CPU2006 Result Flag Description

Base Optimization Flags

C benchmarks

- +Ofaster
- hp_cc,hp_aCC,hp_f90
- COPTIMIZE
- This option selects +Ofast, but with the optimization level increased to +O4.
- Includes:
  - +Ofast
    - +O2
      
      +O1
    - +Onolimit
    - +Ofltacc=relaxed
    - +DSnative
    - +FPD
    - -Wl,+pi,1M
    - -Wl,+pd,1M
    - +mergeseg
  - +O4
    - +O3
      
      +O2
      
      +O1
    - -ipo
- +Otype_safety=ansi
- hp_cc,hp_aCC
- COPTIMIZE
- Enable [disable] aliasing across types.
  - off - The default. Specifies that aliasing can occur freely across types. This is a synonym to +Onoptrs_ansi and +Onoptrs_strongly_typed options in legacy cc.
  - limited - Code follows ANSI aliasing rules. Unnamed objects should be treated as if they had an unknown type.
  - ansi - Code follows ANSI aliasing rules, and unnamed objects should be treated the same as named objects. This option is synonym to +Optrs_ansi option in legacy cc.
  - strong - Code follows ANSI aliasing rules, except that accesses through lvalues of a character type are not permitted to touch objects of other types. This assumes that field addresses are not taken. This option is synonym to +Optrs_strongly_typed option in legacy cc.
- -Wl,-a,archive_shared
- COPTIMIZE
- Specifies whether shared or archive libraries are searched with the -l option. For archive_shared, the archive form is preferred, but the shared form is allowed.
- -Wl,+pd,64M
- COPTIMIZE
- Request a particular virtual memory page size that should be used for data. Sizes of 4K, 16K, 64K, 256K, 1M, 4M, 16M, 64M, 256M, 1G, 4G, D, and L are supported. A size of D allows the kernel to choose what page size should be used. A size of L results in using the largest page size available. The actual page size may vary if the requested size cannot be fulfilled.
- -Wl,+pi,64M
- COPTIMIZE
- Request a particular virtual memory page size that should be used for instructions. Sizes of 4K, 16K, 64K, 256K, 1M, 4M, 16M, 64M, 256M, 1G, 4G, D, and L are supported. A size of D allows the kernel to choose what page size should be used. A size of L results in using the largest page size available. The actual page size may vary if the requested size cannot be fulfilled.
- -Wl,-N
- EXTRA_LDFLAGS
- In 32-bit mode only, cause the data to be placed immediately following the text, and make the text writable. Files of this type cannot be shared.

C++ benchmarks

- +Ofaster
- hp_cc,hp_aCC,hp_f90
- CXXOPTIMIZE
- This option selects +Ofast, but with the optimization level increased to +O4.
- Includes:
  - +Ofast
    - +O2
      
      +O1
    - +Onolimit
    - +Ofltacc=relaxed
    - +DSnative
    - +FPD
    - -Wl,+pi,1M
    - -Wl,+pd,1M
    - +mergeseg
  - +O4
    - +O3
      
      +O2
      
      +O1
    - -ipo
- +Otype_safety=ansi
- hp_cc,hp_aCC
- CXXOPTIMIZE
- Enable [disable] aliasing across types.
  - off - The default. Specifies that aliasing can occur freely across types. This is a synonym to +Onoptrs_ansi and +Onoptrs_strongly_typed options in legacy cc.
  - limited - Code follows ANSI aliasing rules. Unnamed objects should be treated as if they had an unknown type.
  - ansi - Code follows ANSI aliasing rules, and unnamed objects should be treated the same as named objects. This option is synonym to +Optrs_ansi option in legacy cc.
  - strong - Code follows ANSI aliasing rules, except that accesses through lvalues of a character type are not permitted to touch objects of other types. This assumes that field addresses are not taken. This option is synonym to +Optrs_strongly_typed option in legacy cc.
- -Wl,-a,archive_shared
- CXXOPTIMIZE
- Specifies whether shared or archive libraries are searched with the -l option. For archive_shared, the archive form is preferred, but the shared form is allowed.
- -Wl,+pd,64M
- CXXOPTIMIZE
- Request a particular virtual memory page size that should be used for data. Sizes of 4K, 16K, 64K, 256K, 1M, 4M, 16M, 64M, 256M, 1G, 4G, D, and L are supported. A size of D allows the kernel to choose what page size should be used. A size of L results in using the largest page size available. The actual page size may vary if the requested size cannot be fulfilled.
- -Wl,+pi,64M
- CXXOPTIMIZE
- Request a particular virtual memory page size that should be used for instructions. Sizes of 4K, 16K, 64K, 256K, 1M, 4M, 16M, 64M, 256M, 1G, 4G, D, and L are supported. A size of D allows the kernel to choose what page size should be used. A size of L results in using the largest page size available. The actual page size may vary if the requested size cannot be fulfilled.
- -Wl,-N
- EXTRA_LDFLAGS
- In 32-bit mode only, cause the data to be placed immediately following the text, and make the text writable. Files of this type cannot be shared.

