Description of compiler flags for Intel C++ Compiler 8.1
for IA-32, EM64T and Itanium architectures
Architecture specific flags and actions are noted, where appropriate.
--------------------------------------------------------
-O1 Optimize to favor code size and code locality. Disables
loop unrolling. -O1 may improve performance for applica-
tions with very large code size, many branches, and exe-
cution time not dominated by code within loops. In most
cases, -O2 is recommended over -O1.

-O2 (DEFAULT)
Optimize for code speed. This is the generally recom-
mended optimization level.

IPF: Turn software pipelining ON.

-O3 Enable -O2 optimizations and in addition, enable more
aggressive optimizations such as loop and memory access
transformation. The -O3 optimizations may slow down code
in some cases compared to -O2 optimizations. Recommended
for applications that have loops with heavy use of float-
ing point calculations and process large data sets.

-ax<codes> generate code specialized for processor extensions
specified by <codes> while also generating generic code.
<codes> includes one or more of the following characters:
i Pentium Pro and Pentium II processor instructions
M MMX(TM) instructions
K streaming SIMD extensions (implies i and M above)
W Pentium 4 processor with Streaming SIMD Extensions 2
(implies i, M and K)
N Pentium 4 processor with Streaming SIMD Extensions 2
P Pentium 4 processor with Streaming SIMD Extensions 3

-x<codes> generate specialized code to run exclusively on processors
supporting the extensions indicated by <codes> as
described above.

----------------------------------------------------------------------------------
Additional Notes for EM64T:
----------------------------------------------------------------------------------
On Intel� EM64T systems, -axW and -axP are the only valid options.
----------------------------------------------------------------------------------

----------------------------------------------------------------------------------
Additional Notes on -xP:
----------------------------------------------------------------------------------
-xP The -xP option targets your program to run on Intel Pentium 4
and compatible Intel processors. The resulting code might
contain unconditional use of features that are not supported
on other processors. Programs, where the function main() is
compiled with this option, will detect non compatible processors
and generate an error message during execution. This option
also enables new optimizations in addition to Intel processor
specific optimizations.

These options also enable advanced data layout and code restructuring
optimizations to improve memory accesses for Intel processors.
----------------------------------------------------------------------------------

-Ob{0|1|2} Controls the compiler's inline expansion.
0: disable inlining.
1: disables inlining unless -ip or -Ob2 are specified.
2: enables inlining of any function. However, the
compiler decides which functions are inlined. This
option enables interprocedural optimizations and has
the same effect as specifying the -ip option.

-IPF_fp_relaxed[-] (IPF only)
Enable [disable] use of faster but slightly less accurate
code sequences for math functions, such as divide and
square root.

-ip enable single-file IP optimizations
(within files, same as -Ob2)

-ipo multi-file ip optimizations that includes:
- inline function expansion
- interprocedural constant propogation
- dead code elimination
- propagation of function characteristics
- passing arguments in registers
- loop-invariant code motion

-fast The -fast option maximizes speed across the entire pro-
gram. It sets command options that can improve run-time
performance, as follows:

For Itanium-based systems, -fast sets -O3, -ipo, and -static.
For IA-32 and Intel� EM64T systems, -fast sets -O3, -ipo, -static, and -xP.
Note that on IA-32 and Intel� EM64T systems, programs compiled with the -xP
option will detect non-compatible processors and generate an error
message during execution.

-ansi_alias Directs the compiler to assume that the program
adheres to the type-based aliasing rules defined in Section 6.5 of the ISO C
Standard. If your program adheres to these rules, this option will allow
the compiler to optimize more aggressively. If it doesn't adhere to these
rules, it can cause the compiler to generate incorrect code.

-auto_ilp32 specify that the application cannot exceed a 32-bit address space

-cxxlib-icc Directs the Intel compiler to use the C++ run-time libraries and
C++ header files included with the Intel compiler. They include:
libcprts standard C++ headers
libcprts standard C++ library
libcxa and libunwind C++ language support

-prof_gen instrument program for profiling for the first phase of
two-phase profile guided otimization

-prof_use Instructs the compiler to produce a profile-optimized
executable and merges available dynamic information (.dyn)
files into a pgopti.dpi file. If you perform multiple
executions of the instrumented program, -prof_use merges
the dynamic information files again and overwrites the
previous pgopti.dpi file.
Without any other options, the current directory is
searched for .dyn files

-alias_args[-]
Assume arguments may be aliased. (DEFAULT) [not aliased].

