Home
Main website
Display Sidebar
Hide Ads
Recent Changes
View Source:
cc(1)
Edit
PageHistory
Diff
Info
LikePages
GCC !!!GCC NAME SYNOPSIS WARNING DESCRIPTION OPTIONS OVERALL OPTIONS LANGUAGE OPTIONS PREPROCESSOR OPTIONS ASSEMBLER OPTION LINKER OPTIONS DIRECTORY OPTIONS WARNING OPTIONS DEBUGGING OPTIONS OPTIMIZATION OPTIONS TARGET OPTIONS MACHINE DEPENDENT OPTIONS CODE GENERATION OPTIONS PRAGMAS FILES EXIT STATUS SEE ALSO BUGS COPYING AUTHORS ---- !!NAME gcc, g++ - GNU project C and C++ Compiler (gcc-2.95) !!SYNOPSIS __gcc__ [[ ''option'' | ''filename'' ]...__ g++__ [[ ''option'' | ''filename'' ]... !!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. This man page is not kept up to date except when volunteers want to maintain it. If you find a discrepancy between the man page and the software, please check the Info file, which is the authoritative documentation. If we find that the things in this man page that are out of date cause significant confusion or complaints, we will stop distributing the man page. The alternative, updating the man page when we update the Info file, is impossible because the rest of the work of maintaining GNU CC leaves us no time for that. The GNU project regards man pages as obsolete and should not let them take time away from other things. For complete and current documentation, refer to the Info file `__gcc__' or the manual ''Using and Porting GNU CC (for version 2.0)''. Both are made from the Texinfo source file __gcc.texinfo__. !!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: __gcc__ assumes preprocessed (__.i__) files are C and assumes C style linking. __g++__ assumes preprocessed (__.i__) files are C++ and assumes C++ style linking. Suffixes of source file names indicate the language and kind of processing to be done: __.c__ C source; preprocess, compile, assemble __.C__ C++ source; preprocess, compile, assemble __.cc__ C++ source; preprocess, compile, assemble __.cxx__ C++ source; preprocess, compile, assemble __.m__ Objective-C source; preprocess, compile, assemble __.i__ preprocessed C; compile, assemble __.ii__ preprocessed C++; compile, assemble __.s__ Assembler source; assemble __.S__ Assembler source; preprocess, assemble __.h__ Preprocessor file; not usually named on command line Files with other suffixes are passed to the linker. Common cases include: __.o__ Object file __.a__ Archive file Linking is always the last stage unless you use one of the __-c__, __-S__, or __-E__ options to avoid it (or unless compilation errors stop the whole process). For the link stage, all __.o__ files corresponding to source files, __-l__ libraries, unrecognized filenames (including named __.o__ object files and __.a__ archives) are passed to the linker in command-line order. !!OPTIONS Options must be separate: `__-dr__' is quite different from `__-d -r__ '. Most `__-f__' and `__-W__' options have two contrary forms: __-f__''name'' and __-fno-__''name'' (or __-W__''name'' and __-Wno-__''name''). 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. __Overall Options__ -c -S -E -o ''file'' -pipe -v -x ''language'' __Language Options__ -ansi -fall-virtual -fcond-mismatch -fdollars-in-identifiers -fenum-int-equiv -fexternal-templates -fno-asm -fno-builtin -fhosted -fno-hosted -ffreestanding -fno-freestanding -fno-strict-prototype -fsigned-bitfields -fsigned-char -fthis-is-variable -funsigned-bitfields -funsigned-char -fwritable-strings -traditional -traditional-cpp -trigraphs __Warning Options__ -fsyntax-only -pedantic -pedantic-errors -w -W -Wall -Waggregate-return -Wcast-align -Wcast-qual -Wchar-subscript -Wcomment -Wconversion -Wenum-clash -Werror -Wformat -Wid-clash-''len'' -Wimplicit -Wimplicit-int -Wimplicit-function-declaration -Winline -Wlong-long -Wmain -Wmissing-prototypes -Wmissing-declarations -Wnested-externs -Wno-import -Wparentheses -Wpointer-arith -Wredundant-decls -Wreturn-type -Wshadow -Wstrict-prototypes -Wswitch -Wtemplate-debugging -Wtraditional -Wtrigraphs -Wuninitialized -Wunused -Wwrite-strings __Debugging Options__ -a -d''letters'' -fpretend-float -g -g''level'' -gcoff -gxcoff -gxcoff+ -gdwarf -gdwarf+ -gstabs -gstabs+ -ggdb -p -pg -save-temps -print-file-name=''library'' -print-libgcc-file-name -print-prog-name=''program'' __Optimization Options__ -fcaller-saves -fcse-follow-jumps -fcse-skip-blocks -fdelayed-branch -felide-constructors -fexpensive-optimizations -ffast-math -ffloat-store -fforce-addr -fforce-mem -finline-functions -fkeep-inline-functions -fmemoize-lookups -fno-default-inline -fno-defer-pop -fno-function-cse -fno-inline -fno-peephole -fomit-frame-pointer -frerun-cse-after-loop -fschedule-insns -fschedule-insns2 -fstrength-reduce -fthread-jumps -funroll-all-loops -funroll-loops -O -O2 -O3 __Preprocessor Options__ -A''assertion'' -C -dD -dM -dN -D''macro''[[=''defn''] -E -H -idirafter ''dir'' -include ''file'' -imacros ''file'' -iprefix ''file'' -iwithprefix ''dir'' -M -MD -MM -MMD -nostdinc -P -U''macro'' -undef __Assembler Option__ -Wa,''option'' __Linker Options__ -l''library'' -nostartfiles -nostdlib -static -shared -symbolic -Xlinker ''option'' -Wl,''option'' -u ''symbol'' __Directory Options__ -B''prefix'' -I''dir'' -I- -L''dir'' __Target Options__ -b ''machine'' -V ''version'' __Configuration Dependent Options__ ''M680x0 Options'' -m68000 -m68020 -m68020-40 -m68030 -m68040 -m68881 -mbitfield -mc68000 -mc68020 -mfpa -mnobitfield -mrtd -mshort -msoft-float ''VAX Options'' -mg -mgnu -munix ''SPARC Options'' -mepilogue -mfpu -mhard-float -mno-fpu -mno-epilogue -msoft-float -msparclite -mv8 -msupersparc -mcypress ''Convex Options'' -margcount -mc1 -mc2 -mnoargcount ''AMD29K Options'' -m29000 -m29050 -mbw -mdw -mkernel-registers -mlarge -mnbw -mnodw -msmall -mstack-check -muser-registers ''M88K Options'' -m88000 -m88100 -m88110 -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-serialize-volatile -mno-underscores -mocs-debug-info -mocs-frame-position -moptimize-arg-area -mserialize-volatile -mshort-data-''num'' -msvr3 -msvr4 -mtrap-large-shift -muse-div-instruction -mversion-03.00 -mwarn-passed-structs ''RS6000 Options'' -mfp-in-toc -mno-fop-in-toc ''RT Options'' -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 ''MIPS Options'' -mcpu=''cpu type'' -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 ''num'' -nocpp ''i386 Options'' -m486 -mno-486 -msoft-float -mno-fp-ret-in-387 ''HPPA Options'' -mpa-risc-1-0 -mpa-risc-1-1 -mkernel -mshared-libs -mno-shared-libs -mlong-calls -mdisable-fpregs -mdisable-indexing -mtrailing-colon ''i960 Options'' -m''cpu-type'' -mnumerics -msoft-float -mleaf-procedures -mno-leaf-procedures -mtail-call -mno-tail-call -mcomplex-addr -mno-complex-addr -mcode-align -mno-code-align -mic-compat -mic2.0-compat -mic3.0-compat -masm-compat -mintel-asm -mstrict-align -mno-strict-align -mold-align -mno-old-align ''DEC Alpha Options'' -mfp-regs -mno-fp-regs -mno-soft-float -msoft-float ''System V Options'' -G -Qy -Qn -YP,''paths'' -Ym,''dir'' __Code Generation Options__ -fcall-saved-''reg'' -fcall-used-''reg'' -ffixed-''reg'' -finhibit-size-directive -fnonnull-objects -fno-common -fno-ident -fno-gnu-linker -fpcc-struct-return -fpic -fPIC -freg-struct-return -fshared-data -fshort-enums -fshort-double -fvolatile -fvolatile-global -fverbose-asm !!OVERALL OPTIONS __-x__ ''language'' Specify explicitly the ''language'' 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 `__-x__' op- tion. Possible values of ''language'' are `__c__', `__objec- tive-c__', `__c-header__', `__c++__', `__cpp-output__', `__assem- bler__', and `__assembler-with-cpp__'. __-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 `__-x__' has not been used at all). If you want only some of the four stages (preprocess, com- pile, assemble, link), you can use `__-x__' (or filename suf- fixes) to tell __gcc__ where to start, and one of the options `__-c__', `__-S__', or `__-E__' to say where __gcc__ is to stop. Note that some combinations (for example, `__-x cpp-output -E__') instruct __gcc__ to do nothing at all. __-c__ Compile or assemble the source files, but do not link. The compiler output is an object file corre- sponding to each source file. By default, GCC makes the object file name for a source file by replacing the suffix `__.c__', `__.i__', `__.s__', etc., with `__.o__'. Use __-o__ to select another name. GCC ignores any unrecognized input files (those that do not require compilation or assembly) with the __-c__ option. __-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__', `__.i__', etc., with `__.s__'. Use __-o__ to select another name. GCC ignores any input files that don't require com- pilation. __-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 prepro- cessing. __-o__ ''file'' Place output in file ''file''. 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 `__-o__' when compiling more than one input file, unless you are producing an executable file as output. If you do not specify `__-o__', the default is to put an executable file in `__a.out__', the object file for `''source''__.__''suffix''' in `''source''__.o__', its assembler file in `''source''__.s__', and all preprocessed C source on standard output. __-v__ Print (on standard error output) the commands exe- cuted to run the stages of compilation. Also print the version number of the compiler driver program and of the preprocessor and the compiler proper. __-pipe__ Use pipes rather than temporary files for communi- cation between the various stages of compilation. This fails to work on some systems where the assem- bler cannot read from a pipe; but the GNU assembler has no trouble. !!LANGUAGE OPTIONS The following options control the dialect of C that the compiler accepts: __-ansi__ Support all ANSI standard C programs. This turns off certain features of GNU C that are incompatible with ANSI C, such as the __asm__, __inline__ and __typeof__ keywords, and predefined macros such as __unix__ and __vax__ that identify the type of system you are using. It also enables the undesirable and rarely used ANSI trigraph feature, and disallows `__$__' as part of identifiers. The alternate keywords ____asm____, ____extension____, ____inline____ and ____typeof____ continue to work despite `__-ansi__'. 