Fortran benchmarks

- +Ofaster
- hp_cc,hp_aCC,hp_f90
- FOPTIMIZE
- This option selects +Ofast, but with the optimization level increased to +O4.
- Includes:
  - +Ofast
    - +O2
      
      +O1
    - +Onolimit
    - +Ofltacc=relaxed
    - +DSnative
    - +FPD
    - -Wl,+pi,1M
    - -Wl,+pd,1M
    - +mergeseg
  - +O4
    - +O3
      
      +O2
      
      +O1
    - -ipo
- -Wl,-a,archive_shared
- FOPTIMIZE
- Specifies whether shared or archive libraries are searched with the -l option. For archive_shared, the archive form is preferred, but the shared form is allowed.
- -Wl,+pd,64M
- FOPTIMIZE
- Request a particular virtual memory page size that should be used for data. Sizes of 4K, 16K, 64K, 256K, 1M, 4M, 16M, 64M, 256M, 1G, 4G, D, and L are supported. A size of D allows the kernel to choose what page size should be used. A size of L results in using the largest page size available. The actual page size may vary if the requested size cannot be fulfilled.
- -Wl,+pi,64M
- FOPTIMIZE
- Request a particular virtual memory page size that should be used for instructions. Sizes of 4K, 16K, 64K, 256K, 1M, 4M, 16M, 64M, 256M, 1G, 4G, D, and L are supported. A size of D allows the kernel to choose what page size should be used. A size of L results in using the largest page size available. The actual page size may vary if the requested size cannot be fulfilled.
- -Wl,-N
- EXTRA_LDFLAGS
- In 32-bit mode only, cause the data to be placed immediately following the text, and make the text writable. Files of this type cannot be shared.

Benchmarks using both Fortran and C

- +Ofaster
- hp_cc,hp_aCC,hp_f90
- COPTIMIZE, FOPTIMIZE
- This option selects +Ofast, but with the optimization level increased to +O4.
- Includes:
  - +Ofast
    - +O2
      
      +O1
    - +Onolimit
    - +Ofltacc=relaxed
    - +DSnative
    - +FPD
    - -Wl,+pi,1M
    - -Wl,+pd,1M
    - +mergeseg
  - +O4
    - +O3
      
      +O2
      
      +O1
    - -ipo
- +Otype_safety=ansi
- hp_cc,hp_aCC
- COPTIMIZE
- Enable [disable] aliasing across types.
  - off - The default. Specifies that aliasing can occur freely across types. This is a synonym to +Onoptrs_ansi and +Onoptrs_strongly_typed options in legacy cc.
  - limited - Code follows ANSI aliasing rules. Unnamed objects should be treated as if they had an unknown type.
  - ansi - Code follows ANSI aliasing rules, and unnamed objects should be treated the same as named objects. This option is synonym to +Optrs_ansi option in legacy cc.
  - strong - Code follows ANSI aliasing rules, except that accesses through lvalues of a character type are not permitted to touch objects of other types. This assumes that field addresses are not taken. This option is synonym to +Optrs_strongly_typed option in legacy cc.
- -Wl,-a,archive_shared
- COPTIMIZE, FOPTIMIZE
- Specifies whether shared or archive libraries are searched with the -l option. For archive_shared, the archive form is preferred, but the shared form is allowed.
- -Wl,+pd,64M
- COPTIMIZE, FOPTIMIZE
- Request a particular virtual memory page size that should be used for data. Sizes of 4K, 16K, 64K, 256K, 1M, 4M, 16M, 64M, 256M, 1G, 4G, D, and L are supported. A size of D allows the kernel to choose what page size should be used. A size of L results in using the largest page size available. The actual page size may vary if the requested size cannot be fulfilled.
- -Wl,+pi,64M
- COPTIMIZE, FOPTIMIZE
- Request a particular virtual memory page size that should be used for instructions. Sizes of 4K, 16K, 64K, 256K, 1M, 4M, 16M, 64M, 256M, 1G, 4G, D, and L are supported. A size of D allows the kernel to choose what page size should be used. A size of L results in using the largest page size available. The actual page size may vary if the requested size cannot be fulfilled.
- -Wl,-N
- EXTRA_LDFLAGS
- In 32-bit mode only, cause the data to be placed immediately following the text, and make the text writable. Files of this type cannot be shared.