-------------------------------------------------------------
Description of compiler flags for Intel FORTRAN Compiler 8.1
for IA-32, EM64T and Itanium architectures
Architecture specific flags and actions are noted, where appropriate.
--------------------------------------------------------
-O1 Maximize speed; disables some optimizations that increase
code size for a small speed benefit. This option enables
global optimization. This includes data-flow analysis,
code motion, strength reduction and test replacement,
split-lifetime analysis, and instruction scheduling.
Specifying -O2 includes the optimizations performed by -O1.
Note that, on IA-32 systems, -O1 and -O2 are equivalent.

-O2 (DEFAULT)
Minimizes size; optimizes for speed, but disables some
optimizations that increase code size for a small speed
benefit; for the Itanium� compiler, -O1 turns off software
pipelining to reduce code size. This option enables local
optimizations within the source program unit, recognition
of common subexpressions, and expansion of integer multiplication
and division using shifts.

-O3 Maximize speed plus use higher-level optimizations; optimizations
include loop transformation, software pipelining, and (IA-32 only)
prefetching; this option may not improve performance for some
programs. Specifying -O3 includes the optimizations performed
by -O2. This option enables additional global optimizations that
improve speed (at the cost of extra code size).
These optimizations include:
o Loop unrolling, including instruction scheduling
o Code replication to eliminate branches
o Padding the size of certain power-of-two arrays to allow
more efficient cache use.

-x<codes> generate specialized code to run exclusively on processors
supporting the extensions indicated by <codes> as
described above.

-IPF_fp_relaxed (IPF only)
Enables use of faster but slightly less accurate code sequences
for math functions, such as divide and sqrt. When compared to
strict IEEE* precision, this option slightly reduces the accu-
racy of floating-point calculations performed by these func-
tions, usually limited to the least significant digit.

-ip enable single-file IP optimizations
(within files, same as -Ob2)

-fast The -fast option maximizes speed across the entire pro-
gram. It sets command options that can improve run-time
performance, as follows:

-ansi_alias Enables (default) or disables the compiler to assume that the program
adheres to the ANSI Fortran type aliasablility rules. For example, an object
of type real cannot be accessed as an integer. You should see the ANSI
standard for the complete set of rules

-prof_gen instrument program for profiling for the first phase of
two-phase profile guided otimization

-scalar_rep(-) Enables(disables) scalar replacement performed during loop
transformations (requires /O3).

-auto Causes all variables to be allocated on the stack, rather than
in local static storage. Does not affect variables that appear in an
EQUIVALENCE or SAVE statement, or those that are in COMMON. Makes all
local variables AUTOMATIC, same as /4Ya.

Portability options for CPU2000:
-------------------------------

178.galgel:
-FI
Fixed-format F90 source code.

187.facerec:
srcalt=AllocShape
This src.alt adds code to checks that the allocatable array FTemp
is allocated before calling SHAPE on it.

176.gcc:
-Dalloca=_alloca
Replace occurrences of alloca() with _alloca()
-Dalloca=_alloca
Replace occurrences of alloca() with _alloca()
-DUSG
Specify that the programming environment is like System V Unix systems
srcalt=64bitgcc35
This src.alt eliminates the use of the cast as lvalue extension,
which allows 176.gcc to be built on systems using GCC 3.5 or later.

186.crafty:
-DLINUX_i386
Linux Intel system, use "long long" as 64bit variable.
252.eon:
-DHAS_ERRLIST
Tells that the system provides the "sys_nerr" and "sys_errlist[]" variables
srcalt=fmax_errno
This is needed for systems using GNU GLIBC 2.3.2, or any other compilation
environment where errno is not exported by default or where the system
definition of fmax differs from that in 252.eon.
srcalt=stdcpp
This src.alt addresses issues with lack of legacy header files
as well as the old fmax() problem.
253.perlbmk:
-DSPEC_CPU2000_LINUX_I386
Enable the code changes for porting to Linux on i386 architecture to be utilized
-DSPEC_CPU2000_NEED_BOOL
Use SPEC provided definition of the boolean type
-DSPEC_CPU2000_GLIBC22
Compatibility with 2.2 & later versions of glibc
-DSPEC_CPU2000_LP64
Compile using LP64 programming model.

254.gap:
-DSYS_IS_USG
Tells that the operating system is USG compliant
-DSYS_HAS_IOCTL_PROTO
Do not explicitly declare ioctl()
-DSYS_HAS_TIME_PROTO
Do not explicitly declare time().
-DSYS_HAS_SIGNAL_PROTO
Do not explicitly #include <signal.h>
-DSYS_HAS_ANSI
System is ANSI compliant.
-DSYS_HAS_CALLOC_PROTO
Do not supply a prototype for calloc().
-DSPEC_CPU2000_LP64
Compile using LP64 programming model.

255.vortex:
-DSPEC_CPU2000_LP64
Compile using LP64 programming model.
srcalt=closed_files
Original code attempted a final write using file pointers
which had just been closed. Those file pointers are
set to null in the approved source alt.