CPU2006 Result Flag Description

This flags disclosure file describes the compiler flags associated with the following Intel compilers:

Base Compiler Invocation

C benchmarks

- icc
- CC, LD
- icc invokes the Intel C++ compiler . It is invoked as:
  
  icc [ options ] file1 [ file2 ... ]
  
  where,
  - options: represent zero or more compile options
  - fileN: is a C/C++ source (.C .c .cc .cp .cpp .cxx .c++ .i), assembly (.s), object (.o), static library (.a), or other linkable file.
  Invoking the compiler using icc compiles .c and .i files as C. Using icc only links in C++ libraries if C++ source is provided on the command line.

C++ benchmarks

- icpc
- CXX, LD
- The icpc command uses the same compiler options as the icc command. Invoking the compiler using icpc compiles .c, and .i files as C++. Using icpc always links in C++ libraries.

Fortran benchmarks

- ifort
- FC, LD
- ifort invokes the Intel Fortran compiler. It is invoked as:
  
  ifort [ options ] file1 [ file2 ... ]
  
  where,
  - options: represent zero or more compile options
  - fileN: is a Fortran source file, assembly file, object file, object library, or other linkable file.

Benchmarks using both Fortran and C

- icc
- CC
- icc invokes the Intel C++ compiler . It is invoked as:
  
  icc [ options ] file1 [ file2 ... ]
  
  where,
  - options: represent zero or more compile options
  - fileN: is a C/C++ source (.C .c .cc .cp .cpp .cxx .c++ .i), assembly (.s), object (.o), static library (.a), or other linkable file.
  Invoking the compiler using icc compiles .c and .i files as C. Using icc only links in C++ libraries if C++ source is provided on the command line.
- ifort
- FC, LD
- ifort invokes the Intel Fortran compiler. It is invoked as:
  
  ifort [ options ] file1 [ file2 ... ]
  
  where,
  - options: represent zero or more compile options
  - fileN: is a Fortran source file, assembly file, object file, object library, or other linkable file.

Base Portability Flags

410.bwaves

- -DSPEC_CPU_LP64
- PORTABILITY
- This option is used to indicate that the host system's integers are 32-bits wide, and longs and pointers are 64-bits wide. Not all benchmarks recognize this macro, but the preferred practice for data model selection applies the flags to all benchmarks; this flag description is a placeholder for those benchmarks that do not recognize this macro.

416.gamess

- -DSPEC_CPU_LP64
- PORTABILITY
- This option is used to indicate that the host system's integers are 32-bits wide, and longs and pointers are 64-bits wide. Not all benchmarks recognize this macro, but the preferred practice for data model selection applies the flags to all benchmarks; this flag description is a placeholder for those benchmarks that do not recognize this macro.

433.milc

- -DSPEC_CPU_LP64
- PORTABILITY
- This option is used to indicate that the host system's integers are 32-bits wide, and longs and pointers are 64-bits wide. Not all benchmarks recognize this macro, but the preferred practice for data model selection applies the flags to all benchmarks; this flag description is a placeholder for those benchmarks that do not recognize this macro.

434.zeusmp

- -DSPEC_CPU_LP64
- PORTABILITY
- This option is used to indicate that the host system's integers are 32-bits wide, and longs and pointers are 64-bits wide. Not all benchmarks recognize this macro, but the preferred practice for data model selection applies the flags to all benchmarks; this flag description is a placeholder for those benchmarks that do not recognize this macro.

435.gromacs

- -DSPEC_CPU_LP64
- PORTABILITY
- This option is used to indicate that the host system's integers are 32-bits wide, and longs and pointers are 64-bits wide. Not all benchmarks recognize this macro, but the preferred practice for data model selection applies the flags to all benchmarks; this flag description is a placeholder for those benchmarks that do not recognize this macro.
- -nofor_main
- FPORTABILITY
- For mixed-language benchmarks, tell the compiler that the main program is not written in Fortran

436.cactusADM

- -DSPEC_CPU_LP64
- PORTABILITY
- This option is used to indicate that the host system's integers are 32-bits wide, and longs and pointers are 64-bits wide. Not all benchmarks recognize this macro, but the preferred practice for data model selection applies the flags to all benchmarks; this flag description is a placeholder for those benchmarks that do not recognize this macro.
- -nofor_main
- FPORTABILITY
- For mixed-language benchmarks, tell the compiler that the main program is not written in Fortran

437.leslie3d

- -DSPEC_CPU_LP64
- PORTABILITY
- This option is used to indicate that the host system's integers are 32-bits wide, and longs and pointers are 64-bits wide. Not all benchmarks recognize this macro, but the preferred practice for data model selection applies the flags to all benchmarks; this flag description is a placeholder for those benchmarks that do not recognize this macro.