Peak Optimization Flags

C benchmarks

433.milc

- +Oprofile=collect:all
- hp_cc,hp_aCC,hp_f90
- PASS1_CFLAGS, PASS1_LDCFLAGS
- Instrument the application to enable collection of profile based optimization data. Qualifiers are a comma-separated list of profile collection qualifiers. The profile collection qualifiers are:
  - arc - Enable collection of arc counts.
  - dcache - Enable collection of data cache misses via hardware performance counters.
  - stride - Enable collection of stride data.
  - loopiter - Enable collection of loop iteration counts.
  - all - Enable collection of all types of profile data. Equivalent to the command +Oprofile=collect:arc,dcache,stride,loopiter.
- +Oprofile=use
- hp_cc,hp_aCC,hp_f90
- PASS2_CFLAGS, PASS2_LDCFLAGS
- Examine the profile database file produced by an instrumented program to perform profile based optimizations on the code. Some optimizations are only enabled when a feedback data file is available.
- +Ofaster
- hp_cc,hp_aCC,hp_f90
- COPTIMIZE
- This option selects +Ofast, but with the optimization level increased to +O4.
- Includes:
  - +Ofast
    - +O2
      
      +O1
    - +Onolimit
    - +Ofltacc=relaxed
    - +DSnative
    - +FPD
    - -Wl,+pi,1M
    - -Wl,+pd,1M
    - +mergeseg
  - +O4
    - +O3
      
      +O2
      
      +O1
    - -ipo
- +Otype_safety=ansi
- hp_cc,hp_aCC
- COPTIMIZE
- Enable [disable] aliasing across types.
  - off - The default. Specifies that aliasing can occur freely across types. This is a synonym to +Onoptrs_ansi and +Onoptrs_strongly_typed options in legacy cc.
  - limited - Code follows ANSI aliasing rules. Unnamed objects should be treated as if they had an unknown type.
  - ansi - Code follows ANSI aliasing rules, and unnamed objects should be treated the same as named objects. This option is synonym to +Optrs_ansi option in legacy cc.
  - strong - Code follows ANSI aliasing rules, except that accesses through lvalues of a character type are not permitted to touch objects of other types. This assumes that field addresses are not taken. This option is synonym to +Optrs_strongly_typed option in legacy cc.
- -Wl,-a,archive_shared
- COPTIMIZE
- Specifies whether shared or archive libraries are searched with the -l option. For archive_shared, the archive form is preferred, but the shared form is allowed.
- -Wl,+pd,64M
- COPTIMIZE
- Request a particular virtual memory page size that should be used for data. Sizes of 4K, 16K, 64K, 256K, 1M, 4M, 16M, 64M, 256M, 1G, 4G, D, and L are supported. A size of D allows the kernel to choose what page size should be used. A size of L results in using the largest page size available. The actual page size may vary if the requested size cannot be fulfilled.
- -Wl,+pi,64M
- COPTIMIZE
- Request a particular virtual memory page size that should be used for instructions. Sizes of 4K, 16K, 64K, 256K, 1M, 4M, 16M, 64M, 256M, 1G, 4G, D, and L are supported. A size of D allows the kernel to choose what page size should be used. A size of L results in using the largest page size available. The actual page size may vary if the requested size cannot be fulfilled.
- +Onoparmsoverlap
- COPTIMIZE
- Optimize with the assumption that subprogram arguments do [do not] refer to the same memory. This option can be used at optimization levels 2, 3, and 4.
- -Wl,-N
- EXTRA_LDFLAGS
- In 32-bit mode only, cause the data to be placed immediately following the text, and make the text writable. Files of this type cannot be shared.

