oldlinux-files/gnu/gcc/gcc-2.2.2/gcc.1

.\" Copyright (c) 1991 Free Software Foundation              -*-Text-*-
.\" See section COPYING for conditions for redistribution
.\" FIXME: no info here on predefines.  Should there be?  extra for C++...
.TH GCC 1 "27dec1991" "GNU Tools" "GNU Tools"
.de BP
.sp
.ti \-.2i
\(**
..
.SH NAME
gcc, g++ \- GNU project C and C++ Compiler (v2 preliminary)
.SH SYNOPSIS
.RB gcc " [" \c
.IR option " | " filename " ].\|.\|.
.br
.RB g++ " [" \c
.IR option " | " filename " ].\|.\|.
.SH WARNING
The information in this man page is an extract from the full
documentation of the GNU C compiler, and is limited to the meaning of
the options. 

For complete, current documentation, refer to the Info file `\|\c
.B gcc\c
\&\|' or the manual
.I 
Using and Porting GNU CC (for version 2.0)\c
\&.  Both are made from the Texinfo source file
.BR gcc.texinfo .
.SH DESCRIPTION
The C and C++ compilers are integrated.  Both process input files
through one or more of four stages: preprocessing, compilation,
assembly, and linking.  Source filename suffixes identify the source
language, but which name you use for the compiler governs default
assumptions:
.TP
.B gcc
assumes preprocessed (\c
.B .i\c
\&) files are C and assumes C style linking.
.TP
.B g++
assumes preprocessed (\c
.B .i\c
\&) files are C++ and assumes C++ style linking.
.PP 
Suffixes of source file names indicate the language and kind of
processing to be done:

.ta \w'.cxx 'u
.B .c\c
\&	C source; preprocess, compile, assemble 
.br
.B .C\c
\&	C++ source; preprocess, compile, assemble
.br
.B .cc\c
\&	C++ source; preprocess, compile, assemble
.br
.B .cxx\c
\&	C++ source; preprocess, compile, assemble
.br
.B .m\c
\&	Objective-C source; preprocess, compile, assemble
.br
.B .i\c
\&	preprocessed C or C++; compile, assemble
.br
.B .s\c
\&	Assembler source; assemble 
.br
.B .S\c
\&	Assembler source; preprocess, assemble 
.br
.B .h\c
\&	Preprocessor file; not usually named on command line

.I ??\c
\&	Other (unrecognized) files passed to linker. 
.br
\&	Common cases:
.br
.B .o\c
\&	Object file
.br
.B .a\c
\&	Archive file

Linking is always the last stage unless you use one of the 
.BR \-c ,
.BR \-S ,
or 
.B \-E 
options to avoid it (or unless compilation errors stop the whole
process).  For the link stage, all 
.B .o
files corresponding to source files,
.B \-l 
libraries, unrecognized filenames (including named 
.B .o
object files and 
.B .a 
archives)
are passed to the linker in command-line order.

.SH OPTIONS
Options must be separate: `\|\c
.B \-dr\c
\&\|' is quite different from `\|\c
.B \-d \-r
\&\|'. 

Most `\|\c
.B \-f\c
\&\|' and `\|\c
.B \-W\c
\&\|' options have two contrary forms: 
.BI \-f name
and
.BI \-fno\- name\c
\& (or 
.BI \-W name
and
.BI \-Wno\- name\c
\&). Only the non-default forms are shown here.

Here is a summary of all the options, grouped by type.  Explanations are
in the following sections.

.hy 0
.na
.TP
.B Overall Options
.br
\-c
\-S
\-E
.RI "\-o " file
\-pipe
\-v
.RI "\-x " language

.TP
.B Language Options
\-ansi
\-fall\-virtual
\-fcond\-mismatch
\-fdollars\-in\-identifiers
\-fenum\-int\-equiv
\-fno\-asm
\-fno\-builtin
\-fno\-strict\-prototype
\-fsigned\-bitfields
\-fsigned\-char
\-fthis\-is\-variable
\-funsigned\-bitfields
\-funsigned\-char
\-fwritable\-strings
\-traditional
\-traditional\-cpp
\-trigraphs

.TP
.B Warning Options
\-fsyntax\-only
\-pedantic
\-pedantic\-errors
\-w
\-W
\-Wall
\-Waggregate\-return
\-Wcast\-align
\-Wcast\-qual
\-Wcomment
\-Wconversion
\-Wenum\-clash
\-Werror
\-Wformat
.RI \-Wid\-clash\- len
\-Wimplicit
\-Winline
\-Wmissing\-prototypes
\-Wparentheses
\-Wpointer\-arith
\-Wreturn\-type
\-Wshadow
\-Wstrict\-prototypes
\-Wswitch
\-Wtraditional
\-Wtrigraphs
\-Wuninitialized
\-Wunused
\-Wwrite\-strings

.TP
.B Debugging Options
\-a
.RI \-d letters
\-fpretend\-float
\-g
\-gstabs
\-gdwarf
\-ggdb
\-gsdb
\-p
\-pg
\-save\-temps

.TP
.B Optimization Options
\-fcaller\-saves
\-fcse\-follow\-jumps
\-fdelayed\-branch
\-felide\-constructors
\-fexpensive\-optimizations
\-ffloat\-store
\-fforce\-addr
\-fforce\-mem
\-finline
\-finline\-functions
\-fkeep\-inline\-functions
\-fmemoize\-lookups
\-fno\-default\-inline
\-fno\-defer\-pop
\-fno\-function\-cse
\-fomit\-frame\-pointer
\-frerun\-cse\-after\-loop
\-fschedule\-insns
\-fschedule\-insns2
\-fstrength\-reduce
\-fthread\-jumps
\-funroll\-all\-loops
\-funroll\-loops
\-O
\-O2

.TP
.B Preprocessor Options
\-C
\-dD
\-dM
\-dN
.RI \-D macro [\|= defn \|]
\-E
\-H
.RI "\-i " file
\-M
\-MD
\-MM
\-MMD
\-nostdinc
\-P
.RI \-U macro
\-undef

.TP
.B Linker Options
.RI \-l library
\-nostdlib
\-static

.TP
.B Directory Options
.RI \-B prefix
.RI \-I dir
\-I\-
.RI \-L dir

.TP
.B Target Options
.RI "\-b  " machine
.RI "\-V " version

.TP
.B Machine Dependent Options
.I M680x0\ Options
.br
\-m68000
\-m68020
\-m68881
\-mbitfield
\-mc68000
\-mc68020
\-mfpa
\-mnobitfield
\-mrtd
\-mshort
\-msoft\-float

.I VAX Options
.br
\-mg
\-mgnu
\-munix

.I SPARC Options
.br
\-mfpu
\-mno\-epilogue

.I Convex Options
.br
\-margcount
\-mc1
\-mc2
\-mnoargcount

.I AMD29K Options
.br
\-m29000
\-m29050
\-mbw
\-mdw
\-mkernel\-registers
\-mlarge
\-mnbw
\-mnodw
\-msmall
\-mstack\-check
\-muser\-registers

.I M88K Options
.br
\-mbig\-pic
\-mcheck\-zero\-division
\-mhandle\-large\-shift
\-midentify\-revision
\-mno\-check\-zero\-division
\-mno\-ocs\-debug\-info
\-mno\-ocs\-frame\-position
\-mno\-optimize\-arg\-area
\-mno\-underscores
\-mocs\-debug\-info
\-mocs\-frame\-position
\-moptimize\-arg\-area
.RI \-mshort\-data\- num
\-msvr3
\-msvr4
\-mtrap\-large\-shift
\-muse\-div\-instruction
\-mversion\-03.00
\-mwarn\-passed\-structs

.I RS6000 Options
.br
\-mfp\-in\-toc
\-mno\-fop\-in\-toc

.I RT Options
.br
\-mcall\-lib\-mul
\-mfp\-arg\-in\-fpregs
\-mfp\-arg\-in\-gregs
\-mfull\-fp\-blocks
\-mhc\-struct\-return
\-min\-line\-mul
\-mminimum\-fp\-blocks
\-mnohc\-struct\-return

.I MIPS Options
.br
\-mcpu=\fIcpu type\fP
\-mips2
\-mips3
\-mint64
\-mlong64
\-mlonglong128
\-mmips\-as
\-mgas
\-mrnames
\-mno\-rnames
\-mgpopt
\-mno\-gpopt
\-mstats
\-mno\-stats
\-mmemcpy
\-mno\-memcpy
\-mno\-mips\-tfile
\-mmips\-tfile
\-msoft\-float
\-mhard\-float
\-mabicalls
\-mno\-abicalls
\-mhalf\-pic
\-mno\-half\-pic
\-G \fInum\fP

.I i386 Options
.br
\-m486
\-mno486
\-msoft\-float

.TP
.B Code Generation Options
.RI +e N
.RI \-fcall\-saved\- reg
.RI \-fcall\-used\- reg
.RI \-ffixed\- reg
\-fno\-common
\-fno\-gnu\-binutils
\-fnonnull\-objects
\-fpcc\-struct\-return
\-fpic
\-fPIC
\-fshared\-data
\-fshort\-enums
\-fshort\-double
\-fvolatile
.ad b
.hy 1

.SH OVERALL OPTIONS
.TP
.BI "\-x " "language"
Specify explicitly the 
.I language\c
\& for the following input files (rather than choosing a default based
on the file name suffix) .  This option applies to all following input
files until the next `\|\c
.B \-x\c
\&\|' option.  Possible values of \c
.I language\c
\& are
`\|\c
.B c\c
\&\|', `\|\c
.B objective\-c\c
\&\|', `\|\c
.B c\-header\c
\&\|', `\|\c
.B c++\c
\&\|',
`\|\c
.B cpp\-output\c
\&\|', `\|\c
.B assembler\c
\&\|', and `\|\c
.B assembler\-with\-cpp\c
\&\|'.
.TP
.B \-x none
Turn off any specification of a language, so that subsequent files are
handled according to their file name suffixes (as they are if `\|\c
.B \-x\c
\&\|'
has not been used at all).
.PP

If you want only some of the four stages (preprocess, compile,
assemble, link), you can use
`\|\c
.B \-x\c
\&\|' (or filename suffixes) to tell \c
.B gcc\c
\& where to start, and
one of the options `\|\c
.B \-c\c
\&\|', `\|\c
.B \-S\c
\&\|', or `\|\c
.B \-E\c
\&\|' to say where
\c
.B gcc\c
\& is to stop.  Note that some combinations (for example,
`\|\c
.B \-x cpp\-output \-E\c
\&\|') instruct \c
.B gcc\c
\& to do nothing at all.
.TP
.B \-c
Compile or assemble the source files, but do not link.  The compiler
output is an object file corresponding to each source file.

By default, GCC makes the object file name for a source file by replacing
the suffix `\|\c
.B .c\c
\&\|', `\|\c
.B .i\c
\&\|', `\|\c
.B .s\c
\&\|', etc., with `\|\c
.B .o\c
\&\|'.  Use
.B \-o\c
\& to select another name.