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 compila- tions done with `__-ansi__'. Alternate predefined macros such as ____unix____ and ____vax____ are also avail- able, with or without `__-ansi__'. The `__-ansi__' option does not cause non-ANSI programs to be rejected gratuitously. For that, `__-pedantic__' is required in addition to `__-ansi__'. The preprocessor predefines a macro ____STRICT_ANSI____ when you use the `__-ansi__' 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. __-fno-asm__ Do not recognize __asm__, __inline__ or __typeof__ as a key- word. These words may then be used as identifiers. You can use ____asm____, ____inline____ and ____typeof____ in- stead. `__-ansi__' implies `__-fno-asm__'. __-fno-builtin__ Don't recognize built-in functions that do not be- gin with two leading underscores. Currently, the functions affected include ___exit__, __abort__, __abs__, __allo- ca__, __cos__, __exit__, __fabs__, __labs__, __memcmp__, __memcpy__, __sin__, __sqrt__, __strcmp__, __strcpy__, and __strlen__. The `__-ansi__' option prevents __alloca__ and ___exit__ from being builtin functions. __-fhosted__ Compile for a hosted environment; this implies the `__-fbuiltin__' option, and implies that suspicious declarations of __main__ should be warned about. __-ffreestanding__ Compile for a freestanding environment; this im- plies the `__-fno-builtin__' option, and implies that __main__ has no special requirements. __-fno-strict-prototype__ Treat a function declaration with no arguments, such as `__int foo__ ();', as C would treat it--as say- ing nothing about the number of arguments or their types (C++ only). Normally, such a declaration in C++ means that the function __foo__ takes no arguments. __-trigraphs__ Support ANSI C trigraphs. The `__-ansi__' option im- plies `__-trigraphs__'. __-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). `__-traditional__' has one additional effect for C++: assignment to __this__ is permitted. This is the same as the effect of `__-fthis-is-variable__'. __-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 `__-traditional__'. __-fdollars-in-identifiers__ Permit the use of `__$__' in identifiers (C++ only). You can also use `__-fno-dollars-in-identifiers__' to explicitly prohibit use of `__$__'. (GNU C++ allows `__$__' by default on some target systems but not oth- ers.) __-fenum-int-equiv__ Permit implicit conversion of __int__ to enumeration types (C++ only). Normally GNU C++ allows conver- sion of __enum__ to __int__, but not the other way around. __-fexternal-templates__ Produce smaller code for template declarations, by generating only a single copy of each template function where it is defined (C++ only). To use this option successfully, you must also mark all files that use templates with either `__#pragma im- plementation__' (the definition) or `__#pragma inter- face__' (declarations). When your code is compiled with `__-fexternal-tem- plates__', all template instantiations are external. You must arrange for all necessary instantiations to appear in the implementation file; you can do this with a __typedef__ that references each instantia- tion needed. Conversely, when you compile using the default option `__-fno-external-templates__', all template instantiations are explicitly internal. __-fall-virtual__ Treat all possible member functions as virtual, im- plicitly. All member functions (except for con- structor functions and __new__ or __delete__ member opera- tors) are treated as virtual functions of the class where they appear. This does not mean that all calls to these member functions will be made through the internal table of virtual functions. Under some circumstances, the compiler can determine that a call to a given virtual function can be made directly; in these cases the calls are direct in any case. __-fcond-mismatch__ Allow conditional expressions with mismatched types in the second and third arguments. The value of such an expression is void. __-fthis-is-variable__ Permit assignment to __this__ (C++ only). The incorpo- ration of user-defined free store management into C++ has made assignment to `__this__' an anachronism. Therefore, by default it is invalid to assign to __this__ within a class member function. However, for backwards compatibility, you can make it valid with `__-fthis-is-variable__'. __-funsigned-char__ Let the type __char__ be unsigned, like __unsigned char__. Each kind of machine has a default for what __char__ should be. It is either like __unsigned char__ by de- fault or like __signed char__ by default. Ideally, a portable program should always use __signed char__ or __unsigned char__ when it depends on the signedness of an object. But many programs have been written to use plain __char__ 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 __char__ is always a distinct type from each of __signed char__ and __unsigned char__, even though its behavior is always just like one of those two. __-fsigned-char__ Let the type __char__ be signed, like __signed char__. Note that this is equivalent to `__-fno-un- signed-char__', which is the negative form of `__-fun- signed-char__'. Likewise, `__-fno-signed-char__' is equivalent to `__-funsigned-char__'. __-fsigned-bitfields__ __-funsigned-bitfields__ __-fno-signed-bitfields__ __-fno-unsigned-bitfields__ These options control whether a bitfield is signed or unsigned, when declared with no explicit `__signed__' or `__unsigned__' qualifier. By default, such a bitfield is signed, because this is consistent: the basic integer types such as __int__ are signed types. However, when you specify `__-traditional__', bitfields are all unsigned no matter what. __-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. `__-traditional__' also has this ef- fect. Writing into string constants is a very bad idea; ``constants'' should be constant. !!PREPROCESSOR OPTIONS These options control the C preprocessor, which is run on each C source file before actual compilation. If you use the `__-E__' option, GCC does nothing except pre- processing. Some of these options make sense only togeth- er with `__-E__' because they cause the preprocessor output to be unsuitable for actual compilation. __-include__ ''file'' Process ''file'' as input before processing the regular input file. In effect, the contents of ''file'' are compiled first. Any `__-D__' and `__-U__' options on the command line are always processed before `__-include__ ''file''', regardless of the order in which they are written. All the `__-include__' and `__-imacros__' options are processed in the order in which they are writ- ten. __-imacros__ ''file'' Process ''file'' as input, discarding the resulting output, before processing the regular input file. Because the output generated from ''file'' is discard- ed, the only effect of `__-imacros__ ''file''' is to make the macros defined in ''file'' available for use in the main input. The preprocessor evaluates any `__-D__' and `__-U__' options on the command line before pro- cessing `__-imacros__''file''', regardless of the order in which they are written. All the `__-include__' and `__-imacros__' options are processed in the order in which they are written. __-idirafter__ ''dir'' Add the directory ''dir'' to the second include path. The directories on the second include path are searched when a header file is not found in any of the directories in the main include path (the one that `__-I__' adds to). __-iprefix__ ''prefix'' Specify ''prefix'' as the prefix for subsequent `__-iwithprefix__' options. __-iwithprefix__ ''dir'' Add a directory to the second include path. The directory's name is made by concatenating ''prefix'' and ''dir'', where ''prefix'' was specified previously with `__-iprefix__'. __-nostdinc__ Do not search the standard system directories for header files. Only the directories you have speci- fied with `__-I__' options (and the current directory, if appropriate) are searched. By using both `__-nostdinc__' and `__-I-__', you can limit the include-file search file to only those directo- ries you specify explicitly. __-nostdinc++__ Do not search for header files in the C++-specific standard directories, but do still search the other standard directories. (This option is used when building `__libg++__'.) __-undef__ Do not predefine any nonstandard macros. (Includ- ing architecture flags). __-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. __-C__ Tell the preprocessor not to discard comments. Used with the `__-E__' option. __-P__ Tell the preprocessor not to generate `__#line__' com- mands. Used with the `__-E__' option. __-M [[ -MG ]__ Tell the preprocessor to output a rule suitable for __make__ describing the dependencies of each object file. For each source file, the preprocessor out- puts one __make__-rule whose target is the object file name for that source file and whose dependencies are all the files `__#include__'d in it. This rule may be a single line or may be continued with `__\__'-new- line if it is long. The list of rules is printed on standard output instead of the preprocessed C program. `__-M__' implies `__-E__'. `__-MG__' says to treat missing header files as gener- ated files and assume they live in the same direc- tory as the source file. It must be specified in addition to `__-M__'. __-MM [[ -MG ]__ Like `__-M__' but the output mentions only the user header files included with `__#include__ ''file'' ''#include ''''file'' '' __-MD__ Like `__-M__' but the dependency information is written to files with names made by replacing `__.o__' with `__.d__' at the end of the output file names. This is in addition to compiling the file as speci- fied--`__-MD__' does not inhibit ordinary compilation the way `__-M__' does. The Mach utility `__md__' can be used to merge the `__.d__' files into a single dependency file suitable for using with the `__make__' command. __-MMD__ Like `__-MD__' except mention only user header files, not system header files. __-H__ Print the name of each header file used, in addi- tion to other normal activities. __-A__''question''__(__''answer''__)__ Assert the answer ''answer'' for ''question'', in case it is tested with a preprocessor conditional such as `__#if #__''question''__(__''answer''__)__'. `__-A-__' disables the stan- dard assertions that normally describe the target machine. __-A__''question'' (''answer'') Assert the answer ''answer'' for ''question'', in case it is tested with a preprocessor conditional such as `__#if__ #''question''(''answer'')'. `__-A-__' disables the standard assertions that normally describe the target machine. __-D__''macro'' Define macro ''macro'' with the string `__1__' as its defi- nition. __-D__''macro''__=__''defn'' Define macro ''macro'' as ''defn''. All instances of `__-D__' on the command line are processed before any `__-U__' options. __-U__''macro'' Undefine macro ''macro''. `__-U__' options are evaluated after all `__-D__' options, but before any `__-include__' and `__-imacros__' options. __-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 `__-E__' option. __-dD__ Tell the preprocessor to pass all macro definitions into the output, in their proper sequence in the rest of the output. __-dN__ Like `__-dD__' except that the macro arguments and con- tents are omitted. Only `__#define__ ''name''' is included in the output. !!ASSEMBLER OPTION __-Wa,__''option'' Pass ''option'' as an option to the assembler. If ''op- tion'' contains commas, it is split into multiple op- tions at the commas. !!LINKER OPTIONS These options come into play when the compiler links ob- ject files into an executable output file. They are mean- ingless if the compiler is not doing a link step. ''object-file-name'' A file name that does not end in a special recog- nized suffix is considered to name an object file or library. (Object files are distinguished from libraries by the linker according to the file con- tents.) If GCC does a link step, these object files are used as input to the linker. __-l__''library'' Use the library named ''library'' when linking. The linker searches a standard list of directories for the library, which is actually a file named `__lib__''library''.a'. 