470.lbm

- basepeak = yes

482.sphinx3

- +Oprofile=collect:all
- hp_cc,hp_aCC,hp_f90
- PASS1_CFLAGS, PASS1_LDCFLAGS
- Instrument the application to enable collection of profile based optimization data. Qualifiers are a comma-separated list of profile collection qualifiers. The profile collection qualifiers are:
  - arc - Enable collection of arc counts.
  - dcache - Enable collection of data cache misses via hardware performance counters.
  - stride - Enable collection of stride data.
  - loopiter - Enable collection of loop iteration counts.
  - all - Enable collection of all types of profile data. Equivalent to the command +Oprofile=collect:arc,dcache,stride,loopiter.
- +Oprofile=use
- hp_cc,hp_aCC,hp_f90
- PASS2_CFLAGS, PASS2_LDCFLAGS
- Examine the profile database file produced by an instrumented program to perform profile based optimizations on the code. Some optimizations are only enabled when a feedback data file is available.
- +Ofaster
- hp_cc,hp_aCC,hp_f90
- COPTIMIZE
- This option selects +Ofast, but with the optimization level increased to +O4.
- Includes:
  - +Ofast
    - +O2
      
      +O1
    - +Onolimit
    - +Ofltacc=relaxed
    - +DSnative
    - +FPD
    - -Wl,+pi,1M
    - -Wl,+pd,1M
    - +mergeseg
  - +O4
    - +O3
      
      +O2
      
      +O1
    - -ipo
- +Otype_safety=ansi
- hp_cc,hp_aCC
- COPTIMIZE
- Enable [disable] aliasing across types.
  - off - The default. Specifies that aliasing can occur freely across types. This is a synonym to +Onoptrs_ansi and +Onoptrs_strongly_typed options in legacy cc.
  - limited - Code follows ANSI aliasing rules. Unnamed objects should be treated as if they had an unknown type.
  - ansi - Code follows ANSI aliasing rules, and unnamed objects should be treated the same as named objects. This option is synonym to +Optrs_ansi option in legacy cc.
  - strong - Code follows ANSI aliasing rules, except that accesses through lvalues of a character type are not permitted to touch objects of other types. This assumes that field addresses are not taken. This option is synonym to +Optrs_strongly_typed option in legacy cc.
- -Wl,-a,archive_shared
- COPTIMIZE
- Specifies whether shared or archive libraries are searched with the -l option. For archive_shared, the archive form is preferred, but the shared form is allowed.
- -Wl,+pd,64M
- COPTIMIZE
- Request a particular virtual memory page size that should be used for data. Sizes of 4K, 16K, 64K, 256K, 1M, 4M, 16M, 64M, 256M, 1G, 4G, D, and L are supported. A size of D allows the kernel to choose what page size should be used. A size of L results in using the largest page size available. The actual page size may vary if the requested size cannot be fulfilled.
- -Wl,+pi,64M
- COPTIMIZE
- Request a particular virtual memory page size that should be used for instructions. Sizes of 4K, 16K, 64K, 256K, 1M, 4M, 16M, 64M, 256M, 1G, 4G, D, and L are supported. A size of D allows the kernel to choose what page size should be used. A size of L results in using the largest page size available. The actual page size may vary if the requested size cannot be fulfilled.
- +Onoparmsoverlap
- COPTIMIZE
- Optimize with the assumption that subprogram arguments do [do not] refer to the same memory. This option can be used at optimization levels 2, 3, and 4.

C++ benchmarks

444.namd

- basepeak = yes

447.dealII

- basepeak = yes

450.soplex

- +Oprofile=collect:all
- hp_cc,hp_aCC,hp_f90
- PASS1_CXXFLAGS, PASS1_LDCXXFLAGS
- Instrument the application to enable collection of profile based optimization data. Qualifiers are a comma-separated list of profile collection qualifiers. The profile collection qualifiers are:
  - arc - Enable collection of arc counts.
  - dcache - Enable collection of data cache misses via hardware performance counters.
  - stride - Enable collection of stride data.
  - loopiter - Enable collection of loop iteration counts.
  - all - Enable collection of all types of profile data. Equivalent to the command +Oprofile=collect:arc,dcache,stride,loopiter.
- +Oprofile=use
- hp_cc,hp_aCC,hp_f90
- PASS2_CXXFLAGS, PASS2_LDCXXFLAGS
- Examine the profile database file produced by an instrumented program to perform profile based optimizations on the code. Some optimizations are only enabled when a feedback data file is available.
- +Ofaster
- hp_cc,hp_aCC,hp_f90
- CXXOPTIMIZE
- This option selects +Ofast, but with the optimization level increased to +O4.
- Includes:
  - +Ofast
    - +O2
      
      +O1
    - +Onolimit
    - +Ofltacc=relaxed
    - +DSnative
    - +FPD
    - -Wl,+pi,1M
    - -Wl,+pd,1M
    - +mergeseg
  - +O4
    - +O3
      