GCC ignores any unrecognized input files (those that do not require
compilation or assembly) with the
.B \-c
option.
.TP
.B \-S
Stop after the stage of compilation proper; do not assemble.  The output
is an assembler code file for each non-assembler input
file specified.

By default, GCC makes the assembler file name for a source file by
replacing the suffix `\|\c
.B .c\c
\&\|', `\|\c
.B .i\c
\&\|', etc., with `\|\c
.B .s\c
\&\|'.  Use
.B \-o\c
\& to select another name.


GCC ignores any input files that don't require compilation.
.TP
.B \-E
Stop after the preprocessing stage; do not run the compiler proper.  The
output is preprocessed source code, which is sent to the
standard output.

GCC ignores input files which don't require preprocessing.
.TP
.BI "\-o " file\c
\&
Place output in file \c
.I file\c
\&.  This applies regardless to whatever
sort of output GCC is producing, whether it be an executable file,
an object file, an assembler file or preprocessed C code.

Since only one output file can be specified, it does not make sense to
use `\|\c
.B \-o\c
\&\|' when compiling more than one input file, unless you are
producing an executable file as output.

If you do not specify `\|\c
.B \-o\c
\&\|', the default is to put an executable file
in `\|\c
.B a.out\c
\&\|', the object file for `\|\c
.B \c
.I source\c
\&.\c
.I suffix\c
\&\c
\&\|' in
`\|\c
.B \c
.I source\c
\&.o\c
\&\|', its assembler file in `\|\c
.B \c
.I source\c
\&.s\c
\&\|', and
all preprocessed C source on standard output.
.TP
.B \-v
Print (on standard error output) the commands executed to run the stages
of compilation.  Also print the version number of the compiler driver
program and of the preprocessor and the compiler proper.
.TP
.B \-pipe
Use pipes rather than temporary files for communication between the
various stages of compilation.  This fails to work on some systems where
the assembler cannot read from a pipe; but the GNU assembler has
no trouble.
.PP

.SH LANGUAGE OPTIONS
The following options control the dialect of C that the compiler
accepts:
.TP
.B \-ansi
.I
Support all ANSI standard C programs.  
.I

This turns off certain features of GNU C that are incompatible with
ANSI C, such as the \c
.B asm\c
\&, \c
.B inline\c
\& and \c
.B typeof\c
\&
keywords, and predefined macros such as \c
.B unix\c
\& and \c
.B vax\c
\&
that identify the type of system you are using.  It also enables the
undesirable and rarely used ANSI trigraph feature, and makes the
preprocessor accept `\|\c
.B $\c
\&\|' as part of identifiers.

The alternate keywords \c
.B __asm__\c
\&, \c
.B __extension__\c
\&,
\c
.B __inline__\c
\& and \c
.B __typeof__\c
\& continue to work despite
`\|\c
.B \-ansi\c
\&\|'.  You would not want to use them in an ANSI C program, of
course, but it is useful to put them in header files that might be included
in compilations done with `\|\c
.B \-ansi\c
\&\|'.  Alternate predefined macros
such as \c
.B __unix__\c
\& and \c
.B __vax__\c
\& are also available, with or
without `\|\c
.B \-ansi\c
\&\|'.

The `\|\c
.B \-ansi\c
\&\|' option does not cause non-ANSI programs to be
rejected gratuitously.  For that, `\|\c
.B \-pedantic\c
\&\|' is required in
addition to `\|\c
.B \-ansi\c
\&\|'.  

The preprocessor predefines a macro \c
.B __STRICT_ANSI__\c
\& when you use the `\|\c
.B \-ansi\c
\&\|'
option.  Some header files may notice this macro and refrain
from declaring certain functions or defining certain macros that the
ANSI standard doesn't call for; this is to avoid interfering with any
programs that might use these names for other things.
.TP
.B \-fno\-asm
Do not recognize \c
.B asm\c
\&, \c
.B inline\c
\& or \c
.B typeof\c
\& as a
keyword.  These words may then be used as identifiers.  You can
use \c
.B __asm__\c
\&, \c
.B __inline__\c
\& and \c
.B __typeof__\c
\& instead.
`\|\c
.B \-ansi\c
\&\|' implies `\|\c
.B \-fno\-asm\c
\&\|'.
.TP
.B \-fno\-builtin
.I (Ignored for C++.)
Don't recognize non-ANSI built-in functions.  `\|\c
.B \-ansi\c
\&\|' also has this
effect. Currently, the only function affected is 
.B alloca\c
\&.
.TP
.B \-fno\-strict\-prototype
.I
(C++ only.)
Consider the declaration \c
.B int foo ();\c
\&.  In C++, this means that the
function \c
.B foo\c
\& takes no arguments.  In ANSI C, this is declared
.B int foo(void);\c
\&.  With the flag `\|\c
.B \-fno\-strict\-prototype\c
\&\|',
declaring functions with no arguments is equivalent to declaring its
argument list to be untyped, i.e., \c
.B int foo ();\c
\& is equivalent to
saying \c
.B int foo (...);\c
\&.
.TP
.B \-trigraphs
Support ANSI C trigraphs.  The `\|\c
.B \-ansi\c
\&\|' option implies `\|\c
.B \-trigraphs\c
\&\|'.
.TP
.B \-traditional
Attempt to support some aspects of traditional C compilers.
For details, see the GNU C Manual; the duplicate list here
has been deleted so that we won't get complaints when it
is out of date.

But one note about C++ programs only (not C).  `\|\c
.B \-traditional\c
\&\|' has one additional effect for C++: assignment to 
.B this
is permitted.  This is the same as the effect of `\|\c
.B \-fthis\-is\-variable\c
\&\|'.
.TP
.B \-traditional\-cpp
Attempt to support some aspects of traditional C preprocessors.
This includes the items that specifically mention the preprocessor above,
but none of the other effects of `\|\c
.B \-traditional\c
\&\|'.
.TP
.B \-fdollars\-in\-identifiers
.I
(C++ only.)
Permit the use of `\|\c
.B $\c
\&\|' in identifiers.   (For GNU C, this is the default, and you can
forbid it with `\|\c
.B \-ansi\c
\&\|'.)
Traditional C allowed the character `\|\c
.B $\c
\&\|' to form part of identifiers; by default, GNU C also
allows this.  However, ANSI C forbids `\|\c
.B $\c
\&\|' in identifiers, and GNU C++ also forbids it by default on most
platforms (though on some platforms it's enabled by default for GNU
C++ as well).
.TP
.B \-fenum\-int\-equiv
.I
(C++ only.)
Normally GNU C++ allows conversion of 
.B enum
to
.B int\c
\&, but not the other way around.  Use this option if you want GNU C++
to allow conversion of
.B int
to 
.B enum
as well.  
.TP
.B \-fall\-virtual
.I
(C++ only.)
When you use the `\|\c
.B \-fall\-virtual\c
\&\|', all member functions
(except for constructor functions and new/delete member operators)
declared in the same class with a ``method-call'' operator method are
treated as virtual functions of the given class.  In effect, all
of these methods become ``implicitly virtual.''

This does \c
.I not\c
\& mean that all calls to these methods will be made through the
internal table of virtual functions.  There are some circumstances
under which it is obvious that a call to a given virtual function can
be made directly, and in these cases the calls still go direct.

The effect of making all methods of a class with a declared
\&`\|\c
.B
operator->()()\c
\&\|' implicitly virtual using `\|\c
.B \-fall\-virtual\c
\&\|' extends
also to all non-constructor methods of any class derived from such a
class.
.TP
.B \-fcond\-mismatch
Allow conditional expressions with mismatched types in the second and
third arguments.  The value of such an expression is void.
.TP
.B \-fthis\-is\-variable
.I
(C++ only.)
The incorporation of user-defined free store management into C++ has
made assignment to \c
.B this\c
\& an anachronism.  Therefore, by default GNU
C++ treats the type of \c
.B this\c
\& in a member function of \c
.B class X\c
\&
to be \c
.B X *const\c
\&.  In other words, it is illegal to assign to
\c
.B this\c
\& within a class member function.  However, for backwards
compatibility, you can invoke the old behavior by using
\&`\|\c
.B \-fthis\-is\-variable\c
\&\|'.
.TP
.B \-funsigned\-char
Let the type \c
.B char\c
\& be unsigned, like \c
.B unsigned char\c
\&.

Each kind of machine has a default for what \c
.B char\c
\& should
be.  It is either like \c
.B unsigned char\c
\& by default or like
\c
.B signed char\c
\& by default.

Ideally, a portable program should always use \c
.B signed char\c
\& or
\c
.B unsigned char\c
\& when it depends on the signedness of an object.
But many programs have been written to use plain \c
.B char\c
\& and
expect it to be signed, or expect it to be unsigned, depending on the
machines they were written for.  This option, and its inverse, let you
make such a program work with the opposite default.

The type \c
.B char\c
\& is always a distinct type from each of
\c
.B signed char\c
\& and \c
.B unsigned char\c
\&, even though its behavior
is always just like one of those two.

.TP
.B \-fsigned\-char
Let the type \c
.B char\c
\& be signed, like \c
.B signed char\c
\&.

Note that this is equivalent to `\|\c
.B \-fno\-unsigned\-char\c
\&\|', which is
the negative form of `\|\c
.B \-funsigned\-char\c
\&\|'.  Likewise,
`\|\c
.B \-fno\-signed\-char\c
\&\|' is equivalent to `\|\c
.B \-funsigned\-char\c
\&\|'.
.TP
.B \-fsigned\-bitfields
.TP
.B \-funsigned\-bitfields
.TP
.B \-fno\-signed\-bitfields
.TP
.B \-fno\-unsigned\-bitfields
These options control whether a bitfield is
signed or unsigned, when declared with no explicit `\|\c
.B signed\c
\&\|' or `\|\c
.B unsigned\c
\&\|' qualifier.  By default, such a bitfield is
signed, because this is consistent: the basic integer types such as
\c
.B int\c
\& are signed types.

However, when you specify `\|\c
.B \-traditional\c
\&\|', bitfields are all unsigned
no matter what.
.TP
.B \-fwritable\-strings
Store string constants in the writable data segment and don't uniquize
them.  This is for compatibility with old programs which assume they
can write into string constants.  `\|\c
.B \-traditional\c
\&\|' also has this
effect.

Writing into string constants is a very bad idea; ``constants'' should
be constant.
.PP

.SH PREPROCESSOR OPTIONS
These options control the C preprocessor, which is run on each C source
file before actual compilation.

If you use the `\|\c
.B \-E\c
\&\|' option, GCC does nothing except preprocessing.
Some of these options make sense only together with `\|\c
.B \-E\c
\&\|' because
they cause the preprocessor output to be unsuitable for actual
compilation.
.TP
.BI "\-i " file\c
\&
Process \c
.I file\c
\& as input, discarding the resulting output, before
processing the regular input file.  Because the output generated from
\c
.I file\c
\& is discarded, the only effect of `\|\c
.B \-i \c
.I file\c
\&\c
\&\|' is to
make the macros defined in \c
.I file\c
\& available for use in the main
input.  The preprocessor evaluates any `\|\c
.B \-D\c
\&\|' and `\|\c
.B \-U\c
\&\|' options
on the command line before processing `\|\c
.B \-i\c
\&\|' \c
.I file\c
\&.
.TP
.B \-nostdinc
Do not search the standard system directories for header files.  Only
the directories you have specified with `\|\c
.B \-I\c
\&\|' options (and the
current directory, if appropriate) are searched.   

