===================================================
HP-UX Flag Descriptions for CPU2000 - June 2004
===================================================
-----------------------------------------------------------------
Common Flags for HP-UX F90 Compiler, C Compiler and aCC Compiler
Compiler specific flags are mentioned below or in other notes
-----------------------------------------------------------------
+Olevel Invoke optimizations selected by level. Defined
values for level are:
0 Perform minimal optimizations.
1 Perform optimizations within basic
blocks only. This is the default.
2 Perform level 1 and global
optimizations. Same as -O.
3 Perform level 2 as well as
interprocedural global optimizations
within translation units.
4 Perform level 3 as well as doing
interprocedural optimizations across
translation units (link time
optimizations). Requires concurrent use
of the +Oprofile=use option.
NOTE: Object files generated at this
level contain an intermediate
representation of the user code and are
intended to be temporary files. These
intermediate object files are not
guaranteed to be compatible from one
version of the compiler to the next.
+O[no]datalayout
Enables [disables] profile-driven layout of global
and static data items to improve cache memory
utilization. This option is currently ignored in
the absence of the dynamic profile feedback
option, +Oprofile=use. The default is
+Onodatalayout.
+O[no]dataprefetch
Enable [disable] optimizations to generate data
prefetch instructions for data structures
referenced within innermost loops.
+Odataprefetch is the same as
+Odataprefetch=indirect.
+Onodataprefetch is the same as
+Odataprefetch=none.
+Oentrysched Perform instruction scheduling on
a subprogram's entry and exit code sequences.
This option can be used at optimization level 1
and higher. The default is +Onoentrysched.
+Ofast This option selects a combination of compilation
and link options for optimum execution speed and
reasonable build times. Currently: +O2,
+Onolimit, +Olibcalls, +Ofltacc=relaxed,
+DSnative, +FPD, -Wl,+pi,4M, -Wl,+pd,4M and
-Wl,+mergeseg. This option is a synonym for
-fast. Some of the linker settings above can be
changed with chatr(1).
+Ofaster This option selects +Ofast, but with an increased
optimization level. For f90 the level is +O4.
For aCC, if used with +Oprofile=use the optimization
level will be +O4. Otherwise it will be +O3.
+Ofltacc=level Controls the level of floating point optimizations
that the compiler may perform. The defined values
for level are:
default
Allows contractions, such as fused
multiply-add (FMA), but disallows any
other floating point optimization that
can result in numerical differences.
limited
Like default, but also allows floating
point optimizations which may affect the
generation and propagation of
infinities, NaNs, and the sign of zero.
relaxed
In addition to the optimizations allowed
by limited, permits optimizations, such
as reordering of expressions, even if
parenthesized, that may affect rounding
error. This is the same as +Onofltacc.
strict
Disallows any floating point
optimization that can result in
numerical differences. This is the same
as +Ofltacc.
The default is +Ofltacc=default.
+Olibcalls NOTE: This option is deprecated and may not be
supported in future releases. On Itanium(R)-based HP-
UX, including a system header file will cause the
functions declared therein to be eligible for libcalls
transformations, regardless of the state of
+O[no]libcalls.
+O[no]initcheck
Enable [disable] initialization to zero of any
local, scalar, non-static variable that is
uninitialized with respect to at least one path
leading to its use. This optimization can occur
at optimization levels 2, 3, and 4. The default
is to enable initialization if the variable is
uninitialized with respect to every path leading
to its use.
+O[no]inline Request [disable] inlining and cloning. This option can be
used at optimization level 3 and higher. The
default is +Oinline.
+O[no]inline=function1[,function2...]
Enable [disable] optimizer inlining for the named
functions. This optimization can occur at
optimization levels 3 and 4. The default is +Oinline.
+Oinlinebudget=n aCC(1)/cc(1)
+Oinline_budget=n f90(1)
Perform more aggressive inlining, where n
specifies the degree of aggressiveness, as
follows:
100 Default level of inlining.
> 100 More aggressive inlining at the
expense of compilation time and code
size. The maximum for n is 1000000.
2 - 99 Less aggressive inlining. The
optimizer gives more weight to
compilation time and code size when
determining whether to inline.
1 Inline only if it reduces code size.
This option can be used at optimization level 3 or
higher.
+O[no]limit Suppress [do not suppress] optimizations that
significantly increase compile-time or consume
enormous amounts of memory.
+Olimit is the same as +Olimit=min.
+Onolimit is the same as +Olimit=none.
+Olimit=level Controls the amount of compile-time spent
performing optimization. The defined values for
level are:
default
Based on tuning heuristics, the
optimizer will spend a reasonable amount
of time processing large procedures.
min For large procedures, the optimizer will
avoid non-linear time optimizations.
none The optimizer will fully optimize large
procedures, possibly resulting in
significantly increased compile time.
+O[no]loop_unroll[=unroll_factor]
Enable [disable] loop unrolling. This optimization
can occur at optimization levels 2, 3, and 4. The
default is +Oloop_unroll. The default is 4, that
is, four copies of the loop body. The
unroll_factor controls code expansion.
