]>
Commit | Line | Data |
---|---|---|
1 | @c Copyright (C) 1988, 1989, 1992, 1993, 1994, 1995, 1996, 1997, 1998, 1999, | |
2 | @c 2000, 2001, 2002 Free Software Foundation, Inc. | |
3 | @c This is part of the GCC manual. | |
4 | @c For copying conditions, see the file gcc.texi. | |
5 | ||
6 | @ignore | |
7 | @c man begin COPYRIGHT | |
8 | Copyright @copyright{} 1988, 1989, 1992, 1993, 1994, 1995, 1996, 1997, | |
9 | 1998, 1999, 2000, 2001, 2002 Free Software Foundation, Inc. | |
10 | ||
11 | Permission is granted to copy, distribute and/or modify this document | |
12 | under the terms of the GNU Free Documentation License, Version 1.1 or | |
13 | any later version published by the Free Software Foundation; with the | |
14 | Invariant Sections being ``GNU General Public License'' and ``Funding | |
15 | Free Software'', the Front-Cover texts being (a) (see below), and with | |
16 | the Back-Cover Texts being (b) (see below). A copy of the license is | |
17 | included in the gfdl(7) man page. | |
18 | ||
19 | (a) The FSF's Front-Cover Text is: | |
20 | ||
21 | A GNU Manual | |
22 | ||
23 | (b) The FSF's Back-Cover Text is: | |
24 | ||
25 | You have freedom to copy and modify this GNU Manual, like GNU | |
26 | software. Copies published by the Free Software Foundation raise | |
27 | funds for GNU development. | |
28 | @c man end | |
29 | @c Set file name and title for the man page. | |
30 | @setfilename gcc | |
31 | @settitle GNU project C and C++ compiler | |
32 | @c man begin SYNOPSIS | |
33 | gcc [@option{-c}|@option{-S}|@option{-E}] [@option{-std=}@var{standard}] | |
34 | [@option{-g}] [@option{-pg}] [@option{-O}@var{level}] | |
35 | [@option{-W}@var{warn}@dots{}] [@option{-pedantic}] | |
36 | [@option{-I}@var{dir}@dots{}] [@option{-L}@var{dir}@dots{}] | |
37 | [@option{-D}@var{macro}[=@var{defn}]@dots{}] [@option{-U}@var{macro}] | |
38 | [@option{-f}@var{option}@dots{}] [@option{-m}@var{machine-option}@dots{}] | |
39 | [@option{-o} @var{outfile}] @var{infile}@dots{} | |
40 | ||
41 | Only the most useful options are listed here; see below for the | |
42 | remainder. @samp{g++} accepts mostly the same options as @samp{gcc}. | |
43 | @c man end | |
44 | @c man begin SEEALSO | |
45 | gpl(7), gfdl(7), fsf-funding(7), | |
46 | cpp(1), gcov(1), g77(1), as(1), ld(1), gdb(1), adb(1), dbx(1), sdb(1) | |
47 | and the Info entries for @file{gcc}, @file{cpp}, @file{g77}, @file{as}, | |
48 | @file{ld}, @file{binutils} and @file{gdb}. | |
49 | @c man end | |
50 | @c man begin BUGS | |
51 | For instructions on reporting bugs, see | |
52 | @w{@uref{http://gcc.gnu.org/bugs.html}}. Use of the @command{gccbug} | |
53 | script to report bugs is recommended. | |
54 | @c man end | |
55 | @c man begin AUTHOR | |
56 | See the Info entry for @command{gcc}, or | |
57 | @w{@uref{http://gcc.gnu.org/onlinedocs/gcc/Contributors.html}}, | |
58 | for contributors to GCC@. | |
59 | @c man end | |
60 | @end ignore | |
61 | ||
62 | @node Invoking GCC | |
63 | @chapter GCC Command Options | |
64 | @cindex GCC command options | |
65 | @cindex command options | |
66 | @cindex options, GCC command | |
67 | ||
68 | @c man begin DESCRIPTION | |
69 | ||
70 | When you invoke GCC, it normally does preprocessing, compilation, | |
71 | assembly and linking. The ``overall options'' allow you to stop this | |
72 | process at an intermediate stage. For example, the @option{-c} option | |
73 | says not to run the linker. Then the output consists of object files | |
74 | output by the assembler. | |
75 | ||
76 | Other options are passed on to one stage of processing. Some options | |
77 | control the preprocessor and others the compiler itself. Yet other | |
78 | options control the assembler and linker; most of these are not | |
79 | documented here, since you rarely need to use any of them. | |
80 | ||
81 | @cindex C compilation options | |
82 | Most of the command line options that you can use with GCC are useful | |
83 | for C programs; when an option is only useful with another language | |
84 | (usually C++), the explanation says so explicitly. If the description | |
85 | for a particular option does not mention a source language, you can use | |
86 | that option with all supported languages. | |
87 | ||
88 | @cindex C++ compilation options | |
89 | @xref{Invoking G++,,Compiling C++ Programs}, for a summary of special | |
90 | options for compiling C++ programs. | |
91 | ||
92 | @cindex grouping options | |
93 | @cindex options, grouping | |
94 | The @command{gcc} program accepts options and file names as operands. Many | |
95 | options have multi-letter names; therefore multiple single-letter options | |
96 | may @emph{not} be grouped: @option{-dr} is very different from @w{@samp{-d | |
97 | -r}}. | |
98 | ||
99 | @cindex order of options | |
100 | @cindex options, order | |
101 | You can mix options and other arguments. For the most part, the order | |
102 | you use doesn't matter. Order does matter when you use several options | |
103 | of the same kind; for example, if you specify @option{-L} more than once, | |
104 | the directories are searched in the order specified. | |
105 | ||
106 | Many options have long names starting with @samp{-f} or with | |
107 | @samp{-W}---for example, @option{-fforce-mem}, | |
108 | @option{-fstrength-reduce}, @option{-Wformat} and so on. Most of | |
109 | these have both positive and negative forms; the negative form of | |
110 | @option{-ffoo} would be @option{-fno-foo}. This manual documents | |
111 | only one of these two forms, whichever one is not the default. | |
112 | ||
113 | @c man end | |
114 | ||
115 | @xref{Option Index}, for an index to GCC's options. | |
116 | ||
117 | @menu | |
118 | * Option Summary:: Brief list of all options, without explanations. | |
119 | * Overall Options:: Controlling the kind of output: | |
120 | an executable, object files, assembler files, | |
121 | or preprocessed source. | |
122 | * Invoking G++:: Compiling C++ programs. | |
123 | * C Dialect Options:: Controlling the variant of C language compiled. | |
124 | * C++ Dialect Options:: Variations on C++. | |
125 | * Objective-C Dialect Options:: Variations on Objective-C. | |
126 | * Language Independent Options:: Controlling how diagnostics should be | |
127 | formatted. | |
128 | * Warning Options:: How picky should the compiler be? | |
129 | * Debugging Options:: Symbol tables, measurements, and debugging dumps. | |
130 | * Optimize Options:: How much optimization? | |
131 | * Preprocessor Options:: Controlling header files and macro definitions. | |
132 | Also, getting dependency information for Make. | |
133 | * Assembler Options:: Passing options to the assembler. | |
134 | * Link Options:: Specifying libraries and so on. | |
135 | * Directory Options:: Where to find header files and libraries. | |
136 | Where to find the compiler executable files. | |
137 | * Spec Files:: How to pass switches to sub-processes. | |
138 | * Target Options:: Running a cross-compiler, or an old version of GCC. | |
139 | * Submodel Options:: Specifying minor hardware or convention variations, | |
140 | such as 68010 vs 68020. | |
141 | * Code Gen Options:: Specifying conventions for function calls, data layout | |
142 | and register usage. | |
143 | * Environment Variables:: Env vars that affect GCC. | |
144 | * Running Protoize:: Automatically adding or removing function prototypes. | |
145 | @end menu | |
146 | ||
147 | @c man begin OPTIONS | |
148 | ||
149 | @node Option Summary | |
150 | @section Option Summary | |
151 | ||
152 | Here is a summary of all the options, grouped by type. Explanations are | |
153 | in the following sections. | |
154 | ||
155 | @table @emph | |
156 | @item Overall Options | |
157 | @xref{Overall Options,,Options Controlling the Kind of Output}. | |
158 | @gccoptlist{ | |
159 | -c -S -E -o @var{file} -pipe -pass-exit-codes -x @var{language} @gol | |
160 | -v --target-help --help} | |
161 | ||
162 | @item C Language Options | |
163 | @xref{C Dialect Options,,Options Controlling C Dialect}. | |
164 | @gccoptlist{ | |
165 | -ansi -std=@var{standard} -aux-info @var{filename} @gol | |
166 | -fno-asm -fno-builtin -fno-builtin-@var{function} @gol | |
167 | -fhosted -ffreestanding @gol | |
168 | -trigraphs -traditional -traditional-cpp @gol | |
169 | -fallow-single-precision -fcond-mismatch @gol | |
170 | -fsigned-bitfields -fsigned-char @gol | |
171 | -funsigned-bitfields -funsigned-char @gol | |
172 | -fwritable-strings -fshort-wchar} | |
173 | ||
174 | @item C++ Language Options | |
175 | @xref{C++ Dialect Options,,Options Controlling C++ Dialect}. | |
176 | @gccoptlist{ | |
177 | -fno-access-control -fcheck-new -fconserve-space @gol | |
178 | -fno-const-strings -fdollars-in-identifiers @gol | |
179 | -fno-elide-constructors @gol | |
180 | -fno-enforce-eh-specs -fexternal-templates @gol | |
181 | -falt-external-templates @gol | |
182 | -ffor-scope -fno-for-scope -fno-gnu-keywords @gol | |
183 | -fno-implicit-templates @gol | |
184 | -fno-implicit-inline-templates @gol | |
185 | -fno-implement-inlines -fms-extensions @gol | |
186 | -fno-nonansi-builtins -fno-operator-names @gol | |
187 | -fno-optional-diags -fpermissive @gol | |
188 | -frepo -fno-rtti -fstats -ftemplate-depth-@var{n} @gol | |
189 | -fuse-cxa-atexit -fvtable-gc -fno-weak -nostdinc++ @gol | |
190 | -fno-default-inline -Wctor-dtor-privacy @gol | |
191 | -Wnon-virtual-dtor -Wreorder @gol | |
192 | -Weffc++ -Wno-deprecated @gol | |
193 | -Wno-non-template-friend -Wold-style-cast @gol | |
194 | -Woverloaded-virtual -Wno-pmf-conversions @gol | |
195 | -Wsign-promo -Wsynth} | |
196 | ||
197 | @item Objective-C Language Options | |
198 | @xref{Objective-C Dialect Options,,Options Controlling Objective-C Dialect}. | |
199 | @gccoptlist{ | |
200 | -fconstant-string-class=@var{class-name} @gol | |
201 | -fgnu-runtime -fnext-runtime -gen-decls @gol | |
202 | -Wno-protocol -Wselector} | |
203 | ||
204 | @item Language Independent Options | |
205 | @xref{Language Independent Options,,Options to Control Diagnostic Messages Formatting}. | |
206 | @gccoptlist{ | |
207 | -fmessage-length=@var{n} @gol | |
208 | -fdiagnostics-show-location=@r{[}once@r{|}every-line@r{]}} | |
209 | ||
210 | @item Warning Options | |
211 | @xref{Warning Options,,Options to Request or Suppress Warnings}. | |
212 | @gccoptlist{ | |
213 | -fsyntax-only -pedantic -pedantic-errors @gol | |
214 | -w -W -Wall -Waggregate-return @gol | |
215 | -Wcast-align -Wcast-qual -Wchar-subscripts -Wcomment @gol | |
216 | -Wconversion -Wno-deprecated-declarations @gol | |
217 | -Wdisabled-optimization -Wdiv-by-zero -Werror @gol | |
218 | -Wfloat-equal -Wformat -Wformat=2 @gol | |
219 | -Wformat-nonliteral -Wformat-security @gol | |
220 | -Wimplicit -Wimplicit-int @gol | |
221 | -Wimplicit-function-declaration @gol | |
222 | -Werror-implicit-function-declaration @gol | |
223 | -Wimport -Winline @gol | |
224 | -Wlarger-than-@var{len} -Wlong-long @gol | |
225 | -Wmain -Wmissing-braces -Wmissing-declarations @gol | |
226 | -Wmissing-format-attribute -Wmissing-noreturn @gol | |
227 | -Wmultichar -Wno-format-extra-args -Wno-format-y2k @gol | |
228 | -Wno-import -Wpacked -Wpadded @gol | |
229 | -Wparentheses -Wpointer-arith -Wredundant-decls @gol | |
230 | -Wreturn-type -Wsequence-point -Wshadow @gol | |
231 | -Wsign-compare -Wswitch -Wsystem-headers @gol | |
232 | -Wtrigraphs -Wundef -Wuninitialized @gol | |
233 | -Wunknown-pragmas -Wunreachable-code @gol | |
234 | -Wunused -Wunused-function -Wunused-label -Wunused-parameter @gol | |
235 | -Wunused-value -Wunused-variable -Wwrite-strings} | |
236 | ||
237 | @item C-only Warning Options | |
238 | @gccoptlist{ | |
239 | -Wbad-function-cast -Wmissing-prototypes -Wnested-externs @gol | |
240 | -Wstrict-prototypes -Wtraditional} | |
241 | ||
242 | @item Debugging Options | |
243 | @xref{Debugging Options,,Options for Debugging Your Program or GCC}. | |
244 | @gccoptlist{ | |
245 | -d@var{letters} -dumpspecs -dumpmachine -dumpversion @gol | |
246 | -fdump-unnumbered -fdump-translation-unit@r{[}-@var{n}@r{]} @gol | |
247 | -fdump-class-hierarchy@r{[}-@var{n}@r{]} @gol | |
248 | -fdump-tree-original@r{[}-@var{n}@r{]} -fdump-tree-optimized@r{[}-@var{n}@r{]} @gol | |
249 | -fdump-tree-inlined@r{[}-@var{n}@r{]} @gol | |
250 | -fmem-report -fpretend-float @gol | |
251 | -fprofile-arcs -ftest-coverage -ftime-report @gol | |
252 | -g -g@var{level} -gcoff -gdwarf -gdwarf-1 -gdwarf-1+ -gdwarf-2 @gol | |
253 | -ggdb -gstabs -gstabs+ -gvms -gxcoff -gxcoff+ @gol | |
254 | -p -pg -print-file-name=@var{library} -print-libgcc-file-name @gol | |
255 | -print-multi-directory -print-multi-lib @gol | |
256 | -print-prog-name=@var{program} -print-search-dirs -Q @gol | |
257 | -save-temps -time} | |
258 | ||
259 | @item Optimization Options | |
260 | @xref{Optimize Options,,Options that Control Optimization}. | |
261 | @gccoptlist{ | |
262 | -falign-functions=@var{n} -falign-jumps=@var{n} @gol | |
263 | -falign-labels=@var{n} -falign-loops=@var{n} @gol | |
264 | -fbranch-probabilities -fcaller-saves -fcprop-registers @gol | |
265 | -fcse-follow-jumps -fcse-skip-blocks -fdata-sections @gol | |
266 | -fdelayed-branch -fdelete-null-pointer-checks @gol | |
267 | -fexpensive-optimizations -ffast-math -ffloat-store @gol | |
268 | -fforce-addr -fforce-mem -ffunction-sections @gol | |
269 | -fgcse -fgcse-lm -fgcse-sm @gol | |
270 | -finline-functions -finline-limit=@var{n} -fkeep-inline-functions @gol | |
271 | -fkeep-static-consts -fmerge-constants -fmerge-all-constants @gol | |
272 | -fmove-all-movables -fno-default-inline -fno-defer-pop @gol | |
273 | -fno-function-cse -fno-guess-branch-probability @gol | |
274 | -fno-inline -fno-math-errno -fno-peephole -fno-peephole2 @gol | |
275 | -funsafe-math-optimizations -fno-trapping-math @gol | |
276 | -fomit-frame-pointer -foptimize-register-move @gol | |
277 | -foptimize-sibling-calls -fprefetch-loop-arrays @gol | |
278 | -freduce-all-givs -fregmove -frename-registers @gol | |
279 | -frerun-cse-after-loop -frerun-loop-opt @gol | |
280 | -fschedule-insns -fschedule-insns2 @gol | |
281 | -fsingle-precision-constant -fssa -fssa-ccp -fssa-dce @gol | |
282 | -fstrength-reduce -fstrict-aliasing -fthread-jumps -ftrapv @gol | |
283 | -funroll-all-loops -funroll-loops @gol | |
284 | --param @var{name}=@var{value} | |
285 | -O -O0 -O1 -O2 -O3 -Os} | |
286 | ||
287 | @item Preprocessor Options | |
288 | @xref{Preprocessor Options,,Options Controlling the Preprocessor}. | |
289 | @gccoptlist{ | |
290 | -$ -A@var{question}=@var{answer} -A-@var{question}@r{[}=@var{answer}@r{]} @gol | |
291 | -C -dD -dI -dM -dN @gol | |
292 | -D@var{macro}@r{[}=@var{defn}@r{]} -E -H @gol | |
293 | -idirafter @var{dir} @gol | |
294 | -include @var{file} -imacros @var{file} @gol | |
295 | -iprefix @var{file} -iwithprefix @var{dir} @gol | |
296 | -iwithprefixbefore @var{dir} -isystem @var{dir} @gol | |
297 | -M -MM -MF -MG -MP -MQ -MT -nostdinc -P -remap @gol | |
298 | -trigraphs -undef -U@var{macro} -Wp,@var{option}} | |
299 | ||
300 | @item Assembler Option | |
301 | @xref{Assembler Options,,Passing Options to the Assembler}. | |
302 | @gccoptlist{ | |
303 | -Wa,@var{option}} | |
304 | ||
305 | @item Linker Options | |
306 | @xref{Link Options,,Options for Linking}. | |
307 | @gccoptlist{ | |
308 | @var{object-file-name} -l@var{library} @gol | |
309 | -nostartfiles -nodefaultlibs -nostdlib @gol | |
310 | -s -static -static-libgcc -shared -shared-libgcc -symbolic @gol | |
311 | -Wl,@var{option} -Xlinker @var{option} @gol | |
312 | -u @var{symbol}} | |
313 | ||
314 | @item Directory Options | |
315 | @xref{Directory Options,,Options for Directory Search}. | |
316 | @gccoptlist{ | |
317 | -B@var{prefix} -I@var{dir} -I- -L@var{dir} -specs=@var{file}} | |
318 | ||
319 | @item Target Options | |
320 | @c I wrote this xref this way to avoid overfull hbox. -- rms | |
321 | @xref{Target Options}. | |
322 | @gccoptlist{ | |
323 | -b @var{machine} -V @var{version}} | |
324 | ||
325 | @item Machine Dependent Options | |
326 | @xref{Submodel Options,,Hardware Models and Configurations}. | |
327 | ||
328 | @emph{M680x0 Options} | |
329 | @gccoptlist{ | |
330 | -m68000 -m68020 -m68020-40 -m68020-60 -m68030 -m68040 @gol | |
331 | -m68060 -mcpu32 -m5200 -m68881 -mbitfield -mc68000 -mc68020 @gol | |
332 | -mfpa -mnobitfield -mrtd -mshort -msoft-float -mpcrel @gol | |
333 | -malign-int -mstrict-align} | |
334 | ||
335 | @emph{M68hc1x Options} | |
336 | @gccoptlist{ | |
337 | -m6811 -m6812 -m68hc11 -m68hc12 @gol | |
338 | -mauto-incdec -mshort -msoft-reg-count=@var{count}} | |
339 | ||
340 | @emph{VAX Options} | |
341 | @gccoptlist{ | |
342 | -mg -mgnu -munix} | |
343 | ||
344 | @emph{SPARC Options} | |
345 | @gccoptlist{ | |
346 | -mcpu=@var{cpu-type} @gol | |
347 | -mtune=@var{cpu-type} @gol | |
348 | -mcmodel=@var{code-model} @gol | |
349 | -m32 -m64 @gol | |
350 | -mapp-regs -mbroken-saverestore -mcypress @gol | |
351 | -mepilogue -mfaster-structs -mflat @gol | |
352 | -mfpu -mhard-float -mhard-quad-float @gol | |
353 | -mimpure-text -mlive-g0 -mno-app-regs @gol | |
354 | -mno-epilogue -mno-faster-structs -mno-flat -mno-fpu @gol | |
355 | -mno-impure-text -mno-stack-bias -mno-unaligned-doubles @gol | |
356 | -msoft-float -msoft-quad-float -msparclite -mstack-bias @gol | |
357 | -msupersparc -munaligned-doubles -mv8} | |
358 | ||
359 | @emph{Convex Options} | |
360 | @gccoptlist{ | |
361 | -mc1 -mc2 -mc32 -mc34 -mc38 @gol | |
362 | -margcount -mnoargcount @gol | |
363 | -mlong32 -mlong64 @gol | |
364 | -mvolatile-cache -mvolatile-nocache} | |
365 | ||
366 | @emph{AMD29K Options} | |
367 | @gccoptlist{ | |
368 | -m29000 -m29050 -mbw -mnbw -mdw -mndw @gol | |
369 | -mlarge -mnormal -msmall @gol | |
370 | -mkernel-registers -mno-reuse-arg-regs @gol | |
371 | -mno-stack-check -mno-storem-bug @gol | |
372 | -mreuse-arg-regs -msoft-float -mstack-check @gol | |
373 | -mstorem-bug -muser-registers} | |
374 | ||
375 | @emph{ARM Options} | |
376 | @gccoptlist{ | |
377 | -mapcs-frame -mno-apcs-frame @gol | |
378 | -mapcs-26 -mapcs-32 @gol | |
379 | -mapcs-stack-check -mno-apcs-stack-check @gol | |
380 | -mapcs-float -mno-apcs-float @gol | |
381 | -mapcs-reentrant -mno-apcs-reentrant @gol | |
382 | -msched-prolog -mno-sched-prolog @gol | |
383 | -mlittle-endian -mbig-endian -mwords-little-endian @gol | |
384 | -malignment-traps -mno-alignment-traps @gol | |
385 | -msoft-float -mhard-float -mfpe @gol | |
386 | -mthumb-interwork -mno-thumb-interwork @gol | |
387 | -mcpu=@var{name} -march=@var{name} -mfpe=@var{name} @gol | |
388 | -mstructure-size-boundary=@var{n} @gol | |
389 | -mbsd -mxopen -mno-symrename @gol | |
390 | -mabort-on-noreturn @gol | |
391 | -mlong-calls -mno-long-calls @gol | |
392 | -msingle-pic-base -mno-single-pic-base @gol | |
393 | -mpic-register=@var{reg} @gol | |
394 | -mnop-fun-dllimport @gol | |
395 | -mpoke-function-name @gol | |
396 | -mthumb -marm @gol | |
397 | -mtpcs-frame -mtpcs-leaf-frame @gol | |
398 | -mcaller-super-interworking -mcallee-super-interworking } | |
399 | ||
400 | @emph{MN10200 Options} | |
401 | @gccoptlist{ | |
402 | -mrelax} | |
403 | ||
404 | @emph{MN10300 Options} | |
405 | @gccoptlist{ | |
406 | -mmult-bug -mno-mult-bug @gol | |
407 | -mam33 -mno-am33 @gol | |
408 | -mno-crt0 -mrelax} | |
409 | ||
410 | @emph{M32R/D Options} | |
411 | @gccoptlist{ | |
412 | -m32rx -m32r -mcode-model=@var{model-type} -msdata=@var{sdata-type} @gol | |
413 | -G @var{num}} | |
414 | ||
415 | @emph{M88K Options} | |
416 | @gccoptlist{ | |
417 | -m88000 -m88100 -m88110 -mbig-pic @gol | |
418 | -mcheck-zero-division -mhandle-large-shift @gol | |
419 | -midentify-revision -mno-check-zero-division @gol | |
420 | -mno-ocs-debug-info -mno-ocs-frame-position @gol | |
421 | -mno-optimize-arg-area -mno-serialize-volatile @gol | |
422 | -mno-underscores -mocs-debug-info @gol | |
423 | -mocs-frame-position -moptimize-arg-area @gol | |
424 | -mserialize-volatile -mshort-data-@var{num} -msvr3 @gol | |
425 | -msvr4 -mtrap-large-shift -muse-div-instruction @gol | |
426 | -mversion-03.00 -mwarn-passed-structs} | |
427 | ||
428 | @emph{RS/6000 and PowerPC Options} | |
429 | @gccoptlist{ | |
430 | -mcpu=@var{cpu-type} @gol | |
431 | -mtune=@var{cpu-type} @gol | |
432 | -mpower -mno-power -mpower2 -mno-power2 @gol | |
433 | -mpowerpc -mpowerpc64 -mno-powerpc @gol | |
434 | -maltivec -mno-altivec @gol | |
435 | -mpowerpc-gpopt -mno-powerpc-gpopt @gol | |
436 | -mpowerpc-gfxopt -mno-powerpc-gfxopt @gol | |
437 | -mnew-mnemonics -mold-mnemonics @gol | |
438 | -mfull-toc -mminimal-toc -mno-fp-in-toc -mno-sum-in-toc @gol | |
439 | -m64 -m32 -mxl-call -mno-xl-call -mpe @gol | |
440 | -msoft-float -mhard-float -mmultiple -mno-multiple @gol | |
441 | -mstring -mno-string -mupdate -mno-update @gol | |
442 | -mfused-madd -mno-fused-madd -mbit-align -mno-bit-align @gol | |
443 | -mstrict-align -mno-strict-align -mrelocatable @gol | |
444 | -mno-relocatable -mrelocatable-lib -mno-relocatable-lib @gol | |
445 | -mtoc -mno-toc -mlittle -mlittle-endian -mbig -mbig-endian @gol | |
446 | -mcall-aix -mcall-sysv -mcall-netbsd @gol | |
447 | -maix-struct-return -msvr4-struct-return | |
448 | -mabi=altivec @gol | |
449 | -mprototype -mno-prototype @gol | |
450 | -msim -mmvme -mads -myellowknife -memb -msdata @gol | |
451 | -msdata=@var{opt} -mvxworks -G @var{num} -pthread} | |
452 | ||
453 | @emph{RT Options} | |
454 | @gccoptlist{ | |
455 | -mcall-lib-mul -mfp-arg-in-fpregs -mfp-arg-in-gregs @gol | |
456 | -mfull-fp-blocks -mhc-struct-return -min-line-mul @gol | |
457 | -mminimum-fp-blocks -mnohc-struct-return} | |
458 | ||
459 | @emph{MIPS Options} | |
460 | @gccoptlist{ | |
461 | -mabicalls -march=@var{cpu-type} -mtune=@var{cpu=type} @gol | |
462 | -mcpu=@var{cpu-type} -membedded-data -muninit-const-in-rodata @gol | |
463 | -membedded-pic -mfp32 -mfp64 -mfused-madd -mno-fused-madd @gol | |
464 | -mgas -mgp32 -mgp64 @gol | |
465 | -mgpopt -mhalf-pic -mhard-float -mint64 -mips1 @gol | |
466 | -mips2 -mips3 -mips4 -mlong64 -mlong32 -mlong-calls -mmemcpy @gol | |
467 | -mmips-as -mmips-tfile -mno-abicalls @gol | |
468 | -mno-embedded-data -mno-uninit-const-in-rodata @gol | |
469 | -mno-embedded-pic -mno-gpopt -mno-long-calls @gol | |
470 | -mno-memcpy -mno-mips-tfile -mno-rnames -mno-stats @gol | |
471 | -mrnames -msoft-float @gol | |
472 | -m4650 -msingle-float -mmad @gol | |
473 | -mstats -EL -EB -G @var{num} -nocpp @gol | |
474 | -mabi=32 -mabi=n32 -mabi=64 -mabi=eabi @gol | |
475 | -mfix7000 -mno-crt0 -mflush-func=@var{func} -mno-flush-func} | |
476 | ||
477 | @emph{i386 and x86-64 Options} | |
478 | @gccoptlist{ | |
479 | -mcpu=@var{cpu-type} -march=@var{cpu-type} -mfpmath=@var{unit} @gol | |
480 | -masm=@var{dialect} -mno-fancy-math-387 @gol | |
481 | -mno-fp-ret-in-387 -msoft-float -msvr3-shlib @gol | |
482 | -mno-wide-multiply -mrtd -malign-double @gol | |
483 | -mpreferred-stack-boundary=@var{num} @gol | |
484 | -mmmx -msse -msse2 -msse-math -m3dnow @gol | |
485 | -mthreads -mno-align-stringops -minline-all-stringops @gol | |
486 | -mpush-args -maccumulate-outgoing-args -m128bit-long-double @gol | |
487 | -m96bit-long-double -mregparm=@var{num} -momit-leaf-frame-pointer @gol | |
488 | -mno-red-zone@gol | |
489 | -m32 -m64} | |
490 | ||
491 | @emph{HPPA Options} | |
492 | @gccoptlist{ | |
493 | -march=@var{architecture-type} @gol | |
494 | -mbig-switch -mdisable-fpregs -mdisable-indexing @gol | |
495 | -mfast-indirect-calls -mgas -mjump-in-delay @gol | |
496 | -mlong-load-store -mno-big-switch -mno-disable-fpregs @gol | |
497 | -mno-disable-indexing -mno-fast-indirect-calls -mno-gas @gol | |
498 | -mno-jump-in-delay -mno-long-load-store @gol | |
499 | -mno-portable-runtime -mno-soft-float @gol | |
500 | -mno-space-regs -msoft-float -mpa-risc-1-0 @gol | |
501 | -mpa-risc-1-1 -mpa-risc-2-0 -mportable-runtime @gol | |
502 | -mschedule=@var{cpu-type} -mspace-regs} | |
503 | ||
504 | @emph{Intel 960 Options} | |
505 | @gccoptlist{ | |
506 | -m@var{cpu-type} -masm-compat -mclean-linkage @gol | |
507 | -mcode-align -mcomplex-addr -mleaf-procedures @gol | |
508 | -mic-compat -mic2.0-compat -mic3.0-compat @gol | |
509 | -mintel-asm -mno-clean-linkage -mno-code-align @gol | |
510 | -mno-complex-addr -mno-leaf-procedures @gol | |
511 | -mno-old-align -mno-strict-align -mno-tail-call @gol | |
512 | -mnumerics -mold-align -msoft-float -mstrict-align @gol | |
513 | -mtail-call} | |
514 | ||
515 | @emph{DEC Alpha Options} | |
516 | @gccoptlist{ | |
517 | -mno-fp-regs -msoft-float -malpha-as -mgas @gol | |
518 | -mieee -mieee-with-inexact -mieee-conformant @gol | |
519 | -mfp-trap-mode=@var{mode} -mfp-rounding-mode=@var{mode} @gol | |
520 | -mtrap-precision=@var{mode} -mbuild-constants @gol | |
521 | -mcpu=@var{cpu-type} -mtune=@var{cpu-type} @gol | |
522 | -mbwx -mmax -mfix -mcix @gol | |
523 | -mfloat-vax -mfloat-ieee @gol | |
524 | -mexplicit-relocs -msmall-data -mlarge-data @gol | |
525 | -mmemory-latency=@var{time}} | |
526 | ||
527 | @emph{DEC Alpha/VMS Options} | |
528 | @gccoptlist{ | |
529 | -mvms-return-codes} | |
530 | ||
531 | @emph{Clipper Options} | |
532 | @gccoptlist{ | |
533 | -mc300 -mc400} | |
534 | ||
535 | @emph{H8/300 Options} | |
536 | @gccoptlist{ | |
537 | -mrelax -mh -ms -mint32 -malign-300} | |
538 | ||
539 | @emph{SH Options} | |
540 | @gccoptlist{ | |
541 | -m1 -m2 -m3 -m3e @gol | |
542 | -m4-nofpu -m4-single-only -m4-single -m4 @gol | |
543 | -m5-64media -m5-64media-nofpu @gol | |
544 | -m5-32media -m5-32media-nofpu @gol | |
545 | -m5-compact -m5-compact-nofpu @gol | |
546 | -mb -ml -mdalign -mrelax @gol | |
547 | -mbigtable -mfmovd -mhitachi -mnomacsave @gol | |
548 | -mieee -misize -mpadstruct -mspace @gol | |
549 | -mprefergot -musermode} | |
550 | ||
551 | @emph{System V Options} | |
552 | @gccoptlist{ | |
553 | -Qy -Qn -YP,@var{paths} -Ym,@var{dir}} | |
554 | ||
555 | @emph{ARC Options} | |
556 | @gccoptlist{ | |
557 | -EB -EL @gol | |
558 | -mmangle-cpu -mcpu=@var{cpu} -mtext=@var{text-section} @gol | |
