Base Compiler Invocation
C benchmarks
-
- cc
![[user]](http://www.spec.org/auto/cpu2006/flags/user.png)
- CC, LD
-
Invoke the Sun Studio C Compiler.
C++ benchmarks
-
- CC
- CXX, LD
-
Invoke the Sun Studio C++ Compiler
Fortran benchmarks
-
- f90
- FC, LD
-
Invoke the Sun Studio Fortran 90 Compiler
Compilers: Sun Studio 11
Operating systems: Solaris 10
Last updated: 18-Jul-2006 cds
The text for many of the descriptions below was taken from the Sun Studio 11 Compiler Documentation, which is copyright © 2005 Sun Microsystems, Inc. The original documentation can be found at docs.sun.com.
Invoke the Sun Studio C Compiler.
Invoke the Sun Studio C++ Compiler
Invoke the Sun Studio Fortran 90 Compiler
Invoke the Sun Studio C Compiler.
Invoke the Sun Studio C++ Compiler
Invoke the Sun Studio Fortran 90 Compiler
![[suite]](http://www.spec.org/auto/cpu2006/flags/suite.png)
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
![[benchmark]](http://www.spec.org/auto/cpu2006/flags/benchmark.png)
This macro indicates that the system building the benchmark is a little-endian system.
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.
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.
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.
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.
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.
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.
This macro indicates that the system building the benchmark is a little-endian system.
A convenience option, this switch selects several other options that are described in this file.
Perform optimizations across all object files in the link step:
At -xipo=2, the compiler performs inter-procedural aliasing analysis as well as optimization of memory allocation and layout to improve cache performance.
Specifies which instructions can be used. Among the choices are:
A convenience option, this switch selects several other options that are described in this file.
Perform optimizations across all object files in the link step:
At -xipo=2, the compiler performs inter-procedural aliasing analysis as well as optimization of memory allocation and layout to improve cache performance.
Specifies which instructions can be used. Among the choices are:
Use STLport's Standard Library implementation instead of the default libCstd.
A convenience option, this switch selects the following switches that are described in this file:
Perform optimizations across all object files in the link step:
At -xipo=2, the compiler performs inter-procedural aliasing analysis as well as optimization of memory allocation and layout to improve cache performance.
Specifies which instructions can be used. Among the choices are:
A convenience option, this switch selects several other options that are described in this file.
A convenience option, this switch selects the following switches that are described in this file:
Perform optimizations across all object files in the link step:
At -xipo=2, the compiler performs inter-procedural aliasing analysis as well as optimization of memory allocation and layout to improve cache performance.
Specifies which instructions can be used. Among the choices are:
A convenience option, this switch selects several other options that are described in this file.
Specifies which instructions can be used. Among the choices are:
Set the preferred page size for running the program.
[optimizer flag]
Control the optimizer's loop inliner; The inliner is allowed to increase the size of the program by up to n%.
[optimizer flag]
Control the optimizer's loop inliner; Set inline callee size limit to n. The unit roughly corresponds to the number of instructions.
Control the level of searching that the compiler does for prefetch opportunities by setting n to 1, 2, or 3, where higher numbers mean to do more searching. The default for Fortran is 2. The default for C and C++ is 1.
A convenience option, this switch selects several other options that are described in this file.
Specifies which instructions can be used. Among the choices are:
Treat pointer-valued function parameters as restricted pointers.
Perform optimizations across all object files in the link step:
At -xipo=2, the compiler performs inter-procedural aliasing analysis as well as optimization of memory allocation and layout to improve cache performance.
Enable generation of prefetch instructions on those architectures that support prefetch. val may be one of the following:
If only -xprefetch is specified, -xprefetch=auto,explicit is assumed.
Set the preferred page size for running the program.
[global optimizer flag]
Use iropt in the profile phase of the compiler iropt is the Global optimizer.
[optimizer flag]
Convert multiple short memory load operations into single long load operations.
[optimizer flag]
Inliner only considers routines smaller than n pseudo instructions as possible inline candidates.
Reads mapfile as a text file of directives to ld.
Collect profile data for feedback-directed optimization (FDO). If an option directory is named, the feedback will be stored there. When FDO is used, the training run gathers information regarding execution paths. As of the Sun Studio 11 version of the compiler suite, the training run gathers information about data values on SPARC systems, but not on x86 systems. Hardware performance counters are not used. FDO improves existing optimizations but does not introduce new classes of optimization.
Use data collected for profile feedback. If an option directory is named, look for the feedback data there.
A convenience option, this switch selects several other options that are described in this file.
Specifies which instructions can be used. Among the choices are:
Perform optimizations across all object files in the link step:
At -xipo=2, the compiler performs inter-procedural aliasing analysis as well as optimization of memory allocation and layout to improve cache performance.
Use STLport's Standard Library implementation instead of the default libCstd.
A convenience option, this switch selects several other options that are described in this file.