+Oprefetch_latency=cycles
+Oprefetch_latency applies to loops for which the
compiler generates data prefetch instructions.
cycles must be in the range of 0 to 10000. A
value of zero instructs the compiler to use the
default value, which is 480 cycles for loops
containing floating-point accesses and 150 cycles
for loops that do not contain any floating-point
accesses. See the HP aC++ Online Programmer's
Guide or HP C Online Help.
+O[no]procelim Enable [disable] the elimination of functions that
are not referenced by the application. Only
functions with the hidden export class may be
eliminated. The default is +Onoprocelim at
optimization levels 0 and 1; at levels 2, 3 and 4,
the default is +Oprocelim.
+O[no]ptrs_to_globals[=name1,name2,...,nameN]
Tell the optimizer whether global variables are
modified [are not modified] through pointers.
This optimization can occur at levels 2, 3, 4. The
default is +Optrs_to_globals
+O[no]recovery Generate [do not generate] recovery code for
control speculation. The default is +Onorecovery.
+Oshortdata[=size]
All objects of size size bytes or smaller will be
placed in the short data area, and references to such
data will assume it resides in the short data area.
Valid values of n are 0, or a decimal number between 8
and 4,194,304 (4MB). If no size is specified, all data
is placed in the short data area. If size is 0, no
data will be placed in the short data area, and all
data references will use long offsets. The default is
+Oshortdata=8.
+O[no]type_safety=[off|limited|ansi|strong]
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 cc.
limited
Code follows ANSI aliasing rules, and
that 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 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 cc.
-Bprotected[=symbol[,symbol...]]
The named symbols, or all symbols if no symbols are
specified, are assigned the protected export class.
That means these symbols will not be preempted by
symbols from other load modules, so the compiler may
bypass the linkage table for both code and data
references and bind them to locally defined code and
data symbols.
-Bprotected_data
Marks only data symbols as having the protected export
class.
(e.g. +DD32 or +DD64)
+DDdata_model Generate code using either the ILP32 or LP64 data
model. Defined values for data_model are:
32 Use the ILP32 data model. The sizes of
the int, long and pointer data types are
32-bits.
64 Use the LP64 data model. The size of
the int data type is 32-bits, and the
sizes of the long and pointer data types
are 64-bits. Defines __LP64__ to the
preprocessor.
The default is +DD32.
+DSmodel Perform instruction scheduling appropriate for a
e.g. specific implementation of the architecture.
+DSnative
ON IPF the defined values for model are:
blended Tune for best performance on a
combination of processors (i.e.,
Itanium or Itanium 2 processor).
itanium Tune for best performance on an
Itanium processor.
itanium2 Tune for best performance on an
Itanium 2 processor.
native Tune for best performance on the
processor on which the compiler is
running.
The default model is blended.
+FPflag Specify how the environment for floating-point
e.g. operations should be initialized at program
+FPD start-up. By default, all behaviors are disabled.
The following flags are supported (upper case flag
enables; lower case flag disables):
D (d) Enable sudden underflow (flush to zero)
of denormalized values.
-Wl,-asearch
e.g. (ld option -a search) Specifies library search order.
-Wl,-aarchive_shared
Specify whether shared or archive libraries are
searched with the -l option. The value of search
should be one of archive, shared, archive_shared,
shared_archive, or default. This option can
appear more than once, interspersed among -l
options, to control the searching for each
library. The default is to use the shared version
of a library if one is available, or the archive
version if not.
If either archive or shared is active, only the
specified library type is accepted.
If archive_shared is active, the archive form is
preferred, but the shared form is allowed.
If shared_archive is active, the shared form is
preferred but the archive form is allowed.
-dynamic Produces dynamically bound executables. See -minshared
for partially statically bound executables. The
default behavior is dynamic.
-exec Indicates that any object files created will be used to
create an executable file. Constants with a protected
or hidden export class are placed in the read-only data
section. This option also implies -Bprotected_def.
-minshared Indicates that the result of the current
compilation is going into an executable file that
will make minimal use of shared libraries. Equivalent
to -exec -Bprotected -Wl,-a,archive_shared.
[Profile Feedback Related Options]
+Oprofile=collect
+Oprofile=collect[:<qualifiers>]
+I Instrument the application for profile based
optimization. The profile collection <qualifiers>
are:
arc Collect arc counts (equivalent to
+Oprofile=collect). This is the
default.
stride
Collect stride data.
all Collect all types of profile data.
Equivalent to the command
+Oprofile=collect=arc,stride
<qualifiers> are a comma-separated list of profile
collection qualifiers.
+Oprofile=use
+P Optimize the application based on profile data found in
the database file flow.data, produced by compilation
with +I. +P is equivalent to +Oprofile=use or
+Oprofile=use:filename. See ld(1), +I, and +df, for
more details. The +P option is incompatible with the +I
and -S options. It is incompatible with the -g option
only during compile time.
-----------------------------------------------
Specific Flags for HP-UX F90 Compiler
-----------------------------------------------
+cat Concatenates all source files of the same source
form together, then compiles the concatenated
source all at once. This enables inlining at +O3
within the concatenated file.