559 | -mdata=@var{data-section} -mrodata=@var{readonly-data-section}} | |
560 | ||
561 | @emph{TMS320C3x/C4x Options} | |
562 | @gccoptlist{ | |
563 | -mcpu=@var{cpu} -mbig -msmall -mregparm -mmemparm @gol | |
564 | -mfast-fix -mmpyi -mbk -mti -mdp-isr-reload @gol | |
565 | -mrpts=@var{count} -mrptb -mdb -mloop-unsigned @gol | |
566 | -mparallel-insns -mparallel-mpy -mpreserve-float} | |
567 | ||
568 | @emph{V850 Options} | |
569 | @gccoptlist{ | |
570 | -mlong-calls -mno-long-calls -mep -mno-ep @gol | |
571 | -mprolog-function -mno-prolog-function -mspace @gol | |
572 | -mtda=@var{n} -msda=@var{n} -mzda=@var{n} @gol | |
573 | -mv850 -mbig-switch} | |
574 | ||
575 | @emph{NS32K Options} | |
576 | @gccoptlist{ | |
577 | -m32032 -m32332 -m32532 -m32081 -m32381 @gol | |
578 | -mmult-add -mnomult-add -msoft-float -mrtd -mnortd @gol | |
579 | -mregparam -mnoregparam -msb -mnosb @gol | |
580 | -mbitfield -mnobitfield -mhimem -mnohimem} | |
581 | ||
582 | @emph{AVR Options} | |
583 | @gccoptlist{ | |
584 | -mmcu=@var{mcu} -msize -minit-stack=@var{n} -mno-interrupts @gol | |
585 | -mcall-prologues -mno-tablejump -mtiny-stack} | |
586 | ||
587 | @emph{MCore Options} | |
588 | @gccoptlist{ | |
589 | -mhardlit -mno-hardlit -mdiv -mno-div -mrelax-immediates @gol | |
590 | -mno-relax-immediates -mwide-bitfields -mno-wide-bitfields @gol | |
591 | -m4byte-functions -mno-4byte-functions -mcallgraph-data @gol | |
592 | -mno-callgraph-data -mslow-bytes -mno-slow-bytes -mno-lsim @gol | |
593 | -mlittle-endian -mbig-endian -m210 -m340 -mstack-increment} | |
594 | ||
595 | @emph{MMIX Options} | |
596 | @gccoptlist{ | |
597 | -mlibfuncs -mno-libfuncs -mepsilon -mno-epsilon -mabi=gnu @gol | |
598 | -mabi=mmixware -mzero-extend -mknuthdiv -mtoplevel-symbols @gol | |
599 | -melf -mbranch-predict -mno-branch-predict -mbase-addresses @gol | |
600 | -mno-base-addresses} | |
601 | ||
602 | @emph{IA-64 Options} | |
603 | @gccoptlist{ | |
604 | -mbig-endian -mlittle-endian -mgnu-as -mgnu-ld -mno-pic @gol | |
605 | -mvolatile-asm-stop -mb-step -mregister-names -mno-sdata @gol | |
606 | -mconstant-gp -mauto-pic -minline-divide-min-latency @gol | |
607 | -minline-divide-max-throughput -mno-dwarf2-asm @gol | |
608 | -mfixed-range=@var{register-range}} | |
609 | ||
610 | @emph{D30V Options} | |
611 | @gccoptlist{ | |
612 | -mextmem -mextmemory -monchip -mno-asm-optimize -masm-optimize @gol | |
613 | -mbranch-cost=@var{n} -mcond-exec=@var{n}} | |
614 | ||
615 | @emph{S/390 and zSeries Options} | |
616 | @gccoptlist{ | |
617 | -mhard-float -msoft-float -mbackchain -mno-backchain @gol | |
618 | -msmall-exec -mno-small-exec -mmvcle -mno-mvcle @gol | |
619 | -m64 -m31 -mdebug -mno-debug} | |
620 | ||
621 | @emph{CRIS Options} | |
622 | @gccoptlist{ | |
623 | -mcpu=@var{cpu} -march=@var{cpu} -mtune=@var{cpu} @gol | |
624 | -mmax-stack-frame=@var{n} -melinux-stacksize=@var{n} @gol | |
625 | -metrax4 -metrax100 -mpdebug -mcc-init -mno-side-effects @gol | |
626 | -mstack-align -mdata-align -mconst-align @gol | |
627 | -m32-bit -m16-bit -m8-bit -mno-prologue-epilogue -mno-gotplt @gol | |
628 | -melf -maout -melinux -mlinux -sim -sim2} | |
629 | ||
630 | @emph{PDP-11 Options} | |
631 | @gccoptlist{ | |
632 | -mfpu -msoft-float -mac0 -mno-ac0 -m40 -m45 -m10 @gol | |
633 | -mbcopy -mbcopy-builtin -mint32 -mno-int16 @gol | |
634 | -mint16 -mno-int32 -mfloat32 -mno-float64 @gol | |
635 | -mfloat64 -mno-float32 -mabshi -mno-abshi @gol | |
636 | -mbranch-expensive -mbranch-cheap @gol | |
637 | -msplit -mno-split -munix-asm -mdec-asm} | |
638 | ||
639 | @emph{Xstormy16 Options} | |
640 | @gccoptlist{ | |
641 | -msim} | |
642 | ||
643 | @emph{Xtensa Options} | |
644 | @gccoptlist{ | |
645 | -mbig-endian -mlittle-endian @gol | |
646 | -mdensity -mno-density @gol | |
647 | -mmac16 -mno-mac16 @gol | |
648 | -mmul16 -mno-mul16 @gol | |
649 | -mmul32 -mno-mul32 @gol | |
650 | -mnsa -mno-nsa @gol | |
651 | -mminmax -mno-minmax @gol | |
652 | -msext -mno-sext @gol | |
653 | -mbooleans -mno-booleans @gol | |
654 | -mhard-float -msoft-float @gol | |
655 | -mfused-madd -mno-fused-madd @gol | |
656 | -mserialize-volatile -mno-serialize-volatile @gol | |
657 | -mtext-section-literals -mno-text-section-literals @gol | |
658 | -mtarget-align -mno-target-align @gol | |
659 | -mlongcalls -mno-longcalls} | |
660 | ||
661 | @item Code Generation Options | |
662 | @xref{Code Gen Options,,Options for Code Generation Conventions}. | |
663 | @gccoptlist{ | |
664 | -fcall-saved-@var{reg} -fcall-used-@var{reg} @gol | |
665 | -ffixed-@var{reg} -fexceptions @gol | |
666 | -fnon-call-exceptions -funwind-tables @gol | |
667 | -fasynchronous-unwind-tables @gol | |
668 | -finhibit-size-directive -finstrument-functions @gol | |
669 | -fno-common -fno-ident -fno-gnu-linker @gol | |
670 | -fpcc-struct-return -fpic -fPIC @gol | |
671 | -freg-struct-return -fshared-data -fshort-enums @gol | |
672 | -fshort-double -fvolatile @gol | |
673 | -fvolatile-global -fvolatile-static @gol | |
674 | -fverbose-asm -fpack-struct -fstack-check @gol | |
675 | -fstack-limit-register=@var{reg} -fstack-limit-symbol=@var{sym} @gol | |
676 | -fargument-alias -fargument-noalias @gol | |
677 | -fargument-noalias-global -fleading-underscore} | |
678 | @end table | |
679 | ||
680 | @menu | |
681 | * Overall Options:: Controlling the kind of output: | |
682 | an executable, object files, assembler files, | |
683 | or preprocessed source. | |
684 | * C Dialect Options:: Controlling the variant of C language compiled. | |
685 | * C++ Dialect Options:: Variations on C++. | |
686 | * Objective-C Dialect Options:: Variations on Objective-C. | |
687 | * Language Independent Options:: Controlling how diagnostics should be | |
688 | formatted. | |
689 | * Warning Options:: How picky should the compiler be? | |
690 | * Debugging Options:: Symbol tables, measurements, and debugging dumps. | |
691 | * Optimize Options:: How much optimization? | |
692 | * Preprocessor Options:: Controlling header files and macro definitions. | |
693 | Also, getting dependency information for Make. | |
694 | * Assembler Options:: Passing options to the assembler. | |
695 | * Link Options:: Specifying libraries and so on. | |
696 | * Directory Options:: Where to find header files and libraries. | |
697 | Where to find the compiler executable files. | |
698 | * Spec Files:: How to pass switches to sub-processes. | |
699 | * Target Options:: Running a cross-compiler, or an old version of GCC. | |
700 | @end menu | |
701 | ||
702 | @node Overall Options | |
703 | @section Options Controlling the Kind of Output | |
704 | ||
705 | Compilation can involve up to four stages: preprocessing, compilation | |
706 | proper, assembly and linking, always in that order. The first three | |
707 | stages apply to an individual source file, and end by producing an | |
708 | object file; linking combines all the object files (those newly | |
709 | compiled, and those specified as input) into an executable file. | |
710 | ||
711 | @cindex file name suffix | |
712 | For any given input file, the file name suffix determines what kind of | |
713 | compilation is done: | |
714 | ||
715 | @table @gcctabopt | |
716 | @item @var{file}.c | |
717 | C source code which must be preprocessed. | |
718 | ||
719 | @item @var{file}.i | |
720 | C source code which should not be preprocessed. | |
721 | ||
722 | @item @var{file}.ii | |
723 | C++ source code which should not be preprocessed. | |
724 | ||
725 | @item @var{file}.m | |
726 | Objective-C source code. Note that you must link with the library | |
727 | @file{libobjc.a} to make an Objective-C program work. | |
728 | ||
729 | @item @var{file}.mi | |
730 | Objective-C source code which should not be preprocessed. | |
731 | ||
732 | @item @var{file}.h | |
733 | C header file (not to be compiled or linked). | |
734 | ||
735 | @item @var{file}.cc | |
736 | @itemx @var{file}.cp | |
737 | @itemx @var{file}.cxx | |
738 | @itemx @var{file}.cpp | |
739 | @itemx @var{file}.c++ | |
740 | @itemx @var{file}.C | |
741 | C++ source code which must be preprocessed. Note that in @samp{.cxx}, | |
742 | the last two letters must both be literally @samp{x}. Likewise, | |
743 | @samp{.C} refers to a literal capital C@. | |
744 | ||
745 | @item @var{file}.f | |
746 | @itemx @var{file}.for | |
747 | @itemx @var{file}.FOR | |
748 | Fortran source code which should not be preprocessed. | |
749 | ||
750 | @item @var{file}.F | |
751 | @itemx @var{file}.fpp | |
752 | @itemx @var{file}.FPP | |
753 | Fortran source code which must be preprocessed (with the traditional | |
754 | preprocessor). | |
755 | ||
756 | @item @var{file}.r | |
757 | Fortran source code which must be preprocessed with a RATFOR | |
758 | preprocessor (not included with GCC)@. | |
759 | ||
760 | @xref{Overall Options,,Options Controlling the Kind of Output, g77, | |
761 | Using and Porting GNU Fortran}, for more details of the handling of | |
762 | Fortran input files. | |
763 | ||
764 | @c FIXME: Descriptions of Java file types. | |
765 | @c @var{file}.java | |
766 | @c @var{file}.class | |
767 | @c @var{file}.zip | |
768 | @c @var{file}.jar | |
769 | ||
770 | @item @var{file}.ads | |
771 | Ada source code file which contains a library unit declaration (a | |
772 | declaration of a package, subprogram, or generic, or a generic | |
773 | instantiation), or a library unit renaming declaration (a package, | |
774 | generic, or subprogram renaming declaration). Such files are also | |
775 | called @dfn{specs}. | |
776 | ||
777 | @itemx @var{file}.adb | |
778 | Ada source code file containing a library unit body (a subprogram or | |
779 | package body). Such files are also called @dfn{bodies}. | |
780 | ||
781 | @c GCC also knows about some suffixes for languages not yet included: | |
782 | @c Pascal: | |
783 | @c @var{file}.p | |
784 | @c @var{file}.pas | |
785 | ||
786 | @item @var{file}.ch | |
787 | @itemx @var{file}.chi | |
788 | CHILL source code (preprocessed with the traditional preprocessor). | |
789 | ||
790 | @item @var{file}.s | |
791 | Assembler code. | |
792 | ||
793 | @item @var{file}.S | |
794 | Assembler code which must be preprocessed. | |
795 | ||
796 | @item @var{other} | |
797 | An object file to be fed straight into linking. | |
798 | Any file name with no recognized suffix is treated this way. | |
799 | @end table | |
800 | ||
801 | @opindex x | |
802 | You can specify the input language explicitly with the @option{-x} option: | |
803 | ||
804 | @table @gcctabopt | |
805 | @item -x @var{language} | |
806 | Specify explicitly the @var{language} for the following input files | |
807 | (rather than letting the compiler choose a default based on the file | |
808 | name suffix). This option applies to all following input files until | |
809 | the next @option{-x} option. Possible values for @var{language} are: | |
810 | @example | |
811 | c c-header cpp-output | |
812 | c++ c++-cpp-output | |
813 | objective-c objc-cpp-output | |
814 | assembler assembler-with-cpp | |
815 | ada | |
816 | chill | |
817 | f77 f77-cpp-input ratfor | |
818 | java | |
819 | @end example | |
820 | ||
821 | @item -x none | |
822 | Turn off any specification of a language, so that subsequent files are | |
823 | handled according to their file name suffixes (as they are if @option{-x} | |
824 | has not been used at all). | |
825 | ||
826 | @item -pass-exit-codes | |
827 | @opindex pass-exit-codes | |
828 | Normally the @command{gcc} program will exit with the code of 1 if any | |
829 | phase of the compiler returns a non-success return code. If you specify | |
830 | @option{-pass-exit-codes}, the @command{gcc} program will instead return with | |
831 | numerically highest error produced by any phase that returned an error | |
832 | indication. | |
833 | @end table | |
834 | ||
835 | If you only want some of the stages of compilation, you can use | |
836 | @option{-x} (or filename suffixes) to tell @command{gcc} where to start, and | |
837 | one of the options @option{-c}, @option{-S}, or @option{-E} to say where | |
838 | @command{gcc} is to stop. Note that some combinations (for example, | |
839 | @samp{-x cpp-output -E}) instruct @command{gcc} to do nothing at all. | |
840 | ||
841 | @table @gcctabopt | |
842 | @item -c | |
843 | @opindex c | |
844 | Compile or assemble the source files, but do not link. The linking | |
845 | stage simply is not done. The ultimate output is in the form of an | |
846 | object file for each source file. | |
847 | ||
848 | By default, the object file name for a source file is made by replacing | |
849 | the suffix @samp{.c}, @samp{.i}, @samp{.s}, etc., with @samp{.o}. | |
850 | ||
851 | Unrecognized input files, not requiring compilation or assembly, are | |
852 | ignored. | |
853 | ||
854 | @item -S | |
855 | @opindex S | |
856 | Stop after the stage of compilation proper; do not assemble. The output | |
857 | is in the form of an assembler code file for each non-assembler input | |
858 | file specified. | |
859 | ||
860 | By default, the assembler file name for a source file is made by | |
861 | replacing the suffix @samp{.c}, @samp{.i}, etc., with @samp{.s}. | |
862 | ||
863 | Input files that don't require compilation are ignored. | |
864 | ||
865 | @item -E | |
866 | @opindex E | |
867 | Stop after the preprocessing stage; do not run the compiler proper. The | |
868 | output is in the form of preprocessed source code, which is sent to the | |
869 | standard output. | |
870 | ||
871 | Input files which don't require preprocessing are ignored. | |
872 | ||
873 | @cindex output file option | |
874 | @item -o @var{file} | |
875 | @opindex o | |
876 | Place output in file @var{file}. This applies regardless to whatever | |
877 | sort of output is being produced, whether it be an executable file, | |
878 | an object file, an assembler file or preprocessed C code. | |
879 | ||
880 | Since only one output file can be specified, it does not make sense to | |
881 | use @option{-o} when compiling more than one input file, unless you are | |
882 | producing an executable file as output. | |
883 | ||
884 | If @option{-o} is not specified, the default is to put an executable file | |
885 | in @file{a.out}, the object file for @file{@var{source}.@var{suffix}} in | |
886 | @file{@var{source}.o}, its assembler file in @file{@var{source}.s}, and | |
887 | all preprocessed C source on standard output. | |
888 | ||
889 | @item -v | |
890 | @opindex v | |
891 | Print (on standard error output) the commands executed to run the stages | |
892 | of compilation. Also print the version number of the compiler driver | |
893 | program and of the preprocessor and the compiler proper. | |
894 | ||
895 | @item -pipe | |
896 | @opindex pipe | |
897 | Use pipes rather than temporary files for communication between the | |
898 | various stages of compilation. This fails to work on some systems where | |
899 | the assembler is unable to read from a pipe; but the GNU assembler has | |
900 | no trouble. | |
901 | ||
902 | @item --help | |
903 | @opindex help | |
904 | Print (on the standard output) a description of the command line options | |
905 | understood by @command{gcc}. If the @option{-v} option is also specified | |
906 | then @option{--help} will also be passed on to the various processes | |
907 | invoked by @command{gcc}, so that they can display the command line options | |
908 | they accept. If the @option{-W} option is also specified then command | |
909 | line options which have no documentation associated with them will also | |
910 | be displayed. | |
911 | ||
912 | @item --target-help | |
913 | @opindex target-help | |
914 | Print (on the standard output) a description of target specific command | |
915 | line options for each tool. | |
916 | @end table | |
917 | ||
918 | @node Invoking G++ | |
919 | @section Compiling C++ Programs | |
920 | ||
921 | @cindex suffixes for C++ source | |
922 | @cindex C++ source file suffixes | |
923 | C++ source files conventionally use one of the suffixes @samp{.C}, | |
924 | @samp{.cc}, @samp{.cpp}, @samp{.c++}, @samp{.cp}, or @samp{.cxx}; | |
925 | preprocessed C++ files use the suffix @samp{.ii}. GCC recognizes | |
926 | files with these names and compiles them as C++ programs even if you | |
927 | call the compiler the same way as for compiling C programs (usually with | |
928 | the name @command{gcc}). | |
929 | ||
930 | @findex g++ | |
931 | @findex c++ | |
932 | However, C++ programs often require class libraries as well as a | |
933 | compiler that understands the C++ language---and under some | |
934 | circumstances, you might want to compile programs from standard input, | |
935 | or otherwise without a suffix that flags them as C++ programs. | |
936 | @command{g++} is a program that calls GCC with the default language | |
937 | set to C++, and automatically specifies linking against the C++ | |
938 | library. On many systems, @command{g++} is also | |
939 | installed with the name @command{c++}. | |
940 | ||
941 | @cindex invoking @command{g++} | |
942 | When you compile C++ programs, you may specify many of the same | |
943 | command-line options that you use for compiling programs in any | |
944 | language; or command-line options meaningful for C and related | |
945 | languages; or options that are meaningful only for C++ programs. | |
946 | @xref{C Dialect Options,,Options Controlling C Dialect}, for | |
947 | explanations of options for languages related to C@. | |
948 | @xref{C++ Dialect Options,,Options Controlling C++ Dialect}, for | |
949 | explanations of options that are meaningful only for C++ programs. | |
950 | ||
951 | @node C Dialect Options | |
952 | @section Options Controlling C Dialect | |
953 | @cindex dialect options | |
954 | @cindex language dialect options | |
955 | @cindex options, dialect | |
956 | ||
957 | The following options control the dialect of C (or languages derived | |
958 | from C, such as C++ and Objective-C) that the compiler accepts: | |
959 | ||
960 | @table @gcctabopt | |
961 | @cindex ANSI support | |
962 | @cindex ISO support | |
963 | @item -ansi | |
964 | @opindex ansi | |
965 | In C mode, support all ISO C89 programs. In C++ mode, | |
966 | remove GNU extensions that conflict with ISO C++. | |
967 | ||
968 | This turns off certain features of GCC that are incompatible with ISO | |
969 | C89 (when compiling C code), or of standard C++ (when compiling C++ code), | |
970 | such as the @code{asm} and @code{typeof} keywords, and | |
971 | predefined macros such as @code{unix} and @code{vax} that identify the | |
972 | type of system you are using. It also enables the undesirable and | |
973 | rarely used ISO trigraph feature. For the C compiler, | |
974 | it disables recognition of C++ style @samp{//} comments as well as | |
975 | the @code{inline} keyword. | |
976 | ||
977 | The alternate keywords @code{__asm__}, @code{__extension__}, | |
978 | @code{__inline__} and @code{__typeof__} continue to work despite | |
979 | @option{-ansi}. You would not want to use them in an ISO C program, of | |
980 | course, but it is useful to put them in header files that might be included | |
981 | in compilations done with @option{-ansi}. Alternate predefined macros | |
982 | such as @code{__unix__} and @code{__vax__} are also available, with or | |
983 | without @option{-ansi}. | |
984 | ||
985 | The @option{-ansi} option does not cause non-ISO programs to be | |
986 | rejected gratuitously. For that, @option{-pedantic} is required in | |
987 | addition to @option{-ansi}. @xref{Warning Options}. | |
988 | ||
989 | The macro @code{__STRICT_ANSI__} is predefined when the @option{-ansi} | |
990 | option is used. Some header files may notice this macro and refrain | |
991 | from declaring certain functions or defining certain macros that the | |
992 | ISO standard doesn't call for; this is to avoid interfering with any | |
993 | programs that might use these names for other things. | |
994 | ||
995 | Functions which would normally be built in but do not have semantics | |
996 | defined by ISO C (such as @code{alloca} and @code{ffs}) are not built-in | |
997 | functions with @option{-ansi} is used. @xref{Other Builtins,,Other | |
998 | built-in functions provided by GCC}, for details of the functions | |
999 | affected. | |
1000 | ||
1001 | @item -std= | |
1002 | @opindex std | |
1003 | Determine the language standard. This option is currently only | |
1004 | supported when compiling C@. A value for this option must be provided; | |
1005 | possible values are | |
1006 | ||
1007 | @table @samp | |
1008 | @item c89 | |
1009 | @itemx iso9899:1990 | |
1010 | ISO C89 (same as @option{-ansi}). | |
1011 | ||
1012 | @item iso9899:199409 | |
1013 | ISO C89 as modified in amendment 1. | |
1014 | ||
1015 | @item c99 | |
1016 | @itemx c9x | |
1017 | @itemx iso9899:1999 | |
1018 | @itemx iso9899:199x | |
1019 | ISO C99. Note that this standard is not yet fully supported; see | |
1020 | @w{@uref{http://gcc.gnu.org/c99status.html}} for more information. The | |
1021 | names @samp{c9x} and @samp{iso9899:199x} are deprecated. | |
1022 | ||
1023 | @item gnu89 | |
1024 | Default, ISO C89 plus GNU extensions (including some C99 features). | |
1025 | ||
1026 | @item gnu99 | |
1027 | @item gnu9x | |
1028 | ISO C99 plus GNU extensions. When ISO C99 is fully implemented in GCC, | |
1029 | this will become the default. The name @samp{gnu9x} is deprecated. | |
1030 | ||
1031 | @end table | |
1032 | ||
1033 | Even when this option is not specified, you can still use some of the | |
1034 | features of newer standards in so far as they do not conflict with | |
1035 | previous C standards. For example, you may use @code{__restrict__} even | |
1036 | when @option{-std=c99} is not specified. | |
1037 | ||
1038 | The @option{-std} options specifying some version of ISO C have the same | |
1039 | effects as @option{-ansi}, except that features that were not in ISO C89 | |
1040 | but are in the specified version (for example, @samp{//} comments and | |
1041 | the @code{inline} keyword in ISO C99) are not disabled. | |
1042 | ||
1043 | @xref{Standards,,Language Standards Supported by GCC}, for details of | |
1044 | these standard versions. | |
1045 | ||
1046 | @item -aux-info @var{filename} | |
1047 | @opindex aux-info | |
1048 | Output to the given filename prototyped declarations for all functions | |
1049 | declared and/or defined in a translation unit, including those in header | |
1050 | files. This option is silently ignored in any language other than C@. | |
1051 | ||
1052 | Besides declarations, the file indicates, in comments, the origin of | |
1053 | each declaration (source file and line), whether the declaration was | |
1054 | implicit, prototyped or unprototyped (@samp{I}, @samp{N} for new or | |
1055 | @samp{O} for old, respectively, in the first character after the line | |
1056 | number and the colon), and whether it came from a declaration or a | |
1057 | definition (@samp{C} or @samp{F}, respectively, in the following | |
1058 | character). In the case of function definitions, a K&R-style list of | |
1059 | arguments followed by their declarations is also provided, inside | |
1060 | comments, after the declaration. | |
1061 | ||
1062 | @item -fno-asm | |
1063 | @opindex fno-asm | |
1064 | Do not recognize @code{asm}, @code{inline} or @code{typeof} as a | |
1065 | keyword, so that code can use these words as identifiers. You can use | |
1066 | the keywords @code{__asm__}, @code{__inline__} and @code{__typeof__} | |
1067 | instead. @option{-ansi} implies @option{-fno-asm}. | |
1068 | ||
1069 | In C++, this switch only affects the @code{typeof} keyword, since | |
1070 | @code{asm} and @code{inline} are standard keywords. You may want to | |
1071 | use the @option{-fno-gnu-keywords} flag instead, which has the same | |
1072 | effect. In C99 mode (@option{-std=c99} or @option{-std=gnu99}), this | |
1073 | switch only affects the @code{asm} and @code{typeof} keywords, since | |
1074 | @code{inline} is a standard keyword in ISO C99. | |
1075 | ||
1076 | @item -fno-builtin | |
1077 | @itemx -fno-builtin-@var{function} @r{(C and Objective-C only)} | |
1078 | @opindex fno-builtin | |
1079 | @cindex built-in functions | |
1080 | Don't recognize built-in functions that do not begin with | |
1081 | @samp{__builtin_} as prefix. @xref{Other Builtins,,Other built-in | |
1082 | functions provided by GCC}, for details of the functions affected, | |
1083 | including those which are not built-in functions when @option{-ansi} or | |
1084 | @option{-std} options for strict ISO C conformance are used because they | |
1085 | do not have an ISO standard meaning. | |
1086 | ||
1087 | GCC normally generates special code to handle certain built-in functions | |
1088 | more efficiently; for instance, calls to @code{alloca} may become single | |
1089 | instructions that adjust the stack directly, and calls to @code{memcpy} | |
1090 | may become inline copy loops. The resulting code is often both smaller | |
1091 | and faster, but since the function calls no longer appear as such, you | |
1092 | cannot set a breakpoint on those calls, nor can you change the behavior | |
1093 | of the functions by linking with a different library. | |
1094 | ||
1095 | In C++, @option{-fno-builtin} is always in effect. The @option{-fbuiltin} | |
1096 | option has no effect. Therefore, in C++, the only way to get the | |
1097 | optimization benefits of built-in functions is to call the function | |
1098 | using the @samp{__builtin_} prefix. The GNU C++ Standard Library uses | |
1099 | built-in functions to implement many functions (like | |
1100 | @code{std::strchr}), so that you automatically get efficient code. | |
1101 | ||
1102 | With the @option{-fno-builtin-@var{function}} option, not available | |
1103 | when compiling C++, only the built-in function @var{function} is | |
1104 | disabled. @var{function} must not begin with @samp{__builtin_}. If a | |