      +O2
      
      +O1
    - -ipo
- +Otype_safety=ansi
- hp_cc,hp_aCC
- CXXOPTIMIZE
- Enable [disable] aliasing across types.
  - off - The default. Specifies that aliasing can occur freely across types. This is a synonym to +Onoptrs_ansi and +Onoptrs_strongly_typed options in legacy cc.
  - limited - Code follows ANSI aliasing rules. Unnamed objects should be treated as if they had an unknown type.
  - ansi - Code follows ANSI aliasing rules, and unnamed objects should be treated the same as named objects. This option is synonym to +Optrs_ansi option in legacy cc.
  - strong - Code follows ANSI aliasing rules, except that accesses through lvalues of a character type are not permitted to touch objects of other types. This assumes that field addresses are not taken. This option is synonym to +Optrs_strongly_typed option in legacy cc.
- -Wl,-a,archive_shared
- CXXOPTIMIZE
- Specifies whether shared or archive libraries are searched with the -l option. For archive_shared, the archive form is preferred, but the shared form is allowed.
- -Wl,+pd,64M
- CXXOPTIMIZE
- Request a particular virtual memory page size that should be used for data. Sizes of 4K, 16K, 64K, 256K, 1M, 4M, 16M, 64M, 256M, 1G, 4G, D, and L are supported. A size of D allows the kernel to choose what page size should be used. A size of L results in using the largest page size available. The actual page size may vary if the requested size cannot be fulfilled.
- -Wl,+pi,64M
- CXXOPTIMIZE
- Request a particular virtual memory page size that should be used for instructions. Sizes of 4K, 16K, 64K, 256K, 1M, 4M, 16M, 64M, 256M, 1G, 4G, D, and L are supported. A size of D allows the kernel to choose what page size should be used. A size of L results in using the largest page size available. The actual page size may vary if the requested size cannot be fulfilled.
- +Onoparmsoverlap
- CXXOPTIMIZE
- Optimize with the assumption that subprogram arguments do [do not] refer to the same memory. This option can be used at optimization levels 2, 3, and 4.
- -Wl,-N
- EXTRA_LDFLAGS
- In 32-bit mode only, cause the data to be placed immediately following the text, and make the text writable. Files of this type cannot be shared.

453.povray

- +Oprofile=collect:all
- hp_cc,hp_aCC,hp_f90
- PASS1_CXXFLAGS, PASS1_LDCXXFLAGS
- Instrument the application to enable collection of profile based optimization data. Qualifiers are a comma-separated list of profile collection qualifiers. The profile collection qualifiers are:
  - arc - Enable collection of arc counts.
  - dcache - Enable collection of data cache misses via hardware performance counters.
  - stride - Enable collection of stride data.
  - loopiter - Enable collection of loop iteration counts.
  - all - Enable collection of all types of profile data. Equivalent to the command +Oprofile=collect:arc,dcache,stride,loopiter.
- +Oprofile=use
- hp_cc,hp_aCC,hp_f90
- PASS2_CXXFLAGS, PASS2_LDCXXFLAGS
- Examine the profile database file produced by an instrumented program to perform profile based optimizations on the code. Some optimizations are only enabled when a feedback data file is available.
- +Ofaster
- hp_cc,hp_aCC,hp_f90
- CXXOPTIMIZE
- This option selects +Ofast, but with the optimization level increased to +O4.
- Includes:
  - +Ofast
    - +O2
      
      +O1
    - +Onolimit
    - +Ofltacc=relaxed
    - +DSnative
    - +FPD
    - -Wl,+pi,1M
    - -Wl,+pd,1M
    - +mergeseg
  - +O4
    - +O3
      
      +O2
      
      +O1
    - -ipo
- +Otype_safety=ansi
- hp_cc,hp_aCC
- CXXOPTIMIZE
- Enable [disable] aliasing across types.
  - off - The default. Specifies that aliasing can occur freely across types. This is a synonym to +Onoptrs_ansi and +Onoptrs_strongly_typed options in legacy cc.
  - limited - Code follows ANSI aliasing rules. Unnamed objects should be treated as if they had an unknown type.
  - ansi - Code follows ANSI aliasing rules, and unnamed objects should be treated the same as named objects. This option is synonym to +Optrs_ansi option in legacy cc.
  - strong - Code follows ANSI aliasing rules, except that accesses through lvalues of a character type are not permitted to touch objects of other types. This assumes that field addresses are not taken. This option is synonym to +Optrs_strongly_typed option in legacy cc.
- -Wl,-a,archive_shared
- CXXOPTIMIZE
- Specifies whether shared or archive libraries are searched with the -l option. For archive_shared, the archive form is preferred, but the shared form is allowed.
- -Wl,+pd,64M
- CXXOPTIMIZE
- Request a particular virtual memory page size that should be used for data. Sizes of 4K, 16K, 64K, 256K, 1M, 4M, 16M, 64M, 256M, 1G, 4G, D, and L are supported. A size of D allows the kernel to choose what page size should be used. A size of L results in using the largest page size available. The actual page size may vary if the requested size cannot be fulfilled.
- -Wl,+pi,64M
- CXXOPTIMIZE
- Request a particular virtual memory page size that should be used for instructions. Sizes of 4K, 16K, 64K, 256K, 1M, 4M, 16M, 64M, 256M, 1G, 4G, D, and L are supported. A size of D allows the kernel to choose what page size should be used. A size of L results in using the largest page size available. The actual page size may vary if the requested size cannot be fulfilled.