By using both `\|\c
.B \-nostdinc\c
\&\|' and `\|\c
.B \-I\-\c
\&\|', you can limit the include-file search file to only those
directories you specify explicitly.
.TP
.B \-undef
Do not predefine any nonstandard macros.  (Including architecture flags).
.TP
.B \-E
Run only the C preprocessor.  Preprocess all the C source files
specified and output the results to standard output or to the
specified output file.
.TP
.B \-C
Tell the preprocessor not to discard comments.  Used with the
`\|\c
.B \-E\c
\&\|' option.
.TP
.B \-P
Tell the preprocessor not to generate `\|\c
.B #line\c
\&\|' commands.
Used with the `\|\c
.B \-E\c
\&\|' option.
.TP
.B \-M
Tell the preprocessor to output a rule suitable for \c
.B make\c
\&
describing the dependencies of each object file.  For each source file,
the preprocessor outputs one \c
.B make\c
\&-rule whose target is the object
file name for that source file and whose dependencies are all the files
`\|\c
.B #include\c
\&\|'d in it.  This rule may be a single line or may be
continued with `\|\c
.B \e\c
\&\|'-newline if it is long.  The list of rules is
printed on standard output instead of the preprocessed C program.

`\|\c
.B \-M\c
\&\|' implies `\|\c
.B \-E\c
\&\|'.
.TP
.B \-MM
Like `\|\c
.B \-M\c
\&\|' but the output mentions only the user header files
included with `\|\c
.B #include "\c
.I file\c
\&"\c
\&\|'.  System header files
included with `\|\c
.B #include <\c
.I file\c
\&>\c
\&\|' are omitted.
.TP
.B \-MD
Like `\|\c
.B \-M\c
\&\|' but the dependency information is written to files with
names made by replacing `\|\c
.B .c\c
\&\|' with `\|\c
.B .d\c
\&\|' at the end of the
input file names.  This is in addition to compiling the file as
specified\(em\&`\|\c
.B \-MD\c
\&\|' does not inhibit ordinary compilation the way
`\|\c
.B \-M\c
\&\|' does.

The Mach utility `\|\c
.B md\c
\&\|' can be used to merge the `\|\c
.B .d\c
\&\|' files
into a single dependency file suitable for using with the `\|\c
.B make\c
\&\|'
command.
.TP
.B \-MMD
Like `\|\c
.B \-MD\c
\&\|' except mention only user header files, not system
header files.
.TP
.B \-H
Print the name of each header file used, in addition to other normal
activities.
.TP
.BI \-D macro
Define macro \c
.I macro\c
\& with the string `\|\c
.B 1\c
\&\|' as its definition.
.TP
.BI \-D macro = defn
Define macro \c
.I macro\c
\& as \c
.I defn\c
\&.    All instances of `\|\c
.B \-D\c
\&\|' on
the command line are processed before any `\|\c
.B \-U\c
\&\|' or `\|\c
.B \-i\c
\&\|' options.
.TP
.BI \-U macro
Undefine macro \c
.I macro\c
\&.  `\|\c
.B \-U\c
\&\|' options are evaluated after all `\|\c
.B \-D\c
\&\|' options, but before any `\|\c
.B \-i\c
\&\|' options.
.TP
.B \-dM
Tell the preprocessor to output only a list of the macro definitions
that are in effect at the end of preprocessing.  Used with the `\|\c
.B \-E\c
\&\|'
option.
.TP
.B \-dD
Tell the preprocessing to pass all macro definitions into the output, in
their proper sequence in the rest of the output.
.TP
.B \-dN
Like `\|\c
.B \-dD\c
\&\|' except that the macro arguments and contents are omitted.
Only `\|\c
.B #define \c
.I name\c
\&\c
\&\|' is included in the output.
.PP

.SH LINKER OPTIONS
These options come into play when the compiler links object files into
an executable output file.  They are meaningless if the compiler is
not doing a link step.
.TP
.I object-file-name
A file name that does not end in a special recognized suffix is
considered to name an object file or library.  (Object files are
distinguished from libraries by the linker according to the file
contents.)  If GCC does a link step, these object files are used as input
to the linker.
.TP
.BI \-l library\c
\&
Use the library named \c
.I library\c
\& when linking.

The linker searches a standard list of directories for the library,
which is actually a file named `\|\c
.B lib\c
.I library\c
\&.a\c
\&\|'.  The linker
then uses this file as if it had been specified precisely by name.

The directories searched include several standard system directories
plus any that you specify with `\|\c
.B \-L\c
\&\|'.

Normally the files found this way are library files\(em\&archive files
whose members are object files.  The linker handles an archive file by
scanning through it for members which define symbols that have so far
been referenced but not defined.  However, if the linker finds an
ordinary object file rather than a library, the object file is linked
in the usual fashion.  The only difference between using an `\|\c
.B \-l\c
\&\|' option and specifying a file
name is that `\|\c
.B \-l\c
\&\|' surrounds 
.I library
with `\|\c
.B lib\c
\&\|' and `\|\c
.B .a\c
\&\|' and searches several directories.
.TP
.B \-nostdlib
Don't use the standard system libraries and startup files when linking.
Only the files you specify will be passed to the linker.
.TP
.B \-static
On systems that support dynamic linking, this prevents linking with the shared
libraries.  (`\|\c
.B \-g\c
\&\|' also has this effect.)  On other systems, this option has no effect.
.PP

.SH DIRECTORY OPTIONS
These options specify directories to search for header files, for
libraries and for parts of the compiler:
.TP
.BI "\-I" "dir"\c
\&
Append directory \c
.I dir\c
\& to the list of directories searched for include files.
.TP
.B \-I\-
Any directories you specify with `\|\c
.B \-I\c
\&\|' options before the `\|\c
.B \-I\-\c
\&\|'
option are searched only for the case of `\|\c
.B 
#include "\c
.I file\c
.B
\&"\c
\&\|';
they are not searched for `\|\c
.B #include <\c
.I file\c
\&>\c
\&\|'.

If additional directories are specified with `\|\c
.B \-I\c
\&\|' options after
the `\|\c
.B \-I\-\c
\&\|', these directories are searched for all `\|\c
.B #include\c
\&\|'
directives.  (Ordinarily \c
.I all\c
\& `\|\c
.B \-I\c
\&\|' directories are used
this way.)

In addition, the `\|\c
.B \-I\-\c
\&\|' option inhibits the use of the current
directory (where the current input file came from) as the first search
directory for `\|\c
.B 
#include "\c
.I file\c
.B
\&"\c
\&\|'.  There is no way to
override this effect of `\|\c
.B \-I\-\c
\&\|'.  With `\|\c
.B \-I.\c
\&\|' you can specify
searching the directory which was current when the compiler was
invoked.  That is not exactly the same as what the preprocessor does
by default, but it is often satisfactory.

`\|\c
.B \-I\-\c
\&\|' does not inhibit the use of the standard system directories
for header files.  Thus, `\|\c
.B \-I\-\c
\&\|' and `\|\c
.B \-nostdinc\c
\&\|' are
independent.
.TP
.BI "\-L" "dir"\c
\&
Add directory \c
.I dir\c
\& to the list of directories to be searched
for `\|\c
.B \-l\c
\&\|'.
.TP
.BI "\-B" "prefix"\c
\&
This option specifies where to find the executables, libraries and
data files of the compiler itself.

The compiler driver program runs one or more of the subprograms
`\|\c
.B cpp\c
\&\|', `\|\c
.B cc1\c
\&\|' (or, for C++, `\|\c
.B cc1plus\c
\&\|'), `\|\c
.B as\c
\&\|' and `\|\c
.B ld\c
\&\|'.  It tries
\c
.I prefix\c
\& as a prefix for each program it tries to run, both with and
without `\|\c
.B \c
.I machine\c
\&/\c
.I version\c
\&/\c
\&\|'.

For each subprogram to be run, the compiler driver first tries the
`\|\c
.B \-B\c
\&\|' prefix, if any.  If that name is not found, or if `\|\c
.B \-B\c
\&\|'
was not specified, the driver tries two standard prefixes, which are
`\|\c
.B /usr/lib/gcc/\c
\&\|' and `\|\c
.B /usr/local/lib/gcc-lib/\c
\&\|'.  If neither of
those results in a file name that is found, the compiler driver
searches for the unmodified program
name, using the directories specified in your
`\|\c
.B PATH\c
\&\|' environment variable.

The run-time support file `\|\c
.B libgcc.a\c
\&\|' is also searched for using the
`\|\c
.B \-B\c
\&\|' prefix, if needed.  If it is not found there, the two
standard prefixes above are tried, and that is all.  The file is left
out of the link if it is not found by those means.  Most of the time,
on most machines, `\|\c
.B libgcc.a\c
\&\|' is not actually necessary.

You can get a similar result from the environment variable
\c
.B GCC_EXEC_PREFIX\c
\&; if it is defined, its value is used as a prefix
in the same way.  If both the `\|\c
.B \-B\c
\&\|' option and the
\c
.B GCC_EXEC_PREFIX\c
\& variable are present, the `\|\c
.B \-B\c
\&\|' option is
used first and the environment variable value second.
.PP

.SH WARNING OPTIONS
Warnings are diagnostic messages that report constructions which
are not inherently erroneous but which are risky or suggest there
may have been an error.

These options control the amount and kinds of warnings produced by GNU
CC:
.TP
.B \-fsyntax\-only
Check the code for syntax errors, but don't emit any output.
.TP
.B \-w
Inhibit all warning messages.
.TP
.B \-pedantic
Issue all the warnings demanded by strict ANSI standard C; reject
all programs that use forbidden extensions.

Valid ANSI standard C programs should compile properly with or without
this option (though a rare few will require `\|\c
.B \-ansi\c
\&\|').  However,
without this option, certain GNU extensions and traditional C features
are supported as well.  With this option, they are rejected.  There is
no reason to \c
.I use\c
\& this option; it exists only to satisfy pedants.

`\|\c
.B \-pedantic\c
\&\|' does not cause warning messages for use of the
alternate keywords whose names begin and end with `\|\c
.B __\c
\&\|'.  Pedantic
warnings are also disabled in the expression that follows
\c
.B __extension__\c
\&.  However, only system header files should use
these escape routes; application programs should avoid them.
.TP
.B \-pedantic\-errors
Like `\|\c
.B \-pedantic\c
\&\|', except that errors are produced rather than
warnings.
.TP
.B \-W
Print extra warning messages for these events:
.TP
\ \ \ \(bu
A nonvolatile automatic variable might be changed by a call to
\c
.B longjmp\c
\&.  These warnings are possible only in
optimizing compilation.