1105 | function is named this is not built-in in this version of GCC, this | |
1106 | option is ignored. There is no corresponding | |
1107 | @option{-fbuiltin-@var{function}} option; if you wish to enable | |
1108 | built-in functions selectively when using @option{-fno-builtin} or | |
1109 | @option{-ffreestanding}, you may define macros such as: | |
1110 | ||
1111 | @smallexample | |
1112 | #define abs(n) __builtin_abs ((n)) | |
1113 | #define strcpy(d, s) __builtin_strcpy ((d), (s)) | |
1114 | @end smallexample | |
1115 | ||
1116 | @item -fhosted | |
1117 | @opindex fhosted | |
1118 | @cindex hosted environment | |
1119 | ||
1120 | Assert that compilation takes place in a hosted environment. This implies | |
1121 | @option{-fbuiltin}. A hosted environment is one in which the | |
1122 | entire standard library is available, and in which @code{main} has a return | |
1123 | type of @code{int}. Examples are nearly everything except a kernel. | |
1124 | This is equivalent to @option{-fno-freestanding}. | |
1125 | ||
1126 | @item -ffreestanding | |
1127 | @opindex ffreestanding | |
1128 | @cindex hosted environment | |
1129 | ||
1130 | Assert that compilation takes place in a freestanding environment. This | |
1131 | implies @option{-fno-builtin}. A freestanding environment | |
1132 | is one in which the standard library may not exist, and program startup may | |
1133 | not necessarily be at @code{main}. The most obvious example is an OS kernel. | |
1134 | This is equivalent to @option{-fno-hosted}. | |
1135 | ||
1136 | @xref{Standards,,Language Standards Supported by GCC}, for details of | |
1137 | freestanding and hosted environments. | |
1138 | ||
1139 | @item -trigraphs | |
1140 | @opindex trigraphs | |
1141 | Support ISO C trigraphs. The @option{-ansi} option (and @option{-std} | |
1142 | options for strict ISO C conformance) implies @option{-trigraphs}. | |
1143 | ||
1144 | @cindex traditional C language | |
1145 | @cindex C language, traditional | |
1146 | @item -traditional | |
1147 | @itemx -traditional-cpp | |
1148 | @opindex traditional-cpp | |
1149 | @opindex traditional | |
1150 | Formerly, these options caused GCC to attempt to emulate a pre-standard | |
1151 | C compiler. They are now only supported with the @option{-E} switch. | |
1152 | The preprocessor continues to support a pre-standard mode. See the GNU | |
1153 | CPP manual for details. | |
1154 | ||
1155 | @item -fcond-mismatch | |
1156 | @opindex fcond-mismatch | |
1157 | Allow conditional expressions with mismatched types in the second and | |
1158 | third arguments. The value of such an expression is void. This option | |
1159 | is not supported for C++. | |
1160 | ||
1161 | @item -funsigned-char | |
1162 | @opindex funsigned-char | |
1163 | Let the type @code{char} be unsigned, like @code{unsigned char}. | |
1164 | ||
1165 | Each kind of machine has a default for what @code{char} should | |
1166 | be. It is either like @code{unsigned char} by default or like | |
1167 | @code{signed char} by default. | |
1168 | ||
1169 | Ideally, a portable program should always use @code{signed char} or | |
1170 | @code{unsigned char} when it depends on the signedness of an object. | |
1171 | But many programs have been written to use plain @code{char} and | |
1172 | expect it to be signed, or expect it to be unsigned, depending on the | |
1173 | machines they were written for. This option, and its inverse, let you | |
1174 | make such a program work with the opposite default. | |
1175 | ||
1176 | The type @code{char} is always a distinct type from each of | |
1177 | @code{signed char} or @code{unsigned char}, even though its behavior | |
1178 | is always just like one of those two. | |
1179 | ||
1180 | @item -fsigned-char | |
1181 | @opindex fsigned-char | |
1182 | Let the type @code{char} be signed, like @code{signed char}. | |
1183 | ||
1184 | Note that this is equivalent to @option{-fno-unsigned-char}, which is | |
1185 | the negative form of @option{-funsigned-char}. Likewise, the option | |
1186 | @option{-fno-signed-char} is equivalent to @option{-funsigned-char}. | |
1187 | ||
1188 | @item -fsigned-bitfields | |
1189 | @itemx -funsigned-bitfields | |
1190 | @itemx -fno-signed-bitfields | |
1191 | @itemx -fno-unsigned-bitfields | |
1192 | @opindex fsigned-bitfields | |
1193 | @opindex funsigned-bitfields | |
1194 | @opindex fno-signed-bitfields | |
1195 | @opindex fno-unsigned-bitfields | |
1196 | These options control whether a bit-field is signed or unsigned, when the | |
1197 | declaration does not use either @code{signed} or @code{unsigned}. By | |
1198 | default, such a bit-field is signed, because this is consistent: the | |
1199 | basic integer types such as @code{int} are signed types. | |
1200 | ||
1201 | @item -fwritable-strings | |
1202 | @opindex fwritable-strings | |
1203 | Store string constants in the writable data segment and don't uniquize | |
1204 | them. This is for compatibility with old programs which assume they can | |
1205 | write into string constants. | |
1206 | ||
1207 | Writing into string constants is a very bad idea; ``constants'' should | |
1208 | be constant. | |
1209 | ||
1210 | @item -fshort-wchar | |
1211 | @opindex fshort-wchar | |
1212 | Override the underlying type for @samp{wchar_t} to be @samp{short | |
1213 | unsigned int} instead of the default for the target. This option is | |
1214 | useful for building programs to run under WINE@. | |
1215 | @end table | |
1216 | ||
1217 | @node C++ Dialect Options | |
1218 | @section Options Controlling C++ Dialect | |
1219 | ||
1220 | @cindex compiler options, C++ | |
1221 | @cindex C++ options, command line | |
1222 | @cindex options, C++ | |
1223 | This section describes the command-line options that are only meaningful | |
1224 | for C++ programs; but you can also use most of the GNU compiler options | |
1225 | regardless of what language your program is in. For example, you | |
1226 | might compile a file @code{firstClass.C} like this: | |
1227 | ||
1228 | @example | |
1229 | g++ -g -frepo -O -c firstClass.C | |
1230 | @end example | |
1231 | ||
1232 | @noindent | |
1233 | In this example, only @option{-frepo} is an option meant | |
1234 | only for C++ programs; you can use the other options with any | |
1235 | language supported by GCC@. | |
1236 | ||
1237 | Here is a list of options that are @emph{only} for compiling C++ programs: | |
1238 | ||
1239 | @table @gcctabopt | |
1240 | @item -fno-access-control | |
1241 | @opindex fno-access-control | |
1242 | Turn off all access checking. This switch is mainly useful for working | |
1243 | around bugs in the access control code. | |
1244 | ||
1245 | @item -fcheck-new | |
1246 | @opindex fcheck-new | |
1247 | Check that the pointer returned by @code{operator new} is non-null | |
1248 | before attempting to modify the storage allocated. The current Working | |
1249 | Paper requires that @code{operator new} never return a null pointer, so | |
1250 | this check is normally unnecessary. | |
1251 | ||
1252 | An alternative to using this option is to specify that your | |
1253 | @code{operator new} does not throw any exceptions; if you declare it | |
1254 | @samp{throw()}, G++ will check the return value. See also @samp{new | |
1255 | (nothrow)}. | |
1256 | ||
1257 | @item -fconserve-space | |
1258 | @opindex fconserve-space | |
1259 | Put uninitialized or runtime-initialized global variables into the | |
1260 | common segment, as C does. This saves space in the executable at the | |
1261 | cost of not diagnosing duplicate definitions. If you compile with this | |
1262 | flag and your program mysteriously crashes after @code{main()} has | |
1263 | completed, you may have an object that is being destroyed twice because | |
1264 | two definitions were merged. | |
1265 | ||
1266 | This option is no longer useful on most targets, now that support has | |
1267 | been added for putting variables into BSS without making them common. | |
1268 | ||
1269 | @item -fno-const-strings | |
1270 | @opindex fno-const-strings | |
1271 | Give string constants type @code{char *} instead of type @code{const | |
1272 | char *}. By default, G++ uses type @code{const char *} as required by | |
1273 | the standard. Even if you use @option{-fno-const-strings}, you cannot | |
1274 | actually modify the value of a string constant, unless you also use | |
1275 | @option{-fwritable-strings}. | |
1276 | ||
1277 | This option might be removed in a future release of G++. For maximum | |
1278 | portability, you should structure your code so that it works with | |
1279 | string constants that have type @code{const char *}. | |
1280 | ||
1281 | @item -fdollars-in-identifiers | |
1282 | @opindex fdollars-in-identifiers | |
1283 | Accept @samp{$} in identifiers. You can also explicitly prohibit use of | |
1284 | @samp{$} with the option @option{-fno-dollars-in-identifiers}. (GNU C allows | |
1285 | @samp{$} by default on most target systems, but there are a few exceptions.) | |
1286 | Traditional C allowed the character @samp{$} to form part of | |
1287 | identifiers. However, ISO C and C++ forbid @samp{$} in identifiers. | |
1288 | ||
1289 | @item -fno-elide-constructors | |
1290 | @opindex fno-elide-constructors | |
1291 | The C++ standard allows an implementation to omit creating a temporary | |
1292 | which is only used to initialize another object of the same type. | |
1293 | Specifying this option disables that optimization, and forces G++ to | |
1294 | call the copy constructor in all cases. | |
1295 | ||
1296 | @item -fno-enforce-eh-specs | |
1297 | @opindex fno-enforce-eh-specs | |
1298 | Don't check for violation of exception specifications at runtime. This | |
1299 | option violates the C++ standard, but may be useful for reducing code | |
1300 | size in production builds, much like defining @samp{NDEBUG}. The compiler | |
1301 | will still optimize based on the exception specifications. | |
1302 | ||
1303 | @item -fexternal-templates | |
1304 | @opindex fexternal-templates | |
1305 | ||
1306 | Cause @samp{#pragma interface} and @samp{implementation} to apply to | |
1307 | template instantiation; template instances are emitted or not according | |
1308 | to the location of the template definition. @xref{Template | |
1309 | Instantiation}, for more information. | |
1310 | ||
1311 | This option is deprecated. | |
1312 | ||
1313 | @item -falt-external-templates | |
1314 | @opindex falt-external-templates | |
1315 | Similar to @option{-fexternal-templates}, but template instances are | |
1316 | emitted or not according to the place where they are first instantiated. | |
1317 | @xref{Template Instantiation}, for more information. | |
1318 | ||
1319 | This option is deprecated. | |
1320 | ||
1321 | @item -ffor-scope | |
1322 | @itemx -fno-for-scope | |
1323 | @opindex ffor-scope | |
1324 | @opindex fno-for-scope | |
1325 | If @option{-ffor-scope} is specified, the scope of variables declared in | |
1326 | a @i{for-init-statement} is limited to the @samp{for} loop itself, | |
1327 | as specified by the C++ standard. | |
1328 | If @option{-fno-for-scope} is specified, the scope of variables declared in | |
1329 | a @i{for-init-statement} extends to the end of the enclosing scope, | |
1330 | as was the case in old versions of G++, and other (traditional) | |
1331 | implementations of C++. | |
1332 | ||
1333 | The default if neither flag is given to follow the standard, | |
1334 | but to allow and give a warning for old-style code that would | |
1335 | otherwise be invalid, or have different behavior. | |
1336 | ||
1337 | @item -fno-gnu-keywords | |
1338 | @opindex fno-gnu-keywords | |
1339 | Do not recognize @code{typeof} as a keyword, so that code can use this | |
1340 | word as an identifier. You can use the keyword @code{__typeof__} instead. | |
1341 | @option{-ansi} implies @option{-fno-gnu-keywords}. | |
1342 | ||
1343 | @item -fno-implicit-templates | |
1344 | @opindex fno-implicit-templates | |
1345 | Never emit code for non-inline templates which are instantiated | |
1346 | implicitly (i.e.@: by use); only emit code for explicit instantiations. | |
1347 | @xref{Template Instantiation}, for more information. | |
1348 | ||
1349 | @item -fno-implicit-inline-templates | |
1350 | @opindex fno-implicit-inline-templates | |
1351 | Don't emit code for implicit instantiations of inline templates, either. | |
1352 | The default is to handle inlines differently so that compiles with and | |
1353 | without optimization will need the same set of explicit instantiations. | |
1354 | ||
1355 | @item -fno-implement-inlines | |
1356 | @opindex fno-implement-inlines | |
1357 | To save space, do not emit out-of-line copies of inline functions | |
1358 | controlled by @samp{#pragma implementation}. This will cause linker | |
1359 | errors if these functions are not inlined everywhere they are called. | |
1360 | ||
1361 | @item -fms-extensions | |
1362 | @opindex fms-extensions | |
1363 | Disable pedantic warnings about constructs used in MFC, such as implicit | |
1364 | int and getting a pointer to member function via non-standard syntax. | |
1365 | ||
1366 | @item -fno-nonansi-builtins | |
1367 | @opindex fno-nonansi-builtins | |
1368 | Disable built-in declarations of functions that are not mandated by | |
1369 | ANSI/ISO C@. These include @code{ffs}, @code{alloca}, @code{_exit}, | |
1370 | @code{index}, @code{bzero}, @code{conjf}, and other related functions. | |
1371 | ||
1372 | @item -fno-operator-names | |
1373 | @opindex fno-operator-names | |
1374 | Do not treat the operator name keywords @code{and}, @code{bitand}, | |
1375 | @code{bitor}, @code{compl}, @code{not}, @code{or} and @code{xor} as | |
1376 | synonyms as keywords. | |
1377 | ||
1378 | @item -fno-optional-diags | |
1379 | @opindex fno-optional-diags | |
1380 | Disable diagnostics that the standard says a compiler does not need to | |
1381 | issue. Currently, the only such diagnostic issued by G++ is the one for | |
1382 | a name having multiple meanings within a class. | |
1383 | ||
1384 | @item -fpermissive | |
1385 | @opindex fpermissive | |
1386 | Downgrade messages about nonconformant code from errors to warnings. By | |
1387 | default, G++ effectively sets @option{-pedantic-errors} without | |
1388 | @option{-pedantic}; this option reverses that. This behavior and this | |
1389 | option are superseded by @option{-pedantic}, which works as it does for GNU C@. | |
1390 | ||
1391 | @item -frepo | |
1392 | @opindex frepo | |
1393 | Enable automatic template instantiation at link time. This option also | |
1394 | implies @option{-fno-implicit-templates}. @xref{Template | |
1395 | Instantiation}, for more information. | |
1396 | ||
1397 | @item -fno-rtti | |
1398 | @opindex fno-rtti | |
1399 | Disable generation of information about every class with virtual | |
1400 | functions for use by the C++ runtime type identification features | |
1401 | (@samp{dynamic_cast} and @samp{typeid}). If you don't use those parts | |
1402 | of the language, you can save some space by using this flag. Note that | |
1403 | exception handling uses the same information, but it will generate it as | |
1404 | needed. | |
1405 | ||
1406 | @item -fstats | |
1407 | @opindex fstats | |
1408 | Emit statistics about front-end processing at the end of the compilation. | |
1409 | This information is generally only useful to the G++ development team. | |
1410 | ||
1411 | @item -ftemplate-depth-@var{n} | |
1412 | @opindex ftemplate-depth | |
1413 | Set the maximum instantiation depth for template classes to @var{n}. | |
1414 | A limit on the template instantiation depth is needed to detect | |
1415 | endless recursions during template class instantiation. ANSI/ISO C++ | |
1416 | conforming programs must not rely on a maximum depth greater than 17. | |
1417 | ||
1418 | @item -fuse-cxa-atexit | |
1419 | @opindex fuse-cxa-atexit | |
1420 | Register destructors for objects with static storage duration with the | |
1421 | @code{__cxa_atexit} function rather than the @code{atexit} function. | |
1422 | This option is required for fully standards-compliant handling of static | |
1423 | destructors, but will only work if your C library supports | |
1424 | @code{__cxa_atexit}. | |
1425 | ||
1426 | @item -fvtable-gc | |
1427 | @opindex fvtable-gc | |
1428 | Emit special relocations for vtables and virtual function references | |
1429 | so that the linker can identify unused virtual functions and zero out | |
1430 | vtable slots that refer to them. This is most useful with | |
1431 | @option{-ffunction-sections} and @option{-Wl,--gc-sections}, in order to | |
1432 | also discard the functions themselves. | |
1433 | ||
1434 | This optimization requires GNU as and GNU ld. Not all systems support | |
1435 | this option. @option{-Wl,--gc-sections} is ignored without @option{-static}. | |
1436 | ||
1437 | @item -fno-weak | |
1438 | @opindex fno-weak | |
1439 | Do not use weak symbol support, even if it is provided by the linker. | |
1440 | By default, G++ will use weak symbols if they are available. This | |
1441 | option exists only for testing, and should not be used by end-users; | |
1442 | it will result in inferior code and has no benefits. This option may | |
1443 | be removed in a future release of G++. | |
1444 | ||
1445 | @item -nostdinc++ | |
1446 | @opindex nostdinc++ | |
1447 | Do not search for header files in the standard directories specific to | |
1448 | C++, but do still search the other standard directories. (This option | |
1449 | is used when building the C++ library.) | |
1450 | @end table | |
1451 | ||
1452 | In addition, these optimization, warning, and code generation options | |
1453 | have meanings only for C++ programs: | |
1454 | ||
1455 | @table @gcctabopt | |
1456 | @item -fno-default-inline | |
1457 | @opindex fno-default-inline | |
1458 | Do not assume @samp{inline} for functions defined inside a class scope. | |
1459 | @xref{Optimize Options,,Options That Control Optimization}. Note that these | |
1460 | functions will have linkage like inline functions; they just won't be | |
1461 | inlined by default. | |
1462 | ||
1463 | @item -Wctor-dtor-privacy @r{(C++ only)} | |
1464 | @opindex Wctor-dtor-privacy | |
1465 | Warn when a class seems unusable, because all the constructors or | |
1466 | destructors in a class are private and the class has no friends or | |
1467 | public static member functions. | |
1468 | ||
1469 | @item -Wnon-virtual-dtor @r{(C++ only)} | |
1470 | @opindex Wnon-virtual-dtor | |
1471 | Warn when a class declares a non-virtual destructor that should probably | |
1472 | be virtual, because it looks like the class will be used polymorphically. | |
1473 | ||
1474 | @item -Wreorder @r{(C++ only)} | |
1475 | @opindex Wreorder | |
1476 | @cindex reordering, warning | |
1477 | @cindex warning for reordering of member initializers | |
1478 | Warn when the order of member initializers given in the code does not | |
1479 | match the order in which they must be executed. For instance: | |
1480 | ||
1481 | @smallexample | |
1482 | struct A @{ | |
1483 | int i; | |
1484 | int j; | |
1485 | A(): j (0), i (1) @{ @} | |
1486 | @}; | |
1487 | @end smallexample | |
1488 | ||
1489 | Here the compiler will warn that the member initializers for @samp{i} | |
1490 | and @samp{j} will be rearranged to match the declaration order of the | |
1491 | members. | |
1492 | @end table | |
1493 | ||
1494 | The following @option{-W@dots{}} options are not affected by @option{-Wall}. | |
1495 | ||
1496 | @table @gcctabopt | |
1497 | @item -Weffc++ @r{(C++ only)} | |
1498 | @opindex Weffc++ | |
1499 | Warn about violations of the following style guidelines from Scott Meyers' | |
1500 | @cite{Effective C++} book: | |
1501 | ||
1502 | @itemize @bullet | |
1503 | @item | |
1504 | Item 11: Define a copy constructor and an assignment operator for classes | |
1505 | with dynamically allocated memory. | |
1506 | ||
1507 | @item | |
1508 | Item 12: Prefer initialization to assignment in constructors. | |
1509 | ||
1510 | @item | |
1511 | Item 14: Make destructors virtual in base classes. | |
1512 | ||
1513 | @item | |
1514 | Item 15: Have @code{operator=} return a reference to @code{*this}. | |
1515 | ||
1516 | @item | |
1517 | Item 23: Don't try to return a reference when you must return an object. | |
1518 | ||
1519 | @end itemize | |
1520 | ||
1521 | and about violations of the following style guidelines from Scott Meyers' | |
1522 | @cite{More Effective C++} book: | |
1523 | ||
1524 | @itemize @bullet | |
1525 | @item | |
1526 | Item 6: Distinguish between prefix and postfix forms of increment and | |
1527 | decrement operators. | |
1528 | ||
1529 | @item | |
1530 | Item 7: Never overload @code{&&}, @code{||}, or @code{,}. | |
1531 | ||
1532 | @end itemize | |
1533 | ||
1534 | If you use this option, you should be aware that the standard library | |
1535 | headers do not obey all of these guidelines; you can use @samp{grep -v} | |
1536 | to filter out those warnings. | |
1537 | ||
1538 | @item -Wno-deprecated @r{(C++ only)} | |
1539 | @opindex Wno-deprecated | |
1540 | Do not warn about usage of deprecated features. @xref{Deprecated Features}. | |
1541 | ||
1542 | @item -Wno-non-template-friend @r{(C++ only)} | |
1543 | @opindex Wno-non-template-friend | |
1544 | Disable warnings when non-templatized friend functions are declared | |
1545 | within a template. With the advent of explicit template specification | |
1546 | support in G++, if the name of the friend is an unqualified-id (i.e., | |
1547 | @samp{friend foo(int)}), the C++ language specification demands that the | |
1548 | friend declare or define an ordinary, nontemplate function. (Section | |
1549 | 14.5.3). Before G++ implemented explicit specification, unqualified-ids | |
1550 | could be interpreted as a particular specialization of a templatized | |
1551 | function. Because this non-conforming behavior is no longer the default | |
1552 | behavior for G++, @option{-Wnon-template-friend} allows the compiler to | |
1553 | check existing code for potential trouble spots, and is on by default. | |
1554 | This new compiler behavior can be turned off with | |
1555 | @option{-Wno-non-template-friend} which keeps the conformant compiler code | |
1556 | but disables the helpful warning. | |
1557 | ||
1558 | @item -Wold-style-cast @r{(C++ only)} | |
1559 | @opindex Wold-style-cast | |
1560 | Warn if an old-style (C-style) cast to a non-void type is used within | |
1561 | a C++ program. The new-style casts (@samp{static_cast}, | |
1562 | @samp{reinterpret_cast}, and @samp{const_cast}) are less vulnerable to | |
1563 | unintended effects, and much easier to grep for. | |
1564 | ||
1565 | @item -Woverloaded-virtual @r{(C++ only)} | |
1566 | @opindex Woverloaded-virtual | |
1567 | @cindex overloaded virtual fn, warning | |
1568 | @cindex warning for overloaded virtual fn | |
1569 | Warn when a function declaration hides virtual functions from a | |
1570 | base class. For example, in: | |
1571 | ||
1572 | @smallexample | |
1573 | struct A @{ | |
1574 | virtual void f(); | |
1575 | @}; | |
1576 | ||
1577 | struct B: public A @{ | |
1578 | void f(int); | |
1579 | @}; | |
1580 | @end smallexample | |
1581 | ||
1582 | the @code{A} class version of @code{f} is hidden in @code{B}, and code | |
1583 | like this: | |
1584 | ||
1585 | @smallexample | |
1586 | B* b; | |
1587 | b->f(); | |
1588 | @end smallexample | |
1589 | ||
1590 | will fail to compile. | |
1591 | ||
1592 | @item -Wno-pmf-conversions @r{(C++ only)} | |
1593 | @opindex Wno-pmf-conversions | |
1594 | Disable the diagnostic for converting a bound pointer to member function | |
1595 | to a plain pointer. | |
1596 | ||
1597 | @item -Wsign-promo @r{(C++ only)} | |
1598 | @opindex Wsign-promo | |
1599 | Warn when overload resolution chooses a promotion from unsigned or | |
1600 | enumeral type to a signed type over a conversion to an unsigned type of | |
1601 | the same size. Previous versions of G++ would try to preserve | |
1602 | unsignedness, but the standard mandates the current behavior. | |
1603 | ||
1604 | @item -Wsynth @r{(C++ only)} | |
1605 | @opindex Wsynth | |
1606 | @cindex warning for synthesized methods | |
1607 | @cindex synthesized methods, warning | |
1608 | Warn when G++'s synthesis behavior does not match that of cfront. For | |
1609 | instance: | |
1610 | ||
1611 | @smallexample | |
1612 | struct A @{ | |
1613 | operator int (); | |
1614 | A& operator = (int); | |
1615 | @}; | |
1616 | ||
1617 | main () | |
1618 | @{ | |
1619 | A a,b; | |
1620 | a = b; | |
1621 | @} | |
1622 | @end smallexample | |
1623 | ||
1624 | In this example, G++ will synthesize a default @samp{A& operator = | |
1625 | (const A&);}, while cfront will use the user-defined @samp{operator =}. | |
1626 | @end table | |
1627 | ||
1628 | @node Objective-C Dialect Options | |
1629 | @section Options Controlling Objective-C Dialect | |
1630 | ||
1631 | @cindex compiler options, Objective-C | |
1632 | @cindex Objective-C options, command line | |
1633 | @cindex options, Objective-C | |
1634 | This section describes the command-line options that are only meaningful | |
1635 | for Objective-C programs; but you can also use most of the GNU compiler | |
1636 | options regardless of what language your program is in. For example, | |
1637 | you might compile a file @code{some_class.m} like this: | |
1638 | ||
1639 | @example | |
1640 | gcc -g -fgnu-runtime -O -c some_class.m | |
1641 | @end example | |
1642 | ||
1643 | @noindent | |
1644 | In this example, only @option{-fgnu-runtime} is an option meant only for | |
1645 | Objective-C programs; you can use the other options with any language | |
1646 | supported by GCC@. | |
1647 | ||
1648 | Here is a list of options that are @emph{only} for compiling Objective-C | |
1649 | programs: | |
1650 | ||
1651 | @table @gcctabopt | |
1652 | @item -fconstant-string-class=@var{class-name} | |
1653 | @opindex fconstant-string-class | |