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 `__-L__'. Normally the files found this way are library files--archive files whose members are object files. The linker handles an archive file by scan- ning 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 `__-l__' option and specifying a file name is that `__-l__' surrounds ''library'' with `__lib__' and `__.a__' and searches several directories. __-lobjc__ You need this special case of the __-l__ option in or- der to link an Objective C program. __-nostartfiles__ Do not use the standard system startup files when linking. The standard libraries are used normally. __-nostdlib__ Don't use the standard system libraries and startup files when linking. Only the files you specify will be passed to the linker. __-static__ On systems that support dynamic linking, this pre- vents linking with the shared libraries. On other systems, this option has no effect. __-shared__ Produce a shared object which can then be linked with other objects to form an executable. Only a few systems support this option. __-symbolic__ Bind references to global symbols when building a shared object. Warn about any unresolved refer- ences (unless overridden by the link editor option `__-Xlinker -z -Xlinker defs__'). Only a few systems support this option. __-Xlinker__ ''option'' Pass ''option'' as an option to the linker. You can use this to supply system-specific linker options which GNU CC does not know how to recognize. If you want to pass an option that takes an argu- ment, you must use `__-Xlinker__' twice, once for the option and once for the argument. For example, to pass `__-assert definitions__', you must write `__-Xlink- er -assert -Xlinker definitions__'. It does not work to write `__-Xlinker __', because this passes the entire string as a single argument, which is not what the linker expects. __-Wl,__''option'' Pass ''option'' as an option to the linker. If ''option'' contains commas, it is split into multiple options at the commas. __-u__ ''symbol'' Pretend the symbol ''symbol'' is undefined, to force linking of library modules to define it. You can use `__-u__' multiple times with different symbols to force loading of additional library modules. !!DIRECTORY OPTIONS These options specify directories to search for header files, for libraries and for parts of the compiler: __-I__''dir'' Append directory ''dir'' to the list of directories searched for include files. __-I-__ Any directories you specify with `__-I__' options be- fore the `__-I-__' option are searched only for the case of `__#include __''file''____'; they are not searched for `__#include __''file'''' If additional directories are specified with `__-I__' options after the `__-I-__', these directories are searched for all `__#include__' directives. (Ordinari- ly ''all'' `__-I__' directories are used this way.) In addition, the `__-I-__' option inhibits the use of the current directory (where the current input file came from) as the first search directory for `__#in- clude __''file''____'. There is no way to override this effect of `__-I-__'. With `__-I.__' 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. `__-I-__' does not inhibit the use of the standard sys- tem directories for header files. Thus, `__-I-__' and `__-nostdinc__' are independent. __-L__''dir'' Add directory ''dir'' to the list of directories to be searched for `__-l__'. __-B__''prefix'' This option specifies where to find the executa- bles, libraries and data files of the compiler it- self. The compiler driver program runs one or more of the subprograms `__cpp__', `__cc1__' (or, for C++, `__cc1plus__'), `__as__' and `__ld__'. It tries ''prefix'' as a prefix for each program it tries to run, both with and without `''machine''__/__''version''__/__'. For each subprogram to be run, the compiler driver first tries the `__-B__' prefix, if any. If that name is not found, or if `__-B__' was not specified, the driver tries two standard prefixes, which are `__/usr/lib/gcc/__' and `__/usr/local/lib/gcc-lib/__'. If neither of those results in a file name that is found, the compiler driver searches for the unmodi- fied program name, using the directories specified in your `__PATH__' environment variable. The run-time support file `__libgcc.a__' is also searched for using the `__-B__' 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, `__libgcc.a__' is not actually necessary. You can get a similar result from the environment variable __GCC_EXEC_PREFIX__; if it is defined, its value is used as a prefix in the same way. If both the `__-B__' option and the __GCC_EXEC_PREFIX__ variable are present, the `__-B__' option is used first and the environment variable value second. !!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: __-fsyntax-only__ Check the code for syntax errors, but don't emit any output. __-w__ Inhibit all warning messages. __-Wno-import__ Inhibit warning messages about the use of __#import__. __-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 prop- erly with or without this option (though a rare few will require `__-ansi__'). However, without this op- tion, certain GNU extensions and traditional C fea- tures are supported as well. With this option, they are rejected. There is no reason to ''use'' this option; it exists only to satisfy pedants. `__-pedantic__' does not cause warning messages for use of the alternate keywords whose names begin and end with `______'. Pedantic warnings are also disabled in the expression that follows ____extension____. Howev- er, only system header files should use these es- cape routes; application programs should avoid them. __-pedantic-errors__ Like `__-pedantic__', except that errors are produced rather than warnings. __-W__ Print extra warning messages for these events: A nonvolatile automatic variable might be changed by a call to __longjmp__. These warnings are possible only in optimizing compilation. The compiler sees only the calls to __setjmp__. It cannot know where __longjmp__ 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 __longjmp__ cannot in fact be called at the place which would cause a problem. A function can return either with or without a val- ue. (Falling off the end of the function body is considered returning without a value.) For exam- ple, this function would evoke such a warning: foo (a) { if (a Spurious warnings can occur because GNU CC does not realize that certain functions (including __abort__ and __longjmp__) will never return. An expression-statement or the left-hand side of a comma expression contains no side effects. To sup- press the warning, cast the unused expression to void. For example, an expression such as `__x[[i,j]__' will cause a warning, but `__x[[(void)i,j]__' will not. An unsigned value is compared against zero with `____' or `____'. __-Wimplicit-int__ Warn whenever a declaration does not specify a type. __-Wimplicit-function-declaration__ Warn whenever a function is used before being de- clared. __-Wimplicit__ Same as -Wimplicit-int and -Wimplicit-function-dec- laration. __-Wmain__ Warn if the __main__ function is declared or defined with a suspicious type. Typically, it is a func- tion with external linkage, returning __int__, and tak- ing zero or two arguments. __-Wreturn-type__ Warn whenever a function is defined with a re- turn-type that defaults to __int__. Also warn about any __return__ statement with no return-value in a function whose return-type is not __void__. __-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. __-Wswitch__ Warn whenever a __switch__ statement has an index of enumeral type and lacks a __case__ for one or more of the named codes of that enumeration. (The presence of a __default__ label prevents this warning.) __case__ labels outside the enumeration range also provoke warnings when this option is used. __-Wcomment__ Warn whenever a comment-start sequence `__/__' appears in a comment. __-Wtrigraphs__ Warn if any trigraphs are encountered (assuming they are enabled). __-Wformat__ Check calls to __printf__ and __scanf__, etc., to make sure that the arguments supplied have types appropriate to the format string specified. __-Wchar-subscripts__ Warn if an array subscript has type __char__. This is a common cause of error, as programmers often for- get that this type is signed on some machines. __-Wuninitialized__ An automatic variable is used without first being initialized. These warnings are possible only in optimizing com- pilation, because they require data flow informa- tion that is computed only when optimizing. If you don't specify `__-O__', 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 __volatile__, 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 delet- ed 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: { int x; switch (y) { case 1: x = 1; break; case 2: x = 4; break; case 3: x = 5; } foo (x); } If the value of __y__ is always 1, 2 or 3, then __x__ is always initialized, but GNU CC doesn't know this. Here is another common case: { int save_y; if (change_y) save_y = y, y = new_y; ... if (change_y) y = save_y; } This has no bug because __save_y__ is used only if it is set. Some spurious warnings can be avoided if you de- clare as __volatile__ all the functions you use that never return. __-Wparentheses__ Warn if parentheses are omitted in certain con- texts. __-Wtemplate-debugging__ When using templates in a C++ program, warn if de- bugging is not yet fully available (C++ only). __-Wall__ All of the above `__-W__' options combined. These are all the options which pertain to usage that we rec- ommend avoiding and that we believe is easy to avoid, even in conjunction with macros. The remaining `__-W...