Fortran benchmarks

410.bwaves

- basepeak = yes

416.gamess

- +Ofaster
- hp_cc,hp_aCC,hp_f90
- FOPTIMIZE
- This option selects +Ofast, but with the optimization level increased to +O4.
- Includes:
  - +Ofast
    - +O2
      
      +O1
    - +Onolimit
    - +Ofltacc=relaxed
    - +DSnative
    - +FPD
    - -Wl,+pi,1M
    - -Wl,+pd,1M
    - +mergeseg
  - +O4
    - +O3
      
      +O2
      
      +O1
    - -ipo
- -Wl,-a,archive_shared
- FOPTIMIZE
- Specifies whether shared or archive libraries are searched with the -l option. For archive_shared, the archive form is preferred, but the shared form is allowed.
- -Wl,+pd,64M
- FOPTIMIZE
- Request a particular virtual memory page size that should be used for data. Sizes of 4K, 16K, 64K, 256K, 1M, 4M, 16M, 64M, 256M, 1G, 4G, D, and L are supported. A size of D allows the kernel to choose what page size should be used. A size of L results in using the largest page size available. The actual page size may vary if the requested size cannot be fulfilled.
- -Wl,+pi,64M
- FOPTIMIZE
- Request a particular virtual memory page size that should be used for instructions. Sizes of 4K, 16K, 64K, 256K, 1M, 4M, 16M, 64M, 256M, 1G, 4G, D, and L are supported. A size of D allows the kernel to choose what page size should be used. A size of L results in using the largest page size available. The actual page size may vary if the requested size cannot be fulfilled.
- +Odataprefetch=direct
- FOPTIMIZE
- +Odataprefetch=kind
  
  Control generation of data prefetch instructions for data structures referenced within innermost loops. The defined values for kind are:
  - direct - Enable generation of data prefetch instructions for the benefit of direct memory accesses, but not indirect memory accesses.
  - indirect - Enable generation of data prefetch instructions for the benefit of both direct and indirect memory accesses. This is the default at optimization levels +O2 and above.
  - none - Disable generation of data prefetch instructions. This is the default at optimization levels +O1 and below.
- -Wl,-N
- EXTRA_LDFLAGS
- In 32-bit mode only, cause the data to be placed immediately following the text, and make the text writable. Files of this type cannot be shared.

434.zeusmp

- basepeak = yes

437.leslie3d

- basepeak = yes

459.GemsFDTD

- +Oprofile=collect:all
- hp_cc,hp_aCC,hp_f90
- PASS1_FFLAGS, PASS1_LDFFLAGS
- Instrument the application to enable collection of profile based optimization data. Qualifiers are a comma-separated list of profile collection qualifiers. The profile collection qualifiers are:
  - arc - Enable collection of arc counts.
  - dcache - Enable collection of data cache misses via hardware performance counters.
  - stride - Enable collection of stride data.
  - loopiter - Enable collection of loop iteration counts.
  - all - Enable collection of all types of profile data. Equivalent to the command +Oprofile=collect:arc,dcache,stride,loopiter.
- +Oprofile=use
- hp_cc,hp_aCC,hp_f90
- PASS2_FFLAGS, PASS2_LDFFLAGS
- Examine the profile database file produced by an instrumented program to perform profile based optimizations on the code. Some optimizations are only enabled when a feedback data file is available.
- +Ofaster
- hp_cc,hp_aCC,hp_f90
- FOPTIMIZE
- This option selects +Ofast, but with the optimization level increased to +O4.
- Includes:
  - +Ofast
    - +O2
      
      +O1
    - +Onolimit
    - +Ofltacc=relaxed
    - +DSnative
    - +FPD
    - -Wl,+pi,1M
    - -Wl,+pd,1M
    - +mergeseg
  - +O4
    - +O3
      