1654 | Use @var{class-name} as the name of the class to instantiate for each | |
1655 | literal string specified with the syntax @code{@@"@dots{}"}. The default | |
1656 | class name is @code{NXConstantString}. | |
1657 | ||
1658 | @item -fgnu-runtime | |
1659 | @opindex fgnu-runtime | |
1660 | Generate object code compatible with the standard GNU Objective-C | |
1661 | runtime. This is the default for most types of systems. | |
1662 | ||
1663 | @item -fnext-runtime | |
1664 | @opindex fnext-runtime | |
1665 | Generate output compatible with the NeXT runtime. This is the default | |
1666 | for NeXT-based systems, including Darwin and Mac OS X@. | |
1667 | ||
1668 | @item -gen-decls | |
1669 | @opindex gen-decls | |
1670 | Dump interface declarations for all classes seen in the source file to a | |
1671 | file named @file{@var{sourcename}.decl}. | |
1672 | ||
1673 | @item -Wno-protocol | |
1674 | @opindex Wno-protocol | |
1675 | Do not warn if methods required by a protocol are not implemented | |
1676 | in the class adopting it. | |
1677 | ||
1678 | @item -Wselector | |
1679 | @opindex Wselector | |
1680 | Warn if a selector has multiple methods of different types defined. | |
1681 | ||
1682 | @c not documented because only avail via -Wp | |
1683 | @c @item -print-objc-runtime-info | |
1684 | ||
1685 | @end table | |
1686 | ||
1687 | @node Language Independent Options | |
1688 | @section Options to Control Diagnostic Messages Formatting | |
1689 | @cindex options to control diagnostics formatting | |
1690 | @cindex diagnostic messages | |
1691 | @cindex message formatting | |
1692 | ||
1693 | Traditionally, diagnostic messages have been formatted irrespective of | |
1694 | the output device's aspect (e.g.@: its width, @dots{}). The options described | |
1695 | below can be used to control the diagnostic messages formatting | |
1696 | algorithm, e.g.@: how many characters per line, how often source location | |
1697 | information should be reported. Right now, only the C++ front end can | |
1698 | honor these options. However it is expected, in the near future, that | |
1699 | the remaining front ends would be able to digest them correctly. | |
1700 | ||
1701 | @table @gcctabopt | |
1702 | @item -fmessage-length=@var{n} | |
1703 | @opindex fmessage-length | |
1704 | Try to format error messages so that they fit on lines of about @var{n} | |
1705 | characters. The default is 72 characters for @command{g++} and 0 for the rest of | |
1706 | the front ends supported by GCC@. If @var{n} is zero, then no | |
1707 | line-wrapping will be done; each error message will appear on a single | |
1708 | line. | |
1709 | ||
1710 | @opindex fdiagnostics-show-location | |
1711 | @item -fdiagnostics-show-location=once | |
1712 | Only meaningful in line-wrapping mode. Instructs the diagnostic messages | |
1713 | reporter to emit @emph{once} source location information; that is, in | |
1714 | case the message is too long to fit on a single physical line and has to | |
1715 | be wrapped, the source location won't be emitted (as prefix) again, | |
1716 | over and over, in subsequent continuation lines. This is the default | |
1717 | behavior. | |
1718 | ||
1719 | @item -fdiagnostics-show-location=every-line | |
1720 | Only meaningful in line-wrapping mode. Instructs the diagnostic | |
1721 | messages reporter to emit the same source location information (as | |
1722 | prefix) for physical lines that result from the process of breaking | |
1723 | a message which is too long to fit on a single line. | |
1724 | ||
1725 | @end table | |
1726 | ||
1727 | @node Warning Options | |
1728 | @section Options to Request or Suppress Warnings | |
1729 | @cindex options to control warnings | |
1730 | @cindex warning messages | |
1731 | @cindex messages, warning | |
1732 | @cindex suppressing warnings | |
1733 | ||
1734 | Warnings are diagnostic messages that report constructions which | |
1735 | are not inherently erroneous but which are risky or suggest there | |
1736 | may have been an error. | |
1737 | ||
1738 | You can request many specific warnings with options beginning @samp{-W}, | |
1739 | for example @option{-Wimplicit} to request warnings on implicit | |
1740 | declarations. Each of these specific warning options also has a | |
1741 | negative form beginning @samp{-Wno-} to turn off warnings; | |
1742 | for example, @option{-Wno-implicit}. This manual lists only one of the | |
1743 | two forms, whichever is not the default. | |
1744 | ||
1745 | These options control the amount and kinds of warnings produced by GCC: | |
1746 | ||
1747 | @table @gcctabopt | |
1748 | @cindex syntax checking | |
1749 | @item -fsyntax-only | |
1750 | @opindex fsyntax-only | |
1751 | Check the code for syntax errors, but don't do anything beyond that. | |
1752 | ||
1753 | @item -pedantic | |
1754 | @opindex pedantic | |
1755 | Issue all the warnings demanded by strict ISO C and ISO C++; | |
1756 | reject all programs that use forbidden extensions, and some other | |
1757 | programs that do not follow ISO C and ISO C++. For ISO C, follows the | |
1758 | version of the ISO C standard specified by any @option{-std} option used. | |
1759 | ||
1760 | Valid ISO C and ISO C++ programs should compile properly with or without | |
1761 | this option (though a rare few will require @option{-ansi} or a | |
1762 | @option{-std} option specifying the required version of ISO C)@. However, | |
1763 | without this option, certain GNU extensions and traditional C and C++ | |
1764 | features are supported as well. With this option, they are rejected. | |
1765 | ||
1766 | @option{-pedantic} does not cause warning messages for use of the | |
1767 | alternate keywords whose names begin and end with @samp{__}. Pedantic | |
1768 | warnings are also disabled in the expression that follows | |
1769 | @code{__extension__}. However, only system header files should use | |
1770 | these escape routes; application programs should avoid them. | |
1771 | @xref{Alternate Keywords}. | |
1772 | ||
1773 | Some users try to use @option{-pedantic} to check programs for strict ISO | |
1774 | C conformance. They soon find that it does not do quite what they want: | |
1775 | it finds some non-ISO practices, but not all---only those for which | |
1776 | ISO C @emph{requires} a diagnostic, and some others for which | |
1777 | diagnostics have been added. | |
1778 | ||
1779 | A feature to report any failure to conform to ISO C might be useful in | |
1780 | some instances, but would require considerable additional work and would | |
1781 | be quite different from @option{-pedantic}. We don't have plans to | |
1782 | support such a feature in the near future. | |
1783 | ||
1784 | Where the standard specified with @option{-std} represents a GNU | |
1785 | extended dialect of C, such as @samp{gnu89} or @samp{gnu99}, there is a | |
1786 | corresponding @dfn{base standard}, the version of ISO C on which the GNU | |
1787 | extended dialect is based. Warnings from @option{-pedantic} are given | |
1788 | where they are required by the base standard. (It would not make sense | |
1789 | for such warnings to be given only for features not in the specified GNU | |
1790 | C dialect, since by definition the GNU dialects of C include all | |
1791 | features the compiler supports with the given option, and there would be | |
1792 | nothing to warn about.) | |
1793 | ||
1794 | @item -pedantic-errors | |
1795 | @opindex pedantic-errors | |
1796 | Like @option{-pedantic}, except that errors are produced rather than | |
1797 | warnings. | |
1798 | ||
1799 | @item -w | |
1800 | @opindex w | |
1801 | Inhibit all warning messages. | |
1802 | ||
1803 | @item -Wno-import | |
1804 | @opindex Wno-import | |
1805 | Inhibit warning messages about the use of @samp{#import}. | |
1806 | ||
1807 | @item -Wchar-subscripts | |
1808 | @opindex Wchar-subscripts | |
1809 | Warn if an array subscript has type @code{char}. This is a common cause | |
1810 | of error, as programmers often forget that this type is signed on some | |
1811 | machines. | |
1812 | ||
1813 | @item -Wcomment | |
1814 | @opindex Wcomment | |
1815 | Warn whenever a comment-start sequence @samp{/*} appears in a @samp{/*} | |
1816 | comment, or whenever a Backslash-Newline appears in a @samp{//} comment. | |
1817 | ||
1818 | @item -Wformat | |
1819 | @opindex Wformat | |
1820 | Check calls to @code{printf} and @code{scanf}, etc., to make sure that | |
1821 | the arguments supplied have types appropriate to the format string | |
1822 | specified, and that the conversions specified in the format string make | |
1823 | sense. This includes standard functions, and others specified by format | |
1824 | attributes (@pxref{Function Attributes}), in the @code{printf}, | |
1825 | @code{scanf}, @code{strftime} and @code{strfmon} (an X/Open extension, | |
1826 | not in the C standard) families. | |
1827 | ||
1828 | The formats are checked against the format features supported by GNU | |
1829 | libc version 2.2. These include all ISO C89 and C99 features, as well | |
1830 | as features from the Single Unix Specification and some BSD and GNU | |
1831 | extensions. Other library implementations may not support all these | |
1832 | features; GCC does not support warning about features that go beyond a | |
1833 | particular library's limitations. However, if @option{-pedantic} is used | |
1834 | with @option{-Wformat}, warnings will be given about format features not | |
1835 | in the selected standard version (but not for @code{strfmon} formats, | |
1836 | since those are not in any version of the C standard). @xref{C Dialect | |
1837 | Options,,Options Controlling C Dialect}. | |
1838 | ||
1839 | @option{-Wformat} is included in @option{-Wall}. For more control over some | |
1840 | aspects of format checking, the options @option{-Wno-format-y2k}, | |
1841 | @option{-Wno-format-extra-args}, @option{-Wformat-nonliteral}, | |
1842 | @option{-Wformat-security} and @option{-Wformat=2} are available, but are | |
1843 | not included in @option{-Wall}. | |
1844 | ||
1845 | @item -Wno-format-y2k | |
1846 | @opindex Wno-format-y2k | |
1847 | If @option{-Wformat} is specified, do not warn about @code{strftime} | |
1848 | formats which may yield only a two-digit year. | |
1849 | ||
1850 | @item -Wno-format-extra-args | |
1851 | @opindex Wno-format-extra-args | |
1852 | If @option{-Wformat} is specified, do not warn about excess arguments to a | |
1853 | @code{printf} or @code{scanf} format function. The C standard specifies | |
1854 | that such arguments are ignored. | |
1855 | ||
1856 | Where the unused arguments lie between used arguments that are | |
1857 | specified with @samp{$} operand number specifications, normally | |
1858 | warnings are still given, since the implementation could not know what | |
1859 | type to pass to @code{va_arg} to skip the unused arguments. However, | |
1860 | in the case of @code{scanf} formats, this option will suppress the | |
1861 | warning if the unused arguments are all pointers, since the Single | |
1862 | Unix Specification says that such unused arguments are allowed. | |
1863 | ||
1864 | @item -Wformat-nonliteral | |
1865 | @opindex Wformat-nonliteral | |
1866 | If @option{-Wformat} is specified, also warn if the format string is not a | |
1867 | string literal and so cannot be checked, unless the format function | |
1868 | takes its format arguments as a @code{va_list}. | |
1869 | ||
1870 | @item -Wformat-security | |
1871 | @opindex Wformat-security | |
1872 | If @option{-Wformat} is specified, also warn about uses of format | |
1873 | functions that represent possible security problems. At present, this | |
1874 | warns about calls to @code{printf} and @code{scanf} functions where the | |
1875 | format string is not a string literal and there are no format arguments, | |
1876 | as in @code{printf (foo);}. This may be a security hole if the format | |
1877 | string came from untrusted input and contains @samp{%n}. (This is | |
1878 | currently a subset of what @option{-Wformat-nonliteral} warns about, but | |
1879 | in future warnings may be added to @option{-Wformat-security} that are not | |
1880 | included in @option{-Wformat-nonliteral}.) | |
1881 | ||
1882 | @item -Wformat=2 | |
1883 | @opindex Wformat=2 | |
1884 | Enable @option{-Wformat} plus format checks not included in | |
1885 | @option{-Wformat}. Currently equivalent to @samp{-Wformat | |
1886 | -Wformat-nonliteral -Wformat-security}. | |
1887 | ||
1888 | @item -Wimplicit-int | |
1889 | @opindex Wimplicit-int | |
1890 | Warn when a declaration does not specify a type. | |
1891 | ||
1892 | @item -Wimplicit-function-declaration | |
1893 | @itemx -Werror-implicit-function-declaration | |
1894 | @opindex Wimplicit-function-declaration | |
1895 | @opindex Werror-implicit-function-declaration | |
1896 | Give a warning (or error) whenever a function is used before being | |
1897 | declared. | |
1898 | ||
1899 | @item -Wimplicit | |
1900 | @opindex Wimplicit | |
1901 | Same as @option{-Wimplicit-int} and @option{-Wimplicit-function-declaration}. | |
1902 | ||
1903 | @item -Wmain | |
1904 | @opindex Wmain | |
1905 | Warn if the type of @samp{main} is suspicious. @samp{main} should be a | |
1906 | function with external linkage, returning int, taking either zero | |
1907 | arguments, two, or three arguments of appropriate types. | |
1908 | ||
1909 | @item -Wmissing-braces | |
1910 | @opindex Wmissing-braces | |
1911 | Warn if an aggregate or union initializer is not fully bracketed. In | |
1912 | the following example, the initializer for @samp{a} is not fully | |
1913 | bracketed, but that for @samp{b} is fully bracketed. | |
1914 | ||
1915 | @smallexample | |
1916 | int a[2][2] = @{ 0, 1, 2, 3 @}; | |
1917 | int b[2][2] = @{ @{ 0, 1 @}, @{ 2, 3 @} @}; | |
1918 | @end smallexample | |
1919 | ||
1920 | @item -Wparentheses | |
1921 | @opindex Wparentheses | |
1922 | Warn if parentheses are omitted in certain contexts, such | |
1923 | as when there is an assignment in a context where a truth value | |
1924 | is expected, or when operators are nested whose precedence people | |
1925 | often get confused about. | |
1926 | ||
1927 | Also warn about constructions where there may be confusion to which | |
1928 | @code{if} statement an @code{else} branch belongs. Here is an example of | |
1929 | such a case: | |
1930 | ||
1931 | @smallexample | |
1932 | @group | |
1933 | @{ | |
1934 | if (a) | |
1935 | if (b) | |
1936 | foo (); | |
1937 | else | |
1938 | bar (); | |
1939 | @} | |
1940 | @end group | |
1941 | @end smallexample | |
1942 | ||
1943 | In C, every @code{else} branch belongs to the innermost possible @code{if} | |
1944 | statement, which in this example is @code{if (b)}. This is often not | |
1945 | what the programmer expected, as illustrated in the above example by | |
1946 | indentation the programmer chose. When there is the potential for this | |
1947 | confusion, GCC will issue a warning when this flag is specified. | |
1948 | To eliminate the warning, add explicit braces around the innermost | |
1949 | @code{if} statement so there is no way the @code{else} could belong to | |
1950 | the enclosing @code{if}. The resulting code would look like this: | |
1951 | ||
1952 | @smallexample | |
1953 | @group | |
1954 | @{ | |
1955 | if (a) | |
1956 | @{ | |
1957 | if (b) | |
1958 | foo (); | |
1959 | else | |
1960 | bar (); | |
1961 | @} | |
1962 | @} | |
1963 | @end group | |
1964 | @end smallexample | |
1965 | ||
1966 | @item -Wsequence-point | |
1967 | @opindex Wsequence-point | |
1968 | Warn about code that may have undefined semantics because of violations | |
1969 | of sequence point rules in the C standard. | |
1970 | ||
1971 | The C standard defines the order in which expressions in a C program are | |
1972 | evaluated in terms of @dfn{sequence points}, which represent a partial | |
1973 | ordering between the execution of parts of the program: those executed | |
1974 | before the sequence point, and those executed after it. These occur | |
1975 | after the evaluation of a full expression (one which is not part of a | |
1976 | larger expression), after the evaluation of the first operand of a | |
1977 | @code{&&}, @code{||}, @code{? :} or @code{,} (comma) operator, before a | |
1978 | function is called (but after the evaluation of its arguments and the | |
1979 | expression denoting the called function), and in certain other places. | |
1980 | Other than as expressed by the sequence point rules, the order of | |
1981 | evaluation of subexpressions of an expression is not specified. All | |
1982 | these rules describe only a partial order rather than a total order, | |
1983 | since, for example, if two functions are called within one expression | |
1984 | with no sequence point between them, the order in which the functions | |
1985 | are called is not specified. However, the standards committee have | |
1986 | ruled that function calls do not overlap. | |
1987 | ||
1988 | It is not specified when between sequence points modifications to the | |
1989 | values of objects take effect. Programs whose behavior depends on this | |
1990 | have undefined behavior; the C standard specifies that ``Between the | |
1991 | previous and next sequence point an object shall have its stored value | |
1992 | modified at most once by the evaluation of an expression. Furthermore, | |
1993 | the prior value shall be read only to determine the value to be | |
1994 | stored.''. If a program breaks these rules, the results on any | |
1995 | particular implementation are entirely unpredictable. | |
1996 | ||
1997 | Examples of code with undefined behavior are @code{a = a++;}, @code{a[n] | |
1998 | = b[n++]} and @code{a[i++] = i;}. Some more complicated cases are not | |
1999 | diagnosed by this option, and it may give an occasional false positive | |
2000 | result, but in general it has been found fairly effective at detecting | |
2001 | this sort of problem in programs. | |
2002 | ||
2003 | The present implementation of this option only works for C programs. A | |
2004 | future implementation may also work for C++ programs. | |
2005 | ||
2006 | The C standard is worded confusingly, therefore there is some debate | |
2007 | over the precise meaning of the sequence point rules in subtle cases. | |
2008 | Links to discussions of the problem, including proposed formal | |
2009 | definitions, may be found on our readings page, at | |
2010 | @w{@uref{http://gcc.gnu.org/readings.html}}. | |
2011 | ||
2012 | @item -Wreturn-type | |
2013 | @opindex Wreturn-type | |
2014 | Warn whenever a function is defined with a return-type that defaults to | |
2015 | @code{int}. Also warn about any @code{return} statement with no | |
2016 | return-value in a function whose return-type is not @code{void}. | |
2017 | ||
2018 | For C++, a function without return type always produces a diagnostic | |
2019 | message, even when @option{-Wno-return-type} is specified. The only | |
2020 | exceptions are @samp{main} and functions defined in system headers. | |
2021 | ||
2022 | @item -Wswitch | |
2023 | @opindex Wswitch | |
2024 | Warn whenever a @code{switch} statement has an index of enumeral type | |
2025 | and lacks a @code{case} for one or more of the named codes of that | |
2026 | enumeration. (The presence of a @code{default} label prevents this | |
2027 | warning.) @code{case} labels outside the enumeration range also | |
2028 | provoke warnings when this option is used. | |
2029 | ||
2030 | @item -Wtrigraphs | |
2031 | @opindex Wtrigraphs | |
2032 | Warn if any trigraphs are encountered that might change the meaning of | |
2033 | the program (trigraphs within comments are not warned about). | |
2034 | ||
2035 | @item -Wunused-function | |
2036 | @opindex Wunused-function | |
2037 | Warn whenever a static function is declared but not defined or a | |
2038 | non\-inline static function is unused. | |
2039 | ||
2040 | @item -Wunused-label | |
2041 | @opindex Wunused-label | |
2042 | Warn whenever a label is declared but not used. | |
2043 | ||
2044 | To suppress this warning use the @samp{unused} attribute | |
2045 | (@pxref{Variable Attributes}). | |
2046 | ||
2047 | @item -Wunused-parameter | |
2048 | @opindex Wunused-parameter | |
2049 | Warn whenever a function parameter is unused aside from its declaration. | |
2050 | ||
2051 | To suppress this warning use the @samp{unused} attribute | |
2052 | (@pxref{Variable Attributes}). | |
2053 | ||
2054 | @item -Wunused-variable | |
2055 | @opindex Wunused-variable | |
2056 | Warn whenever a local variable or non-constant static variable is unused | |
2057 | aside from its declaration | |
2058 | ||
2059 | To suppress this warning use the @samp{unused} attribute | |
2060 | (@pxref{Variable Attributes}). | |
2061 | ||
2062 | @item -Wunused-value | |
2063 | @opindex Wunused-value | |
2064 | Warn whenever a statement computes a result that is explicitly not used. | |
2065 | ||
2066 | To suppress this warning cast the expression to @samp{void}. | |
2067 | ||
2068 | @item -Wunused | |
2069 | @opindex Wunused | |
2070 | All all the above @option{-Wunused} options combined. | |
2071 | ||
2072 | In order to get a warning about an unused function parameter, you must | |
2073 | either specify @samp{-W -Wunused} or separately specify | |
2074 | @option{-Wunused-parameter}. | |
2075 | ||
2076 | @item -Wuninitialized | |
2077 | @opindex Wuninitialized | |
2078 | Warn if an automatic variable is used without first being initialized or | |
2079 | if a variable may be clobbered by a @code{setjmp} call. | |
2080 | ||
2081 | These warnings are possible only in optimizing compilation, | |
2082 | because they require data flow information that is computed only | |
2083 | when optimizing. If you don't specify @option{-O}, you simply won't | |
2084 | get these warnings. | |
2085 | ||
2086 | These warnings occur only for variables that are candidates for | |
2087 | register allocation. Therefore, they do not occur for a variable that | |
2088 | is declared @code{volatile}, or whose address is taken, or whose size | |
2089 | is other than 1, 2, 4 or 8 bytes. Also, they do not occur for | |
2090 | structures, unions or arrays, even when they are in registers. | |
2091 | ||
2092 | Note that there may be no warning about a variable that is used only | |
2093 | to compute a value that itself is never used, because such | |
2094 | computations may be deleted by data flow analysis before the warnings | |
2095 | are printed. | |
2096 | ||
2097 | These warnings are made optional because GCC is not smart | |
2098 | enough to see all the reasons why the code might be correct | |
2099 | despite appearing to have an error. Here is one example of how | |
2100 | this can happen: | |
2101 | ||
2102 | @smallexample | |
2103 | @group | |
2104 | @{ | |
2105 | int x; | |
2106 | switch (y) | |
2107 | @{ | |
2108 | case 1: x = 1; | |
2109 | break; | |
2110 | case 2: x = 4; | |
2111 | break; | |
2112 | case 3: x = 5; | |
2113 | @} | |
2114 | foo (x); | |
2115 | @} | |
2116 | @end group | |
2117 | @end smallexample | |
2118 | ||
2119 | @noindent | |
2120 | If the value of @code{y} is always 1, 2 or 3, then @code{x} is | |
2121 | always initialized, but GCC doesn't know this. Here is | |
2122 | another common case: | |
2123 | ||
2124 | @smallexample | |
2125 | @{ | |
2126 | int save_y; | |
2127 | if (change_y) save_y = y, y = new_y; | |
2128 | @dots{} | |
2129 | if (change_y) y = save_y; | |
2130 | @} | |
2131 | @end smallexample | |
2132 | ||
2133 | @noindent | |
2134 | This has no bug because @code{save_y} is used only if it is set. | |
2135 | ||
2136 | @cindex @code{longjmp} warnings | |
2137 | This option also warns when a non-volatile automatic variable might be | |
2138 | changed by a call to @code{longjmp}. These warnings as well are possible | |
2139 | only in optimizing compilation. | |
2140 | ||
2141 | The compiler sees only the calls to @code{setjmp}. It cannot know | |
2142 | where @code{longjmp} will be called; in fact, a signal handler could | |
2143 | call it at any point in the code. As a result, you may get a warning | |
2144 | even when there is in fact no problem because @code{longjmp} cannot | |
2145 | in fact be called at the place which would cause a problem. | |
2146 | ||
2147 | Some spurious warnings can be avoided if you declare all the functions | |
2148 | you use that never return as @code{noreturn}. @xref{Function | |
2149 | Attributes}. | |
2150 | ||
2151 | @item -Wreorder @r{(C++ only)} | |
2152 | @opindex Wreorder | |
2153 | @cindex reordering, warning | |
2154 | @cindex warning for reordering of member initializers | |
2155 | Warn when the order of member initializers given in the code does not | |
2156 | match the order in which they must be executed. For instance: | |
2157 | ||
2158 | @item -Wunknown-pragmas | |
2159 | @opindex Wunknown-pragmas | |
2160 | @cindex warning for unknown pragmas | |
2161 | @cindex unknown pragmas, warning | |
2162 | @cindex pragmas, warning of unknown | |
2163 | Warn when a #pragma directive is encountered which is not understood by | |
2164 | GCC@. If this command line option is used, warnings will even be issued | |
2165 | for unknown pragmas in system header files. This is not the case if | |
2166 | the warnings were only enabled by the @option{-Wall} command line option. | |
2167 | ||
2168 | @item -Wall | |
2169 | @opindex Wall | |
2170 | All of the above @samp{-W} options combined. This enables all the | |
2171 | warnings about constructions that some users consider questionable, and | |
2172 | that are easy to avoid (or modify to prevent the warning), even in | |
2173 | conjunction with macros. | |
2174 | ||
2175 | @item -Wdiv-by-zero | |
2176 | @opindex Wno-div-by-zero | |
2177 | @opindex Wdiv-by-zero | |
2178 | Warn about compile-time integer division by zero. This is default. To | |
2179 | inhibit the warning messages, use @option{-Wno-div-by-zero}. Floating | |