The compiler sees only the calls to \c
.B setjmp\c
\&.  It cannot know
where \c
.B longjmp\c
\& will be called; in fact, a signal handler could
call it at any point in the code.  As a result, you may get a warning
even when there is in fact no problem because \c
.B longjmp\c
\& cannot
in fact be called at the place which would cause a problem.
.TP
\ \ \ \(bu
A function can return either with or without a value.  (Falling
off the end of the function body is considered returning without
a value.)  For example, this function would evoke such a
warning:
.sp
.br
foo\ (a)
.br
{
.br
\ \ if\ (a\ >\ 0)
.br
\ \ \ \ return\ a;
.br
}
.br
.sp

Spurious warnings can occur because GNU CC does not realize that
certain functions (including \c
.B abort\c
\& and \c
.B longjmp\c
\&)
will never return.
.TP
\ \ \ \(bu
An expression-statement contains no side effects.
.TP
\ \ \ \(bu
An unsigned value is compared against zero with `\|\c
.B >\c
\&\|' or `\|\c
.B <=\c
\&\|'.
.PP
.TP
.B \-Wimplicit
Warn whenever a function or parameter is implicitly declared.
.TP
.B \-Wreturn\-type
Warn whenever a function is defined with a return-type that defaults
to \c
.B int\c
\&.  Also warn about any \c
.B return\c
\& statement with no
return-value in a function whose return-type is not \c
.B void\c
\&.
.TP
.B \-Wunused
Warn whenever a local variable is unused aside from its declaration,
whenever a function is declared static but never defined, and whenever
a statement computes a result that is explicitly not used.
.TP
.B \-Wswitch
Warn whenever a \c
.B switch\c
\& statement has an index of enumeral type
and lacks a \c
.B case\c
\& for one or more of the named codes of that
enumeration.  (The presence of a \c
.B default\c
\& label prevents this
warning.)  \c
.B case\c
\& labels outside the enumeration range also
provoke warnings when this option is used.
.TP
.B \-Wcomment
Warn whenever a comment-start sequence `\|\c
.B /*\c
\&\|' appears in a comment.
.TP
.B \-Wtrigraphs
Warn if any trigraphs are encountered (assuming they are enabled).
.TP
.B \-Wformat
Check calls to \c
.B printf\c
\& and \c
.B scanf\c
\&, etc., to make sure that
the arguments supplied have types appropriate to the format string
specified.
.TP
.B \-Wuninitialized
An automatic variable is used without first being initialized.

These warnings are possible only in optimizing compilation,
because they require data flow information that is computed only
when optimizing.  If you don't specify `\|\c
.B \-O\c
\&\|', you simply won't
get these warnings.

These warnings occur only for variables that are candidates for
register allocation.  Therefore, they do not occur for a variable that
is declared \c
.B volatile\c
\&, or whose address is taken, or whose size
is other than 1, 2, 4 or 8 bytes.  Also, they do not occur for
structures, unions or arrays, even when they are in registers.

Note that there may be no warning about a variable that is used only
to compute a value that itself is never used, because such
computations may be deleted by data flow analysis before the warnings
are printed.

These warnings are made optional because GNU CC is not smart
enough to see all the reasons why the code might be correct
despite appearing to have an error.  Here is one example of how
this can happen:

.sp
.br
{
.br
\ \ int\ x;
.br
\ \ switch\ (y)
.br
\ \ \ \ {
.br
\ \ \ \ case\ 1:\ x\ =\ 1;
.br
\ \ \ \ \ \ break;
.br
\ \ \ \ case\ 2:\ x\ =\ 4;
.br
\ \ \ \ \ \ break;
.br
\ \ \ \ case\ 3:\ x\ =\ 5;
.br
\ \ \ \ }
.br
\ \ foo\ (x);
.br
}
.br
.sp


If the value of \c
.B y\c
\& is always 1, 2 or 3, then \c
.B x\c
\& is
always initialized, but GNU CC doesn't know this.  Here is
another common case:

.sp
.br
{
.br
\ \ int\ save_y;
.br
\ \ if\ (change_y)\ save_y\ =\ y,\ y\ =\ new_y;
.br
\ \ .\|.\|.
.br
\ \ if\ (change_y)\ y\ =\ save_y;
.br
}
.br
.sp


This has no bug because \c
.B save_y\c
\& is used only if it is set.

Some spurious warnings can be avoided if you declare as
\c
.B volatile\c
\& all the functions you use that never return.
.TP
.B \-Wparentheses
Warn if parentheses are omitted in certain contexts.
.TP
.B \-Wall
All of the above `\|\c
.B \-W\c
\&\|' options combined.  These are all the
options which pertain to usage that we recommend avoiding and that we
believe is easy to avoid, even in conjunction with macros.
.PP

The remaining `\|\c
.B \-W.\|.\|.\c
\&\|' options are not implied by `\|\c
.B \-Wall\c
\&\|'
because they warn about constructions that we consider reasonable to
use, on occasion, in clean programs.
.TP
.B \-Wtraditional
Warn about certain constructs that behave differently in traditional and
ANSI C.
.TP
\ \ \ \(bu
Macro arguments occurring within string constants in the macro body.
These would substitute the argument in traditional C, but are part of
the constant in ANSI C.
.TP
\ \ \ \(bu
A function declared external in one block and then used after the end of
the block.
.TP
\ \ \ \(bu
A \c
.B switch\c
\& statement has an operand of type \c
.B long\c
\&.
.PP
.TP
.B \-Wshadow
Warn whenever a local variable shadows another local variable.
.TP
.BI "\-Wid\-clash\-" "len"\c
\&
Warn whenever two distinct identifiers match in the first \c
.I len\c
\&
characters.  This may help you prepare a program that will compile
with certain obsolete, brain-damaged compilers.
.TP
.B \-Wpointer\-arith
Warn about anything that depends on the ``size of'' a function type or
of \c
.B void\c
\&.  GNU C assigns these types a size of 1, for
convenience in calculations with \c
.B void *\c
\& pointers and pointers
to functions.
.TP
.B \-Wcast\-qual
Warn whenever a pointer is cast so as to remove a type qualifier from
the target type.  For example, warn if a \c
.B const char *\c
\& is cast
to an ordinary \c
.B char *\c
\&.
.TP
.B \-Wcast\-align
Warn whenever a pointer is cast such that the required alignment of the
target is increased.  For example, warn if a \c
.B char *\c
\& is cast to
an \c
.B int *\c
\& on machines where integers can only be accessed at
two- or four-byte boundaries.
.TP
.B \-Wwrite\-strings
Give string constants the type \c
.B const char[\c
.I length\c
\&]\c
\& so that
copying the address of one into a non-\c
.B const\c
\& \c
.B char *\c
\&
pointer will get a warning.  These warnings will help you find at
compile time code that can try to write into a string constant, but
only if you have been very careful about using \c
.B const\c
\& in
declarations and prototypes.  Otherwise, it will just be a nuisance;
this is why we did not make `\|\c
.B \-Wall\c
\&\|' request these warnings.
.TP
.B \-Wconversion
Warn if a prototype causes a type conversion that is different from what
would happen to the same argument in the absence of a prototype.  This
includes conversions of fixed point to floating and vice versa, and
conversions changing the width or signedness of a fixed point argument
except when the same as the default promotion.
.TP
.B \-Waggregate\-return
Warn if any functions that return structures or unions are defined or
called.  (In languages where you can return an array, this also elicits
a warning.)
.TP
.B \-Wstrict\-prototypes
Warn if a function is declared or defined without specifying the
argument types.  (An old-style function definition is permitted without
a warning if preceded by a declaration which specifies the argument
types.)
.TP
.B \-Wmissing\-prototypes
Warn if a global function is defined without a previous prototype
declaration.  This warning is issued even if the definition itself
provides a prototype.  The aim is to detect global functions that fail
to be declared in header files.
.TP
.B \-Wenum\-clash
.I (C++ only.)
Warn when converting between different enumeration types.
.TP
.B \-Woverloaded\-virtual
.I (C++ only.)
In a derived class, the definitions of virtual functions must match
the type signature of a virtual function declared in the base class.
Use this option to request warnings when a derived class declares a
function that may be an erroneous attempt to define a virtual
function: that is, warn when a function with the same name as a
virtual function in the base class, but with a type signature that
doesn't match any virtual functions from the base class.
.TP
.B \-Winline
Warn if a function can not be inlined, and either it was declared as inline,
or else the
.B \-finline\-functions
option was given.
.TP
.B \-Werror
Treat warnings as errors; abort compilation after any warning.
.PP

.SH DEBUGGING OPTIONS
GNU CC has various special options that are used for debugging
either your program or GCC:
.TP
.B \-g
Produce debugging information in the operating system's native format
(for DBX or SDB or DWARF).  GDB also can work with this debugging
information.  On most systems that use DBX format, `\|\c
.B \-g\c
\&\|' enables use
of extra debugging information that only GDB can use; if you want to
control for certain whether to generate this information, use
`\|\c
.B \-ggdb\c
\&\|' or `\|\c
.B \-gdbx\c
\&\|'.

Unlike most other C compilers, GNU CC allows you to use `\|\c
.B \-g\c
\&\|' with
`\|\c
.B \-O\c
\&\|'.  The shortcuts taken by optimized code may occasionally
produce surprising results: some variables you declared may not exist
at all; flow of control may briefly move where you did not expect it;
some statements may not be executed because they compute constant
results or their values were already at hand; some statements may
execute in different places because they were moved out of loops.

Nevertheless it proves possible to debug optimized output.  This makes
it reasonable to use the optimizer for programs that might have bugs.

The following options are useful when GNU CC is configured and
compiled with the capability for more than one debugging format. 
.TP
.B \-ggdb
Produce debugging information in DBX format (if that is supported),
including GDB extensions.
.TP
.B \-gdbx
Produce debugging information in DBX format (if that is supported),
without GDB extensions.
.TP
.B \-gsdb
Produce debugging information in SDB format (if that is supported).
.TP
.B \-gdwarf
Produce debugging information in DWARF format (if that is supported).
.PP
.BI "\-g" "level"
.br
.BI "\-ggdb" "level"
.br
.BI "\-gdbx" "level"
.br
.BI "\-gsdb" "level"
.TP
.BI "\-gdwarf" "level"
Request debugging information and also use \c
.I level\c
\& to specify how
much information.  The default level is 2.