__' options are not implied by `__-Wall__' because they warn about constructions that we consider reasonable to use, on occasion, in clean programs. __-Wtraditional__ Warn about certain constructs that behave differ- ently in traditional and ANSI C. Macro arguments occurring within string constants in the macro body. These would substitute the ar- gument in traditional C, but are part of the con- stant in ANSI C. A function declared external in one block and then used after the end of the block. A __switch__ statement has an operand of type __long__. __-Wshadow__ Warn whenever a local variable shadows another lo- cal variable. __-Wid-clash-__''len'' Warn whenever two distinct identifiers match in the first ''len'' characters. This may help you prepare a program that will compile with certain obsolete, brain-damaged compilers. __-Wpointer-arith__ Warn about anything that depends on the ``size of'' a function type or of __void__. GNU C assigns these types a size of 1, for convenience in calculations with __void__ pointers and pointers to functions. __-Wcast-qual__ Warn whenever a pointer is cast so as to remove a type qualifier from the target type. For example, warn if a __const char__ is cast to an ordinary __char__ . __-Wcast-align__ Warn whenever a pointer is cast such that the re- quired alignment of the target is increased. For example, warn if a __char__ is cast to an __int__ on ma- chines where integers can only be accessed at two- or four-byte boundaries. __-Wwrite-strings__ Give string constants the type __const char[[__''length''__]__ so that copying the address of one into a non-__const char__ 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 __const__ in declarations and prototypes. Otherwise, it will just be a nuisance; this is why we did not make `__-Wall__' request these warnings. __-Wconversion__ Warn if a prototype causes a type conversion that is different from what would happen to the same ar- gument in the absence of a prototype. This in- cludes 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. __-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 warn- ing.) __-Wstrict-prototypes__ Warn if a function is declared or defined without specifying the argument types. (An old-style func- tion definition is permitted without a warning if preceded by a declaration which specifies the argu- ment types.) __-Wmissing-prototypes__ Warn if a global function is defined without a pre- vious prototype declaration. This warning is is- sued even if the definition itself provides a pro- totype. The aim is to detect global functions that fail to be declared in header files. __-Wmissing-declarations__ Warn if a global function is defined without a pre- vious declaration. Do so even if the definition itself provides a prototype. Use this option to detect global functions that are not declared in header files. __-Wredundant-decls__ Warn if anything is declared more than once in the same scope, even in cases where multiple declara- tion is valid and changes nothing. __-Wnested-externs__ Warn if an __extern__ declaration is encountered within an function. __-Wenum-clash__ Warn about conversion between different enumeration types (C++ only). __-Wlong-long__ Warn if __long long__ type is used. This is default. To inhibit the warning messages, use flag `__-Wno-long-long__'. Flags `__-W-long-long__' and `__-Wno-long-long__' are taken into account only when flag `__-pedantic__' is used. __-Woverloaded-virtual__ (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 signa- ture that doesn't match any virtual functions from the base class. __-Winline__ Warn if a function can not be inlined, and either it was declared as inline, or else the __-fin- line-functions__ option was given. __-Werror__ Treat warnings as errors; abort compilation after any warning. !!DEBUGGING OPTIONS GNU CC has various special options that are used for de- bugging either your program or GCC: __-g__ Produce debugging information in the operating sys- tem's native format (stabs, COFF, XCOFF, or DWARF). GDB can work with this debugging information. On most systems that use stabs format, `__-g__' enables use of extra debugging information that only GDB can use; this extra information makes debugging work better in GDB but will probably make other de- buggers crash or refuse to read the program. If you want to control for certain whether to generate the extra information, use `__-gstabs+__', `__-gstabs__', `__-gxcoff+__', `__-gxcoff__', `__-gdwarf+__', or `__-gdwarf__' (see below). Unlike most other C compilers, GNU CC allows you to use `__-g__' with `__-O__'. The shortcuts taken by opti- mized code may occasionally produce surprising re- sults: 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 val- ues were already at hand; some statements may exe- cute 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 opti- mizer for programs that might have bugs. The following options are useful when GNU CC is generated with the capability for more than one debugging format. __-ggdb__ Produce debugging information in the native format (if that is supported), including GDB extensions if at all possible. __-gstabs__ Produce debugging information in stabs format (if that is supported), without GDB extensions. This is the format used by DBX on most BSD systems. __-gstabs+__ Produce debugging information in stabs format (if that is supported), using GNU extensions understood only by the GNU debugger (GDB). The use of these extensions is likely to make other debuggers crash or refuse to read the program. __-gcoff__ Produce debugging information in COFF format (if that is supported). This is the format used by SDB on most System V systems prior to System V Release 4. __-gxcoff__ Produce debugging information in XCOFF format (if that is supported). This is the format used by the DBX debugger on IBM RS/6000 systems. __-gxcoff+__ Produce debugging information in XCOFF format (if that is supported), using GNU extensions understood only by the GNU debugger (GDB). The use of these extensions is likely to make other debuggers crash or refuse to read the program. __-gdwarf__ Produce debugging information in DWARF format (if that is supported). This is the format used by SDB on most System V Release 4 systems. __-gdwarf+__ Produce debugging information in DWARF format (if that is supported), using GNU extensions understood only by the GNU debugger (GDB). The use of these extensions is likely to make other debuggers crash or refuse to read the program. __-g__''level''__ -ggdb__''level''__ -gstabs__''level''__ -gcoff__''level'' __-gxcoff__''level'' __-gdwarf__''level'' Request debugging information and also use ''level'' 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 informa- tion about local variables and no line numbers. Level 3 includes extra information, such as all the macro definitions present in the program. Some de- buggers support macro expansion when you use `__-g3__'. __-p__ Generate extra code to write profile information suitable for the analysis program __prof__. __-pg__ Generate extra code to write profile information suitable for the analysis program __gprof__. __-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 __tcov__. Note, however, that the format of the data is not what __tcov__ expects. Eventually GNU __gprof__ should be ex- tended to process this data. __-ax__ Generate extra code to read basic block profiling parameters from file `bb.in' and write profiling results to file `bb.out'. `bb.in' contains a list of functions. Whenever a function on the list is entered, profiling is turned on. When the outmost function is left, profiling is turned off. If a function name is prefixed with `-' the function is excluded from profiling. If a function name is not unique it can be disambiguated by writing `/path/filename.d:functionname'. `bb.out' will list some available filenames. Four function names have a special meaning: `__bb_jumps__' will cause jump frequencies to be written to `bb.out'. `__bb_trace__' will cause the sequence of basic blocks to be piped into `gzip' and written to file `bbtrace.gz'. `__bb_hidecall__' will cause call instructions to be excluded from the trace. `__bb_showret__' will cause return instructions to be included in the trace. __-d__''letters'' Says to make debugging dumps during compilation at times specified by ''letters''. This is used for de- bugging the compiler. The file names for most of the dumps are made by appending a word to the source file name (e.g. `__foo.c.rtl__' or `__foo.c.jump__'). __-dM__ Dump all macro definitions, at the end of prepro- cessing, and write no output. __-dN__ Dump all macro names, at the end of preprocessing. __-dD__ Dump all macro definitions, at the end of prepro- cessing, in addition to normal output. __-dy__ Dump debugging information during parsing, to stan- dard error. __-dr__ Dump after RTL generation, to `''file''__.rtl__'. __-dx__ Just generate RTL for a function instead of compil- ing it. Usually used with `__r__'. __-dj__ Dump after first jump optimization, to `''file''__.jump__'. __-ds__ Dump after CSE (including the jump optimization that sometimes follows CSE), to `''file''__.cse__'. __-dL__ Dump after loop optimization, to `''file''__.loop__'. __-dt__ Dump after the second CSE pass (including the jump optimization that sometimes follows CSE), to `''file''__.cse2__'. __-df__ Dump after flow analysis, to `''file''__.flow__'. __-dc__ Dump after instruction combination, to `''file''__.com- bine__'. __-dS__ Dump after the first instruction scheduling pass, to `''file''__.sched__'. __-dl__ Dump after local register allocation, to `''file''__.lreg__'. __-dg__ Dump after global register allocation, to `''file''__.greg__'. __-dR__ Dump after the second instruction scheduling pass, to `''file''__.sched2__'. __-dJ__ Dump after last jump optimization, to `''file''__.jump2__'. __-dd__ Dump after delayed branch scheduling, to `''file''__.dbr__'. __-dk__ Dump after conversion from registers to stack, to `''file''__.stack__'. __-da__ Produce all the dumps listed above. __-dm__ Print statistics on memory usage, at the end of the run, to standard error. __-dp__ Annotate the assembler output with a comment indi- cating which pattern and alternative was used. __-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 ma- chine. __-save-temps__ Store the usual ``temporary'' intermediate files permanently; place them in the current directory and name them based on the source file. Thus, com- piling `__foo.c__' with `__-c -save-temps__' would produce files `__foo.cpp__' and `__foo.s__', as well as `__foo.o__'. __-print-file-name=__''library'' Print the full absolute name of the library file ''library'' that would be used when linking--and do not do anything else. With this option, GNU CC does not compile or link anything; it just prints the file name. __-print-libgcc-file-name__ Same as `__-print-file-name=libgcc.a__'. __-print-prog-name=__''program'' Like `__-print-file-name__', but searches for a program such as `cpp'. !!OPTIMIZATION OPTIONS These options control various sorts of optimizations: __-O__ __-O1__ Optimize. Optimizing compilation takes somewhat more time, and a lot more memory for a large func- tion. Without `__-O__', 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 re- sults you would expect from the source code. Without `__-O__', only variables declared __register__ are allocated in registers. The resulting compiled code is a little worse than produced by PCC without `__-O__'. With `__-O__', the compiler tries to reduce code size and execution time. When you specify `__-O__', the two options `__-fthread-jumps__' and `__-fdefer-pop__' are turned on. On machines that have delay slots, the `__-fde- layed-branch__' option is turned on. For those ma- chines that can support debugging even without a frame pointer, the `__-fomit-frame-pointer__' option is turned on. On some machines other flags may also be turned on. __-O2__ Optimize even more. Nearly all supported optimiza- tions that do not involve a space-speed tradeoff are performed. Loop unrolling and function inlin- ing are not done, for example. As compared to __-O__, this option increases both compilation time and the performance of the generated code. __-O3__ Optimize yet more. This turns on everything __-O2__ does, along with also turning on __-finline-func- tions.