      +O2
      
      +O1
    - -ipo
- -Wl,-a,archive_shared
- FOPTIMIZE
- Specifies whether shared or archive libraries are searched with the -l option. For archive_shared, the archive form is preferred, but the shared form is allowed.
- -Wl,+pd,64M
- FOPTIMIZE
- Request a particular virtual memory page size that should be used for data. Sizes of 4K, 16K, 64K, 256K, 1M, 4M, 16M, 64M, 256M, 1G, 4G, D, and L are supported. A size of D allows the kernel to choose what page size should be used. A size of L results in using the largest page size available. The actual page size may vary if the requested size cannot be fulfilled.
- -Wl,+pi,64M
- FOPTIMIZE
- Request a particular virtual memory page size that should be used for instructions. Sizes of 4K, 16K, 64K, 256K, 1M, 4M, 16M, 64M, 256M, 1G, 4G, D, and L are supported. A size of D allows the kernel to choose what page size should be used. A size of L results in using the largest page size available. The actual page size may vary if the requested size cannot be fulfilled.
- +Odataprefetch=direct
- FOPTIMIZE
- +Odataprefetch=kind
  
  Control generation of data prefetch instructions for data structures referenced within innermost loops. The defined values for kind are:
  - direct - Enable generation of data prefetch instructions for the benefit of direct memory accesses, but not indirect memory accesses.
  - indirect - Enable generation of data prefetch instructions for the benefit of both direct and indirect memory accesses. This is the default at optimization levels +O2 and above.
  - none - Disable generation of data prefetch instructions. This is the default at optimization levels +O1 and below.
- -Wl,-N
- EXTRA_LDFLAGS
- In 32-bit mode only, cause the data to be placed immediately following the text, and make the text writable. Files of this type cannot be shared.

465.tonto

- +Oprofile=collect:all
- hp_cc,hp_aCC,hp_f90
- PASS1_FFLAGS, PASS1_LDFFLAGS
- Instrument the application to enable collection of profile based optimization data. Qualifiers are a comma-separated list of profile collection qualifiers. The profile collection qualifiers are:
  - arc - Enable collection of arc counts.
  - dcache - Enable collection of data cache misses via hardware performance counters.
  - stride - Enable collection of stride data.
  - loopiter - Enable collection of loop iteration counts.
  - all - Enable collection of all types of profile data. Equivalent to the command +Oprofile=collect:arc,dcache,stride,loopiter.
- +Oprofile=use
- hp_cc,hp_aCC,hp_f90
- PASS2_FFLAGS, PASS2_LDFFLAGS
- Examine the profile database file produced by an instrumented program to perform profile based optimizations on the code. Some optimizations are only enabled when a feedback data file is available.
- +Ofaster
- hp_cc,hp_aCC,hp_f90
- FOPTIMIZE
- This option selects +Ofast, but with the optimization level increased to +O4.
- Includes:
  - +Ofast
    - +O2
      
      +O1
    - +Onolimit
    - +Ofltacc=relaxed
    - +DSnative
    - +FPD
    - -Wl,+pi,1M
    - -Wl,+pd,1M
    - +mergeseg
  - +O4
    - +O3
      
      +O2
      
      +O1
    - -ipo
- -Wl,-a,archive_shared
- FOPTIMIZE
- Specifies whether shared or archive libraries are searched with the -l option. For archive_shared, the archive form is preferred, but the shared form is allowed.
- -Wl,+pd,64M
- FOPTIMIZE
- Request a particular virtual memory page size that should be used for data. Sizes of 4K, 16K, 64K, 256K, 1M, 4M, 16M, 64M, 256M, 1G, 4G, D, and L are supported. A size of D allows the kernel to choose what page size should be used. A size of L results in using the largest page size available. The actual page size may vary if the requested size cannot be fulfilled.
- -Wl,+pi,64M
- FOPTIMIZE
- Request a particular virtual memory page size that should be used for instructions. Sizes of 4K, 16K, 64K, 256K, 1M, 4M, 16M, 64M, 256M, 1G, 4G, D, and L are supported. A size of D allows the kernel to choose what page size should be used. A size of L results in using the largest page size available. The actual page size may vary if the requested size cannot be fulfilled.
- +Odataprefetch=direct
- FOPTIMIZE
- +Odataprefetch=kind
  
  Control generation of data prefetch instructions for data structures referenced within innermost loops. The defined values for kind are:
  - direct - Enable generation of data prefetch instructions for the benefit of direct memory accesses, but not indirect memory accesses.
  - indirect - Enable generation of data prefetch instructions for the benefit of both direct and indirect memory accesses. This is the default at optimization levels +O2 and above.
  - none - Disable generation of data prefetch instructions. This is the default at optimization levels +O1 and below.