Level 1 produces minimal information, enough for making backtraces in
parts of the program that you don't plan to debug.  This includes
descriptions of functions and external variables, but no information
about local variables and no line numbers.
.TP
.B \-p
Generate extra code to write profile information suitable for the
analysis program \c
.B prof\c
\&.
.TP
.B \-pg
Generate extra code to write profile information suitable for the
analysis program \c
.B gprof\c
\&.
.TP
.B \-a
Generate extra code to write profile information for basic blocks,
which will record the number of times each basic block is executed.
This data could be analyzed by a program like \c
.B tcov\c
\&.  Note,
however, that the format of the data is not what \c
.B tcov\c
\& expects.
Eventually GNU \c
.B gprof\c
\& should be extended to process this data.
.TP
.BI "\-d" "letters"\c
\&
Says to make debugging dumps during compilation at times specified by
\c
.I letters\c
\&.  This is used for debugging the compiler.  The file names
for most of the dumps are made by appending a word to the source file
name (e.g.  `\|\c
.B foo.c.rtl\c
\&\|' or `\|\c
.B foo.c.jump\c
\&\|').
.TP
.B \-dM
Dump all macro definitions, at the end of preprocessing, and write no
output.
.TP
.B \-dN
Dump all macro names, at the end of preprocessing.
.TP
.B \-dD
Dump all macro definitions, at the end of preprocessing, in addition to
normal output.
.TP
.B \-dy
Dump debugging information during parsing, to standard error.
.TP
.B \-dr
Dump after RTL generation, to `\|\c
.B \c
.I file\c
\&.rtl\c
\&\|'.
.TP
.B \-dx
Just generate RTL for a function instead of compiling it.  Usually used
with `\|\c
.B r\c
\&\|'.
.TP
.B \-dj
Dump after first jump optimization, to `\|\c
.B \c
.I file\c
\&.jump\c
\&\|'.
.TP
.B \-ds
Dump after CSE (including the jump optimization that sometimes
follows CSE), to `\|\c
.B \c
.I file\c
\&.cse\c
\&\|'.
.TP
.B \-dL
Dump after loop optimization, to `\|\c
.B \c
.I file\c
\&.loop\c
\&\|'.
.TP
.B \-dt
Dump after the second CSE pass (including the jump optimization that
sometimes follows CSE), to `\|\c
.B \c
.I file\c
\&.cse2\c
\&\|'.
.TP
.B \-df
Dump after flow analysis, to `\|\c
.B \c
.I file\c
\&.flow\c
\&\|'.
.TP
.B \-dc
Dump after instruction combination, to `\|\c
.B \c
.I file\c
\&.combine\c
\&\|'.
.TP
.B \-dS
Dump after the first instruction scheduling pass, to
`\|\c
.B \c
.I file\c
\&.sched\c
\&\|'.
.TP
.B \-dl
Dump after local register allocation, to `\|\c
.B \c
.I file\c
\&.lreg\c
\&\|'.
.TP
.B \-dg
Dump after global register allocation, to `\|\c
.B \c
.I file\c
\&.greg\c
\&\|'.
.TP
.B \-dR
Dump after the second instruction scheduling pass, to
`\|\c
.B \c
.I file\c
\&.sched2\c
\&\|'.
.TP
.B \-dJ
Dump after last jump optimization, to `\|\c
.B \c
.I file\c
\&.jump2\c
\&\|'.
.TP
.B \-dd
Dump after delayed branch scheduling, to `\|\c
.B \c
.I file\c
\&.dbr\c
\&\|'.
.TP
.B \-dk
Dump after conversion from registers to stack, to `\|\c
.B \c
.I file\c
\&.stack\c
\&\|'.
.TP
.B \-dm
Print statistics on memory usage, at the end of the run, to
standard error.
.TP
.B \-dp
Annotate the assembler output with a comment indicating which
pattern and alternative was used.
.TP
.B \-fpretend\-float
When running a cross-compiler, pretend that the target machine uses the
same floating point format as the host machine.  This causes incorrect
output of the actual floating constants, but the actual instruction
sequence will probably be the same as GNU CC would make when running on
the target machine.
.TP
.B \-save\-temps
Store the usual ``temporary'' intermediate files permanently; place them
in the current directory and name them based on the source file.  Thus,
compiling `\|\c
.B foo.c\c
\&\|' with `\|\c
.B \-c \-save\-temps\c
\&\|' would produce files
`\|\c
.B foo.cpp\c
\&\|' and `\|\c
.B foo.s\c
\&\|', as well as `\|\c
.B foo.o\c
\&\|'.
.PP

.SH OPTIMIZATION OPTIONS
These options control various sorts of optimizations:
.TP
.B \-O
Optimize.  Optimizing compilation takes somewhat more time, and a lot
more memory for a large function.

Without `\|\c
.B \-O\c
\&\|', the compiler's goal is to reduce the cost of
compilation and to make debugging produce the expected results.
Statements are independent: if you stop the program with a breakpoint
between statements, you can then assign a new value to any variable or
change the program counter to any other statement in the function and
get exactly the results you would expect from the source code.

Without `\|\c
.B \-O\c
\&\|', only variables declared \c
.B register\c
\& are
allocated in registers.  The resulting compiled code is a little worse
than produced by PCC without `\|\c
.B \-O\c
\&\|'.

With `\|\c
.B \-O\c
\&\|', the compiler tries to reduce code size and execution
time.

When you specify `\|\c
.B \-O\c
\&\|', `\|\c
.B \-fthread\-jumps\c
\&\|' and
`\|\c
.B \-fdelayed\-branch\c
\&\|' are turned on.  On some machines other
flags may also be turned on.
.TP
.B \-O2
Highly optimize.  As compared to `\|\c
.B \-O\c
\&\|', this
option will increase both compilation time and the performance of the
generated code.

All `\|\c
.B \-f\c
.I flag\c
\&\c
\&\|' options that control optimization are turned on
when you specify `\|\c
.B \-O2\c
\&\|', except `\|\c
.B \-funroll\-loops\c
\&\|'
and `\|\c
.B \-funroll\-all\-loops\c
\&\|'.  
.PP

Options of the form `\|\c
.B \-f\c
.I flag\c
\&\c
\&\|' specify machine-independent
flags.  Most flags have both positive and negative forms; the negative
form of `\|\c
.B \-ffoo\c
\&\|' would be `\|\c
.B \-fno\-foo\c
\&\|'.  The following list shows
only one form\(em\&the one which is not the default.
You can figure out the other form by either removing `\|\c
.B no\-\c
\&\|' or
adding it.
.TP
.B \-ffloat\-store
Do not store floating point variables in registers.  This
prevents undesirable excess precision on machines such as the
68000 where the floating registers (of the 68881) keep more
precision than a \c
.B double\c
\& is supposed to have.

For most programs, the excess precision does only good, but a few
programs rely on the precise definition of IEEE floating point.
Use `\|\c
.B \-ffloat\-store\c
\&\|' for such programs.
.TP
.B \-fmemoize\-lookups
.TP
.B \-fsave\-memoized
.I
(C++ only.)
These flags are used to get the compiler to compile programs faster
using heuristics.  They are not on by default since they are only effective
about half the time.  The other half of the time programs compile more
slowly (and take more memory).

The first time the compiler must build a call to a member function (or
reference to a data member), it must (1) determine whether the class
implements member functions of that name; (2) resolve which member
function to call (which involves figuring out what sorts of type
conversions need to be made); and (3) check the visibility of the member
function to the caller.  All of this adds up to slower compilation.
Normally, the second time a call is made to that member function (or
reference to that data member), it must go through the same lengthy
process again.  This means that code like this
.sp
.br
\ \ cout\ <<\ "This\ "\ <<\ p\ <<\ "\ has\ "\ <<\ n\ <<\ "\ legs.\en";
.br
.sp
makes six passes through all three steps.  By using a software cache,
a ``hit'' significantly reduces this cost.  Unfortunately, using the
cache introduces another layer of mechanisms which must be implemented,
and so incurs its own overhead.  `\|\c
.B \-fmemoize\-lookups\c
\&\|' enables
the software cache.

Because access privileges (visibility) to members and member functions
may differ from one function context to the next, 
.B g++
may need to flush the cache. With the `\|\c
.B \-fmemoize\-lookups\c
\&\|' flag, the cache is flushed after every
function that is compiled.  The `\|\c
\-fsave\-memoized\c
\&\|' flag enables the same software cache, but when the compiler
determines that the context of the last function compiled would yield
the same access privileges of the next function to compile, it
preserves the cache. 
This is most helpful when defining many member functions for the same
class: with the exception of member functions which are friends of
other classes, each member function has exactly the same access
privileges as every other, and the cache need not be flushed.
.TP
.B \-fno\-default\-inline
.I
(C++ only.)
If `\|\c
.B \-fdefault\-inline\c
\&\|' is enabled then member functions defined inside class
scope are compiled inline by default; i.e., you don't need to
add `\|\c
.B inline\c
\&\|' in front of the member function name.  By popular
demand, this option is now the default.  To keep GNU C++ from inlining
these member functions, specify `\|\c
.B \-fno\-default\-inline\c
\&\|'.
.TP
.B \-fno\-defer\-pop
Always pop the arguments to each function call as soon as that
function returns.  For machines which must pop arguments after a
function call, the compiler normally lets arguments accumulate on the
stack for several function calls and pops them all at once.
.TP
.B \-fforce\-mem
Force memory operands to be copied into registers before doing
arithmetic on them.  This may produce better code by making all
memory references potential common subexpressions.  When they are
not common subexpressions, instruction combination should
eliminate the separate register-load.  I am interested in hearing
about the difference this makes.
.TP
.B \-fforce\-addr
Force memory address constants to be copied into registers before
doing arithmetic on them.  This may produce better code just as
`\|\c
.B \-fforce\-mem\c
\&\|' may.  I am interested in hearing about the
difference this makes.
.TP
.B \-fomit\-frame\-pointer
Don't keep the frame pointer in a register for functions that
don't need one.  This avoids the instructions to save, set up and
restore frame pointers; it also makes an extra register available
in many functions.  \c
.I It also makes debugging impossible on
most machines.\c
\&

On some machines, such as the Vax, this flag has no effect, because
the standard calling sequence automatically handles the frame pointer
and nothing is saved by pretending it doesn't exist.  The
machine-description macro \c
.B FRAME_POINTER_REQUIRED\c
\& controls
whether a target machine supports this flag.  
.TP
.B \-finline
Pay attention the \c
.B inline\c
\& keyword.  Normally the negation of this
option `\|\c
.B \-fno\-inline\c
\&\|' is used to keep the compiler from expanding
any functions inline.  However, the opposite effect may be desirable
when compiling with `\|\c
.B \-g\c
\&\|', since `\|\c
.B \-g\c
\&\|' normally turns off all
inline function expansion.
.TP
.B \-finline\-functions
Integrate all simple functions into their callers.  The compiler
heuristically decides which functions are simple enough to be worth
integrating in this way.

If all calls to a given function are integrated, and the function is
declared \c
.B static\c
\&, then GCC normally does not output the function as
assembler code in its own right.
.TP
.B \-fcaller\-saves
Enable values to be allocated in registers that will be clobbered by
function calls, by emitting extra instructions to save and restore the
registers around such calls.  Such allocation is done only when it
seems to result in better code than would otherwise be produced.

This option is enabled by default on certain machines, usually those
which have no call-preserved registers to use instead.
.TP
.B \-fkeep\-inline\-functions
Even if all calls to a given function are integrated, and the function
is declared \c
.B static\c
\&, nevertheless output a separate run-time
callable version of the function.
.TP
.B \-fno\-function\-cse
Do not put function addresses in registers; make each instruction that
calls a constant function contain the function's address explicitly.

This option results in less efficient code, but some strange hacks
that alter the assembler output may be confused by the optimizations
performed when this option is not used.
.PP

The following options control specific optimizations.  The `\|\c
.B \-O2\c
\&\|'
option turns on all of these optimizations except `\|\c
.B \-funroll\-loops\c
\&\|'
and `\|\c
.B \-funroll\-all\-loops\c
\&\|'.  