__ __-O0__ Do not optimize. If you use multiple __-O__ options, with or without level numbers, the last such option is the one that is effective. Options of the form `__-f__''flag''' specify machine-independent flags. Most flags have both positive and negative forms; the negative form of `__-ffoo__' would be `__-fno-foo__'. The following list shows only one form--the one which is not the default. You can figure out the other form by either removing `__no-__' or adding it. __-ffloat-store__ Do not store floating point variables in registers. This prevents undesirable excess precision on ma- chines such as the 68000 where the floating regis- ters (of the 68881) keep more precision than a __dou- ble__ is supposed to have. For most programs, the excess precision does only good, but a few programs rely on the precise defi- nition of IEEE floating point. Use `__-ffloat-store__' for such programs. __-fmemoize-lookups__ __-fsave-memoized__ Use heuristics to compile faster (C++ only). These heuristics are not enabled by default, since they are only effective for certain input files. Other input files compile more slowly. 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 cout makes six passes through all three steps. By using a software cache, a ``hit'' significantly reduces this cost. Unfortunately, using the cache intro- duces another layer of mechanisms which must be im- plemented, and so incurs its own overhead. `__-fmem- oize-lookups__' enables the software cache. Because access privileges (visibility) to members and member functions may differ from one function context to the next, __g++__ may need to flush the cache. With the `__-fmemoize-lookups__' flag, the cache is flushed after every function that is com- piled. The `-fsave-memoized' 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 privi- leges as every other, and the cache need not be flushed. __-fno-default-inline__ Don't make member functions inline by default mere- ly because they are defined inside the class scope (C++ only). __-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 com- piler normally lets arguments accumulate on the stack for several function calls and pops them all at once. __-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 poten- tial common subexpressions. When they are not com- mon subexpressions, instruction combination should eliminate the separate register-load. I am inter- ested in hearing about the difference this makes. __-fforce-addr__ Force memory address constants to be copied into registers before doing arithmetic on them. This may produce better code just as `__-fforce-mem__' may. I am interested in hearing about the difference this makes. __-fomit-frame-pointer__ Don't keep the frame pointer in a register for functions that don't need one. This avoids the in- structions to save, set up and restore frame point- ers; it also makes an extra register available in many functions. ''It also makes debugging impossible on most machines''. On some machines, such as the Vax, this flag has no effect, because the standard calling sequence auto- matically handles the frame pointer and nothing is saved by pretending it doesn't exist. The ma- chine-description macro __FRAME_POINTER_REQUIRED__ con- trols whether a target machine supports this flag. __-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 __static__, then GCC nor- mally does not output the function as assembler code in its own right. __-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 regis- ters 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 ma- chines, usually those which have no call-preserved registers to use instead. __-fkeep-inline-functions__ Even if all calls to a given function are integrat- ed, and the function is declared __static__, neverthe- less output a separate run-time callable version of the function. __-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. __-fno-peephole__ Disable any machine-specific peephole optimiza- tions. __-ffast-math__ This option allows GCC to violate some ANSI or IEEE rules/specifications in the interest of optimizing code for speed. For example, it allows the compil- er to assume arguments to the __sqrt__ function are non-negative numbers. This option should never be turned on by any `__-O__' option since it can result in incorrect output for programs which depend on an exact implementation of IEEE or ANSI rules/specifications for math func- tions. The following options control specific optimizations. The `__-O2__' option turns on all of these optimizations except `__-funroll-loops__' and `__-funroll-all-loops__'. The `__-O__' option usually turns on the `__-fthread-jumps__' and `__-fdelayed-branch__' 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 de- sired. __-fstrength-reduce__ Perform the optimizations of loop strength reduc- tion and elimination of iteration variables. __-fthread-jumps__ Perform optimizations where we check to see if a jump branches to a location where another compari- son subsumed by the first is found. If so, the first branch is redirected to either the destina- tion of the second branch or a point immediately following it, depending on whether the condition is known to be true or false. __-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. __-funroll-all-loops__ Perform the optimization of loop unrolling. This is done for all loops. This usually makes programs run more slowly. __-fcse-follow-jumps__ In common subexpression elimination, scan through jump instructions when the target of the jump is not reached by any other path. For example, when CSE encounters an __if__ statement with an __else__ clause, CSE will follow the jump when the condition tested is false. __-fcse-skip-blocks__ This is similar to `__-fcse-follow-jumps__', but causes CSE to follow jumps which conditionally skip over blocks. When CSE encounters a simple __if__ statement with no else clause, `__-fcse-skip-blocks__' causes CSE to follow the jump around the body of the __if__. __-frerun-cse-after-loop__ Re-run common subexpression elimination after loop optimizations has been performed. __-felide-constructors__ Elide constructors when this seems plausible (C++ only). With this flag, GNU C++ initializes __y__ di- rectly from the call to __foo__ without going through a temporary in the following code: A foo (); A y = foo (); Without this option, GNU C++ first initializes __y__ by calling the appropriate constructor for type __A__; then assigns the result of __foo__ to a temporary; and, finally, replaces the initial value of `__y__' with the temporary. The default behavior (`__-fno-elide-constructors__') is specified by the draft ANSI C++ standard. If your program's constructors have side effects, using `__-felide-constructors__' can make your program act differently, since some constructor calls may be omitted. __-fexpensive-optimizations__ Perform a number of minor optimizations that are relatively expensive. __-fdelayed-branch__ If supported for the target machine, attempt to re- order instructions to exploit instruction slots available after delayed branch instructions. __-fschedule-insns__ If supported for the target machine, attempt to re- order 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. __-fschedule-insns2__ Similar to `__-fschedule-insns__', but requests an ad- ditional pass of instruction scheduling after reg- ister 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. !!TARGET OPTIONS By default, GNU CC compiles code for the same type of ma- chine that you are using. However, it can also be in- stalled as a cross-compiler, to compile for some other type of machine. In fact, several different configura- tions of GNU CC, for different target machines, can be in- stalled side by side. Then you specify which one to use with the `__-b__' option. In addition, older and newer versions of GNU CC can be in- stalled side by side. One of them (probably the newest) will be the default, but you may sometimes wish to use an- other. __-b__ ''machine'' The argument ''machine'' specifies the target machine for compilation. This is useful when you have in- stalled GNU CC as a cross-compiler. The value to use for ''machine'' 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 `__configure__ i386v', meaning to compile for an 80386 running System V, then you would specify `__-b i386v__' to run that cross compiler. When you do not specify `__-b__', it normally means to compile for the same type of machine that you are using. __-V__ ''version'' The argument ''version'' specifies which version of GNU CC to run. This is useful when multiple versions are installed. For example, ''version'' might be `__2.0__', meaning to run GNU CC version 2.0. The default version, when you do not specify `__-V__', is controlled by the way GNU CC is installed. Nor- mally, it will be a version that is recommended for general use. !!MACHINE DEPENDENT OPTIONS Each of the target machine types can have its own special options, starting with `__-m__', to choose among various hard- ware models or configurations--for example, 68010 vs 68020, floating coprocessor or none. A single installed version of the compiler can compile for any model or con- figuration, according to the options specified. Some configurations of the compiler also support addition- al special options, usually for command-line compatibility with other compilers on