2180 | point division by zero is not warned about, as it can be a legitimate | |
2181 | way of obtaining infinities and NaNs. | |
2182 | ||
2183 | @item -Wmultichar | |
2184 | @opindex Wno-multichar | |
2185 | @opindex Wmultichar | |
2186 | Warn if a multicharacter constant (@samp{'FOOF'}) is used. This is | |
2187 | default. To inhibit the warning messages, use @option{-Wno-multichar}. | |
2188 | Usually they indicate a typo in the user's code, as they have | |
2189 | implementation-defined values, and should not be used in portable code. | |
2190 | ||
2191 | @item -Wsystem-headers | |
2192 | @opindex Wsystem-headers | |
2193 | @cindex warnings from system headers | |
2194 | @cindex system headers, warnings from | |
2195 | Print warning messages for constructs found in system header files. | |
2196 | Warnings from system headers are normally suppressed, on the assumption | |
2197 | that they usually do not indicate real problems and would only make the | |
2198 | compiler output harder to read. Using this command line option tells | |
2199 | GCC to emit warnings from system headers as if they occurred in user | |
2200 | code. However, note that using @option{-Wall} in conjunction with this | |
2201 | option will @emph{not} warn about unknown pragmas in system | |
2202 | headers---for that, @option{-Wunknown-pragmas} must also be used. | |
2203 | @end table | |
2204 | ||
2205 | The following @option{-W@dots{}} options are not implied by @option{-Wall}. | |
2206 | Some of them warn about constructions that users generally do not | |
2207 | consider questionable, but which occasionally you might wish to check | |
2208 | for; others warn about constructions that are necessary or hard to avoid | |
2209 | in some cases, and there is no simple way to modify the code to suppress | |
2210 | the warning. | |
2211 | ||
2212 | @table @gcctabopt | |
2213 | @item -W | |
2214 | @opindex W | |
2215 | Print extra warning messages for these events: | |
2216 | ||
2217 | @itemize @bullet | |
2218 | @item | |
2219 | A function can return either with or without a value. (Falling | |
2220 | off the end of the function body is considered returning without | |
2221 | a value.) For example, this function would evoke such a | |
2222 | warning: | |
2223 | ||
2224 | @smallexample | |
2225 | @group | |
2226 | foo (a) | |
2227 | @{ | |
2228 | if (a > 0) | |
2229 | return a; | |
2230 | @} | |
2231 | @end group | |
2232 | @end smallexample | |
2233 | ||
2234 | @item | |
2235 | An expression-statement or the left-hand side of a comma expression | |
2236 | contains no side effects. | |
2237 | To suppress the warning, cast the unused expression to void. | |
2238 | For example, an expression such as @samp{x[i,j]} will cause a warning, | |
2239 | but @samp{x[(void)i,j]} will not. | |
2240 | ||
2241 | @item | |
2242 | An unsigned value is compared against zero with @samp{<} or @samp{<=}. | |
2243 | ||
2244 | @item | |
2245 | A comparison like @samp{x<=y<=z} appears; this is equivalent to | |
2246 | @samp{(x<=y ? 1 : 0) <= z}, which is a different interpretation from | |
2247 | that of ordinary mathematical notation. | |
2248 | ||
2249 | @item | |
2250 | Storage-class specifiers like @code{static} are not the first things in | |
2251 | a declaration. According to the C Standard, this usage is obsolescent. | |
2252 | ||
2253 | @item | |
2254 | The return type of a function has a type qualifier such as @code{const}. | |
2255 | Such a type qualifier has no effect, since the value returned by a | |
2256 | function is not an lvalue. (But don't warn about the GNU extension of | |
2257 | @code{volatile void} return types. That extension will be warned about | |
2258 | if @option{-pedantic} is specified.) | |
2259 | ||
2260 | @item | |
2261 | If @option{-Wall} or @option{-Wunused} is also specified, warn about unused | |
2262 | arguments. | |
2263 | ||
2264 | @item | |
2265 | A comparison between signed and unsigned values could produce an | |
2266 | incorrect result when the signed value is converted to unsigned. | |
2267 | (But don't warn if @option{-Wno-sign-compare} is also specified.) | |
2268 | ||
2269 | @item | |
2270 | An aggregate has a partly bracketed initializer. | |
2271 | For example, the following code would evoke such a warning, | |
2272 | because braces are missing around the initializer for @code{x.h}: | |
2273 | ||
2274 | @smallexample | |
2275 | struct s @{ int f, g; @}; | |
2276 | struct t @{ struct s h; int i; @}; | |
2277 | struct t x = @{ 1, 2, 3 @}; | |
2278 | @end smallexample | |
2279 | ||
2280 | @item | |
2281 | An aggregate has an initializer which does not initialize all members. | |
2282 | For example, the following code would cause such a warning, because | |
2283 | @code{x.h} would be implicitly initialized to zero: | |
2284 | ||
2285 | @smallexample | |
2286 | struct s @{ int f, g, h; @}; | |
2287 | struct s x = @{ 3, 4 @}; | |
2288 | @end smallexample | |
2289 | @end itemize | |
2290 | ||
2291 | @item -Wfloat-equal | |
2292 | @opindex Wfloat-equal | |
2293 | Warn if floating point values are used in equality comparisons. | |
2294 | ||
2295 | The idea behind this is that sometimes it is convenient (for the | |
2296 | programmer) to consider floating-point values as approximations to | |
2297 | infinitely precise real numbers. If you are doing this, then you need | |
2298 | to compute (by analysing the code, or in some other way) the maximum or | |
2299 | likely maximum error that the computation introduces, and allow for it | |
2300 | when performing comparisons (and when producing output, but that's a | |
2301 | different problem). In particular, instead of testing for equality, you | |
2302 | would check to see whether the two values have ranges that overlap; and | |
2303 | this is done with the relational operators, so equality comparisons are | |
2304 | probably mistaken. | |
2305 | ||
2306 | @item -Wtraditional @r{(C only)} | |
2307 | @opindex Wtraditional | |
2308 | Warn about certain constructs that behave differently in traditional and | |
2309 | ISO C@. Also warn about ISO C constructs that have no traditional C | |
2310 | equivalent, and/or problematic constructs which should be avoided. | |
2311 | ||
2312 | @itemize @bullet | |
2313 | @item | |
2314 | Macro parameters that appear within string literals in the macro body. | |
2315 | In traditional C macro replacement takes place within string literals, | |
2316 | but does not in ISO C@. | |
2317 | ||
2318 | @item | |
2319 | In traditional C, some preprocessor directives did not exist. | |
2320 | Traditional preprocessors would only consider a line to be a directive | |
2321 | if the @samp{#} appeared in column 1 on the line. Therefore | |
2322 | @option{-Wtraditional} warns about directives that traditional C | |
2323 | understands but would ignore because the @samp{#} does not appear as the | |
2324 | first character on the line. It also suggests you hide directives like | |
2325 | @samp{#pragma} not understood by traditional C by indenting them. Some | |
2326 | traditional implementations would not recognize @samp{#elif}, so it | |
2327 | suggests avoiding it altogether. | |
2328 | ||
2329 | @item | |
2330 | A function-like macro that appears without arguments. | |
2331 | ||
2332 | @item | |
2333 | The unary plus operator. | |
2334 | ||
2335 | @item | |
2336 | The @samp{U} integer constant suffix, or the @samp{F} or @samp{L} floating point | |
2337 | constant suffixes. (Traditional C does support the @samp{L} suffix on integer | |
2338 | constants.) Note, these suffixes appear in macros defined in the system | |
2339 | headers of most modern systems, e.g.@: the @samp{_MIN}/@samp{_MAX} macros in @code{<limits.h>}. | |
2340 | Use of these macros in user code might normally lead to spurious | |
2341 | warnings, however gcc's integrated preprocessor has enough context to | |
2342 | avoid warning in these cases. | |
2343 | ||
2344 | @item | |
2345 | A function declared external in one block and then used after the end of | |
2346 | the block. | |
2347 | ||
2348 | @item | |
2349 | A @code{switch} statement has an operand of type @code{long}. | |
2350 | ||
2351 | @item | |
2352 | A non-@code{static} function declaration follows a @code{static} one. | |
2353 | This construct is not accepted by some traditional C compilers. | |
2354 | ||
2355 | @item | |
2356 | The ISO type of an integer constant has a different width or | |
2357 | signedness from its traditional type. This warning is only issued if | |
2358 | the base of the constant is ten. I.e.@: hexadecimal or octal values, which | |
2359 | typically represent bit patterns, are not warned about. | |
2360 | ||
2361 | @item | |
2362 | Usage of ISO string concatenation is detected. | |
2363 | ||
2364 | @item | |
2365 | Initialization of automatic aggregates. | |
2366 | ||
2367 | @item | |
2368 | Identifier conflicts with labels. Traditional C lacks a separate | |
2369 | namespace for labels. | |
2370 | ||
2371 | @item | |
2372 | Initialization of unions. If the initializer is zero, the warning is | |
2373 | omitted. This is done under the assumption that the zero initializer in | |
2374 | user code appears conditioned on e.g.@: @code{__STDC__} to avoid missing | |
2375 | initializer warnings and relies on default initialization to zero in the | |
2376 | traditional C case. | |
2377 | ||
2378 | @item | |
2379 | Conversions by prototypes between fixed/floating point values and vice | |
2380 | versa. The absence of these prototypes when compiling with traditional | |
2381 | C would cause serious problems. This is a subset of the possible | |
2382 | conversion warnings, for the full set use @option{-Wconversion}. | |
2383 | @end itemize | |
2384 | ||
2385 | @item -Wundef | |
2386 | @opindex Wundef | |
2387 | Warn if an undefined identifier is evaluated in an @samp{#if} directive. | |
2388 | ||
2389 | @item -Wshadow | |
2390 | @opindex Wshadow | |
2391 | Warn whenever a local variable shadows another local variable, parameter or | |
2392 | global variable or whenever a built-in function is shadowed. | |
2393 | ||
2394 | @item -Wlarger-than-@var{len} | |
2395 | @opindex Wlarger-than | |
2396 | Warn whenever an object of larger than @var{len} bytes is defined. | |
2397 | ||
2398 | @item -Wpointer-arith | |
2399 | @opindex Wpointer-arith | |
2400 | Warn about anything that depends on the ``size of'' a function type or | |
2401 | of @code{void}. GNU C assigns these types a size of 1, for | |
2402 | convenience in calculations with @code{void *} pointers and pointers | |
2403 | to functions. | |
2404 | ||
2405 | @item -Wbad-function-cast @r{(C only)} | |
2406 | @opindex Wbad-function-cast | |
2407 | Warn whenever a function call is cast to a non-matching type. | |
2408 | For example, warn if @code{int malloc()} is cast to @code{anything *}. | |
2409 | ||
2410 | @item -Wcast-qual | |
2411 | @opindex Wcast-qual | |
2412 | Warn whenever a pointer is cast so as to remove a type qualifier from | |
2413 | the target type. For example, warn if a @code{const char *} is cast | |
2414 | to an ordinary @code{char *}. | |
2415 | ||
2416 | @item -Wcast-align | |
2417 | @opindex Wcast-align | |
2418 | Warn whenever a pointer is cast such that the required alignment of the | |
2419 | target is increased. For example, warn if a @code{char *} is cast to | |
2420 | an @code{int *} on machines where integers can only be accessed at | |
2421 | two- or four-byte boundaries. | |
2422 | ||
2423 | @item -Wwrite-strings | |
2424 | @opindex Wwrite-strings | |
2425 | When compiling C, give string constants the type @code{const | |
2426 | char[@var{length}]} so that | |
2427 | copying the address of one into a non-@code{const} @code{char *} | |
2428 | pointer will get a warning; when compiling C++, warn about the | |
2429 | deprecated conversion from string constants to @code{char *}. | |
2430 | These warnings will help you find at | |
2431 | compile time code that can try to write into a string constant, but | |
2432 | only if you have been very careful about using @code{const} in | |
2433 | declarations and prototypes. Otherwise, it will just be a nuisance; | |
2434 | this is why we did not make @option{-Wall} request these warnings. | |
2435 | ||
2436 | @item -Wconversion | |
2437 | @opindex Wconversion | |
2438 | Warn if a prototype causes a type conversion that is different from what | |
2439 | would happen to the same argument in the absence of a prototype. This | |
2440 | includes conversions of fixed point to floating and vice versa, and | |
2441 | conversions changing the width or signedness of a fixed point argument | |
2442 | except when the same as the default promotion. | |
2443 | ||
2444 | Also, warn if a negative integer constant expression is implicitly | |
2445 | converted to an unsigned type. For example, warn about the assignment | |
2446 | @code{x = -1} if @code{x} is unsigned. But do not warn about explicit | |
2447 | casts like @code{(unsigned) -1}. | |
2448 | ||
2449 | @item -Wsign-compare | |
2450 | @opindex Wsign-compare | |
2451 | @cindex warning for comparison of signed and unsigned values | |
2452 | @cindex comparison of signed and unsigned values, warning | |
2453 | @cindex signed and unsigned values, comparison warning | |
2454 | Warn when a comparison between signed and unsigned values could produce | |
2455 | an incorrect result when the signed value is converted to unsigned. | |
2456 | This warning is also enabled by @option{-W}; to get the other warnings | |
2457 | of @option{-W} without this warning, use @samp{-W -Wno-sign-compare}. | |
2458 | ||
2459 | @item -Waggregate-return | |
2460 | @opindex Waggregate-return | |
2461 | Warn if any functions that return structures or unions are defined or | |
2462 | called. (In languages where you can return an array, this also elicits | |
2463 | a warning.) | |
2464 | ||
2465 | @item -Wstrict-prototypes @r{(C only)} | |
2466 | @opindex Wstrict-prototypes | |
2467 | Warn if a function is declared or defined without specifying the | |
2468 | argument types. (An old-style function definition is permitted without | |
2469 | a warning if preceded by a declaration which specifies the argument | |
2470 | types.) | |
2471 | ||
2472 | @item -Wmissing-prototypes @r{(C only)} | |
2473 | @opindex Wmissing-prototypes | |
2474 | Warn if a global function is defined without a previous prototype | |
2475 | declaration. This warning is issued even if the definition itself | |
2476 | provides a prototype. The aim is to detect global functions that fail | |
2477 | to be declared in header files. | |
2478 | ||
2479 | @item -Wmissing-declarations | |
2480 | @opindex Wmissing-declarations | |
2481 | Warn if a global function is defined without a previous declaration. | |
2482 | Do so even if the definition itself provides a prototype. | |
2483 | Use this option to detect global functions that are not declared in | |
2484 | header files. | |
2485 | ||
2486 | @item -Wmissing-noreturn | |
2487 | @opindex Wmissing-noreturn | |
2488 | Warn about functions which might be candidates for attribute @code{noreturn}. | |
2489 | Note these are only possible candidates, not absolute ones. Care should | |
2490 | be taken to manually verify functions actually do not ever return before | |
2491 | adding the @code{noreturn} attribute, otherwise subtle code generation | |
2492 | bugs could be introduced. You will not get a warning for @code{main} in | |
2493 | hosted C environments. | |
2494 | ||
2495 | @item -Wmissing-format-attribute | |
2496 | @opindex Wmissing-format-attribute | |
2497 | @opindex Wformat | |
2498 | If @option{-Wformat} is enabled, also warn about functions which might be | |
2499 | candidates for @code{format} attributes. Note these are only possible | |
2500 | candidates, not absolute ones. GCC will guess that @code{format} | |
2501 | attributes might be appropriate for any function that calls a function | |
2502 | like @code{vprintf} or @code{vscanf}, but this might not always be the | |
2503 | case, and some functions for which @code{format} attributes are | |
2504 | appropriate may not be detected. This option has no effect unless | |
2505 | @option{-Wformat} is enabled (possibly by @option{-Wall}). | |
2506 | ||
2507 | @item -Wno-deprecated-declarations | |
2508 | @opindex Wno-deprecated-declarations | |
2509 | Do not warn about uses of functions, variables, and types marked as | |
2510 | deprecated by using the @code{deprecated} attribute. | |
2511 | (@pxref{Function Attributes}, @pxref{Variable Attributes}, | |
2512 | @pxref{Type Attributes}.) | |
2513 | ||
2514 | @item -Wpacked | |
2515 | @opindex Wpacked | |
2516 | Warn if a structure is given the packed attribute, but the packed | |
2517 | attribute has no effect on the layout or size of the structure. | |
2518 | Such structures may be mis-aligned for little benefit. For | |
2519 | instance, in this code, the variable @code{f.x} in @code{struct bar} | |
2520 | will be misaligned even though @code{struct bar} does not itself | |
2521 | have the packed attribute: | |
2522 | ||
2523 | @smallexample | |
2524 | @group | |
2525 | struct foo @{ | |
2526 | int x; | |
2527 | char a, b, c, d; | |
2528 | @} __attribute__((packed)); | |
2529 | struct bar @{ | |
2530 | char z; | |
2531 | struct foo f; | |
2532 | @}; | |
2533 | @end group | |
2534 | @end smallexample | |
2535 | ||
2536 | @item -Wpadded | |
2537 | @opindex Wpadded | |
2538 | Warn if padding is included in a structure, either to align an element | |
2539 | of the structure or to align the whole structure. Sometimes when this | |
2540 | happens it is possible to rearrange the fields of the structure to | |
2541 | reduce the padding and so make the structure smaller. | |
2542 | ||
2543 | @item -Wredundant-decls | |
2544 | @opindex Wredundant-decls | |
2545 | Warn if anything is declared more than once in the same scope, even in | |
2546 | cases where multiple declaration is valid and changes nothing. | |
2547 | ||
2548 | @item -Wnested-externs @r{(C only)} | |
2549 | @opindex Wnested-externs | |
2550 | Warn if an @code{extern} declaration is encountered within a function. | |
2551 | ||
2552 | @item -Wunreachable-code | |
2553 | @opindex Wunreachable-code | |
2554 | Warn if the compiler detects that code will never be executed. | |
2555 | ||
2556 | This option is intended to warn when the compiler detects that at | |
2557 | least a whole line of source code will never be executed, because | |
2558 | some condition is never satisfied or because it is after a | |
2559 | procedure that never returns. | |
2560 | ||
2561 | It is possible for this option to produce a warning even though there | |
2562 | are circumstances under which part of the affected line can be executed, | |
2563 | so care should be taken when removing apparently-unreachable code. | |
2564 | ||
2565 | For instance, when a function is inlined, a warning may mean that the | |
2566 | line is unreachable in only one inlined copy of the function. | |
2567 | ||
2568 | This option is not made part of @option{-Wall} because in a debugging | |
2569 | version of a program there is often substantial code which checks | |
2570 | correct functioning of the program and is, hopefully, unreachable | |
2571 | because the program does work. Another common use of unreachable | |
2572 | code is to provide behavior which is selectable at compile-time. | |
2573 | ||
2574 | @item -Winline | |
2575 | @opindex Winline | |
2576 | Warn if a function can not be inlined and it was declared as inline. | |
2577 | ||
2578 | @item -Wlong-long | |
2579 | @opindex Wlong-long | |
2580 | @opindex Wno-long-long | |
2581 | Warn if @samp{long long} type is used. This is default. To inhibit | |
2582 | the warning messages, use @option{-Wno-long-long}. Flags | |
2583 | @option{-Wlong-long} and @option{-Wno-long-long} are taken into account | |
2584 | only when @option{-pedantic} flag is used. | |
2585 | ||
2586 | @item -Wdisabled-optimization | |
2587 | @opindex Wdisabled-optimization | |
2588 | Warn if a requested optimization pass is disabled. This warning does | |
2589 | not generally indicate that there is anything wrong with your code; it | |
2590 | merely indicates that GCC's optimizers were unable to handle the code | |
2591 | effectively. Often, the problem is that your code is too big or too | |
2592 | complex; GCC will refuse to optimize programs when the optimization | |
2593 | itself is likely to take inordinate amounts of time. | |
2594 | ||
2595 | @item -Werror | |
2596 | @opindex Werror | |
2597 | Make all warnings into errors. | |
2598 | @end table | |
2599 | ||
2600 | @node Debugging Options | |
2601 | @section Options for Debugging Your Program or GCC | |
2602 | @cindex options, debugging | |
2603 | @cindex debugging information options | |
2604 | ||
2605 | GCC has various special options that are used for debugging | |
2606 | either your program or GCC: | |
2607 | ||
2608 | @table @gcctabopt | |
2609 | @item -g | |
2610 | @opindex g | |
2611 | Produce debugging information in the operating system's native format | |
2612 | (stabs, COFF, XCOFF, or DWARF)@. GDB can work with this debugging | |
2613 | information. | |
2614 | ||
2615 | On most systems that use stabs format, @option{-g} enables use of extra | |
2616 | debugging information that only GDB can use; this extra information | |
2617 | makes debugging work better in GDB but will probably make other debuggers | |
2618 | crash or | |
2619 | refuse to read the program. If you want to control for certain whether | |
2620 | to generate the extra information, use @option{-gstabs+}, @option{-gstabs}, | |
2621 | @option{-gxcoff+}, @option{-gxcoff}, @option{-gdwarf-1+}, @option{-gdwarf-1}, | |
2622 | or @option{-gvms} (see below). | |
2623 | ||
2624 | Unlike most other C compilers, GCC allows you to use @option{-g} with | |
2625 | @option{-O}. The shortcuts taken by optimized code may occasionally | |
2626 | produce surprising results: some variables you declared may not exist | |
2627 | at all; flow of control may briefly move where you did not expect it; | |
2628 | some statements may not be executed because they compute constant | |
2629 | results or their values were already at hand; some statements may | |
2630 | execute in different places because they were moved out of loops. | |
2631 | ||
2632 | Nevertheless it proves possible to debug optimized output. This makes | |
2633 | it reasonable to use the optimizer for programs that might have bugs. | |
2634 | ||
2635 | The following options are useful when GCC is generated with the | |
2636 | capability for more than one debugging format. | |
2637 | ||
2638 | @item -ggdb | |
2639 | @opindex ggdb | |
2640 | Produce debugging information for use by GDB@. This means to use the | |
2641 | most expressive format available (DWARF 2, stabs, or the native format | |
2642 | if neither of those are supported), including GDB extensions if at all | |
2643 | possible. | |
2644 | ||
2645 | @item -gstabs | |
2646 | @opindex gstabs | |
2647 | Produce debugging information in stabs format (if that is supported), | |
2648 | without GDB extensions. This is the format used by DBX on most BSD | |
2649 | systems. On MIPS, Alpha and System V Release 4 systems this option | |
2650 | produces stabs debugging output which is not understood by DBX or SDB@. | |
2651 | On System V Release 4 systems this option requires the GNU assembler. | |
2652 | ||
2653 | @item -gstabs+ | |
2654 | @opindex gstabs+ | |
2655 | Produce debugging information in stabs format (if that is supported), | |
2656 | using GNU extensions understood only by the GNU debugger (GDB)@. The | |
2657 | use of these extensions is likely to make other debuggers crash or | |
2658 | refuse to read the program. | |
2659 | ||
2660 | @item -gcoff | |
2661 | @opindex gcoff | |
2662 | Produce debugging information in COFF format (if that is supported). | |
2663 | This is the format used by SDB on most System V systems prior to | |
2664 | System V Release 4. | |
2665 | ||
2666 | @item -gxcoff | |
2667 | @opindex gxcoff | |
2668 | Produce debugging information in XCOFF format (if that is supported). | |
2669 | This is the format used by the DBX debugger on IBM RS/6000 systems. | |
2670 | ||
2671 | @item -gxcoff+ | |
2672 | @opindex gxcoff+ | |
2673 | Produce debugging information in XCOFF format (if that is supported), | |
2674 | using GNU extensions understood only by the GNU debugger (GDB)@. The | |
2675 | use of these extensions is likely to make other debuggers crash or | |
2676 | refuse to read the program, and may cause assemblers other than the GNU | |
2677 | assembler (GAS) to fail with an error. | |
2678 | ||
2679 | @item -gdwarf | |
2680 | @opindex gdwarf | |
2681 | Produce debugging information in DWARF version 1 format (if that is | |
2682 | supported). This is the format used by SDB on most System V Release 4 | |
2683 | systems. | |
2684 | ||
2685 | @item -gdwarf+ | |
2686 | @opindex gdwarf+ | |
2687 | Produce debugging information in DWARF version 1 format (if that is | |
2688 | supported), using GNU extensions understood only by the GNU debugger | |
2689 | (GDB)@. The use of these extensions is likely to make other debuggers | |
2690 | crash or refuse to read the program. | |
2691 | ||
2692 | @item -gdwarf-2 | |
2693 | @opindex gdwarf-2 | |
2694 | Produce debugging information in DWARF version 2 format (if that is | |
2695 | supported). This is the format used by DBX on IRIX 6. | |
2696 | ||
2697 | @item -gvms | |
2698 | @opindex gvms | |
2699 | Produce debugging information in VMS debug format (if that is | |
2700 | supported). This is the format used by DEBUG on VMS systems. | |
2701 | ||
2702 | @item -g@var{level} | |
2703 | @itemx -ggdb@var{level} | |
2704 | @itemx -gstabs@var{level} | |
2705 | @itemx -gcoff@var{level} | |
2706 | @itemx -gxcoff@var{level} | |
2707 | @itemx -gdwarf@var{level} | |
2708 | @itemx -gdwarf-2@var{level} | |
2709 | @itemx -gvms@var{level} | |