Specifies which instructions can be used. Among the choices are:
Perform optimizations across all object files in the link step:
At -xipo=2, the compiler performs inter-procedural aliasing analysis as well as optimization of memory allocation and layout to improve cache performance.
Allow the compiler to transform math library calls within loops into calls to the vector math library. Specifying
[optimizer flag]
Increase the probability that the compiler will perform memcpy/memset transformations.
Allows the compiler to perform type-based alias analysis:
Treat pointer-valued function parameters as restricted pointers.
Allow the compiler to use the frame-pointer register (%ebp on IA32, %rbp on AMD64) as an unallocated callee-saves register.
Use STLport's Standard Library implementation instead of the default libCstd.
A convenience option, this switch selects several other options that are described in this file.
Perform optimizations across all object files in the link step:
At -xipo=2, the compiler performs inter-procedural aliasing analysis as well as optimization of memory allocation and layout to improve cache performance.
Specifies which instructions can be used. Among the choices are:
Use STLport's Standard Library implementation instead of the default libCstd.
A convenience option, this switch selects the following switches that are described in this file:
Specifies which instructions can be used. Among the choices are:
Enable unrolling loops n times where possible.
Allocate routine local variables on the stack.
Specify optimization level n:
Perform optimizations across all object files in the link step:
At -xipo=2, the compiler performs inter-procedural aliasing analysis as well as optimization of memory allocation and layout to improve cache performance.
Control the level of searching that the compiler does for prefetch opportunities by setting n to 1, 2, or 3, where higher numbers mean to do more searching. The default for Fortran is 2. The default for C and C++ is 1.
[optimizer flag]
Do not perform loop distribution transformations.
A convenience option, this switch selects the following switches that are described in this file:
Specifies which instructions can be used. Among the choices are:
Perform optimizations across all object files in the link step:
At -xipo=2, the compiler performs inter-procedural aliasing analysis as well as optimization of memory allocation and layout to improve cache performance.
Control the level of searching that the compiler does for prefetch opportunities by setting n to 1, 2, or 3, where higher numbers mean to do more searching. The default for Fortran is 2. The default for C and C++ is 1.
A convenience option, this switch selects the following switches that are described in this file:
Specifies which instructions can be used. Among the choices are:
Perform optimizations across all object files in the link step:
At -xipo=2, the compiler performs inter-procedural aliasing analysis as well as optimization of memory allocation and layout to improve cache performance.
Allow the compiler to use the frame-pointer register (%ebp on IA32, %rbp on AMD64) as an unallocated callee-saves register.
A convenience option, this switch selects the following switches that are described in this file:
Perform optimizations across all object files in the link step:
At -xipo=2, the compiler performs inter-procedural aliasing analysis as well as optimization of memory allocation and layout to improve cache performance.
Specifies which instructions can be used. Among the choices are:
Collect profile data for feedback-directed optimization (FDO). If an option directory is named, the feedback will be stored there. When FDO is used, the training run gathers information regarding execution paths. As of the Sun Studio 11 version of the compiler suite, the training run gathers information about data values on SPARC systems, but not on x86 systems. Hardware performance counters are not used. FDO improves existing optimizations but does not introduce new classes of optimization.
Use data collected for profile feedback. If an option directory is named, look for the feedback data there.
A convenience option, this switch selects the following switches that are described in this file:
Specifies which instructions can be used. Among the choices are:
Perform optimizations across all object files in the link step:
At -xipo=2, the compiler performs inter-procedural aliasing analysis as well as optimization of memory allocation and layout to improve cache performance.
Allow the compiler to transform math library calls within loops into calls to the vector math library. Specifying
Specify degree of aliasing to be assumed by the compiler. Specifying -xalias without a list gives the best performance for most programs that do not violate Fortran aliasing rules, and corresponds to: -xalias=no%dummy,no%craypointer,no%actual,no%overindex,no%ftnpointer
Analyze loops for inter-iteration data dependencies, and do loop restructuring.
Links in a library of general purpose memory allocation routines which can be faster than those found in libc, at the expense of more virtual memory consumed.
A convenience option, this switch selects several other options that are described in this file.
A convenience option, this switch selects the following switches that are described in this file:
Perform optimizations across all object files in the link step:
At -xipo=2, the compiler performs inter-procedural aliasing analysis as well as optimization of memory allocation and layout to improve cache performance.
Specifies which instructions can be used. Among the choices are:
Allow optimizer to use x87 hardware instructions for sine, cosine, and rsqrt. The precision and rounding effects are determined by the underlying hardware implementation, rather than by standard IEEE754 semantics.
Allow optimizer to use x87 hardware instructions for sine, cosine, and rsqrt. The precision and rounding effects are determined by the underlying hardware implementation, rather than by standard IEEE754 semantics.
A convenience option, this switch selects several other options that are described in this file.
