusively on processors with the extensions T. This option can generate SSSE3, SSE3, SSE2, and SSE instructions for Intel processors, and it can optimize for the Intel (R) Core (TM) 2 Duo processor family.