444.namd

- -DSPEC_CPU_LP64
- PORTABILITY
- This option is used to indicate that the host system's integers are 32-bits wide, and longs and pointers are 64-bits wide. Not all benchmarks recognize this macro, but the preferred practice for data model selection applies the flags to all benchmarks; this flag description is a placeholder for those benchmarks that do not recognize this macro.

447.dealII

- -DSPEC_CPU_LP64
- PORTABILITY
- This option is used to indicate that the host system's integers are 32-bits wide, and longs and pointers are 64-bits wide. Not all benchmarks recognize this macro, but the preferred practice for data model selection applies the flags to all benchmarks; this flag description is a placeholder for those benchmarks that do not recognize this macro.

450.soplex

- -DSPEC_CPU_LP64
- PORTABILITY
- This option is used to indicate that the host system's integers are 32-bits wide, and longs and pointers are 64-bits wide. Not all benchmarks recognize this macro, but the preferred practice for data model selection applies the flags to all benchmarks; this flag description is a placeholder for those benchmarks that do not recognize this macro.

453.povray

- -DSPEC_CPU_LP64
- PORTABILITY
- This option is used to indicate that the host system's integers are 32-bits wide, and longs and pointers are 64-bits wide. Not all benchmarks recognize this macro, but the preferred practice for data model selection applies the flags to all benchmarks; this flag description is a placeholder for those benchmarks that do not recognize this macro.

454.calculix

- -DSPEC_CPU_LP64
- PORTABILITY
- This option is used to indicate that the host system's integers are 32-bits wide, and longs and pointers are 64-bits wide. Not all benchmarks recognize this macro, but the preferred practice for data model selection applies the flags to all benchmarks; this flag description is a placeholder for those benchmarks that do not recognize this macro.
- -nofor_main
- FPORTABILITY
- For mixed-language benchmarks, tell the compiler that the main program is not written in Fortran

459.GemsFDTD

- -DSPEC_CPU_LP64
- PORTABILITY
- This option is used to indicate that the host system's integers are 32-bits wide, and longs and pointers are 64-bits wide. Not all benchmarks recognize this macro, but the preferred practice for data model selection applies the flags to all benchmarks; this flag description is a placeholder for those benchmarks that do not recognize this macro.

465.tonto

- -DSPEC_CPU_LP64
- PORTABILITY
- This option is used to indicate that the host system's integers are 32-bits wide, and longs and pointers are 64-bits wide. Not all benchmarks recognize this macro, but the preferred practice for data model selection applies the flags to all benchmarks; this flag description is a placeholder for those benchmarks that do not recognize this macro.

470.lbm

- -DSPEC_CPU_LP64
- PORTABILITY
- This option is used to indicate that the host system's integers are 32-bits wide, and longs and pointers are 64-bits wide. Not all benchmarks recognize this macro, but the preferred practice for data model selection applies the flags to all benchmarks; this flag description is a placeholder for those benchmarks that do not recognize this macro.

481.wrf

- -DSPEC_CPU_LP64
- PORTABILITY
- This option is used to indicate that the host system's integers are 32-bits wide, and longs and pointers are 64-bits wide. Not all benchmarks recognize this macro, but the preferred practice for data model selection applies the flags to all benchmarks; this flag description is a placeholder for those benchmarks that do not recognize this macro.
- -DSPEC_CPU_MACOSX
- CPORTABILITY
- This macro indicates that the benchmark is being compiled on a MacOS X system.
- -DSPEC_CPU_MACOSX_INTEL
- CPORTABILITY
- This macro indicates that the benchmark is being compiled on an Intel-architecture MacOS X system.
- -DSPEC_CPU_CASE_FLAG
- CPORTABILITY
- This macro indicates that Fortran functions called from C should have their names lower-cased.

482.sphinx3

- -DSPEC_CPU_LP64
- PORTABILITY
- This option is used to indicate that the host system's integers are 32-bits wide, and longs and pointers are 64-bits wide. Not all benchmarks recognize this macro, but the preferred practice for data model selection applies the flags to all benchmarks; this flag description is a placeholder for those benchmarks that do not recognize this macro.