The `\|\c
.B \-O\c
\&\|' option usually turns on
the `\|\c
.B \-fthread\-jumps\c
\&\|' and `\|\c
.B \-fdelayed\-branch\c
\&\|' options, but
specific machines may change the default optimizations.

You can use the following flags in the rare cases when ``fine-tuning''
of optimizations to be performed is desired.
.TP
.B \-fstrength\-reduce
Perform the optimizations of loop strength reduction and
elimination of iteration variables.
.TP
.B \-fthread\-jumps
Perform optimizations where we check to see if a jump branches to a
location where another comparison subsumed by the first is found.  If
so, the first branch is redirected to either the destination of the
second branch or a point immediately following it, depending on whether
the condition is known to be true or false.
.TP
.B \-funroll\-loops
Perform the optimization of loop unrolling.  This is only done for loops
whose number of iterations can be determined at compile time or run time.
.TP
.B \-funroll\-all\-loops
Perform the optimization of loop unrolling.  This is done for all loops.
This usually makes programs run more slowly.
.TP
.B \-fcse\-follow\-jumps
In common subexpression elimination, scan through jump instructions in
certain cases.  This is not as powerful as completely global CSE, but
not as slow either.
.TP
.B \-frerun\-cse\-after\-loop
Re-run common subexpression elimination after loop optimizations has been
performed.  
.TP
.B \-felide\-constructors
.I
(C++ only.)
Use this option to instruct the compiler to be smarter about when it can
elide constructors.  Without this flag, GNU C++ and cfront both
generate effectively the same code for:
.sp
.br
A\ foo\ ();
.br
A\ x\ (foo\ ());\ \ \ //\ x\ initialized\ by\ `foo\ ()',\ no\ ctor\ called
.br
A\ y\ =\ foo\ ();\ \ \ //\ call\ to\ `foo\ ()'\ heads\ to\ temporary,
.br
\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ //\ y\ is\ initialized\ from\ the\ temporary.
.br
.sp
Note the difference!  With this flag, GNU C++ initializes `\|\c
.B y\c
\&\|' directly
from the call to 
.B foo ()
without going through a temporary.
.TP
.B \-fexpensive\-optimizations
Perform a number of minor optimizations that are relatively expensive.
.TP
.B \-fdelayed\-branch
If supported for the target machine, attempt to reorder instructions
to exploit instruction slots available after delayed branch
instructions.
.TP
.B \-fschedule\-insns
If supported for the target machine, attempt to reorder instructions to
eliminate execution stalls due to required data being unavailable.  This
helps machines that have slow floating point or memory load instructions
by allowing other instructions to be issued until the result of the load
or floating point instruction is required.
.TP
.B \-fschedule\-insns2
Similar to `\|\c
.B \-fschedule\-insns\c
\&\|', but requests an additional pass of
instruction scheduling after register allocation has been done.  This is
especially useful on machines with a relatively small number of
registers and where memory load instructions take more than one cycle.
.PP

.SH TARGET OPTIONS
By default, GNU CC compiles code for the same type of machine that you
are using.  However, it can also be installed as a cross-compiler, to
compile for some other type of machine.  In fact, several different
configurations of GNU CC, for different target machines, can be
installed side by side.  Then you specify which one to use with the
`\|\c
.B \-b\c
\&\|' option.

In addition, older and newer versions of GNU CC can be installed side
by side.  One of them (probably the newest) will be the default, but
you may sometimes wish to use another.
.TP
.BI "\-b " "machine"\c
\&
The argument \c
.I machine\c
\& specifies the target machine for compilation.
This is useful when you have installed GNU CC as a cross-compiler.

The value to use for \c
.I machine\c
\& is the same as was specified as the
machine type when configuring GNU CC as a cross-compiler.  For
example, if a cross-compiler was configured with `\|\c
.B configure
i386v\c
\&\|', meaning to compile for an 80386 running System V, then you
would specify `\|\c
.B \-b i386v\c
\&\|' to run that cross compiler.

When you do not specify `\|\c
.B \-b\c
\&\|', it normally means to compile for
the same type of machine that you are using.
.TP
.BI "\-V " "version"\c
\&
The argument \c
.I version\c
\& specifies which version of GNU CC to run.
This is useful when multiple versions are installed.  For example,
\c
.I version\c
\& might be `\|\c
.B 2.0\c
\&\|', meaning to run GNU CC version 2.0.

The default version, when you do not specify `\|\c
.B \-V\c
\&\|', is controlled
by the way GNU CC is installed.  Normally, it will be a version that
is recommended for general use.
.PP

.SH MACHINE DEPENDENT OPTIONS
Each of the target machine types can have its own special options,
starting with `\|\c
.B \-m\c
\&\|', to choose among various hardware models or
configurations\(em\&for example, 68010 vs 68020, floating coprocessor or
none.  A single installed version of the compiler can compile for any
model or configuration, according to the options specified.

These are the `\|\c
.B \-m\c
\&\|' options defined for the 68000 series:
.TP
.B \-m68020
.TP
.B \-mc68020
Generate output for a 68020 (rather than a 68000).  This is the
default if you use the unmodified sources.
.TP
.B \-m68000
.TP
.B \-mc68000
Generate output for a 68000 (rather than a 68020).
.TP
.B \-m68881
Generate output containing 68881 instructions for floating point.
This is the default if you use the unmodified sources.
.TP
.B \-mfpa
Generate output containing Sun FPA instructions for floating point.
.TP
.B \-msoft\-float
Generate output containing library calls for floating point.
.I
WARNING: 
the requisite libraries are not part of GNU CC.  Normally the
facilities of the machine's usual C compiler are used, but this can't
be done directly in cross-compilation.  You must make your own
arrangements to provide suitable library functions for cross-compilation.
.TP
.B \-mshort
Consider type \c
.B int\c
\& to be 16 bits wide, like \c
.B short int\c
\&.
.TP
.B \-mnobitfield
Do not use the bit-field instructions.  `\|\c
.B \-m68000\c
\&\|' implies
`\|\c
.B \-mnobitfield\c
\&\|'.
.TP
.B \-mbitfield
Do use the bit-field instructions.  `\|\c
.B \-m68020\c
\&\|' implies
`\|\c
.B \-mbitfield\c
\&\|'.  This is the default if you use the unmodified
sources.
.TP
.B \-mrtd
Use a different function-calling convention, in which functions
that take a fixed number of arguments return with the \c
.B rtd\c
\&
instruction, which pops their arguments while returning.  This
saves one instruction in the caller since there is no need to pop
the arguments there.

This calling convention is incompatible with the one normally
used on Unix, so you cannot use it if you need to call libraries
compiled with the Unix compiler.

Also, you must provide function prototypes for all functions that
take variable numbers of arguments (including \c
.B printf\c
\&);
otherwise incorrect code will be generated for calls to those
functions.

In addition, seriously incorrect code will result if you call a
function with too many arguments.  (Normally, extra arguments are
harmlessly ignored.)

The \c
.B rtd\c
\& instruction is supported by the 68010 and 68020
processors, but not by the 68000.
.PP

These `\|\c
.B \-m\c
\&\|' options are defined for the Vax:
.TP
.B \-munix
Do not output certain jump instructions (\c
.B aobleq\c
\& and so on)
that the Unix assembler for the Vax cannot handle across long
ranges.
.TP
.B \-mgnu
Do output those jump instructions, on the assumption that you
will assemble with the GNU assembler.
.TP
.B \-mg
Output code for g-format floating point numbers instead of d-format.
.PP

These `\|\c
.B \-m\c
\&\|' switches are supported on the Sparc:
.TP
.B \-mfpu
Generate output containing floating point instructions.  This is the
default if you use the unmodified sources.
.TP
.B \-mno\-epilogue
Generate separate return instructions for \c
.B return\c
\& statements.
This has both advantages and disadvantages; I don't recall what they
are.
.PP

These `\|\c
.B \-m\c
\&\|' options are defined for the Convex:
.TP
.B \-mc1
Generate output for a C1.  This is the default when the compiler is
configured for a C1.
.TP
.B \-mc2
Generate output for a C2.  This is the default when the compiler is
configured for a C2.
.TP
.B \-margcount
Generate code which puts an argument count in the word preceding each
argument list.  Some nonportable Convex and Vax programs need this word.
(Debuggers don't, except for functions with variable-length argument
lists; this info is in the symbol table.)
.TP
.B \-mnoargcount
Omit the argument count word.  This is the default if you use the
unmodified sources.
.PP

These `\|\c
.B \-m\c
\&\|' options are defined for the AMD Am29000:
.TP
.B \-mdw
Generate code that assumes the DW bit is set, i.e., that byte and
halfword operations are directly supported by the hardware.  This is the
default.
.TP
.B \-mnodw
Generate code that assumes the DW bit is not set.
.TP
.B \-mbw
Generate code that assumes the system supports byte and halfword write
operations.  This is the default.
.TP
.B \-mnbw
Generate code that assumes the systems does not support byte and
halfword write operations.  This implies `\|\c
.B \-mnodw\c
\&\|'.
.TP
.B \-msmall
Use a small memory model that assumes that all function addresses are
either within a single 256 KB segment or at an absolute address of less
than 256K.  This allows the \c
.B call\c
\& instruction to be used instead
of a \c
.B const\c
\&, \c
.B consth\c
\&, \c
.B calli\c
\& sequence.
.TP
.B \-mlarge
Do not assume that the \c
.B call\c
\& instruction can be used; this is the
default.
.TP
.B \-m29050
Generate code for the Am29050.
.TP
.B \-m29000
Generate code for the Am29000.  This is the default.
.TP
.B \-mkernel\-registers
Generate references to registers \c
.B gr64-gr95\c
\& instead of
\c
.B gr96-gr127\c
\&.  This option can be used when compiling kernel code
that wants a set of global registers disjoint from that used by
user-mode code.