2710 | Request debugging information and also use @var{level} to specify how | |
2711 | much information. The default level is 2. | |
2712 | ||
2713 | Level 1 produces minimal information, enough for making backtraces in | |
2714 | parts of the program that you don't plan to debug. This includes | |
2715 | descriptions of functions and external variables, but no information | |
2716 | about local variables and no line numbers. | |
2717 | ||
2718 | Level 3 includes extra information, such as all the macro definitions | |
2719 | present in the program. Some debuggers support macro expansion when | |
2720 | you use @option{-g3}. | |
2721 | ||
2722 | @cindex @code{prof} | |
2723 | @item -p | |
2724 | @opindex p | |
2725 | Generate extra code to write profile information suitable for the | |
2726 | analysis program @code{prof}. You must use this option when compiling | |
2727 | the source files you want data about, and you must also use it when | |
2728 | linking. | |
2729 | ||
2730 | @cindex @code{gprof} | |
2731 | @item -pg | |
2732 | @opindex pg | |
2733 | Generate extra code to write profile information suitable for the | |
2734 | analysis program @code{gprof}. You must use this option when compiling | |
2735 | the source files you want data about, and you must also use it when | |
2736 | linking. | |
2737 | ||
2738 | @cindex @code{tcov} | |
2739 | @item -a | |
2740 | @opindex a | |
2741 | Generate extra code to write profile information for basic blocks, which will | |
2742 | record the number of times each basic block is executed, the basic block start | |
2743 | address, and the function name containing the basic block. If @option{-g} is | |
2744 | used, the line number and filename of the start of the basic block will also be | |
2745 | recorded. If not overridden by the machine description, the default action is | |
2746 | to append to the text file @file{bb.out}. | |
2747 | ||
2748 | This data could be analyzed by a program like @code{tcov}. Note, | |
2749 | however, that the format of the data is not what @code{tcov} expects. | |
2750 | Eventually GNU @code{gprof} should be extended to process this data. | |
2751 | ||
2752 | @item -Q | |
2753 | @opindex Q | |
2754 | Makes the compiler print out each function name as it is compiled, and | |
2755 | print some statistics about each pass when it finishes. | |
2756 | ||
2757 | @item -ftime-report | |
2758 | @opindex ftime-report | |
2759 | Makes the compiler print some statistics about the time consumed by each | |
2760 | pass when it finishes. | |
2761 | ||
2762 | @item -fmem-report | |
2763 | @opindex fmem-report | |
2764 | Makes the compiler print some statistics about permanent memory | |
2765 | allocation when it finishes. | |
2766 | ||
2767 | @item -fprofile-arcs | |
2768 | @opindex fprofile-arcs | |
2769 | Instrument @dfn{arcs} during compilation to generate coverage data | |
2770 | or for profile-directed block ordering. During execution the program | |
2771 | records how many times each branch is executed and how many times it is | |
2772 | taken. When the compiled program exits it saves this data to a file | |
2773 | called @file{@var{sourcename}.da} for each source file. | |
2774 | ||
2775 | For profile-directed block ordering, compile the program with | |
2776 | @option{-fprofile-arcs} plus optimization and code generation options, | |
2777 | generate the arc profile information by running the program on a | |
2778 | selected workload, and then compile the program again with the same | |
2779 | optimization and code generation options plus | |
2780 | @option{-fbranch-probabilities} (@pxref{Optimize Options,,Options that | |
2781 | Control Optimization}). | |
2782 | ||
2783 | The other use of @option{-fprofile-arcs} is for use with @code{gcov}, | |
2784 | when it is used with the @option{-ftest-coverage} option. GCC | |
2785 | supports two methods of determining code coverage: the options that | |
2786 | support @code{gcov}, and options @option{-a} and @option{-ax}, which | |
2787 | write information to text files. The options that support @code{gcov} | |
2788 | do not need to instrument every arc in the program, so a program compiled | |
2789 | with them runs faster than a program compiled with @option{-a}, which | |
2790 | adds instrumentation code to every basic block in the program. The | |
2791 | tradeoff: since @code{gcov} does not have execution counts for all | |
2792 | branches, it must start with the execution counts for the instrumented | |
2793 | branches, and then iterate over the program flow graph until the entire | |
2794 | graph has been solved. Hence, @code{gcov} runs a little more slowly than | |
2795 | a program which uses information from @option{-a} and @option{-ax}. | |
2796 | ||
2797 | With @option{-fprofile-arcs}, for each function of your program GCC | |
2798 | creates a program flow graph, then finds a spanning tree for the graph. | |
2799 | Only arcs that are not on the spanning tree have to be instrumented: the | |
2800 | compiler adds code to count the number of times that these arcs are | |
2801 | executed. When an arc is the only exit or only entrance to a block, the | |
2802 | instrumentation code can be added to the block; otherwise, a new basic | |
2803 | block must be created to hold the instrumentation code. | |
2804 | ||
2805 | This option makes it possible to estimate branch probabilities and to | |
2806 | calculate basic block execution counts. In general, basic block | |
2807 | execution counts as provided by @option{-a} do not give enough | |
2808 | information to estimate all branch probabilities. | |
2809 | ||
2810 | @need 2000 | |
2811 | @item -ftest-coverage | |
2812 | @opindex ftest-coverage | |
2813 | Create data files for the @code{gcov} code-coverage utility | |
2814 | (@pxref{Gcov,, @code{gcov}: a GCC Test Coverage Program}). | |
2815 | The data file names begin with the name of your source file: | |
2816 | ||
2817 | @table @gcctabopt | |
2818 | @item @var{sourcename}.bb | |
2819 | A mapping from basic blocks to line numbers, which @code{gcov} uses to | |
2820 | associate basic block execution counts with line numbers. | |
2821 | ||
2822 | @item @var{sourcename}.bbg | |
2823 | A list of all arcs in the program flow graph. This allows @code{gcov} | |
2824 | to reconstruct the program flow graph, so that it can compute all basic | |
2825 | block and arc execution counts from the information in the | |
2826 | @code{@var{sourcename}.da} file. | |
2827 | @end table | |
2828 | ||
2829 | Use @option{-ftest-coverage} with @option{-fprofile-arcs}; the latter | |
2830 | option adds instrumentation to the program, which then writes | |
2831 | execution counts to another data file: | |
2832 | ||
2833 | @table @gcctabopt | |
2834 | @item @var{sourcename}.da | |
2835 | Runtime arc execution counts, used in conjunction with the arc | |
2836 | information in the file @code{@var{sourcename}.bbg}. | |
2837 | @end table | |
2838 | ||
2839 | Coverage data will map better to the source files if | |
2840 | @option{-ftest-coverage} is used without optimization. | |
2841 | ||
2842 | @item -d@var{letters} | |
2843 | @opindex d | |
2844 | Says to make debugging dumps during compilation at times specified by | |
2845 | @var{letters}. This is used for debugging the compiler. The file names | |
2846 | for most of the dumps are made by appending a pass number and a word to | |
2847 | the source file name (e.g. @file{foo.c.00.rtl} or @file{foo.c.01.sibling}). | |
2848 | Here are the possible letters for use in @var{letters}, and their meanings: | |
2849 | ||
2850 | @table @samp | |
2851 | @item A | |
2852 | @opindex dA | |
2853 | Annotate the assembler output with miscellaneous debugging information. | |
2854 | @item b | |
2855 | @opindex db | |
2856 | Dump after computing branch probabilities, to @file{@var{file}.14.bp}. | |
2857 | @item B | |
2858 | @opindex dB | |
2859 | Dump after block reordering, to @file{@var{file}.29.bbro}. | |
2860 | @item c | |
2861 | @opindex dc | |
2862 | Dump after instruction combination, to the file @file{@var{file}.16.combine}. | |
2863 | @item C | |
2864 | @opindex dC | |
2865 | Dump after the first if conversion, to the file @file{@var{file}.17.ce}. | |
2866 | @item d | |
2867 | @opindex dd | |
2868 | Dump after delayed branch scheduling, to @file{@var{file}.31.dbr}. | |
2869 | @item D | |
2870 | @opindex dD | |
2871 | Dump all macro definitions, at the end of preprocessing, in addition to | |
2872 | normal output. | |
2873 | @item e | |
2874 | @opindex de | |
2875 | Dump after SSA optimizations, to @file{@var{file}.04.ssa} and | |
2876 | @file{@var{file}.07.ussa}. | |
2877 | @item E | |
2878 | @opindex dE | |
2879 | Dump after the second if conversion, to @file{@var{file}.26.ce2}. | |
2880 | @item f | |
2881 | @opindex df | |
2882 | Dump after life analysis, to @file{@var{file}.15.life}. | |
2883 | @item F | |
2884 | @opindex dF | |
2885 | Dump after purging @code{ADDRESSOF} codes, to @file{@var{file}.09.addressof}. | |
2886 | @item g | |
2887 | @opindex dg | |
2888 | Dump after global register allocation, to @file{@var{file}.21.greg}. | |
2889 | @item h | |
2890 | @opindex dh | |
2891 | Dump after finalization of EH handling code, to @file{@var{file}.02.eh}. | |
2892 | @item k | |
2893 | @opindex dk | |
2894 | Dump after reg-to-stack conversion, to @file{@var{file}.28.stack}. | |
2895 | @item o | |
2896 | @opindex do | |
2897 | Dump after post-reload optimizations, to @file{@var{file}.22.postreload}. | |
2898 | @item G | |
2899 | @opindex dG | |
2900 | Dump after GCSE, to @file{@var{file}.10.gcse}. | |
2901 | @item i | |
2902 | @opindex di | |
2903 | Dump after sibling call optimizations, to @file{@var{file}.01.sibling}. | |
2904 | @item j | |
2905 | @opindex dj | |
2906 | Dump after the first jump optimization, to @file{@var{file}.03.jump}. | |
2907 | @item k | |
2908 | @opindex dk | |
2909 | Dump after conversion from registers to stack, to @file{@var{file}.32.stack}. | |
2910 | @item l | |
2911 | @opindex dl | |
2912 | Dump after local register allocation, to @file{@var{file}.20.lreg}. | |
2913 | @item L | |
2914 | @opindex dL | |
2915 | Dump after loop optimization, to @file{@var{file}.11.loop}. | |
2916 | @item M | |
2917 | @opindex dM | |
2918 | Dump after performing the machine dependent reorganisation pass, to | |
2919 | @file{@var{file}.30.mach}. | |
2920 | @item n | |
2921 | @opindex dn | |
2922 | Dump after register renumbering, to @file{@var{file}.25.rnreg}. | |
2923 | @item N | |
2924 | @opindex dN | |
2925 | Dump after the register move pass, to @file{@var{file}.18.regmove}. | |
2926 | @item r | |
2927 | @opindex dr | |
2928 | Dump after RTL generation, to @file{@var{file}.00.rtl}. | |
2929 | @item R | |
2930 | @opindex dR | |
2931 | Dump after the second scheduling pass, to @file{@var{file}.27.sched2}. | |
2932 | @item s | |
2933 | @opindex ds | |
2934 | Dump after CSE (including the jump optimization that sometimes follows | |
2935 | CSE), to @file{@var{file}.08.cse}. | |
2936 | @item S | |
2937 | @opindex dS | |
2938 | Dump after the first scheduling pass, to @file{@var{file}.19.sched}. | |
2939 | @item t | |
2940 | @opindex dt | |
2941 | Dump after the second CSE pass (including the jump optimization that | |
2942 | sometimes follows CSE), to @file{@var{file}.12.cse2}. | |
2943 | @item w | |
2944 | @opindex dw | |
2945 | Dump after the second flow pass, to @file{@var{file}.23.flow2}. | |
2946 | @item X | |
2947 | @opindex dX | |
2948 | Dump after SSA dead code elimination, to @file{@var{file}.06.ssadce}. | |
2949 | @item z | |
2950 | @opindex dz | |
2951 | Dump after the peephole pass, to @file{@var{file}.24.peephole2}. | |
2952 | @item a | |
2953 | @opindex da | |
2954 | Produce all the dumps listed above. | |
2955 | @item m | |
2956 | @opindex dm | |
2957 | Print statistics on memory usage, at the end of the run, to | |
2958 | standard error. | |
2959 | @item p | |
2960 | @opindex dp | |
2961 | Annotate the assembler output with a comment indicating which | |
2962 | pattern and alternative was used. The length of each instruction is | |
2963 | also printed. | |
2964 | @item P | |
2965 | @opindex dP | |
2966 | Dump the RTL in the assembler output as a comment before each instruction. | |
2967 | Also turns on @option{-dp} annotation. | |
2968 | @item v | |
2969 | @opindex dv | |
2970 | For each of the other indicated dump files (except for | |
2971 | @file{@var{file}.00.rtl}), dump a representation of the control flow graph | |
2972 | suitable for viewing with VCG to @file{@var{file}.@var{pass}.vcg}. | |
2973 | @item x | |
2974 | @opindex dx | |
2975 | Just generate RTL for a function instead of compiling it. Usually used | |
2976 | with @samp{r}. | |
2977 | @item y | |
2978 | @opindex dy | |
2979 | Dump debugging information during parsing, to standard error. | |
2980 | @end table | |
2981 | ||
2982 | @item -fdump-unnumbered | |
2983 | @opindex fdump-unnumbered | |
2984 | When doing debugging dumps (see @option{-d} option above), suppress instruction | |
2985 | numbers and line number note output. This makes it more feasible to | |
2986 | use diff on debugging dumps for compiler invocations with different | |
2987 | options, in particular with and without @option{-g}. | |
2988 | ||
2989 | @item -fdump-translation-unit @r{(C and C++ only)} | |
2990 | @itemx -fdump-translation-unit-@var{options} @r{(C and C++ only)} | |
2991 | @opindex fdump-translation-unit | |
2992 | Dump a representation of the tree structure for the entire translation | |
2993 | unit to a file. The file name is made by appending @file{.tu} to the | |
2994 | source file name. If the @samp{-@var{options}} form is used, @var{options} | |
2995 | controls the details of the dump as described for the | |
2996 | @option{-fdump-tree} options. | |
2997 | ||
2998 | @item -fdump-class-hierarchy @r{(C++ only)} | |
2999 | @itemx -fdump-class-hierarchy-@var{options} @r{(C++ only)} | |
3000 | @opindex fdump-class-hierarchy | |
3001 | Dump a representation of each class's hierarchy and virtual function | |
3002 | table layout to a file. The file name is made by appending @file{.class} | |
3003 | to the source file name. If the @samp{-@var{options}} form is used, | |
3004 | @var{options} controls the details of the dump as described for the | |
3005 | @option{-fdump-tree} options. | |
3006 | ||
3007 | @item -fdump-tree-@var{switch} @r{(C++ only)} | |
3008 | @itemx -fdump-tree-@var{switch}-@var{options} @r{(C++ only)} | |
3009 | @opindex fdump-tree | |
3010 | Control the dumping at various stages of processing the intermediate | |
3011 | language tree to a file. The file name is generated by appending a switch | |
3012 | specific suffix to the source file name. If the @samp{-@var{options}} | |
3013 | form is used, @var{options} is a list of @samp{-} separated options that | |
3014 | control the details of the dump. Not all options are applicable to all | |
3015 | dumps, those which are not meaningful will be ignored. The following | |
3016 | options are available | |
3017 | ||
3018 | @table @samp | |
3019 | @item address | |
3020 | Print the address of each node. Usually this is not meaningful as it | |
3021 | changes according to the environment and source file. Its primary use | |
3022 | is for tying up a dump file with a debug environment. | |
3023 | @item slim | |
3024 | Inhibit dumping of members of a scope or body of a function merely | |
3025 | because that scope has been reached. Only dump such items when they | |
3026 | are directly reachable by some other path. | |
3027 | @item all | |
3028 | Turn on all options. | |
3029 | @end table | |
3030 | ||
3031 | The following tree dumps are possible: | |
3032 | @table @samp | |
3033 | @item original | |
3034 | Dump before any tree based optimization, to @file{@var{file}.original}. | |
3035 | @item optimized | |
3036 | Dump after all tree based optimization, to @file{@var{file}.optimized}. | |
3037 | @item inlined | |
3038 | Dump after function inlining, to @file{@var{file}.inlined}. | |
3039 | @end table | |
3040 | ||
3041 | @item -fpretend-float | |
3042 | @opindex fpretend-float | |
3043 | When running a cross-compiler, pretend that the target machine uses the | |
3044 | same floating point format as the host machine. This causes incorrect | |
3045 | output of the actual floating constants, but the actual instruction | |
3046 | sequence will probably be the same as GCC would make when running on | |
3047 | the target machine. | |
3048 | ||
3049 | @item -save-temps | |
3050 | @opindex save-temps | |
3051 | Store the usual ``temporary'' intermediate files permanently; place them | |
3052 | in the current directory and name them based on the source file. Thus, | |
3053 | compiling @file{foo.c} with @samp{-c -save-temps} would produce files | |
3054 | @file{foo.i} and @file{foo.s}, as well as @file{foo.o}. This creates a | |
3055 | preprocessed @file{foo.i} output file even though the compiler now | |
3056 | normally uses an integrated preprocessor. | |
3057 | ||
3058 | @item -time | |
3059 | @opindex time | |
3060 | Report the CPU time taken by each subprocess in the compilation | |
3061 | sequence. For C source files, this is the compiler proper and assembler | |
3062 | (plus the linker if linking is done). The output looks like this: | |
3063 | ||
3064 | @smallexample | |
3065 | # cc1 0.12 0.01 | |
3066 | # as 0.00 0.01 | |
3067 | @end smallexample | |
3068 | ||
3069 | The first number on each line is the ``user time,'' that is time spent | |
3070 | executing the program itself. The second number is ``system time,'' | |
3071 | time spent executing operating system routines on behalf of the program. | |
3072 | Both numbers are in seconds. | |
3073 | ||
3074 | @item -print-file-name=@var{library} | |
3075 | @opindex print-file-name | |
3076 | Print the full absolute name of the library file @var{library} that | |
3077 | would be used when linking---and don't do anything else. With this | |
3078 | option, GCC does not compile or link anything; it just prints the | |
3079 | file name. | |
3080 | ||
3081 | @item -print-multi-directory | |
3082 | @opindex print-multi-directory | |
3083 | Print the directory name corresponding to the multilib selected by any | |
3084 | other switches present in the command line. This directory is supposed | |
3085 | to exist in @env{GCC_EXEC_PREFIX}. | |
3086 | ||
3087 | @item -print-multi-lib | |
3088 | @opindex print-multi-lib | |
3089 | Print the mapping from multilib directory names to compiler switches | |
3090 | that enable them. The directory name is separated from the switches by | |
3091 | @samp{;}, and each switch starts with an @samp{@@} instead of the | |
3092 | @samp{-}, without spaces between multiple switches. This is supposed to | |
3093 | ease shell-processing. | |
3094 | ||
3095 | @item -print-prog-name=@var{program} | |
3096 | @opindex print-prog-name | |
3097 | Like @option{-print-file-name}, but searches for a program such as @samp{cpp}. | |
3098 | ||
3099 | @item -print-libgcc-file-name | |
3100 | @opindex print-libgcc-file-name | |
3101 | Same as @option{-print-file-name=libgcc.a}. | |
3102 | ||
3103 | This is useful when you use @option{-nostdlib} or @option{-nodefaultlibs} | |
3104 | but you do want to link with @file{libgcc.a}. You can do | |
3105 | ||
3106 | @example | |
3107 | gcc -nostdlib @var{files}@dots{} `gcc -print-libgcc-file-name` | |
3108 | @end example | |
3109 | ||
3110 | @item -print-search-dirs | |
3111 | @opindex print-search-dirs | |
3112 | Print the name of the configured installation directory and a list of | |
3113 | program and library directories gcc will search---and don't do anything else. | |
3114 | ||
3115 | This is useful when gcc prints the error message | |
3116 | @samp{installation problem, cannot exec cpp0: No such file or directory}. | |
3117 | To resolve this you either need to put @file{cpp0} and the other compiler | |
3118 | components where gcc expects to find them, or you can set the environment | |
3119 | variable @env{GCC_EXEC_PREFIX} to the directory where you installed them. | |
3120 | Don't forget the trailing '/'. | |
3121 | @xref{Environment Variables}. | |
3122 | ||
3123 | @item -dumpmachine | |
3124 | @opindex dumpmachine | |
3125 | Print the compiler's target machine (for example, | |
3126 | @samp{i686-pc-linux-gnu})---and don't do anything else. | |
3127 | ||
3128 | @item -dumpversion | |
3129 | @opindex dumpversion | |
3130 | Print the compiler version (for example, @samp{3.0})---and don't do | |
3131 | anything else. | |
3132 | ||
3133 | @item -dumpspecs | |
3134 | @opindex dumpspecs | |
3135 | Print the compiler's built-in specs---and don't do anything else. (This | |
3136 | is used when GCC itself is being built.) @xref{Spec Files}. | |
3137 | @end table | |
3138 | ||
3139 | @node Optimize Options | |
3140 | @section Options That Control Optimization | |
3141 | @cindex optimize options | |
3142 | @cindex options, optimization | |
3143 | ||
3144 | These options control various sorts of optimizations: | |
3145 | ||
3146 | @table @gcctabopt | |
3147 | @item -O | |
3148 | @itemx -O1 | |
3149 | @opindex O | |
3150 | @opindex O1 | |
3151 | Optimize. Optimizing compilation takes somewhat more time, and a lot | |
3152 | more memory for a large function. | |
3153 | ||
3154 | Without @option{-O}, the compiler's goal is to reduce the cost of | |
3155 | compilation and to make debugging produce the expected results. | |
3156 | Statements are independent: if you stop the program with a breakpoint | |
3157 | between statements, you can then assign a new value to any variable or | |
3158 | change the program counter to any other statement in the function and | |
3159 | get exactly the results you would expect from the source code. | |
3160 | ||
3161 | With @option{-O}, the compiler tries to reduce code size and execution | |
3162 | time, without performing any optimizations that take a great deal of | |
3163 | compilation time. | |
3164 | ||
3165 | @item -O2 | |
3166 | @opindex O2 | |
3167 | Optimize even more. GCC performs nearly all supported optimizations | |
3168 | that do not involve a space-speed tradeoff. The compiler does not | |
3169 | perform loop unrolling or function inlining when you specify @option{-O2}. | |
3170 | As compared to @option{-O}, this option increases both compilation time | |
3171 | and the performance of the generated code. | |
3172 | ||
3173 | @option{-O2} turns on all optional optimizations except for loop unrolling, | |
3174 | function inlining, and register renaming. It also turns on the | |
3175 | @option{-fforce-mem} option on all machines and frame pointer elimination | |
3176 | on machines where doing so does not interfere with debugging. | |
3177 | ||
3178 | Please note the warning under @option{-fgcse} about | |
3179 | invoking @option{-O2} on programs that use computed gotos. | |
3180 | ||
3181 | @item -O3 | |
3182 | @opindex O3 | |
3183 | Optimize yet more. @option{-O3} turns on all optimizations specified by | |
3184 | @option{-O2} and also turns on the @option{-finline-functions} and | |
3185 | @option{-frename-registers} options. | |
3186 | ||
3187 | @item -O0 | |
3188 | @opindex O0 | |
3189 | Do not optimize. | |
3190 | ||
3191 | @item -Os | |
3192 | @opindex Os | |
3193 | Optimize for size. @option{-Os} enables all @option{-O2} optimizations that | |
3194 | do not typically increase code size. It also performs further | |
3195 | optimizations designed to reduce code size. | |
3196 | ||
3197 | If you use multiple @option{-O} options, with or without level numbers, | |
3198 | the last such option is the one that is effective. | |
3199 | @end table | |
3200 | ||
3201 | Options of the form @option{-f@var{flag}} specify machine-independent | |
3202 | flags. Most flags have both positive and negative forms; the negative | |
3203 | form of @option{-ffoo} would be @option{-fno-foo}. In the table below, | |
3204 | only one of the forms is listed---the one which is not the default. | |
3205 | You can figure out the other form by either removing @samp{no-} or | |
3206 | adding it. | |
3207 | ||
3208 | @table @gcctabopt | |
3209 | @item -ffloat-store | |
3210 | @opindex ffloat-store | |
3211 | Do not store floating point variables in registers, and inhibit other | |
3212 | options that might change whether a floating point value is taken from a | |
3213 | register or memory. | |
3214 | ||
3215 | @cindex floating point precision | |
3216 | This option prevents undesirable excess precision on machines such as | |
3217 | the 68000 where the floating registers (of the 68881) keep more | |
3218 | precision than a @code{double} is supposed to have. Similarly for the | |
3219 | x86 architecture. For most programs, the excess precision does only | |
3220 | good, but a few programs rely on the precise definition of IEEE floating | |
3221 | point. Use @option{-ffloat-store} for such programs, after modifying | |
3222 | them to store all pertinent intermediate computations into variables. | |
3223 | ||
3224 | @item -fno-default-inline | |
3225 | @opindex fno-default-inline | |
3226 | Do not make member functions inline by default merely because they are | |
3227 | defined inside the class scope (C++ only). Otherwise, when you specify | |
3228 | @w{@option{-O}}, member functions defined inside class scope are compiled | |
3229 | inline by default; i.e., you don't need to add @samp{inline} in front of | |
3230 | the member function name. | |
3231 | ||
3232 | @item -fno-defer-pop | |
3233 | @opindex fno-defer-pop | |
3234 | Always pop the arguments to each function call as soon as that function | |
3235 | returns. For machines which must pop arguments after a function call, | |
3236 | the compiler normally lets arguments accumulate on the stack for several | |
3237 | function calls and pops them all at once. | |
3238 | ||
3239 | @item -fforce-mem | |
3240 | @opindex fforce-mem | |
3241 | Force memory operands to be copied into registers before doing | |