Base Optimization Flags

C benchmarks

- -static-intel
- CC, LD
- This option causes the Intel-provided libraries to be linked in statically. It is the opposite of -shared-intel. Note that when this option is provided, libguide is also linked in statically.
- -fast
- COPTIMIZE
- The -fast option enhances execution speed across the entire program by including the following options that can improve run-time performance:
  - -xT (optimizations for Intel Core 2 Duo processor family)
  - -O3 (maximum speed and high-level optimizations)
  - -ipo (enables interprocedural optimizations across files)
  - -no-prec-div (disable -prec-div), where -prec-div improves precision of FP divides (some speed impact)
  - -mdynamic-no-pic, where -mydynamic-no-pic indicates that code is not relocatable
  Options set by -fast cannot be overidden, list options separately to change behavior. The options set by -fast may change from release to release.
- Includes:
  - -xT
  - -O3
  - -ipo
  - -no-prec-div
  - -mdynamic-no-pic
- -parallel
- COPTIMIZE
- Enables the compiler to generate runtime control code for effective automatic parallelization

C++ benchmarks

- -static-intel
- CXX, LD
- This option causes the Intel-provided libraries to be linked in statically. It is the opposite of -shared-intel. Note that when this option is provided, libguide is also linked in statically.
- -xT
- CXXOPTIMIZE
- The -xT option tells the compiler to generate optimized code for the Intel Core 2 Duo processor family. It can generate SSSE3, SSE3, SSE2, and SSE instructions for the Intel processors.
- -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 IA-32 and Intel EM64T processors, when O3 is used with options -ax or -x (Linux/Mac OSX), the compiler performs more aggressive data dependency analysis than for O2, which may result in longer compilation times.
  
  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.
- -no-prec-div
- CXXOPTIMIZE
- This option improves precision of floating-point divides. It has a slight impact on speed.
  
  With some optimizations, such as -xN and -xB (Linux) or /QxN and /QxB (Windows), 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.
  
  However, sometimes the value produced by this transformation is not as accurate as full IEEE division. When it is important to have fully precise IEEE division, use this option to disable the floating-point division-to-multiplication optimization. The result is more accurate, with some loss of performance.
  
  If you specify -no-prec-div (Linux and Mac OSX), it enables optimizations that give slightly less precise results than full IEEE division. The default is -prec-div.
- -mdynamic-no-pic
- CXXOPTIMIZE
- Code is not relocatable, but external references are relocatable.
- -parallel
- CXXOPTIMIZE
- Enables the compiler to generate runtime control code for effective automatic parallelization

Fortran benchmarks

- -static-intel
- FC, LD
- This option causes the Intel-provided libraries to be linked in statically. It is the opposite of -shared-intel. Note that when this option is provided, libguide is also linked in statically.
- -fast
- FOPTIMIZE
- The -fast option enhances execution speed across the entire program by including the following options that can improve run-time performance:
  - -xT (optimizations for Intel Core 2 Duo processor family)
  - -O3 (maximum speed and high-level optimizations)
  - -ipo (enables interprocedural optimizations across files)
  - -no-prec-div (disable -prec-div), where -prec-div improves precision of FP divides (some speed impact)
  - -mdynamic-no-pic, where -mydynamic-no-pic indicates that code is not relocatable
  Options set by -fast cannot be overidden, list options separately to change behavior. The options set by -fast may change from release to release.
- Includes:
  - -xT
  - -O3
  - -ipo
  - -no-prec-div
  - -mdynamic-no-pic
- -parallel
- FOPTIMIZE
- Enables the compiler to generate runtime control code for effective automatic parallelization

Benchmarks using both Fortran and C

- -static-intel
- CC, FC, LD
- This option causes the Intel-provided libraries to be linked in statically. It is the opposite of -shared-intel. Note that when this option is provided, libguide is also linked in statically.
- -fast
- COPTIMIZE, FOPTIMIZE
- The -fast option enhances execution speed across the entire program by including the following options that can improve run-time performance:
  - -xT (optimizations for Intel Core 2 Duo processor family)
  - -O3 (maximum speed and high-level optimizations)
  - -ipo (enables interprocedural optimizations across files)
  - -no-prec-div (disable -prec-div), where -prec-div improves precision of FP divides (some speed impact)
  - -mdynamic-no-pic, where -mydynamic-no-pic indicates that code is not relocatable
  Options set by -fast cannot be overidden, list options separately to change behavior. The options set by -fast may change from release to release.
- Includes:
  - -xT
  - -O3
  - -ipo
  - -no-prec-div
  - -mdynamic-no-pic
- -parallel
- COPTIMIZE, FOPTIMIZE
- Enables the compiler to generate runtime control code for effective automatic parallelization

Base Other Flags

C benchmarks

- -Wl,-stack_addr,0xF10000000
- EXTRA_LDFLAGS
- Specifies the initial address of the stack pointer value, where value is a hexadecimal number rounded to the segment alignment. The default segment alignment is the target pagesize (currently, 1000 hexadecimal for the PowerPC and for i386). If -stack_size is specified and -stack_addr is not, a default stack address specific for the architecture being linked will be used and its value printed as a warning message. This creates a segment named __UNIXSTACK. Note that the initial stack address will be either at the high address of the segment or the low address of the segment depending on which direction the stack grows for the architecture being linked.
- -Wl,-stack_size,0x64000000
- EXTRA_LDFLAGS
- Specifies the size of the stack segment value, where value is a hexadecimal number rounded to the segment alignment. The default segment alignment is the target pagesize (currently, 1000 hexadecimal for the PowerPC and for i386). If -stack_addr is specified and -stack_size is not, a default stack size specific for the architecture being linked will be used and its value printed as a warning message. This creates a segment named __UNIXSTACK .