Note that when this option is used, register names in `\|\c
.B \-f\c
\&\|' flags
must use the normal, user-mode, names.
.TP
.B \-muser\-registers
Use the normal set of global registers, \c
.B gr96-gr127\c
\&.  This is the
default.
.TP
.B \-mstack\-check
Insert a call to \c
.B __msp_check\c
\& after each stack adjustment.  This
is often used for kernel code.
.PP

These `\|\c
.B \-m\c
\&\|' options are defined for Motorola 88K architectures:
.TP
.B \-mbig\-pic
Emit position-independent code, suitable for dynamic linking, even if
branches need large displacements.  Equivalent to the general-use option `\|\c
.B \-fPIC\c
\&\|'.
The general-use option `\|\c
.B \-fpic\c
\&\|',
by contrast, only emits valid 88k code if all branches involve small
displacements. 
GCC does not emit position-independent code by default.
.TP
.B \-midentify\-revision
Include an \c
.B ident\c
\& directive in the assembler output recording the
source file name, compiler name and version, timestamp, and compilation
flags used.
.TP
.B \-mno\-underscores
In assembler output, emit symbol names without adding an underscore
character at the beginning of each name.  The default is to use an
underscore as prefix on each name.
.TP
.B \-mno\-check\-zero\-division
.TP
.B \-mcheck\-zero\-division
Early models of the 88K architecture had problems with division by zero;
in particular, many of them didn't trap.  Use these options to avoid
including (or to include explicitly) additional code to detect division
by zero and signal an exception.  All GCC configurations for the 88K use
`\|\c
.B \-mcheck\-zero\-division\c
\&\|' by default.
.TP
.B \-mocs\-debug\-info
.TP
.B \-mno\-ocs\-debug\-info
Include (or omit) additional debugging information (about
registers used in each stack frame) as specified in the 88Open Object
Compatibility Standard, ``OCS''.  This extra information is not needed
by GDB.  The default for DG/UX, SVr4, and Delta 88 SVr3.2 is to
include this information; other 88k configurations omit this information
by default.
.TP
.B \-mocs\-frame\-position
.TP
.B \-mno\-ocs\-frame\-position
Force (or do not require) register values to be stored in a particular
place in stack frames, as specified in OCS.  The DG/UX, Delta88 SVr3.2,
and BCS configurations use `\|\c
.B \-mocs\-frame\-position\c
\&\|'; other 88k
configurations have the default `\|\c
.B \-mno\-ocs\-frame\-position\c
\&\|'.
.TP
.B \-moptimize\-arg\-area
.TP
.B \-mno\-optimize\-arg\-area
Control how to store function arguments in stack frames.
`\|\c
.B \-moptimize\-arg\-area\c
\&\|' saves space, but may break some
debuggers (not GDB).  `\|\c
.B \-mno\-optimize\-arg\-area\c
\&\|' conforms better to
standards.   By default GCC does not optimize the argument area.
.TP
.BI "\-mshort\-data\-" "num"\c
\&
.I num\c
\&
Generate smaller data references by making them relative to \c
.B r0\c
\&,
which allows loading a value using a single instruction (rather than the
usual two).  You control which data references are affected by
specifying \c
.I num\c
\& with this option.  For example, if you specify
`\|\c
.B \-mshort\-data\-512\c
\&\|', then the data references affected are those
involving displacements of less than 512 bytes.
`\|\c
.B \-mshort\-data\-\c
.I num\c
\&\c
\&\|' is not effective for \c
.I num\c
\& greater
than 64K.
.TP
.B \-msvr4
.TP
.B \-msvr3
Turn on (`\|\c
.B \-msvr4\c
\&\|') or off (`\|\c
.B \-msvr3\c
\&\|') compiler extensions
related to System V release 4 (SVr4).  This controls the following:
.TP
\ \ \ \(bu 
Which variant of the assembler syntax to emit (which you can select
independently using `\|\c
.B \-mversion03.00\c
\&\|').  
.TP
\ \ \ \(bu
`\|\c
.B \-msvr4\c
\&\|' makes the C preprocessor recognize `\|\c
.B #pragma weak\c
\&\|'
.TP
\ \ \ \(bu
`\|\c
.B \-msvr4\c
\&\|' makes GCC issue additional declaration directives used in
SVr4.  
.PP
`\|\c
.B \-msvr3\c
\&\|' is the default for all m88K configurations except
the SVr4 configuration.
.TP
.B \-mtrap\-large\-shift
.TP
.B \-mhandle\-large\-shift
Include code to detect bit-shifts of more than 31 bits; respectively,
trap such shifts or emit code to handle them properly.  By default GCC
makes no special provision for large bit shifts.
.TP
.B \-muse\-div\-instruction
Very early models of the 88K architecture didn't have a divide
instruction, so GCC avoids that instruction by default.  Use this option
to specify that it's safe to use the divide instruction.  
.TP
.B \-mversion\-03.00
Use alternative assembler syntax for the assembler version corresponding
to SVr4, but without enabling the other features triggered by
`\|\c
.B \-svr4\c
\&\|'.  This is implied by `\|\c
.B \-svr4\c
\&\|', is the default for the
SVr4 configuration of GCC, and is permitted by the DG/UX configuration
only if `\|\c
.B \-svr4\c
\&\|' is also specified.  The Delta 88 SVr3.2
configuration ignores this option.
.TP
.B \-mwarn\-passed\-structs
Warn when a function passes a struct as an argument or result.
Structure-passing conventions have changed during the evolution of the C
language, and are often the source of portability problems.  By default,
GCC issues no such warning.
.PP
These options are defined for the IBM RS6000:

.PP
.B \-mfp\-in\-toc
.TP
.B \-mno\-fp\-in\-toc
Control whether or not floating-point constants go in the Table of
Contents (TOC), a table of all global variable and function addresses.  By
default GCC puts floating-point constants there; if the TOC overflows,
`\|\c
.B \-mno\-fp\-in\-toc\c
\&\|' will reduce the size of the TOC, which may avoid
the overflow.

.PP
These `\|\c
.B \-m\c
\&\|' options are defined for the IBM RT PC:
.TP
.B \-min\-line\-mul
Use an in-line code sequence for integer multiplies.  This is the
default.
.TP
.B \-mcall\-lib\-mul
Call \c
.B lmul$$\c
\& for integer multiples.
.TP
.B \-mfull\-fp\-blocks
Generate full-size floating point data blocks, including the minimum
amount of scratch space recommended by IBM.  This is the default.
.TP
.B \-mminimum\-fp\-blocks
Do not include extra scratch space in floating point data blocks.  This
results in smaller code, but slower execution, since scratch space must
be allocated dynamically.
.TP
.B \-mfp\-arg\-in\-fpregs
Use a calling sequence incompatible with the IBM calling convention in
which floating point arguments are passed in floating point registers.
Note that \c
.B varargs.h\c
\& and \c
.B stdargs.h\c
\& will not work with
floating point operands if this option is specified.
.TP
.B \-mfp\-arg\-in\-gregs
Use the normal calling convention for floating point arguments.  This is
the default.
.TP
.B \-mhc\-struct\-return
Return structures of more than one word in memory, rather than in a
register.  This provides compatibility with the MetaWare HighC (hc)
compiler.  Use `\|\c
.B \-fpcc\-struct\-return\c
\&\|' for compatibility with the
Portable C Compiler (pcc).
.TP
.B \-mnohc\-struct\-return
Return some structures of more than one word in registers, when
convenient.  This is the default.  For compatibility with the
IBM-supplied compilers, use either `\|\c
.B \-fpcc\-struct\-return\c
\&\|' or
`\|\c
.B \-mhc\-struct\-return\c
\&\|'.
.PP
These `\|\c
.B \-m\c
\&\|' options are defined for the MIPS family of computers:
.TP
.BI "\-mcpu=" "cpu-type"
Assume the defaults for the machine type 
.I cpu-type
when
scheduling instructions.  The default
.I cpu-type
is
.BR default ,
which picks the longest cycles times for any of the machines, in order
that the code run at reasonable rates on all MIPS cpu's.  Other
choices for
.I cpu-type
are
.BR r2000 ,
.BR r3000 ,
.BR r4000 ,
and
.BR r6000 .
While picking a specific
.I cpu-type
will schedule things appropriately for that particular chip, the
compiler will not generate any code that does not meet level 1 of the
MIPS ISA (instruction set architecture) without the
.B \-mips2
or
.B \-mips3
switches being used.
.TP
.B \-mips2
Issue instructions from level 2 of the MIPS ISA (branch likely, square
root instructions).  The 
.B \-mcpu=r4000
or
.B \-mcpu=r6000
switch must be used in conjunction with
.BR \-mips2 .
.TP
.B \-mips3
Issue instructions from level 3 of the MIPS ISA (64 bit instructions).
The
.B \-mcpu=r4000
switch must be used in conjunction with
.BR \-mips2 .
.TP
.B \-mint64
.TP
.B \-mlong64
.TP
.B \-mlonglong128
These options don't work at present.
.TP
.B \-mmips\-as
Generate code for the MIPS assembler, and invoke
.B mips\-tfile
to add normal debug information.  This is the default for all
platforms except for the OSF/1 reference platform, using the OSF/rose
object format.  If any of the
.BR \-ggdb ,
.BR \-gstabs ,
or
.B \-gstabs+
switches are used, the
.B mips\-tfile
program will encapsulate the stabs within MIPS ECOFF.
.TP
.B \-mgas
Generate code for the GNU assembler.  This is the default on the OSF/1
reference platform, using the OSF/rose object format.
.TP
.B \-mrnames
.TP
.B \-mno\-rnames
The
.B \-mrnames
switch says to output code using the MIPS software names for the
registers, instead of the hardware names (ie, 
.B a0
instead of
.BR $4 ).
The GNU assembler does not support the
.B \-mrnames
switch, and the MIPS assembler will be instructed to run the MIPS C
preprocessor over the source file.  The
.B \-mno\-rnames
switch is default.
.TP
.B \-mgpopt
.TP
.B \-mno\-gpopt
The
.B \-mgpopt
switch says to write all of the data declarations before the
instructions in the text section, to all the MIPS assembler to
generate one word memory references instead of using two words for
short global or static data items.  This is on by default if
optimization is selected.
.TP
.B \-mstats
.TP
.B \-mno\-stats
For each non-inline function processed, the
.B \-mstats
switch causes the compiler to emit one line to the standard error file
to print statistics about the program (number of registers saved,
stack size, etc.).
.TP
.B \-mmemcpy
.TP
.B \-mno\-memcpy
The
.B \-mmemcpy
switch makes all block moves call the appropriate string function
.RB ( memcpy
or
.BR bcopy )
instead of possibly generating inline code.
.TP
.B \-mmips\-tfile
.TP
.B \-mno\-mips\-tfile
The
.B \-mno\-mips\-tfile
switch causes the compiler not postprocess the object file with the
.B mips\-tfile
program, after the MIPS assembler has generated it to add debug
support.  If
.B mips\-tfile
is not run, then no local variables will be available to the debugger.
In addition,
.B stage2
and
.B stage3
objects will have the temporary file names passed to the assembler
embedded in the object file, which means the objects will not compare
the same.
.TP
.B \-msoft\-float
Generate output containing library calls for floating point.
.I
WARNING: 
the requisite libraries are not part of GNU CC.  Normally the
facilities of the machine's usual C compiler are used, but this can't
be done directly in cross-compilation.  You must make your own
arrangements to provide suitable library functions for cross-compilation.
.TP
.B \-mhard\-float
Generate output containing floating point instructions.  This is the
default if you use the unmodified sources.
.TP
.B \-mfp64
Assume that the
.B FR
bit in the status word is on, and that there are 32 64-bit floating
point registers, instead of 32 32-bit floating point registers.  You
must also specify the
.B \-mcpu=r4000
and
.B \-mips3
switches.
.TP
.B \-mfp32
Assume that there are 32 32-bit floating point registers.  This is the
default.
.TP
.B \-mabicalls
The
.B \-mabicalls
switch says to emit the
.BR \&.abicalls ,
.BR \&.cpload ,
and
.B \&.cprestore
pseudo operations that some System V.4 ports use for position
independent code.
.TP
.B \-mhalf\-pic
.TP
.B \-mno\-half\-pic
The
.B \-mhalf\-pic
switch says to put pointers to extern references into the data section
and load them up, rather than put the references in the text section.
This option does not work at present.
.B
.BI \-G num
Put global and static items less than or equal to 
.I num
bytes into the small data or bss sections instead of the normal data
or bss section.  This allows the assembler to emit one word memory
reference instructions based on the global pointer
.RB ( gp
or
.BR $28 ),
instead of the normal two words used.  By default,
.I num
is 8 when the MIPS assembler is used, and 0 when the GNU
assembler is used.  The
.BI \-G num
switch is also passed to the assembler and linker.  All modules should
be compiled with the same
.BI \-G num
value.
.PP

.SH CODE GENERATION OPTIONS
These machine-independent options control the interface conventions
used in code generation.