the same platform. These are the `__-m__' options defined for the 68000 series: __-m68000__ __-mc68000__ Generate output for a 68000. This is the default when the compiler is configured for 68000-based systems. __-m68020__ __-mc68020__ Generate output for a 68020 (rather than a 68000). This is the default when the compiler is configured for 68020-based systems. __-m68881__ Generate output containing 68881 instructions for floating point. This is the default for most 68020-based systems unless __-nfp__ was specified when the compiler was configured. __-m68030__ Generate output for a 68030. This is the default when the compiler is configured for 68030-based systems. __-m68040__ Generate output for a 68040. This is the default when the compiler is configured for 68040-based systems. __-m68020-40__ Generate output for a 68040, without using any of the new instructions. This results in code which can run relatively efficiently on either a 68020/68881 or a 68030 or a 68040. __-mfpa__ Generate output containing Sun FPA instructions for floating point. __-msoft-float__ Generate output containing library calls for float- ing point. ''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 li- brary functions for cross-compilation. __-mshort__ Consider type __int__ to be 16 bits wide, like __short int__. __-mnobitfield__ Do not use the bit-field instructions. `__-m68000__' implies `__-mnobitfield__'. __-mbitfield__ Do use the bit-field instructions. `__-m68020__' im- plies `__-mbitfield__'. This is the default if you use the unmodified sources. __-mrtd__ Use a different function-calling convention, in which functions that take a fixed number of argu- ments return with the __rtd__ 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 __printf__); 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 __rtd__ instruction is supported by the 68010 and 68020 processors, but not by the 68000. These `__-m__' options are defined for the Vax: __-munix__ Do not output certain jump instructions (__aobleq__ and so on) that the Unix assembler for the Vax cannot handle across long ranges. __-mgnu__ Do output those jump instructions, on the assump- tion that you will assemble with the GNU assembler. __-mg__ Output code for g-format floating point numbers in- stead of d-format. These `__-m__' switches are supported on the SPARC: __-mfpu__ __-mhard-float__ Generate output containing floating point instruc- tions. This is the default. __-mno-fpu__ __-msoft-float__ Generate output containing library calls for float- ing point. ''Warning:'' there is no GNU floating-point library for SPARC. Normally the facilities of the machine's usual C compiler are used, but this can- not be done directly in cross-compilation. You must make your own arrangements to provide suitable library functions for cross-compilation. __-msoft-float__ changes the calling convention in the output file; therefore, it is only useful if you compile ''all'' of a program with this option. __-mno-epilogue__ __-mepilogue__ With __-mepilogue__ (the default), the compiler always emits code for function exit at the end of each function. Any function exit in the middle of the function (such as a return statement in C) will generate a jump to the exit code at the end of the function. With __-mno-epilogue__, the compiler tries to emit exit code inline at every function exit. __-mno-v8__ __-mv8__ __-msparclite__ These three options select variations on the SPARC architecture. By default (unless specifically configured for the Fujitsu SPARClite), GCC generates code for the v7 variant of the SPARC architecture. __-mv8__ will give you SPARC v8 code. The only differ- ence from v7 code is that the compiler emits the integer multiply and integer divide instructions which exist in SPARC v8 but not in SPARC v7. __-msparclite__ will give you SPARClite code. This adds the integer multiply, integer divide step and scan (ffs) instructions which exist in SPARClite but not in SPARC v7. __-mcypress__ __-msupersparc__ These two options select the processor for which the code is optimised. With __-mcypress__ (the default), the compiler optimis- es code for the Cypress CY7C602 chip, as used in the !SparcStation/!SparcServer 3xx series. This is also appropriate for the older !SparcStation 1, 2, IPX etc. With __-msupersparc__ the compiler optimises code for the !SuperSparc cpu, as used in the !SparcStation 10, 1000 and 2000 series. This flag also enables use of the full SPARC v8 instruction set. These `__-m__' options are defined for the Convex: __-mc1__ Generate output for a C1. This is the default when the compiler is configured for a C1. __-mc2__ Generate output for a C2. This is the default when the compiler is configured for a C2. __-margcount__ Generate code which puts an argument count in the word preceding each argument list. Some non- portable Convex and Vax programs need this word. (Debuggers don't, except for functions with vari- able-length argument lists; this info is in the symbol table.) __-mnoargcount__ Omit the argument count word. This is the default if you use the unmodified sources. These `__-m__' options are defined for the AMD Am29000: __-mdw__ Generate code that assumes the DW bit is set, i.e., that byte and halfword operations are directly sup- ported by the hardware. This is the default. __-mnodw__ Generate code that assumes the DW bit is not set. __-mbw__ Generate code that assumes the system supports byte and halfword write operations. This is the de- fault. __-mnbw__ Generate code that assumes the systems does not support byte and halfword write operations. This implies `__-mnodw__'. __-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 __call__ instruction to be used instead of a __const__, __consth__, __calli__ sequence. __-mlarge__ Do not assume that the __call__ instruction can be used; this is the default. __-m29050__ Generate code for the Am29050. __-m29000__ Generate code for the Am29000. This is the de- fault. __-mkernel-registers__ Generate references to registers __gr64-gr95__ instead of __gr96-gr127__. This option can be used when com- piling kernel code that wants a set of global reg- isters disjoint from that used by user-mode code. Note that when this option is used, register names in `__-f__' flags must use the normal, user-mode, names. __-muser-registers__ Use the normal set of global registers, __gr96-gr127__. This is the default. __-mstack-check__ Insert a call to ____msp_check__ after each stack ad- justment. This is often used for kernel code. These `__-m__' options are defined for Motorola 88K architec- tures: __-m88000__ Generate code that works well on both the m88100 and the m88110. __-m88100__ Generate code that works best for the m88100, but that also runs on the m88110. __-m88110__ Generate code that works best for the m88110, and may not run on the m88100. __-midentify-revision__ Include an __ident__ directive in the assembler output recording the source file name, compiler name and version, timestamp, and compilation flags used. __-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. __-mno-check-zero-division__ __-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 de- tect division by zero and signal an exception. All GCC configurations for the 88K use `__-mcheck-ze- ro-division__' by default. __-mocs-debug-info__ __-mno-ocs-debug-info__ Include (or omit) additional debugging information (about registers used in each stack frame) as spec- ified 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. __-mocs-frame-position__ __-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 `__-mocs-frame-position__'; oth- er 88k configurations have the default `__-mno-ocs-frame-position__'. __-moptimize-arg-area__ __-mno-optimize-arg-area__ Control how to store function arguments in stack frames. `__-moptimize-arg-area__' saves space, but may break some debuggers (not GDB). `__-mno-opti- mize-arg-area__' conforms better to standards. By default GCC does not optimize the argument area. __-mshort-data-__''num'' ''num'' Generate smaller data references by making them relative to __r0__, which allows loading a value using a single instruction (rather than the usual two). You control which data references are affected by specifying ''num'' with this option. For example, if you specify `__-mshort-data-512__', then the data ref- erences affected are those involving displacements of less than 512 bytes. `__-mshort-data-__''num''' is not effective for ''num'' greater than 64K. __-mserialize-volatile__ __-mno-serialize-volatile__ Do, or do not, generate code to guarantee sequen- tial consistency of volatile memory references. GNU CC always guarantees consistency by default, for the preferred processor submodel. How this is done depends on the submodel. The m88100 processor does not reorder memory refer- ences and so always provides sequential consisten- cy. If you use `__-m88100__', GNU CC does not generate any special instructions for sequential consisten- cy. The order of memory references made by the m88110 processor does not always match the order of the instructions requesting those references. In par- ticular, a load instruction may execute before a preceding store instruction. Such reordering vio- lates sequential consistency of volatile memory references, when there are multiple processors. When you use `__-m88000__' or `__-m88110__', GNU CC gener- ates special instructions when appropriate, to force execution in the proper order. The extra code generated to guarantee consistency may affect the performance of your application. If you know that you can safely forgo this guarantee, you may use the option `__-mno-serialize-volatile__'. If you use the `__-m88100__' option but require sequen- tial consistency when running on the m88110 proces- sor, you should use `__-mserialize-volatile__'. __-msvr4__ __-msvr3__ Turn on (`__-msvr4__') or off (`__-msvr3__') compiler ex- tensions related to System V release 4 (SVr4). This controls the following: Which variant of the assembler syntax to emit (which you can select independently using `__-mver- sion-03.00__'). `__-msvr4__' makes the C preprocessor recognize `__#prag- ma weak__' `__-msvr4__' makes GCC issue additional declaration di- rectives used in SVr4. `__-msvr3__' is the default for all m88K configurations except the SVr4 configuration. __-mtrap-large-shift__ __-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. __-muse-div-instruction__ Very early models of the 88K architecture didn't have a divide instruction, so GCC avoids that in- struction by default. Use this option to specify that it's safe to use the divide instruction. __-mversion-03.00__ In the DG/UX configuration, there are two flavors of SVr4. This option modifies __-msvr4__ to select whether the hybrid-COFF or