3242 | arithmetic on them. This produces better code by making all memory | |
3243 | references potential common subexpressions. When they are not common | |
3244 | subexpressions, instruction combination should eliminate the separate | |
3245 | register-load. The @option{-O2} option turns on this option. | |
3246 | ||
3247 | @item -fforce-addr | |
3248 | @opindex fforce-addr | |
3249 | Force memory address constants to be copied into registers before | |
3250 | doing arithmetic on them. This may produce better code just as | |
3251 | @option{-fforce-mem} may. | |
3252 | ||
3253 | @item -fomit-frame-pointer | |
3254 | @opindex fomit-frame-pointer | |
3255 | Don't keep the frame pointer in a register for functions that | |
3256 | don't need one. This avoids the instructions to save, set up and | |
3257 | restore frame pointers; it also makes an extra register available | |
3258 | in many functions. @strong{It also makes debugging impossible on | |
3259 | some machines.} | |
3260 | ||
3261 | On some machines, such as the VAX, this flag has no effect, because | |
3262 | the standard calling sequence automatically handles the frame pointer | |
3263 | and nothing is saved by pretending it doesn't exist. The | |
3264 | machine-description macro @code{FRAME_POINTER_REQUIRED} controls | |
3265 | whether a target machine supports this flag. @xref{Registers,,Register | |
3266 | Usage, gccint, GNU Compiler Collection (GCC) Internals}. | |
3267 | ||
3268 | @item -foptimize-sibling-calls | |
3269 | @opindex foptimize-sibling-calls | |
3270 | Optimize sibling and tail recursive calls. | |
3271 | ||
3272 | @item -ftrapv | |
3273 | @opindex ftrapv | |
3274 | This option generates traps for signed overflow on addition, subtraction, | |
3275 | multiplication operations. | |
3276 | ||
3277 | @item -fno-inline | |
3278 | @opindex fno-inline | |
3279 | Don't pay attention to the @code{inline} keyword. Normally this option | |
3280 | is used to keep the compiler from expanding any functions inline. | |
3281 | Note that if you are not optimizing, no functions can be expanded inline. | |
3282 | ||
3283 | @item -finline-functions | |
3284 | @opindex finline-functions | |
3285 | Integrate all simple functions into their callers. The compiler | |
3286 | heuristically decides which functions are simple enough to be worth | |
3287 | integrating in this way. | |
3288 | ||
3289 | If all calls to a given function are integrated, and the function is | |
3290 | declared @code{static}, then the function is normally not output as | |
3291 | assembler code in its own right. | |
3292 | ||
3293 | @item -finline-limit=@var{n} | |
3294 | @opindex finline-limit | |
3295 | By default, gcc limits the size of functions that can be inlined. This flag | |
3296 | allows the control of this limit for functions that are explicitly marked as | |
3297 | inline (ie marked with the inline keyword or defined within the class | |
3298 | definition in c++). @var{n} is the size of functions that can be inlined in | |
3299 | number of pseudo instructions (not counting parameter handling). The default | |
3300 | value of @var{n} is 600. | |
3301 | Increasing this value can result in more inlined code at | |
3302 | the cost of compilation time and memory consumption. Decreasing usually makes | |
3303 | the compilation faster and less code will be inlined (which presumably | |
3304 | means slower programs). This option is particularly useful for programs that | |
3305 | use inlining heavily such as those based on recursive templates with C++. | |
3306 | ||
3307 | @emph{Note:} pseudo instruction represents, in this particular context, an | |
3308 | abstract measurement of function's size. In no way, it represents a count | |
3309 | of assembly instructions and as such its exact meaning might change from one | |
3310 | release to an another. | |
3311 | ||
3312 | @item -fkeep-inline-functions | |
3313 | @opindex fkeep-inline-functions | |
3314 | Even if all calls to a given function are integrated, and the function | |
3315 | is declared @code{static}, nevertheless output a separate run-time | |
3316 | callable version of the function. This switch does not affect | |
3317 | @code{extern inline} functions. | |
3318 | ||
3319 | @item -fkeep-static-consts | |
3320 | @opindex fkeep-static-consts | |
3321 | Emit variables declared @code{static const} when optimization isn't turned | |
3322 | on, even if the variables aren't referenced. | |
3323 | ||
3324 | GCC enables this option by default. If you want to force the compiler to | |
3325 | check if the variable was referenced, regardless of whether or not | |
3326 | optimization is turned on, use the @option{-fno-keep-static-consts} option. | |
3327 | ||
3328 | @item -fmerge-constants | |
3329 | Attempt to merge identical constants (string constants and floating point | |
3330 | constants) accross compilation units. | |
3331 | ||
3332 | This option is default for optimized compilation if assembler and linker | |
3333 | support it. Use @option{-fno-merge-constants} to inhibit this behavior. | |
3334 | ||
3335 | @item -fmerge-all-constants | |
3336 | Attempt to merge identical constants and identical variables. | |
3337 | ||
3338 | This option implies @option{-fmerge-constants}. In addition to | |
3339 | @option{-fmerge-constants} this considers e.g. even constant initialized | |
3340 | arrays or initialized constant variables with integral or floating point | |
3341 | types. Languages like C or C++ require each non-automatic variable to | |
3342 | have distinct location, so using this option will result in non-conforming | |
3343 | behavior. | |
3344 | ||
3345 | @item -fno-function-cse | |
3346 | @opindex fno-function-cse | |
3347 | Do not put function addresses in registers; make each instruction that | |
3348 | calls a constant function contain the function's address explicitly. | |
3349 | ||
3350 | This option results in less efficient code, but some strange hacks | |
3351 | that alter the assembler output may be confused by the optimizations | |
3352 | performed when this option is not used. | |
3353 | ||
3354 | @item -ffast-math | |
3355 | @opindex ffast-math | |
3356 | Sets @option{-fno-math-errno}, @option{-funsafe-math-optimizations}, and @* | |
3357 | @option{-fno-trapping-math}. | |
3358 | ||
3359 | This option causes the preprocessor macro @code{__FAST_MATH__} to be defined. | |
3360 | ||
3361 | This option should never be turned on by any @option{-O} option since | |
3362 | it can result in incorrect output for programs which depend on | |
3363 | an exact implementation of IEEE or ISO rules/specifications for | |
3364 | math functions. | |
3365 | ||
3366 | @item -fno-math-errno | |
3367 | @opindex fno-math-errno | |
3368 | Do not set ERRNO after calling math functions that are executed | |
3369 | with a single instruction, e.g., sqrt. A program that relies on | |
3370 | IEEE exceptions for math error handling may want to use this flag | |
3371 | for speed while maintaining IEEE arithmetic compatibility. | |
3372 | ||
3373 | This option should never be turned on by any @option{-O} option since | |
3374 | it can result in incorrect output for programs which depend on | |
3375 | an exact implementation of IEEE or ISO rules/specifications for | |
3376 | math functions. | |
3377 | ||
3378 | The default is @option{-fmath-errno}. | |
3379 | ||
3380 | @item -funsafe-math-optimizations | |
3381 | @opindex funsafe-math-optimizations | |
3382 | Allow optimizations for floating-point arithmetic that (a) assume | |
3383 | that arguments and results are valid and (b) may violate IEEE or | |
3384 | ANSI standards. When used at link-time, it may include libraries | |
3385 | or startup files that change the default FPU control word or other | |
3386 | similar optimizations. | |
3387 | ||
3388 | This option should never be turned on by any @option{-O} option since | |
3389 | it can result in incorrect output for programs which depend on | |
3390 | an exact implementation of IEEE or ISO rules/specifications for | |
3391 | math functions. | |
3392 | ||
3393 | The default is @option{-fno-unsafe-math-optimizations}. | |
3394 | ||
3395 | @item -fno-trapping-math | |
3396 | @opindex fno-trapping-math | |
3397 | Compile code assuming that floating-point operations cannot generate | |
3398 | user-visible traps. Setting this option may allow faster code | |
3399 | if one relies on ``non-stop'' IEEE arithmetic, for example. | |
3400 | ||
3401 | This option should never be turned on by any @option{-O} option since | |
3402 | it can result in incorrect output for programs which depend on | |
3403 | an exact implementation of IEEE or ISO rules/specifications for | |
3404 | math functions. | |
3405 | ||
3406 | The default is @option{-ftrapping-math}. | |
3407 | @end table | |
3408 | ||
3409 | The following options control specific optimizations. The @option{-O2} | |
3410 | option turns on all of these optimizations except @option{-funroll-loops} | |
3411 | and @option{-funroll-all-loops}. On most machines, the @option{-O} option | |
3412 | turns on the @option{-fthread-jumps} and @option{-fdelayed-branch} options, | |
3413 | but specific machines may handle it differently. | |
3414 | ||
3415 | You can use the following flags in the rare cases when ``fine-tuning'' | |
3416 | of optimizations to be performed is desired. | |
3417 | ||
3418 | Not all of the optimizations performed by GCC have @option{-f} options | |
3419 | to control them. | |
3420 | ||
3421 | @table @gcctabopt | |
3422 | @item -fstrength-reduce | |
3423 | @opindex fstrength-reduce | |
3424 | Perform the optimizations of loop strength reduction and | |
3425 | elimination of iteration variables. | |
3426 | ||
3427 | @item -fthread-jumps | |
3428 | @opindex fthread-jumps | |
3429 | Perform optimizations where we check to see if a jump branches to a | |
3430 | location where another comparison subsumed by the first is found. If | |
3431 | so, the first branch is redirected to either the destination of the | |
3432 | second branch or a point immediately following it, depending on whether | |
3433 | the condition is known to be true or false. | |
3434 | ||
3435 | @item -fcse-follow-jumps | |
3436 | @opindex fcse-follow-jumps | |
3437 | In common subexpression elimination, scan through jump instructions | |
3438 | when the target of the jump is not reached by any other path. For | |
3439 | example, when CSE encounters an @code{if} statement with an | |
3440 | @code{else} clause, CSE will follow the jump when the condition | |
3441 | tested is false. | |
3442 | ||
3443 | @item -fcse-skip-blocks | |
3444 | @opindex fcse-skip-blocks | |
3445 | This is similar to @option{-fcse-follow-jumps}, but causes CSE to | |
3446 | follow jumps which conditionally skip over blocks. When CSE | |
3447 | encounters a simple @code{if} statement with no else clause, | |
3448 | @option{-fcse-skip-blocks} causes CSE to follow the jump around the | |
3449 | body of the @code{if}. | |
3450 | ||
3451 | @item -frerun-cse-after-loop | |
3452 | @opindex frerun-cse-after-loop | |
3453 | Re-run common subexpression elimination after loop optimizations has been | |
3454 | performed. | |
3455 | ||
3456 | @item -frerun-loop-opt | |
3457 | @opindex frerun-loop-opt | |
3458 | Run the loop optimizer twice. | |
3459 | ||
3460 | @item -fgcse | |
3461 | @opindex fgcse | |
3462 | Perform a global common subexpression elimination pass. | |
3463 | This pass also performs global constant and copy propagation. | |
3464 | ||
3465 | @emph{Note:} When compiling a program using computed gotos, a GCC | |
3466 | extension, you may get better runtime performance if you disable | |
3467 | the global common subexpression elmination pass by adding | |
3468 | @option{-fno-gcse} to the command line. | |
3469 | ||
3470 | @item -fgcse-lm | |
3471 | @opindex fgcse-lm | |
3472 | When @option{-fgcse-lm} is enabled, global common subexpression elimination will | |
3473 | attempt to move loads which are only killed by stores into themselves. This | |
3474 | allows a loop containing a load/store sequence to be changed to a load outside | |
3475 | the loop, and a copy/store within the loop. | |
3476 | ||
3477 | @item -fgcse-sm | |
3478 | @opindex fgcse-sm | |
3479 | When @option{-fgcse-sm} is enabled, A store motion pass is run after global common | |
3480 | subexpression elimination. This pass will attempt to move stores out of loops. | |
3481 | When used in conjunction with @option{-fgcse-lm}, loops containing a load/store sequence | |
3482 | can be changed to a load before the loop and a store after the loop. | |
3483 | ||
3484 | @item -fdelete-null-pointer-checks | |
3485 | @opindex fdelete-null-pointer-checks | |
3486 | Use global dataflow analysis to identify and eliminate useless checks | |
3487 | for null pointers. The compiler assumes that dereferencing a null | |
3488 | pointer would have halted the program. If a pointer is checked after | |
3489 | it has already been dereferenced, it cannot be null. | |
3490 | ||
3491 | In some environments, this assumption is not true, and programs can | |
3492 | safely dereference null pointers. Use | |
3493 | @option{-fno-delete-null-pointer-checks} to disable this optimization | |
3494 | for programs which depend on that behavior. | |
3495 | ||
3496 | @item -fexpensive-optimizations | |
3497 | @opindex fexpensive-optimizations | |
3498 | Perform a number of minor optimizations that are relatively expensive. | |
3499 | ||
3500 | @item -foptimize-register-move | |
3501 | @itemx -fregmove | |
3502 | @opindex foptimize-register-move | |
3503 | @opindex fregmove | |
3504 | Attempt to reassign register numbers in move instructions and as | |
3505 | operands of other simple instructions in order to maximize the amount of | |
3506 | register tying. This is especially helpful on machines with two-operand | |
3507 | instructions. GCC enables this optimization by default with @option{-O2} | |
3508 | or higher. | |
3509 | ||
3510 | Note @option{-fregmove} and @option{-foptimize-register-move} are the same | |
3511 | optimization. | |
3512 | ||
3513 | @item -fdelayed-branch | |
3514 | @opindex fdelayed-branch | |
3515 | If supported for the target machine, attempt to reorder instructions | |
3516 | to exploit instruction slots available after delayed branch | |
3517 | instructions. | |
3518 | ||
3519 | @item -fschedule-insns | |
3520 | @opindex fschedule-insns | |
3521 | If supported for the target machine, attempt to reorder instructions to | |
3522 | eliminate execution stalls due to required data being unavailable. This | |
3523 | helps machines that have slow floating point or memory load instructions | |
3524 | by allowing other instructions to be issued until the result of the load | |
3525 | or floating point instruction is required. | |
3526 | ||
3527 | @item -fschedule-insns2 | |
3528 | @opindex fschedule-insns2 | |
3529 | Similar to @option{-fschedule-insns}, but requests an additional pass of | |
3530 | instruction scheduling after register allocation has been done. This is | |
3531 | especially useful on machines with a relatively small number of | |
3532 | registers and where memory load instructions take more than one cycle. | |
3533 | ||
3534 | @item -ffunction-sections | |
3535 | @itemx -fdata-sections | |
3536 | @opindex ffunction-sections | |
3537 | @opindex fdata-sections | |
3538 | Place each function or data item into its own section in the output | |
3539 | file if the target supports arbitrary sections. The name of the | |
3540 | function or the name of the data item determines the section's name | |
3541 | in the output file. | |
3542 | ||
3543 | Use these options on systems where the linker can perform optimizations | |
3544 | to improve locality of reference in the instruction space. HPPA | |
3545 | processors running HP-UX and Sparc processors running Solaris 2 have | |
3546 | linkers with such optimizations. Other systems using the ELF object format | |
3547 | as well as AIX may have these optimizations in the future. | |
3548 | ||
3549 | Only use these options when there are significant benefits from doing | |
3550 | so. When you specify these options, the assembler and linker will | |
3551 | create larger object and executable files and will also be slower. | |
3552 | You will not be able to use @code{gprof} on all systems if you | |
3553 | specify this option and you may have problems with debugging if | |
3554 | you specify both this option and @option{-g}. | |
3555 | ||
3556 | @item -fcaller-saves | |
3557 | @opindex fcaller-saves | |
3558 | Enable values to be allocated in registers that will be clobbered by | |
3559 | function calls, by emitting extra instructions to save and restore the | |
3560 | registers around such calls. Such allocation is done only when it | |
3561 | seems to result in better code than would otherwise be produced. | |
3562 | ||
3563 | This option is always enabled by default on certain machines, usually | |
3564 | those which have no call-preserved registers to use instead. | |
3565 | ||
3566 | For all machines, optimization level 2 and higher enables this flag by | |
3567 | default. | |
3568 | ||
3569 | @item -funroll-loops | |
3570 | @opindex funroll-loops | |
3571 | Unroll loops whose number of iterations can be determined at compile | |
3572 | time or upon entry to the loop. @option{-funroll-loops} implies both | |
3573 | @option{-fstrength-reduce} and @option{-frerun-cse-after-loop}. This | |
3574 | option makes code larger, and may or may not make it run faster. | |
3575 | ||
3576 | @item -funroll-all-loops | |
3577 | @opindex funroll-all-loops | |
3578 | Unroll all loops, even if their number of iterations is uncertain when | |
3579 | the loop is entered. This usually makes programs run more slowly. | |
3580 | @option{-funroll-all-loops} implies the same options as | |
3581 | @option{-funroll-loops}, | |
3582 | ||
3583 | @item -fprefetch-loop-arrays | |
3584 | @opindex fprefetch-loop-arrays | |
3585 | If supported by the target machine, generate instructions to prefetch | |
3586 | memory to improve the performance of loops that access large arrays. | |
3587 | ||
3588 | @item -fmove-all-movables | |
3589 | @opindex fmove-all-movables | |
3590 | Forces all invariant computations in loops to be moved | |
3591 | outside the loop. | |
3592 | ||
3593 | @item -freduce-all-givs | |
3594 | @opindex freduce-all-givs | |
3595 | Forces all general-induction variables in loops to be | |
3596 | strength-reduced. | |
3597 | ||
3598 | @emph{Note:} When compiling programs written in Fortran, | |
3599 | @option{-fmove-all-movables} and @option{-freduce-all-givs} are enabled | |
3600 | by default when you use the optimizer. | |
3601 | ||
3602 | These options may generate better or worse code; results are highly | |
3603 | dependent on the structure of loops within the source code. | |
3604 | ||
3605 | These two options are intended to be removed someday, once | |
3606 | they have helped determine the efficacy of various | |
3607 | approaches to improving loop optimizations. | |
3608 | ||
3609 | Please let us (@w{@email{gcc@@gcc.gnu.org}} and @w{@email{fortran@@gnu.org}}) | |
3610 | know how use of these options affects | |
3611 | the performance of your production code. | |
3612 | We're very interested in code that runs @emph{slower} | |
3613 | when these options are @emph{enabled}. | |
3614 | ||
3615 | @item -fno-peephole | |
3616 | @itemx -fno-peephole2 | |
3617 | @opindex fno-peephole | |
3618 | @opindex fno-peephole2 | |
3619 | Disable any machine-specific peephole optimizations. The difference | |
3620 | between @option{-fno-peephole} and @option{-fno-peephole2} is in how they | |
3621 | are implemented in the compiler; some targets use one, some use the | |
3622 | other, a few use both. | |
3623 | ||
3624 | @item -fbranch-probabilities | |
3625 | @opindex fbranch-probabilities | |
3626 | After running a program compiled with @option{-fprofile-arcs} | |
3627 | (@pxref{Debugging Options,, Options for Debugging Your Program or | |
3628 | @command{gcc}}), you can compile it a second time using | |
3629 | @option{-fbranch-probabilities}, to improve optimizations based on | |
3630 | the number of times each branch was taken. When the program | |
3631 | compiled with @option{-fprofile-arcs} exits it saves arc execution | |
3632 | counts to a file called @file{@var{sourcename}.da} for each source | |
3633 | file The information in this data file is very dependent on the | |
3634 | structure of the generated code, so you must use the same source code | |
3635 | and the same optimization options for both compilations. | |
3636 | ||
3637 | With @option{-fbranch-probabilities}, GCC puts a @samp{REG_EXEC_COUNT} | |
3638 | note on the first instruction of each basic block, and a | |
3639 | @samp{REG_BR_PROB} note on each @samp{JUMP_INSN} and @samp{CALL_INSN}. | |
3640 | These can be used to improve optimization. Currently, they are only | |
3641 | used in one place: in @file{reorg.c}, instead of guessing which path a | |
3642 | branch is mostly to take, the @samp{REG_BR_PROB} values are used to | |
3643 | exactly determine which path is taken more often. | |
3644 | ||
3645 | @item -fno-guess-branch-probability | |
3646 | @opindex fno-guess-branch-probability | |
3647 | Do not guess branch probabilities using a randomized model. | |
3648 | ||
3649 | Sometimes gcc will opt to use a randomized model to guess branch | |
3650 | probabilities, when none are available from either profiling feedback | |
3651 | (@option{-fprofile-arcs}) or @samp{__builtin_expect}. This means that | |
3652 | different runs of the compiler on the same program may produce different | |
3653 | object code. | |
3654 | ||
3655 | In a hard real-time system, people don't want different runs of the | |
3656 | compiler to produce code that has different behavior; minimizing | |
3657 | non-determinism is of paramount import. This switch allows users to | |
3658 | reduce non-determinism, possibly at the expense of inferior | |
3659 | optimization. | |
3660 | ||
3661 | @item -fstrict-aliasing | |
3662 | @opindex fstrict-aliasing | |
3663 | Allows the compiler to assume the strictest aliasing rules applicable to | |
3664 | the language being compiled. For C (and C++), this activates | |
3665 | optimizations based on the type of expressions. In particular, an | |
3666 | object of one type is assumed never to reside at the same address as an | |
3667 | object of a different type, unless the types are almost the same. For | |
3668 | example, an @code{unsigned int} can alias an @code{int}, but not a | |
3669 | @code{void*} or a @code{double}. A character type may alias any other | |
3670 | type. | |
3671 | ||
3672 | Pay special attention to code like this: | |
3673 | @example | |
3674 | union a_union @{ | |
3675 | int i; | |
3676 | double d; | |
3677 | @}; | |
3678 | ||
3679 | int f() @{ | |
3680 | a_union t; | |
3681 | t.d = 3.0; | |
3682 | return t.i; | |
3683 | @} | |
3684 | @end example | |
3685 | The practice of reading from a different union member than the one most | |
3686 | recently written to (called ``type-punning'') is common. Even with | |
3687 | @option{-fstrict-aliasing}, type-punning is allowed, provided the memory | |
3688 | is accessed through the union type. So, the code above will work as | |
3689 | expected. However, this code might not: | |
3690 | @example | |
3691 | int f() @{ | |
3692 | a_union t; | |
3693 | int* ip; | |
3694 | t.d = 3.0; | |
3695 | ip = &t.i; | |
3696 | return *ip; | |
3697 | @} | |
3698 | @end example | |
3699 | ||
3700 | Every language that wishes to perform language-specific alias analysis | |
3701 | should define a function that computes, given an @code{tree} | |
3702 | node, an alias set for the node. Nodes in different alias sets are not | |
3703 | allowed to alias. For an example, see the C front-end function | |
3704 | @code{c_get_alias_set}. | |
3705 | ||
3706 | @item -falign-functions | |
3707 | @itemx -falign-functions=@var{n} | |
3708 | @opindex falign-functions | |
3709 | Align the start of functions to the next power-of-two greater than | |
3710 | @var{n}, skipping up to @var{n} bytes. For instance, | |
3711 | @option{-falign-functions=32} aligns functions to the next 32-byte | |
3712 | boundary, but @option{-falign-functions=24} would align to the next | |
3713 | 32-byte boundary only if this can be done by skipping 23 bytes or less. | |
3714 | ||
3715 | @option{-fno-align-functions} and @option{-falign-functions=1} are | |
3716 | equivalent and mean that functions will not be aligned. | |
3717 | ||
3718 | Some assemblers only support this flag when @var{n} is a power of two; | |
3719 | in that case, it is rounded up. | |
3720 | ||
3721 | If @var{n} is not specified, use a machine-dependent default. | |
3722 | ||
3723 | @item -falign-labels | |
3724 | @itemx -falign-labels=@var{n} | |
3725 | @opindex falign-labels | |
3726 | Align all branch targets to a power-of-two boundary, skipping up to | |
3727 | @var{n} bytes like @option{-falign-functions}. This option can easily | |
3728 | make code slower, because it must insert dummy operations for when the | |
3729 | branch target is reached in the usual flow of the code. | |
3730 | ||
3731 | If @option{-falign-loops} or @option{-falign-jumps} are applicable and | |
3732 | are greater than this value, then their values are used instead. | |
3733 | ||
3734 | If @var{n} is not specified, use a machine-dependent default which is | |
3735 | very likely to be @samp{1}, meaning no alignment. | |
3736 | ||
3737 | @item -falign-loops | |
3738 | @itemx -falign-loops=@var{n} | |
3739 | @opindex falign-loops | |
3740 | Align loops to a power-of-two boundary, skipping up to @var{n} bytes | |
3741 | like @option{-falign-functions}. The hope is that the loop will be | |
3742 | executed many times, which will make up for any execution of the dummy | |
3743 | operations. | |
3744 | ||
3745 | If @var{n} is not specified, use a machine-dependent default. | |