Most of them begin with `\|\c
\-f\c
\&\|'.  These options have both positive and negative forms; the negative form
of `\|\c
.B \-ffoo\c
\&\|' would be `\|\c
.B \-fno\-foo\c
\&\|'.  In the table below, only
one of the forms is listed\(em\&the one which is not the default.  You
can figure out the other form by either removing `\|\c
.B no\-\c
\&\|' or adding
it.
.TP
.BI +e N
.I (C++ only.)
control whether virtual function definitions in classes
are used to generate code, or only to define interfaces for their
callers.  These options are provided for compatibility with cfront
1.x usage; the recommended GNU C++ usage is to use 
.B #pragma interface
and 
.B
#pragma implementation\c
\&, instead.

With `\|\c
.B +e0\c
\&\|', virtual function definitions in classes are declared extern;
the declaration is used only as an interface specification, not to
generate code for the virtual functions (in this compilation).

With `\|\c
.B +e1\c
\&\|', 
.B g++
actually generates the code implementing virtual functions
defined in the code, and makes them publicly visible.
.TP
.B \-fnonnull\-objects
.I
(C++ only.)
Normally, GNU C++ makes conservative assumptions about objects reached
through references.  For example, the compiler must check that `\|\c
.B a\c
\&\|' is not null in code like the following:
.br
\ \ \ \ obj\ &a\ =\ g\ ();
.br
\ \ \ \ a.f\ (2);
.br
Checking that references of this sort have non-null values requires
extra code, however, and it is unnecessary for many programs.  You can
use `\|\c
.B \-fnonnull\-objects\c
\&\|' to omit the checks for null, if your program doesn't require the
default checking.
.TP
.B \-fpcc\-struct\-return
Use the same convention for returning \c
.B struct\c
\& and \c
.B union\c
\&
values that is used by the usual C compiler on your system.  This
convention is less efficient for small structures, and on many
machines it fails to be reentrant; but it has the advantage of
allowing intercallability between GCC-compiled code and PCC-compiled
code.
.TP
.B \-fshort\-enums
Allocate to an \c
.B enum\c
\& type only as many bytes as it needs for the
declared range of possible values.  Specifically, the \c
.B enum\c
\& type
will be equivalent to the smallest integer type which has enough room.
.TP
.B \-fshort\-double
Use the same size for
.B double
as for
.B float
\&.
.TP
.B \-fshared\-data
Requests that the data and non-\c
.B const\c
\& variables of this
compilation be shared data rather than private data.  The distinction
makes sense only on certain operating systems, where shared data is
shared between processes running the same program, while private data
exists in one copy per process.
.TP
.B \-fno\-common
Allocate even uninitialized global variables in the bss section of the
object file, rather than generating them as common blocks.  This has the
effect that if the same variable is declared (without \c
.B extern\c
\&) in
two different compilations, you will get an error when you link them.
The only reason this might be useful is if you wish to verify that the
program will work on other systems which always work this way.
.TP
.B \-fvolatile
Consider all memory references through pointers to be volatile.
.TP
.B \-fpic
If supported for the target machines, generate position-independent code,
suitable for use in a shared library.
.TP
.B \-fPIC
If supported for the target machine, emit position-independent code,
suitable for dynamic linking, even if branches need large displacements.
.TP
.BI "\-ffixed\-" "reg"\c
\&
Treat the register named \c
.I reg\c
\& as a fixed register; generated code
should never refer to it (except perhaps as a stack pointer, frame
pointer or in some other fixed role).

\c
.I reg\c
\& must be the name of a register.  The register names accepted
are machine-specific and are defined in the \c
.B REGISTER_NAMES\c
\&
macro in the machine description macro file.

This flag does not have a negative form, because it specifies a
three-way choice.
.TP
.BI "\-fcall\-used\-" "reg"\c
\&
Treat the register named \c
.I reg\c
\& as an allocatable register that is
clobbered by function calls.  It may be allocated for temporaries or
variables that do not live across a call.  Functions compiled this way
will not save and restore the register \c
.I reg\c
\&.

Use of this flag for a register that has a fixed pervasive role in the
machine's execution model, such as the stack pointer or frame pointer,
will produce disastrous results.

This flag does not have a negative form, because it specifies a
three-way choice.
.TP
.BI "\-fcall\-saved\-" "reg"\c
\&
Treat the register named \c
.I reg\c
\& as an allocatable register saved by
functions.  It may be allocated even for temporaries or variables that
live across a call.  Functions compiled this way will save and restore
the register \c
.I reg\c
\& if they use it.

Use of this flag for a register that has a fixed pervasive role in the
machine's execution model, such as the stack pointer or frame pointer,
will produce disastrous results.

A different sort of disaster will result from the use of this flag for
a register in which function values may be returned.

This flag does not have a negative form, because it specifies a
three-way choice.
.TP
.B \-fgnu\-binutils
.TP
.B \-fno\-gnu\-binutils
.I
(C++ only.)
`\|\c
.B \-fgnu\-binutils
\&\|' (the default for most, but not all, platforms) makes GNU C++
emit extra information for static initialization and finalization.
This information has to be passed from the assembler to the GNU
linker.  Some assemblers won't pass this information; you must either
use GNU
.B as
or specify the option `\|\c
.B \-fno\-gnu\-binutils\c
\&\|'.

With `\|\c
.B \-fno\-gnu\-binutils\c
\&\|', you must use the program
.B collect
(part of the GCC distribution) for linking.
.PP

.SH PRAGMAS
Two `\|\c
.B #pragma\c
\&\|' directives are supported for GNU C++, to permit using the same
header file for two purposes: as a definition of interfaces to a given
object class, and as the full definition of the contents of that object class.
.TP
.B #pragma interface
.I (C++ only.)
Use this directive in header files that define object classes, to save
space in most of the object files that use those classes.  Normally,
local copies of certain information (backup copies of inline member
functions, debugging information, and the internal tables that
implement virtual functions) must be kept in each object file that
includes class definitions.  You can use this pragma to avoid such
duplication.  When a header file containing `\|\c
.B #pragma interface\c
\&\|' is included in a compilation, this auxiliary information
will not be generated (unless the main input source file itself uses
`\|\c
.B #pragma implementation\c
\&\|').  Instead, the object files will contain references to be
resolved at link time.  
.tr !"
.TP
.B #pragma implementation
.TP
.BI "#pragma implementation !" objects .h!
.I (C++ only.)
Use this pragma in a main input file, when you want full output from
included header files to be generated (and made globally visible).
The included header file, in turn, should use `\|\c
.B #pragma interface\c
\&\|'.  
Backup copies of inline member functions, debugging information, and
the internal tables used to implement virtual functions are all
generated in implementation files.

If you use `\|\c
.B #pragma implementation\c
\&\|' with no argument, it applies to an include file with the same
basename as your source file; for example, in `\|\c
.B allclass.cc\c
\&\|', `\|\c
.B #pragma implementation\c
\&\|' by itself is equivalent to `\|\c
.B 
#pragma implementation "allclass.h"\c
\&\|'.  Use the string argument if you want a single implementation
file to include code from multiple header files.  

There is no way to split up the contents of a single header file into
multiple implementation files. 
.SH FILES
.ta \w'LIBDIR/g++\-include 'u
file.c	C source file
.br
file.h	C header (preprocessor) file
.br
file.i	preprocessed C source file
.br
file.C	C++ source file
.br
file.cc	C++ source file
.br
file.cxx	C++ source file
.br
file.m	Objective-C source file
.br
file.s	assembly language file
.br
file.o	object file
.br
a.out	link edited output
.br
\fITMPDIR\fR/cc\(**	temporary files
.br
\fILIBDIR\fR/cpp	preprocessor
.br
\fILIBDIR\fR/cc1	compiler for C
.br
\fILIBDIR\fR/cc1plus	compiler for C++
.br
\fILIBDIR\fR/collect	linker front end needed on some machines
.br
\fILIBDIR\fR/libgcc.a	GCC subroutine library
.br
/lib/crt[01n].o	start-up routine
.br
\fILIBDIR\fR/ccrt0	additional start-up routine for C++
.br
/lib/libc.a	standard C library, see
.IR intro (3)
.br
/usr/include	standard directory for 
.B #include
files
.br
\fILIBDIR\fR/include	standard gcc directory for
.B #include
files
.br
\fILIBDIR\fR/g++\-include	additional g++ directory for
.B #include
.sp
.I LIBDIR
is usually
.B /usr/local/lib/\c
.IR machine / version .
.br
.I TMPDIR
comes from the environment variable 
.B TMPDIR
(default
.B /usr/tmp
if available, else
.B /tmp\c
\&).
.SH "SEE ALSO"
cpp(1), as(1), ld(1), gdb(1), adb(1), dbx(1), sdb(1).
.br
.RB "`\|" gcc "\|', `\|" cpp \|', 
.RB `\| as \|', `\| ld \|',
and 
.RB `\| gdb \|'
entries in
.B info\c
\&.
.br
.I 
Using and Porting GNU CC (for version 2.0)\c
, Richard M. Stallman, November 1990; 
.I
The C Preprocessor\c
, Richard M. Stallman, July 1990;
.I 
Using GDB: A Guide to the GNU Source-Level Debugger\c
, Richard M. Stallman and Roland H. Pesch, December 1991;
.I
Using as: the GNU Assembler\c
, Dean Elsner, Jay Fenlason & friends, March 1991;
.I
gld: the GNU linker\c
, Steve Chamberlain and Roland Pesch, April 1991.

.SH BUGS
Report bugs to
.BR bug\-gcc@prep.ai.mit.edu .
Bugs tend actually to be fixed if they can be isolated, so it is in your
interest to report them in such a way that they can be easily reproduced.
.SH COPYING
Copyright (c) 1991 Free Software Foundation, Inc.
.PP
Permission is granted to make and distribute verbatim copies of
this manual provided the copyright notice and this permission notice
are preserved on all copies.
.PP
Permission is granted to copy and distribute modified versions of this
manual under the conditions for verbatim copying, provided that the
entire resulting derived work is distributed under the terms of a
permission notice identical to this one.
.PP
Permission is granted to copy and distribute translations of this
manual into another language, under the above conditions for modified
versions, except that this permission notice may be included in
translations approved by the Free Software Foundation instead of in
the original English.
.SH AUTHORS
See the GNU CC Manual for the contributors to GNU CC.