real-ELF flavor is used. All other configurations ignore this option. __-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. These options are defined for the IBM RS6000: __-mfp-in-toc__ __-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, `__-mno-fp-in-toc__' will reduce the size of the TOC, which may avoid the overflow. These `__-m__' options are defined for the IBM RT PC: __-min-line-mul__ Use an in-line code sequence for integer multi- plies. This is the default. __-mcall-lib-mul__ Call __lmul$$__ for integer multiples. __-mfull-fp-blocks__ Generate full-size floating point data blocks, in- cluding the minimum amount of scratch space recom- mended by IBM. This is the default. __-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. __-mfp-arg-in-fpregs__ Use a calling sequence incompatible with the IBM calling convention in which floating point argu- ments are passed in floating point registers. Note that __varargs.h__ and __stdargs.h__ will not work with floating point operands if this option is speci- fied. __-mfp-arg-in-gregs__ Use the normal calling convention for floating point arguments. This is the default. __-mhc-struct-return__ Return structures of more than one word in memory, rather than in a register. This provides compati- bility with the !MetaWare HighC (hc) compiler. Use `__-fpcc-struct-return__' for compatibility with the Portable C Compiler (pcc). __-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 `__-fpcc-struct-return__' or `__-mhc-struct-return__'. These `__-m__' options are defined for the MIPS family of com- puters: __-mcpu=__''cpu-type'' Assume the defaults for the machine type ''cpu-type'' when scheduling instructions. The default ''cpu-type'' is __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 ''cpu-type'' are __r2000__, __r3000__, __r4000__, and __r6000__. While picking a specific ''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 __-mips2__ or __-mips3__ switches being used. __-mips2__ Issue instructions from level 2 of the MIPS ISA (branch likely, square root instructions). The __-mcpu=r4000__ or __-mcpu=r6000__ switch must be used in conjunction with __-mips2__. __-mips3__ Issue instructions from level 3 of the MIPS ISA (64 bit instructions). The __-mcpu=r4000__ switch must be used in conjunction with __-mips2__. __-mint64__ __-mlong64__ __-mlonglong128__ These options don't work at present. __-mmips-as__ Generate code for the MIPS assembler, and invoke __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 __-ggdb__, __-gstabs__, or __-gstabs+__ switches are used, the __mips-tfile__ program will en- capsulate the stabs within MIPS ECOFF. __-mgas__ Generate code for the GNU assembler. This is the default on the OSF/1 reference platform, using the OSF/rose object format. __-mrnames__ __-mno-rnames__ The __-mrnames__ switch says to output code using the MIPS software names for the registers, instead of the hardware names (ie, __a0__ instead of __$4__). The GNU assembler does not support the __-mrnames__ switch, and the MIPS assembler will be instructed to run the MIPS C preprocessor over the source file. The __-mno-rnames__ switch is default. __-mgpopt__ __-mno-gpopt__ The __-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. __-mstats__ __-mno-stats__ For each non-inline function processed, the __-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.). __-mmemcpy__ __-mno-memcpy__ The __-mmemcpy__ switch makes all block moves call the appropriate string function (__memcpy__ or __bcopy__) in- stead of possibly generating inline code. __-mmips-tfile__ __-mno-mips-tfile__ The __-mno-mips-tfile__ switch causes the compiler not postprocess the object file with the __mips-tfile__ program, after the MIPS assembler has generated it to add debug support. If __mips-tfile__ is not run, then no local variables will be available to the debugger. In addition, __stage2__ and __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. __-msoft-float__ Generate output containing library calls for float- ing point. ''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 li- brary functions for cross-compilation. __-mhard-float__ Generate output containing floating point instruc- tions. This is the default if you use the unmodi- fied sources. __-mfp64__ Assume that the __FR__ bit in the status word is on, and that there are 32 64-bit floating point regis- ters, instead of 32 32-bit floating point regis- ters. You must also specify the __-mcpu=r4000__ and __-mips3__ switches. __-mfp32__ Assume that there are 32 32-bit floating point reg- isters. This is the default. __-mabicalls__ __-mno-abicalls__ Emit (or do not emit) the __.abicalls__, __.cpload__, and __.cprestore__ pseudo operations that some System V.4 ports use for position independent code. __-mhalf-pic__ __-mno-half-pic__ The __-mhalf-pic__ switch says to put pointers to ex- tern references into the data section and load them up, rather than put the references in the text sec- tion. This option does not work at present. __-G__''num'' Put global and static items less than or equal to ''num'' bytes into the small data or bss sections in- stead of the normal data or bss section. This al- lows the assembler to emit one word memory refer- ence instructions based on the global pointer (__gp__ or __$28__), instead of the normal two words used. By default, ''num'' is 8 when the MIPS assembler is used, and 0 when the GNU assembler is used. The __-G__''num'' switch is also passed to the assembler and linker. All modules should be compiled with the same __-G__''num'' value. __-nocpp__ Tell the MIPS assembler to not run its preprocessor over user assembler files (with a `__.s__' suffix) when assembling them. These `__-m__' options are defined for the Intel 80386 family of computers: __-m486__ __-mno-486__ Control whether or not code is optimized for a 486 instead of an 386. Code generated for a 486 will run on a 386 and vice versa. __-msoft-float__ Generate output containing library calls for float- ing point. ''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 li- brary functions for cross-compilation. On machines where a function returns floating point results in the 80387 register stack, some floating point opcodes may be emitted even if `__-msoft-float__' is used. __-mno-fp-ret-in-387__ Do not use the FPU registers for return values of functions. The usual calling convention has functions return values of types __float__ and __double__ in an FPU regis- ter, even if there is no FPU. The idea is that the operating system should emulate an FPU. The option `__-mno-fp-ret-in-387__' causes such values to be returned in ordinary CPU registers instead. These `__-m__' options are defined for the HPPA family of com- puters: __-mpa-risc-1-0__ Generate code for a PA 1.0 processor. __-mpa-risc-1-1__ Generate code for a PA 1.1 processor. __-mkernel__ Generate code which is suitable for use in kernels. Specifically, avoid __add__ instructions in which one of the arguments is the DP register; generate __addil__ instructions instead. This avoids a rather serious bug in the HP-UX linker. __-mshared-libs__ Generate code that can be linked against HP-UX shared libraries. This option is not fully func- tion yet, and is not on by default for any PA tar- get. Using this option can cause incorrect code to be generated by the compiler. __-mno-shared-libs__ Don't generate code that will be linked against shared libraries. This is the default for all PA targets. __-mlong-calls__ Generate code which allows calls to functions greater than 256K away from the caller when the caller and callee are in the same source file. Do not turn this option on unless code refuses to link with ``branch out of range errors from the linker. __-mdisable-fpregs__ Prevent floating point registers from being used in any manner. This is necessary for compiling ker- nels which perform lazy context switching of float- ing point registers. If you use this option and attempt to perform floating point operations, the compiler will abort. __-mdisable-indexing__ Prevent the compiler from using indexing address modes. This avoids some rather obscure problems when compiling MIG generated code under MACH. __-mtrailing-colon__ Add a colon to the end of label definitions (for ELF assemblers). These `__-m__' options are defined for the Intel 80960 family of computers: __-m__''cpu-type'' Assume the defaults for the machine type ''cpu-type'' for instruction and addressing-mode availability and alignment. The default ''cpu-type'' is __kb__; other choices are __ka__, __mc__, __ca__, __cf__, __sa__, and __sb__. __-mnumerics__ __-msoft-float__ The __-mnumerics__ option indicates that the processor does support floating-point instructions. The __-msoft-float__ option indicates that floating-point support should not be assumed. __-mleaf-procedures__ __-mno-leaf-procedures__ Do (or do not) attempt to alter leaf procedures to be callable with the ''bal'' instruction as well as ''call''. This will result in more efficient code for explicit calls when the ''bal'' instruction can be sub- stituted by the assembler or linker, but less effi- cient code in other cases, such as calls via func- tion pointers, or using a linker that doesn't sup- port this optimization. __-mtail-call__ __-mno-tail-call__ Do (or do not) make additional attempts (beyond those of the machine-independent portions of the compiler) to optimize tail-recursive calls into branches. You may not want to do this because the detection of cases where this is not valid is not totally complete. The default is __-mno-tail-call__. __-mcomplex-addr__ __-mno-complex-addr__ Assume (or do not assume) that the use of a complex addressing mode is a win on this implementation of the i960. Complex addressing modes may not be worthwhile on the K-series, but they definitely are on the C-series. The default is currently __-mcom- plex-addr__ for all processors except the CB and CC. __-mcode-align__ __-mno-code-align__ Align code to 8-byte boundaries for faster fetching (or don't bother). Currently turned on by default for C-series implementations only. __-mic-compat__ __-mic2.0-compat__ __-mic3.0-compat__ Enable compatibility with iC960 v2.0 or v3.0. __-masm-compat__ __-mintel-asm__ Enable compatibility with the iC960 assembler. __-mstrict-align__ __-mno-strict-align__ Do not permit (do permit) unaligned accesses. __-mold-align__ Enable structure-alignment compatibility with In- tel's gcc release version 1.3 (based on gcc 1.37). Currently this is buggy in that __#pragma align 1__ is always assumed as well, and cannot be turned off. These `__-m__' options are defined for the DEC Alpha implemen- tations: __-mno-soft-float__ __-msoft-float__ Use (do not use) the hardware floating-point in- structions for floating-point