A convenience option, this switch selects the following switches that are described in this file:
Perform optimizations across all object files in the link step:
At -xipo=2, the compiler performs inter-procedural aliasing analysis as well as optimization of memory allocation and layout to improve cache performance.
Specifies which instructions can be used. Among the choices are:
A convenience option, this switch selects several other options that are described in this file.
A convenience option, this switch selects the following switches that are described in this file:
Specifies which instructions can be used. Among the choices are:
Set the preferred page size for running the program.
Enable generation of prefetch instructions on those architectures that support prefetch. val may be one of the following:
If only -xprefetch is specified, -xprefetch=auto,explicit is assumed.
Control the level of searching that the compiler does for prefetch opportunities by setting n to 1, 2, or 3, where higher numbers mean to do more searching. The default for Fortran is 2. The default for C and C++ is 1.
[optimizer flag]
Control the optimizer's loop inliner; The inliner is allowed to increase the size of the program by up to n%.
[optimizer flag]
Control the optimizer's loop inliner; Set inline callee size limit to n. The unit roughly corresponds to the number of instructions.
Enable unrolling loops n times where possible.
Allocate routine local variables on the stack.
Specify optimization level n:
Perform optimizations across all object files in the link step:
At -xipo=2, the compiler performs inter-procedural aliasing analysis as well as optimization of memory allocation and layout to improve cache performance.
[optimizer flag]
Do not perform loop distribution transformations.
Directs the compiler to print the name and version ID of each component as the compiler executes.
Controls compiler verbosity. There are several values that can be used with this flag:
The default is -verbose=%none.
Directs the compiler to print the name and version ID of each component as the compiler executes.
Directs the compiler to print the name and version ID of each component as the compiler executes.
Directs the compiler to print the name and version ID of each component as the compiler executes.
Controls compiler verbosity. There are several values that can be used with this flag:
The default is -verbose=%none.
Directs the compiler to print the name and version ID of each component as the compiler executes.
Directs the compiler to print the name and version ID of each component as the compiler executes.
This section contains descriptions of flags that were included implicitly by other flags, but which do not have a permanent home at SPEC.
Allows the compiler to assume that your code does not rely on setting of the errno variable.
Assume data is naturally aligned.
Selects faster (but nonstandard) handling of floating point arithmetic exceptions and gradual underflow.
Controls simplifying assumptions for floating point arithmetic:
Evaluate float expressions as single precision.
Turns off all IEEE 754 trapping modes.
Allows the compiler to perform type-based alias analysis at the specified alias level:
Substitute intrinsic functions or inline system functions where profitable for performance.
Analyze loops for inter-iteration data dependencies, and do loop restructuring.
Use inline expansion for math library, libm.
Cancels forcing expressions to have the precision of the result.
Select the optimized math library.
Specify optimization level n:
Allow the compiler to use the frame-pointer register (%ebp on IA32, %rbp on AMD64) as an unallocated callee-saves register.
Selects options appropriate for the system where the compile is taking place, including architecture, chip, and cache sizes. (These can also be controlled separately, via -xarch, -xchip, and -xcache, respectively.)
Sets the IEEE 754 trapping mode to common exceptions (invalid, division by zero, and overflow).
Enable generation of prefetch instructions on those architectures that support prefetch. val may be one of the following:
If only -xprefetch is specified, -xprefetch=auto,explicit is assumed.
Control the level of searching that the compiler does for prefetch opportunities by setting n to 1, 2, or 3, where higher numbers mean to do more searching. The default for Fortran is 2. The default for C and C++ is 1.
Platform settings
One or more of the following settings may have been applied to the testbed. If so, the "Platform Notes" section of the report will say so; and you can read below to find out more about what these settings mean.
Node Interleaving Enabled (Default = Disabled):
This BIOS option allows the enabling/disabling of memory interleaving across CPU nodes. When disabled, each CPU chip can only access memory within its own node.
autoup=<n> (Unix)
When the file system flush daemon fsflush runs, it writes to disk all modified file buffers that are more
than n seconds old.
PARALLEL=<n> (Unix)
If programs have been compiled with -xautopar, this environment variable can be set to the number of
processors that programs should use.
STACKSIZE=<n> (Unix)
Set the size of the stack (temporary storage area) for each slave thread of a multithreaded program.
submit=echo 'pbind -b...' > dobmk; sh dobmk (SPEC tools, Unix)
When running multiple copies of benchmarks, the SPEC config file feature submit is sometimes used to
cause individual jobs to be bound to specific processors. If so, the specific command may be found in the config file; here
is a brief guide to understanding that command:
tune_t_fsflushr=<n> (Unix)
Controls the number of seconds between runs of the file system flush daemon, fsflush.
ulimit -s <n> (Unix)
Sets the stack size to n kbytes, or "unlimited" to allow the stack size to grow without limit.
Flag description origin markings:
|
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. |
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.
Report generated on Tue Jul 22 10:43:30 2014 by SPEC CPU2006 flags formatter v6906.