-----------------------------------------------
Specific Flags for HP-UX C and aCC Compiler
-----------------------------------------------
-AOe In addition to specifying the extended ANSI C language dialect
as per -Ae (the default), allows the optimizer to aggressively
exploit the assumption that the source code conforms to the ANSI
programming language C standard ISO 9899:1990 plus the
extensions. At present, the effect is to make
+Otype_safety=ansi the default (it can of course be overridden).
As new independently-controllable optimizations are developed
that depend on the extended ANSI C standard, the flags that
enable those optimizations may also become the default under
-AOe.
-Ae Turns on ANSI C c89 mode. This option allows
compilation of c89 compatible C source programs just
like C compiler.
+inline_level [i]num
This option controls how C/C++ inlining hints influence
aCC or cc. Specify num as 0, 1, 2, or 3.
num Meaning
0 No inlining is done (same effect as the +d option).
1 Only small functions are inlined.
2 Only large functions are not inlined.
3 Inlining hints are respected in all cases,
except when the called function is recursive or
when it has a variable number of arguments.
The default level depends on +Olevel as shown in the
following table:
level num
0 1
1 1
2 2
3 2
4 2
If i is also specified, then implicit inlining is
invoked for "small" functions without the inline
keyword.
NOTE: This option controls functions declared with the
inline keyword or within the class declaration and is
effective at all optimization levels. The options
+Oinline and +Oinlinebudget control the high level
optimizer that recognizes other opportunities in the
same source file (+O3) or amongst all source files
(+O4).
-----------------------------------------------
Other descriptions
-----------------------------------------------
-llapack Link in highly tuned math library functions found in the
LAPACK library. B6061AA (HP MLIB) is an optional HP
product which contains the LAPACK library.
effmem.o Replacement for malloc/free that assumes ANSI compliance
and improves spatial locality and minimizes memory usage
by not maintaining a free list.
fastmem.o Replacement for malloc/free that assumes ANSI compliance.
-----------------------------------------------
Descriptions of Portability Flags
-----------------------------------------------
+source={fixed|free|default}
Accept source files in fixed format
(+source=fixed) or free format (+source=free).
The default, +source=default, is free for .f90
files and fixed for .f and .F source files.
176.gcc
-DHOST_WORDS_BIG_ENDIAN : controls how bytes are numbered within
a word.
181.mcf
-DWANT_STDC_PROTO : allows use of the designated prototype.
186.crafty
-DHP : selects header files and code paths compatible with HPUX.
252.eon
-DFMAX_IS_DOUBLE : function fmax returns a double
-DNDEBUG : do not include debug code
-DSPEC_CPU2000_LP64 : use code to make longs and pointers 64 bit
253.perlbmk
-DSPEC_CPU2000_HP : Compile the SPEC CPU2000 modified perl for an
HPUX system.
254.gap:
-DSYS_HAS_CALLOC_PROTO : allows use of the designated prototype
-DSYS_HAS_IOCTL_PROTO : allows use of the designated prototype
-DSYS_HAS_TIME_PROTO : allows use of the designated prototype
-DSPEC_CPU2000_HP : selects header files and code paths compatible
with HPUX.
-DSYS_IS_USG : Compile for a USGish system.
-----------------------------------------------
Descriptions of Kernel Tunables
-----------------------------------------------
(Unless otherwise noted, units are in bytes)
dbc_max_pct Maximum dynamic buffer cache size as a percent of system memory
dbc_min_pct Minimum dynamic buffer cache size as a percent of system memory
maxdsiz Maximum data size
maxdsiz_64bit Maximum data size for 64 bit applications
maxssiz Maximum stack size
maxssiz_64bit Maximum stack size for 64 bit applications
maxtsiz Maximum thread data size
maxtsiz_64bit Maximum thread data size for 64 bit applications
vps_ceiling Maximum System-Selected Page Size (in Kbytes)
vps_pagesize Default user page size (in Kbytes)
swapmem_on Swap to memory flag.
-----------------------------------------------
Descriptions of Other Options and Commands
-----------------------------------------------
mpsched Control the processor or locality domain on which a
specific process executes. (HP-UX command)
tmplog In tmplog mode, the intent log is almost always delayed.
This improves performance, but recent changes may
disappear if the system crashes. This mode is only
recommended for temporary file systems.
nolog nolog is an alias for tmplog.
convosync=delay
Alters the caching behavior of the file system for O_SYNC
and O_DSYNC I/O operations.
The delay value delays O_SYNC or O_DSYNC writes so that
they do not take effect immediately. With this option,
VxFS changes O_SYNC or O_DSYNC writes into delayed
writes. No special action is performed when closing a
file. This option effectively cancels data integrity
guarantees normally provided by opening a file with
O_SYNC or O_DSYNC.
mincache=tmpcache
Alter the caching behavior of the file system.
The tmpcache value disables delayed extending writes,
trading off integrity for performance. When this option
is chosen, VxFS does not zero out new extents allocated
as files are sequentially written. Uninitialized data
may appear in files being written at the time of a system
crash.