operations. When __-msoft-float__ is specified, functions in `__libgcc1.c__' will be used to perform floating-point operations. Unless they are replaced by routines that emulate the floating-point operations, or compiled in such a way as to call such emulations routines, these routines will issue floating-point operations. If you are compiling for an Alpha without float- ing-point operations, you must ensure that the li- brary is built so as not to call them. Note that Alpha implementations without float- ing-point operations are required to have float- ing-point registers. __-mfp-reg__ __-mno-fp-regs__ Generate code that uses (does not use) the float- ing-point register set. __-mno-fp-regs__ implies __-msoft-float__. If the floating-point register set is not used, floating point operands are passed in integer registers as if they were integers and floating-point results are passed in $0 instead of $f0. This is a non-standard calling sequence, so any function with a floating-point argument or re- turn value called by code compiled with __-mno-fp-regs__ must also be compiled with that op- tion. A typical use of this option is building a kernel that does not use, and hence need not save and re- store, any floating-point registers. These additional options are available on System V Release 4 for compatibility with other compilers on those systems: __-G__ On SVr4 systems, __gcc__ accepts the option `__-G__' (and passes it to the system linker), for compatibility with other compilers. However, we suggest you use `__-symbolic__' or `__-shared__' as appropriate, instead of supplying linker options on the __gcc__ command line. __-Qy__ Identify the versions of each tool used by the com- piler, in a __.ident__ assembler directive in the out- put. __-Qn__ Refrain from adding __.ident__ directives to the output file (this is the default). __-YP,__''dirs'' Search the directories ''dirs'', and no others, for li- braries specified with `__-l__'. You can separate di- rectory entries in ''dirs'' from one another with colons. __-Ym,__''dir'' Look in the directory ''dir'' to find the M4 preproces- sor. The assembler uses this option. !!CODE GENERATION OPTIONS These machine-independent options control the interface conventions used in code generation. Most of them begin with `-f'. These options have both positive and negative forms; the negative form of `__-ffoo__' would be `__-fno-foo__'. In the table below, only one of the forms is listed--the one which is not the default. You can figure out the other form by either removing `__no-__' or adding it. __-fnonnull-objects__ Assume that objects reached through references are not null (C++ only). Normally, GNU C++ makes conservative assumptions about objects reached through references. For ex- ample, the compiler must check that __a__ is not null in code like the following: obj Checking that references of this sort have non-null values requires extra code, however, and it is un- necessary for many programs. You can use `__-fnon- null-objects__' to omit the checks for null, if your program doesn't require checking. __-fpcc-struct-return__ Use the same convention for returning __struct__ and __union__ 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 allow- ing intercallability between GCC-compiled code and PCC-compiled code. __-freg-struct-return__ Use the convention that __struct__ and __union__ values are returned in registers when possible. This is more efficient for small structures than __-fpcc-struct-return__. If you specify neither __-fpcc-struct-return__ nor __-freg-struct-return__, GNU CC defaults to whichever convention is standard for the target. If there is no standard convention, GNU CC defaults to __-fpcc-struct-return__. __-fshort-enums__ Allocate to an __enum__ type only as many bytes as it needs for the declared range of possible values. Specifically, the __enum__ type will be equivalent to the smallest integer type which has enough room. __-fshort-double__ Use the same size for __double__ as for __float__ . __-fshared-data__ Requests that the data and non-__const__ 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 be- tween processes running the same program, while private data exists in one copy per process. __-fno-common__ Allocate even uninitialized global variables in the bss section of the object file, rather than gener- ating them as common blocks. This has the effect that if the same variable is declared (without __ex- tern__) 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. __-fno-ident__ Ignore the `__#ident__' directive. __-fno-gnu-linker__ Do not output global initializations (such as C++ constructors and destructors) in the form used by the GNU linker (on systems where the GNU linker is the standard method of handling them). Use this option when you want to use a non-GNU linker, which also requires using the __collect2__ program to make sure the system linker includes constructors and destructors. (__collect2__ is included in the GNU CC distribution.) For systems which ''must'' use __col- lect2__, the compiler driver __gcc__ is configured to do this automatically. __-finhibit-size-directive__ Don't output a __.size__ assembler directive, or any- thing else that would cause trouble if the function is split in the middle, and the two halves are placed at locations far apart in memory. This op- tion is used when compiling `__crtstuff.c__'; you should not need to use it for anything else. __-fverbose-asm__ Put extra commentary information in the generated assembly code to make it more readable. This op- tion is generally only of use to those who actually need to read the generated assembly code (perhaps while debugging the compiler itself). __-fvolatile__ Consider all memory references through pointers to be volatile. __-fvolatile-global__ Consider all memory references to extern and global data items to be volatile. __-fpic__ If supported for the target machines, generate po- sition-independent code, suitable for use in a shared library. __-fPIC__ If supported for the target machine, emit posi- tion-independent code, suitable for dynamic link- ing, even if branches need large displacements. __-ffixed-__''reg'' Treat the register named ''reg'' 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). ''reg'' must be the name of a register. The register names accepted are machine-specific and are defined in the __REGISTER_NAMES__ macro in the machine descrip- tion macro file. This flag does not have a negative form, because it specifies a three-way choice. __-fcall-used-__''reg'' Treat the register named ''reg'' as an allocable regis- ter 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 ''reg''. 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. __-fcall-saved-__''reg'' Treat the register named ''reg'' as an allocable regis- ter 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 ''reg'' 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. !!PRAGMAS Two `__#pragma__' 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. __#pragma interface__ (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. Nor- mally, local copies of certain information (backup copies of inline member functions, debugging infor- mation, and the internal tables that implement vir- tual 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 `__#pragma interface__' is included in a compilation, this auxiliary information will not be generated (unless the main input source file it- self uses `__#pragma implementation__'). Instead, the object files will contain references to be resolved at link time. __#pragma implementation__ __#pragma implementation __''objects''__.h__ (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 `__#pragma interface__'. Backup copies of inline mem- ber functions, debugging information, and the in- ternal tables used to implement virtual functions are all generated in implementation files. If you use `__#pragma implementation__' with no argu- ment, it applies to an include file with the same basename as your source file; for example, in `__all- class.cc__', `__#pragma implementation__' by itself is equivalent to `__#pragma implementation __'. 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 sin- gle header file into multiple implementation files. !!FILES file.c C source file file.h C header (preprocessor) file file.i preprocessed C source file file.C C++ source file file.cc C++ source file file.cxx C++ source file file.m Objective-C source file file.s assembly language file file.o object file a.out link edited output ''TMPDIR''/cc temporary files ''LIBDIR''/cpp preprocessor ''LIBDIR''/cc1 compiler for C ''LIBDIR''/cc1plus compiler for C++ ''LIBDIR''/collect linker front end needed on some machines ''LIBDIR''/libgcc.a GCC subroutine library /lib/crt[[01n].o start-up routine ''LIBDIR''/ccrt0 additional start-up routine for C++ /lib/libc.a standard C library, see intro(3) /usr/include standard directory for __#include__ files ''LIBDIR''/include standard gcc directory for __#include__ files ''LIBDIR''/g++-include additional g++ directory for __#include __''LIBDIR'' is usually __/usr/local/lib/__''machine''/''version''.'' TMPDIR'' comes from the environment variable __TMPDIR__ (default __/usr/tmp__ if available, else __/tmp__). !!EXIT STATUS Normally the exit status is 0, if compilation or link edit are successful, and nonzero else. The option __-Werror__ treats each warning as an error. !!SEE ALSO cpp(1), as(1), ld(1), gdb(1), adb(1), dbx(1), sdb(1). `__gcc__', `__cpp__', `__as__', `__ld__', and `__gdb__' entries in __info__.'' Using and Porting GNU CC (for version 2.0)'', Richard M. Stallman; ''The C Preprocessor'', Richard M. Stallman; ''Debug- ging with GDB: the GNU Source-Level Debugger'', Richard M. Stallman and Roland H. Pesch; ''Using as: the GNU Assembler'', Dean Elsner, Jay Fenlason ''ld: the GNU linker'', Steve Chamberlain and Roland Pesch. !!BUGS For instructions on reporting bugs, see the GCC manual. !!COPYING Copyright 1991, 1992, 1993 Free Software Foundation, Inc. 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. Permission is granted to copy and distribute modified ver- sions 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. Permission is granted to copy and distribute translations of this manual into another language, under the above con- ditions for modified versions, except that this permission notice may be included in translations approved by the Free Software Foundation instead of in the original En- glish. !!AUTHORS See the GNU CC Manual for the contributors to GNU CC. ----
6 pages link to
cc(1)
:
intro(3)
gprof(1)
Man1c
c89(1)
rpcgen(1)
stdio(3)
This page is a man page (or other imported legacy content). We are unable to automatically determine the license status of this page.