Benchmarks using both Fortran and C

435.gromacs

- +Oprofile=collect:all
- hp_cc,hp_aCC,hp_f90
- PASS1_CFLAGS, PASS1_FFLAGS, PASS1_LDFFLAGS
- Instrument the application to enable collection of profile based optimization data. Qualifiers are a comma-separated list of profile collection qualifiers. The profile collection qualifiers are:
  - arc - Enable collection of arc counts.
  - dcache - Enable collection of data cache misses via hardware performance counters.
  - stride - Enable collection of stride data.
  - loopiter - Enable collection of loop iteration counts.
  - all - Enable collection of all types of profile data. Equivalent to the command +Oprofile=collect:arc,dcache,stride,loopiter.
- +Oprofile=use
- hp_cc,hp_aCC,hp_f90
- PASS2_CFLAGS, PASS2_FFLAGS, PASS2_LDFFLAGS
- Examine the profile database file produced by an instrumented program to perform profile based optimizations on the code. Some optimizations are only enabled when a feedback data file is available.
- +Ofaster
- hp_cc,hp_aCC,hp_f90
- COPTIMIZE, FOPTIMIZE
- This option selects +Ofast, but with the optimization level increased to +O4.
- Includes:
  - +Ofast
    - +O2
      
      +O1
    - +Onolimit
    - +Ofltacc=relaxed
    - +DSnative
    - +FPD
    - -Wl,+pi,1M
    - -Wl,+pd,1M
    - +mergeseg
  - +O4
    - +O3
      
      +O2
      
      +O1
    - -ipo
- +Otype_safety=ansi
- hp_cc,hp_aCC
- COPTIMIZE
- Enable [disable] aliasing across types.
  - off - The default. Specifies that aliasing can occur freely across types. This is a synonym to +Onoptrs_ansi and +Onoptrs_strongly_typed options in legacy cc.
  - limited - Code follows ANSI aliasing rules. Unnamed objects should be treated as if they had an unknown type.
  - ansi - Code follows ANSI aliasing rules, and unnamed objects should be treated the same as named objects. This option is synonym to +Optrs_ansi option in legacy cc.
  - strong - Code follows ANSI aliasing rules, except that accesses through lvalues of a character type are not permitted to touch objects of other types. This assumes that field addresses are not taken. This option is synonym to +Optrs_strongly_typed option in legacy cc.
- -Wl,-a,archive_shared
- COPTIMIZE, FOPTIMIZE
- Specifies whether shared or archive libraries are searched with the -l option. For archive_shared, the archive form is preferred, but the shared form is allowed.
- -Wl,+pd,64M
- COPTIMIZE, FOPTIMIZE
- Request a particular virtual memory page size that should be used for data. Sizes of 4K, 16K, 64K, 256K, 1M, 4M, 16M, 64M, 256M, 1G, 4G, D, and L are supported. A size of D allows the kernel to choose what page size should be used. A size of L results in using the largest page size available. The actual page size may vary if the requested size cannot be fulfilled.
- -Wl,+pi,64M
- COPTIMIZE, FOPTIMIZE
- Request a particular virtual memory page size that should be used for instructions. Sizes of 4K, 16K, 64K, 256K, 1M, 4M, 16M, 64M, 256M, 1G, 4G, D, and L are supported. A size of D allows the kernel to choose what page size should be used. A size of L results in using the largest page size available. The actual page size may vary if the requested size cannot be fulfilled.
- +Onoparmsoverlap
- COPTIMIZE
- Optimize with the assumption that subprogram arguments do [do not] refer to the same memory. This option can be used at optimization levels 2, 3, and 4.

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.

System and Other Tuning Information

maxdsiz, maxdsiz_64bit - maximum size (in bytes) of the data segment for any user process

maxssiz, maxssiz_64bit - maximum size (in bytes) of the stack for any user process

dbc_min_pct - minimum percentage of memory used for caching file I/O data and metadata

dbc_max_pct - maximum percentage of memory to be used for caching file I/O data and metadata

	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 DescriptionHewlett-Packard Company HP Integrity rx2660 (1.6GHz/18MB Dual-Core Intel Itanium 2)

Base Compiler Invocation

Peak Compiler Invocation

Base Portability Flags

Peak Portability Flags

Base Optimization Flags

Peak Optimization Flags

Implicitly Included Flags

CPU2006 Flag Description
Hewlett-Packard Company HP Integrity rx2660 (1.6GHz/18MB Dual-Core Intel Itanium 2)