C++ benchmarks

- -Wl,-stack_addr,0xF10000000
- EXTRA_LDFLAGS
- Specifies the initial address of the stack pointer value, where value is a hexadecimal number rounded to the segment alignment. The default segment alignment is the target pagesize (currently, 1000 hexadecimal for the PowerPC and for i386). If -stack_size is specified and -stack_addr is not, a default stack address specific for the architecture being linked will be used and its value printed as a warning message. This creates a segment named __UNIXSTACK. Note that the initial stack address will be either at the high address of the segment or the low address of the segment depending on which direction the stack grows for the architecture being linked.
- -Wl,-stack_size,0x64000000
- EXTRA_LDFLAGS
- Specifies the size of the stack segment value, where value is a hexadecimal number rounded to the segment alignment. The default segment alignment is the target pagesize (currently, 1000 hexadecimal for the PowerPC and for i386). If -stack_addr is specified and -stack_size is not, a default stack size specific for the architecture being linked will be used and its value printed as a warning message. This creates a segment named __UNIXSTACK .

Fortran benchmarks

- -Wl,-stack_addr,0xF10000000
- EXTRA_LDFLAGS
- Specifies the initial address of the stack pointer value, where value is a hexadecimal number rounded to the segment alignment. The default segment alignment is the target pagesize (currently, 1000 hexadecimal for the PowerPC and for i386). If -stack_size is specified and -stack_addr is not, a default stack address specific for the architecture being linked will be used and its value printed as a warning message. This creates a segment named __UNIXSTACK. Note that the initial stack address will be either at the high address of the segment or the low address of the segment depending on which direction the stack grows for the architecture being linked.
- -Wl,-stack_size,0x64000000
- EXTRA_LDFLAGS
- Specifies the size of the stack segment value, where value is a hexadecimal number rounded to the segment alignment. The default segment alignment is the target pagesize (currently, 1000 hexadecimal for the PowerPC and for i386). If -stack_addr is specified and -stack_size is not, a default stack size specific for the architecture being linked will be used and its value printed as a warning message. This creates a segment named __UNIXSTACK .

Benchmarks using both Fortran and C

- -Wl,-stack_addr,0xF10000000
- EXTRA_LDFLAGS
- Specifies the initial address of the stack pointer value, where value is a hexadecimal number rounded to the segment alignment. The default segment alignment is the target pagesize (currently, 1000 hexadecimal for the PowerPC and for i386). If -stack_size is specified and -stack_addr is not, a default stack address specific for the architecture being linked will be used and its value printed as a warning message. This creates a segment named __UNIXSTACK. Note that the initial stack address will be either at the high address of the segment or the low address of the segment depending on which direction the stack grows for the architecture being linked.
- -Wl,-stack_size,0x64000000
- EXTRA_LDFLAGS
- Specifies the size of the stack segment value, where value is a hexadecimal number rounded to the segment alignment. The default segment alignment is the target pagesize (currently, 1000 hexadecimal for the PowerPC and for i386). If -stack_addr is specified and -stack_size is not, a default stack size specific for the architecture being linked will be used and its value printed as a warning message. This creates a segment named __UNIXSTACK .

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

The system under test is deemed reasonably quiet by turning off the following from the System Preferences panel:

Sets the maximum number of threads to use for OpenMP* parallel regions if no other value is specified in the application. This environment variable applies to both -openmp and -parallel (Linux and Mac OS X).
Example syntax on a Mac OS X system with 8 cores:
export OMP_NUM_THREADS=8

For questions about the meanings of these flags, please contact the tester.
For other inquiries, please contact webmaster@spec.org
Copyright 2006-2014 Standard Performance Evaluation Corporation
Tested with SPEC CPU2006 v1.0.1.
Report generated on Tue Jul 22 16:40:33 2014 by SPEC CPU2006 flags formatter v6906.

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

CPU2006 Flag DescriptionApple, Inc. 3.0 GHz 8-Core Xserve

Base Compiler Invocation

Base Portability Flags

Base Optimization Flags

Base Other Flags

Implicitly Included Flags

CPU2006 Flag Description
Apple, Inc. 3.0 GHz 8-Core Xserve