3746 | ||
3747 | @item -falign-jumps | |
3748 | @itemx -falign-jumps=@var{n} | |
3749 | @opindex falign-jumps | |
3750 | Align branch targets to a power-of-two boundary, for branch targets | |
3751 | where the targets can only be reached by jumping, skipping up to @var{n} | |
3752 | bytes like @option{-falign-functions}. In this case, no dummy operations | |
3753 | need be executed. | |
3754 | ||
3755 | If @var{n} is not specified, use a machine-dependent default. | |
3756 | ||
3757 | @item -fssa | |
3758 | @opindex fssa | |
3759 | Perform optimizations in static single assignment form. Each function's | |
3760 | flow graph is translated into SSA form, optimizations are performed, and | |
3761 | the flow graph is translated back from SSA form. Users should not | |
3762 | specify this option, since it is not yet ready for production use. | |
3763 | ||
3764 | @item -fssa-ccp | |
3765 | @opindex fssa-ccp | |
3766 | Perform Sparse Conditional Constant Propagation in SSA form. Requires | |
3767 | @option{-fssa}. Like @option{-fssa}, this is an experimental feature. | |
3768 | ||
3769 | @item -fssa-dce | |
3770 | @opindex fssa-dce | |
3771 | Perform aggressive dead-code elimination in SSA form. Requires @option{-fssa}. | |
3772 | Like @option{-fssa}, this is an experimental feature. | |
3773 | ||
3774 | @item -fsingle-precision-constant | |
3775 | @opindex fsingle-precision-constant | |
3776 | Treat floating point constant as single precision constant instead of | |
3777 | implicitly converting it to double precision constant. | |
3778 | ||
3779 | @item -frename-registers | |
3780 | @opindex frename-registers | |
3781 | Attempt to avoid false dependencies in scheduled code by making use | |
3782 | of registers left over after register allocation. This optimization | |
3783 | will most benefit processors with lots of registers. It can, however, | |
3784 | make debugging impossible, since variables will no longer stay in | |
3785 | a ``home register''. | |
3786 | ||
3787 | @item -fno-cprop-registers | |
3788 | @opindex fno-cprop-registers | |
3789 | After register allocation and post-register allocation instruction splitting, | |
3790 | we perform a copy-propagation pass to try to reduce scheduling dependencies | |
3791 | and occasionally eliminate the copy. | |
3792 | ||
3793 | @item --param @var{name}=@var{value} | |
3794 | @opindex param | |
3795 | In some places, GCC uses various constants to control the amount of | |
3796 | optimization that is done. For example, GCC will not inline functions | |
3797 | that contain more that a certain number of instructions. You can | |
3798 | control some of these constants on the command-line using the | |
3799 | @option{--param} option. | |
3800 | ||
3801 | In each case, the @var{value} is an integer. The allowable choices for | |
3802 | @var{name} are given in the following table: | |
3803 | ||
3804 | @table @gcctabopt | |
3805 | @item max-delay-slot-insn-search | |
3806 | The maximum number of instructions to consider when looking for an | |
3807 | instruction to fill a delay slot. If more than this arbitrary number of | |
3808 | instructions is searched, the time savings from filling the delay slot | |
3809 | will be minimal so stop searching. Increasing values mean more | |
3810 | aggressive optimization, making the compile time increase with probably | |
3811 | small improvement in executable run time. | |
3812 | ||
3813 | @item max-delay-slot-live-search | |
3814 | When trying to fill delay slots, the maximum number of instructions to | |
3815 | consider when searching for a block with valid live register | |
3816 | information. Increasing this arbitrarily chosen value means more | |
3817 | aggressive optimization, increasing the compile time. This parameter | |
3818 | should be removed when the delay slot code is rewritten to maintain the | |
3819 | control-flow graph. | |
3820 | ||
3821 | @item max-gcse-memory | |
3822 | The approximate maximum amount of memory that will be allocated in | |
3823 | order to perform the global common subexpression elimination | |
3824 | optimization. If more memory than specified is required, the | |
3825 | optimization will not be done. | |
3826 | ||
3827 | @item max-gcse-passes | |
3828 | The maximum number of passes of GCSE to run. | |
3829 | ||
3830 | @item max-pending-list-length | |
3831 | The maximum number of pending dependencies scheduling will allow | |
3832 | before flushing the current state and starting over. Large functions | |
3833 | with few branches or calls can create excessively large lists which | |
3834 | needlessly consume memory and resources. | |
3835 | ||
3836 | @item max-inline-insns | |
3837 | If an function contains more than this many instructions, it | |
3838 | will not be inlined. This option is precisely equivalent to | |
3839 | @option{-finline-limit}. | |
3840 | ||
3841 | @end table | |
3842 | @end table | |
3843 | ||
3844 | @node Preprocessor Options | |
3845 | @section Options Controlling the Preprocessor | |
3846 | @cindex preprocessor options | |
3847 | @cindex options, preprocessor | |
3848 | ||
3849 | These options control the C preprocessor, which is run on each C source | |
3850 | file before actual compilation. | |
3851 | ||
3852 | If you use the @option{-E} option, nothing is done except preprocessing. | |
3853 | Some of these options make sense only together with @option{-E} because | |
3854 | they cause the preprocessor output to be unsuitable for actual | |
3855 | compilation. | |
3856 | ||
3857 | @table @gcctabopt | |
3858 | @item -include @var{file} | |
3859 | @opindex include | |
3860 | Process @var{file} as input before processing the regular input file. | |
3861 | In effect, the contents of @var{file} are compiled first. Any @option{-D} | |
3862 | and @option{-U} options on the command line are always processed before | |
3863 | @option{-include @var{file}}, regardless of the order in which they are | |
3864 | written. All the @option{-include} and @option{-imacros} options are | |
3865 | processed in the order in which they are written. | |
3866 | ||
3867 | @item -imacros @var{file} | |
3868 | @opindex imacros | |
3869 | Process @var{file} as input, discarding the resulting output, before | |
3870 | processing the regular input file. Because the output generated from | |
3871 | @var{file} is discarded, the only effect of @option{-imacros @var{file}} | |
3872 | is to make the macros defined in @var{file} available for use in the | |
3873 | main input. All the @option{-include} and @option{-imacros} options are | |
3874 | processed in the order in which they are written. | |
3875 | ||
3876 | @item -idirafter @var{dir} | |
3877 | @opindex idirafter | |
3878 | @cindex second include path | |
3879 | Add the directory @var{dir} to the second include path. The directories | |
3880 | on the second include path are searched when a header file is not found | |
3881 | in any of the directories in the main include path (the one that | |
3882 | @option{-I} adds to). | |
3883 | ||
3884 | @item -iprefix @var{prefix} | |
3885 | @opindex iprefix | |
3886 | Specify @var{prefix} as the prefix for subsequent @option{-iwithprefix} | |
3887 | options. | |
3888 | ||
3889 | @item -iwithprefix @var{dir} | |
3890 | @opindex iwithprefix | |
3891 | Add a directory to the second include path. The directory's name is | |
3892 | made by concatenating @var{prefix} and @var{dir}, where @var{prefix} was | |
3893 | specified previously with @option{-iprefix}. If you have not specified a | |
3894 | prefix yet, the directory containing the installed passes of the | |
3895 | compiler is used as the default. | |
3896 | ||
3897 | @item -iwithprefixbefore @var{dir} | |
3898 | @opindex iwithprefixbefore | |
3899 | Add a directory to the main include path. The directory's name is made | |
3900 | by concatenating @var{prefix} and @var{dir}, as in the case of | |
3901 | @option{-iwithprefix}. | |
3902 | ||
3903 | @item -isystem @var{dir} | |
3904 | @opindex isystem | |
3905 | Add a directory to the beginning of the second include path, marking it | |
3906 | as a system directory, so that it gets the same special treatment as | |
3907 | is applied to the standard system directories. | |
3908 | ||
3909 | @item -nostdinc | |
3910 | @opindex nostdinc | |
3911 | Do not search the standard system directories for header files. Only | |
3912 | the directories you have specified with @option{-I} options (and the | |
3913 | current directory, if appropriate) are searched. @xref{Directory | |
3914 | Options}, for information on @option{-I}. | |
3915 | ||
3916 | By using both @option{-nostdinc} and @option{-I-}, you can limit the include-file | |
3917 | search path to only those directories you specify explicitly. | |
3918 | ||
3919 | @item -remap | |
3920 | @opindex remap | |
3921 | When searching for a header file in a directory, remap file names if a | |
3922 | file named @file{header.gcc} exists in that directory. This can be used | |
3923 | to work around limitations of file systems with file name restrictions. | |
3924 | The @file{header.gcc} file should contain a series of lines with two | |
3925 | tokens on each line: the first token is the name to map, and the second | |
3926 | token is the actual name to use. | |
3927 | ||
3928 | @item -undef | |
3929 | @opindex undef | |
3930 | Do not predefine any nonstandard macros. (Including architecture flags). | |
3931 | ||
3932 | @item -E | |
3933 | @opindex E | |
3934 | Run only the C preprocessor. Preprocess all the C source files | |
3935 | specified and output the results to standard output or to the | |
3936 | specified output file. | |
3937 | ||
3938 | @item -C | |
3939 | @opindex C | |
3940 | Tell the preprocessor not to discard comments. Used with the | |
3941 | @option{-E} option. | |
3942 | ||
3943 | @item -P | |
3944 | @opindex P | |
3945 | Tell the preprocessor not to generate @samp{#line} directives. | |
3946 | Used with the @option{-E} option. | |
3947 | ||
3948 | @cindex make | |
3949 | @cindex dependencies, make | |
3950 | @item -M | |
3951 | @opindex M | |
3952 | Instead of outputting the result of preprocessing, output a rule | |
3953 | suitable for @command{make} describing the dependencies of the main | |
3954 | source file. The preprocessor outputs one @command{make} rule containing | |
3955 | the object file name for that source file, a colon, and the names of all | |
3956 | the included files, including those coming from @option{-include} or | |
3957 | @option{-imacros} command line options. | |
3958 | ||
3959 | Unless specified explicitly (with @option{-MT} or @option{-MQ}), the | |
3960 | object file name consists of the basename of the source file with any | |
3961 | suffix replaced with object file suffix. If there are many included | |
3962 | files then the rule is split into several lines using @samp{\}-newline. | |
3963 | The rule has no commands. | |
3964 | ||
3965 | Passing @option{-M} to the driver implies @option{-E}. | |
3966 | ||
3967 | @item -MM | |
3968 | @opindex MM | |
3969 | Like @option{-M} but do not mention header files that are found in | |
3970 | system header directories, nor header files that are included, | |
3971 | directly or indirectly, from such a header. | |
3972 | ||
3973 | This implies that the choice of angle brackets or double quotes in an | |
3974 | @samp{#include} directive does not in itself determine whether that | |
3975 | header will appear in @option{-MM} dependency output. This is a | |
3976 | slight change in semantics from GCC versions 3.0 and earlier. | |
3977 | ||
3978 | @item -MD | |
3979 | @opindex MD | |
3980 | @option{-MD} is equivalent to @option{-M -MF @var{file}}, except that | |
3981 | @option{-E} is not implied. The driver determines @var{file} based on | |
3982 | whether an @option{-o} option is given. If it is, the driver uses its | |
3983 | argument but with a suffix of @file{.d}, otherwise it take the | |
3984 | basename of the input file and applies a @file{.d} suffix. | |
3985 | ||
3986 | If @option{-MD} is used in conjunction with @option{-E}, any | |
3987 | @option{-o} switch is understood to specify the dependency output file | |
3988 | (but @pxref{-MF}), but if used without @option{-E}, each @option{-o} | |
3989 | is understood to specify a target object file. | |
3990 | ||
3991 | Since @option{-E} is not implied, @option{-MD} can be used to generate | |
3992 | a dependency output file as a side-effect of the compilation process. | |
3993 | ||
3994 | With Mach, you can use the utility @code{md} to merge multiple | |
3995 | dependency files into a single dependency file suitable for using with | |
3996 | the @samp{make} command. | |
3997 | ||
3998 | @item -MMD | |
3999 | @opindex MMD | |
4000 | Like @option{-MD} except mention only user header files, not system | |
4001 | -header files. | |
4002 | ||
4003 | @item -MF @var{file} | |
4004 | @opindex MF | |
4005 | @anchor{-MF} | |
4006 | When used with @option{-M} or @option{-MM}, specifies a | |
4007 | file to write the dependencies to. If no @option{-MF} switch is given | |
4008 | the preprocessor sends the rules to the same place it would have sent | |
4009 | preprocessed output. | |
4010 | ||
4011 | When used with the driver options @option{-MD} or @option{-MMD}, | |
4012 | @option{-MF} overrides the default dependency output file. | |
4013 | ||
4014 | Another way to specify output of a @code{make} rule is by setting | |
4015 | the environment variable @env{DEPENDENCIES_OUTPUT} (@pxref{Environment | |
4016 | Variables}). | |
4017 | ||
4018 | @item -MG | |
4019 | @opindex MG | |
4020 | When used with @option{-M} or @option{-MM}, @option{-MG} says to treat missing | |
4021 | header files as generated files and assume they live in the same | |
4022 | directory as the source file. It suppresses preprocessed output, as a | |
4023 | missing header file is ordinarily an error. | |
4024 | ||
4025 | This feature is used in automatic updating of makefiles. | |
4026 | ||
4027 | @item -MP | |
4028 | @opindex MP | |
4029 | This option instructs CPP to add a phony target for each dependency | |
4030 | other than the main file, causing each to depend on nothing. These | |
4031 | dummy rules work around errors @code{make} gives if you remove header | |
4032 | files without updating the @code{Makefile} to match. | |
4033 | ||
4034 | This is typical output:- | |
4035 | ||
4036 | @smallexample | |
4037 | /tmp/test.o: /tmp/test.c /tmp/test.h | |
4038 | ||
4039 | /tmp/test.h: | |
4040 | @end smallexample | |
4041 | ||
4042 | @item -MQ @var{target} | |
4043 | @item -MT @var{target} | |
4044 | @opindex MQ | |
4045 | @opindex MT | |
4046 | By default CPP uses the main file name, including any path, and appends | |
4047 | the object suffix, normally ``.o'', to it to obtain the name of the | |
4048 | target for dependency generation. With @option{-MT} you can specify a | |
4049 | target yourself, overriding the default one. | |
4050 | ||
4051 | If you want multiple targets, you can specify them as a single argument | |
4052 | to @option{-MT}, or use multiple @option{-MT} options. | |
4053 | ||
4054 | The targets you specify are output in the order they appear on the | |
4055 | command line. @option{-MQ} is identical to @option{-MT}, except that the | |
4056 | target name is quoted for Make, but with @option{-MT} it isn't. For | |
4057 | example, @option{-MT '$(objpfx)foo.o'} gives | |
4058 | ||
4059 | @smallexample | |
4060 | $(objpfx)foo.o: /tmp/foo.c | |
4061 | @end smallexample | |
4062 | ||
4063 | but @option{-MQ '$(objpfx)foo.o'} gives | |
4064 | ||
4065 | @smallexample | |
4066 | $$(objpfx)foo.o: /tmp/foo.c | |
4067 | @end smallexample | |
4068 | ||
4069 | The default target is automatically quoted, as if it were given with | |
4070 | @option{-MQ}. | |
4071 | ||
4072 | @item -H | |
4073 | @opindex H | |
4074 | Print the name of each header file used, in addition to other normal | |
4075 | activities. | |
4076 | ||
4077 | @item -A@var{question}(@var{answer}) | |
4078 | @opindex A | |
4079 | Assert the answer @var{answer} for @var{question}, in case it is tested | |
4080 | with a preprocessing conditional such as @samp{#if | |
4081 | #@var{question}(@var{answer})}. @option{-A-} disables the standard | |
4082 | assertions that normally describe the target machine. | |
4083 | ||
4084 | @item -D@var{macro} | |
4085 | @opindex D | |
4086 | Define macro @var{macro} with the string @samp{1} as its definition. | |
4087 | ||
4088 | @item -D@var{macro}=@var{defn} | |
4089 | Define macro @var{macro} as @var{defn}. All instances of @option{-D} on | |
4090 | the command line are processed before any @option{-U} options. | |
4091 | ||
4092 | Any @option{-D} and @option{-U} options on the command line are processed in | |
4093 | order, and always before @option{-imacros @var{file}}, regardless of the | |
4094 | order in which they are written. | |
4095 | ||
4096 | @item -U@var{macro} | |
4097 | @opindex U | |
4098 | Undefine macro @var{macro}. @option{-U} options are evaluated after all | |
4099 | @option{-D} options, but before any @option{-include} and @option{-imacros} | |
4100 | options. | |
4101 | ||
4102 | Any @option{-D} and @option{-U} options on the command line are processed in | |
4103 | order, and always before @option{-imacros @var{file}}, regardless of the | |
4104 | order in which they are written. | |
4105 | ||
4106 | @item -dM | |
4107 | @opindex dM | |
4108 | Tell the preprocessor to output only a list of the macro definitions | |
4109 | that are in effect at the end of preprocessing. Used with the @option{-E} | |
4110 | option. | |
4111 | ||
4112 | @item -dD | |
4113 | @opindex dD | |
4114 | Tell the preprocessing to pass all macro definitions into the output, in | |
4115 | their proper sequence in the rest of the output. | |
4116 | ||
4117 | @item -dN | |
4118 | @opindex dN | |
4119 | Like @option{-dD} except that the macro arguments and contents are omitted. | |
4120 | Only @samp{#define @var{name}} is included in the output. | |
4121 | ||
4122 | @item -dI | |
4123 | @opindex dI | |
4124 | Output @samp{#include} directives in addition to the result of | |
4125 | preprocessing. | |
4126 | ||
4127 | @item -fpreprocessed | |
4128 | @opindex fpreprocessed | |
4129 | Indicate to the preprocessor that the input file has already been | |
4130 | preprocessed. This suppresses things like macro expansion, trigraph | |
4131 | conversion, escaped newline splicing, and processing of most directives. | |
4132 | The preprocessor still recognizes and removes comments, so that you can | |
4133 | pass a file preprocessed with @option{-C} to the compiler without | |
4134 | problems. In this mode the integrated preprocessor is little more than | |
4135 | a tokenizer for the front ends. | |
4136 | ||
4137 | @option{-fpreprocessed} is implicit if the input file has one of the | |
4138 | extensions @samp{i}, @samp{ii} or @samp{mi}. These are the extensions | |
4139 | that GCC uses for preprocessed files created by @option{-save-temps}. | |
4140 | ||
4141 | @item -trigraphs | |
4142 | @opindex trigraphs | |
4143 | Process ISO standard trigraph sequences. These are three-character | |
4144 | sequences, all starting with @samp{??}, that are defined by ISO C to | |
4145 | stand for single characters. For example, @samp{??/} stands for | |
4146 | @samp{\}, so @samp{'??/n'} is a character constant for a newline. By | |
4147 | default, GCC ignores trigraphs, but in standard-conforming modes it | |
4148 | converts them. See the @option{-std} and @option{-ansi} options. | |
4149 | ||
4150 | The nine trigraph sequences are | |
4151 | @table @samp | |
4152 | @item ??( | |
4153 | @expansion{} @samp{[} | |
4154 | ||
4155 | @item ??) | |
4156 | @expansion{} @samp{]} | |
4157 | ||
4158 | @item ??< | |
4159 | @expansion{} @samp{@{} | |
4160 | ||
4161 | @item ??> | |
4162 | @expansion{} @samp{@}} | |
4163 | ||
4164 | @item ??= | |
4165 | @expansion{} @samp{#} | |
4166 | ||
4167 | @item ??/ | |
4168 | @expansion{} @samp{\} | |
4169 | ||
4170 | @item ??' | |
4171 | @expansion{} @samp{^} | |
4172 | ||
4173 | @item ??! | |
4174 | @expansion{} @samp{|} | |
4175 | ||
4176 | @item ??- | |
4177 | @expansion{} @samp{~} | |
4178 | ||
4179 | @end table | |
4180 | ||
4181 | Trigraph support is not popular, so many compilers do not implement it | |
4182 | properly. Portable code should not rely on trigraphs being either | |
4183 | converted or ignored. | |
4184 | ||
4185 | @item -Wp,@var{option} | |
4186 | @opindex Wp | |
4187 | Pass @var{option} as an option to the preprocessor. If @var{option} | |
4188 | contains commas, it is split into multiple options at the commas. | |
4189 | @end table | |
4190 | ||
4191 | @node Assembler Options | |
4192 | @section Passing Options to the Assembler | |
4193 | ||
4194 | @c prevent bad page break with this line | |
4195 | You can pass options to the assembler. | |
4196 | ||
4197 | @table @gcctabopt | |
4198 | @item -Wa,@var{option} | |
4199 | @opindex Wa | |
4200 | Pass @var{option} as an option to the assembler. If @var{option} | |
4201 | contains commas, it is split into multiple options at the commas. | |
4202 | @end table | |
4203 | ||
4204 | @node Link Options | |
4205 | @section Options for Linking | |
4206 | @cindex link options | |
4207 | @cindex options, linking | |
4208 | ||
4209 | These options come into play when the compiler links object files into | |
4210 | an executable output file. They are meaningless if the compiler is | |
4211 | not doing a link step. | |
4212 | ||
4213 | @table @gcctabopt | |
4214 | @cindex file names | |
4215 | @item @var{object-file-name} | |
4216 | A file name that does not end in a special recognized suffix is | |
4217 | considered to name an object file or library. (Object files are | |
4218 | distinguished from libraries by the linker according to the file | |
4219 | contents.) If linking is done, these object files are used as input | |
4220 | to the linker. | |
4221 | ||
4222 | @item -c | |
4223 | @itemx -S | |
4224 | @itemx -E | |
4225 | @opindex c | |
4226 | @opindex S | |
4227 | @opindex E | |
4228 | If any of these options is used, then the linker is not run, and | |
4229 | object file names should not be used as arguments. @xref{Overall | |
4230 | Options}. | |
4231 | ||
4232 | @cindex Libraries | |
4233 | @item -l@var{library} | |
4234 | @itemx -l @var{library} | |
4235 | @opindex l | |
4236 | Search the library named @var{library} when linking. (The second | |
4237 | alternative with the library as a separate argument is only for | |
4238 | POSIX compliance and is not recommended.) | |
4239 | ||
4240 | It makes a difference where in the command you write this option; the | |
4241 | linker searches and processes libraries and object files in the order they | |
4242 | are specified. Thus, @samp{foo.o -lz bar.o} searches library @samp{z} | |
4243 | after file @file{foo.o} but before @file{bar.o}. If @file{bar.o} refers | |
4244 | to functions in @samp{z}, those functions may not be loaded. | |
4245 | ||
4246 | The linker searches a standard list of directories for the library, | |
4247 | which is actually a file named @file{lib@var{library}.a}. The linker | |
4248 | then uses this file as if it had been specified precisely by name. | |
4249 | ||
4250 | The directories searched include several standard system directories | |
4251 | plus any that you specify with @option{-L}. | |
4252 | ||
4253 | Normally the files found this way are library files---archive files | |
4254 | whose members are object files. The linker handles an archive file by | |
4255 | scanning through it for members which define symbols that have so far | |
4256 | been referenced but not defined. But if the file that is found is an | |
4257 | ordinary object file, it is linked in the usual fashion. The only | |
4258 | difference between using an @option{-l} option and specifying a file name | |
4259 | is that @option{-l} surrounds @var{library} with @samp{lib} and @samp{.a} | |
4260 | and searches several directories. | |
4261 | ||
4262 | @item -lobjc | |
4263 | @opindex lobjc | |
4264 | You need this special case of the @option{-l} option in order to | |
4265 | link an Objective-C program. | |
4266 | ||
4267 | @item -nostartfiles | |
4268 | @opindex nostartfiles | |
4269 | Do not use the standard system startup files when linking. | |
4270 | The standard system libraries are used normally, unless @option{-nostdlib} | |
4271 | or @option{-nodefaultlibs} is used. | |
4272 | ||
4273 | @item -nodefaultlibs | |
4274 | @opindex nodefaultlibs | |
4275 | Do not use the standard system libraries when linking. | |
4276 | Only the libraries you specify will be passed to the linker. | |
4277 | The standard startup files are used normally, unless @option{-nostartfiles} | |
4278 | is used. The compiler may generate calls to memcmp, memset, and memcpy | |
4279 | for System V (and ISO C) environments or to bcopy and bzero for | |
4280 | BSD environments. These entries are usually resolved by entries in | |
4281 | libc. These entry points should be supplied through some other | |
4282 | mechanism when this option is specified. | |
4283 | ||
4284 | @item -nostdlib | |
4285 | @opindex nostdlib | |
4286 | Do not use the standard system startup files or libraries when linking. | |
4287 | No startup files and only the libraries you specify will be passed to | |
4288 | the linker. The compiler may generate calls to memcmp, memset, and memcpy | |
4289 | for System V (and ISO C) environments or to bcopy and bzero for | |
4290 | BSD environments. These entries are usually resolved by entries in | |
4291 | libc. These entry points should be supplied through some other | |
4292 | mechanism when this option is specified. | |
4293 | ||
4294 | @cindex @option{-lgcc}, use with @option{-nostdlib} | |
4295 | @cindex @option{-nostdlib} and unresolved references | |
4296 | @cindex unresolved references and @option{-nostdlib} | |
4297 | @cindex @option{-lgcc}, use with @option{-nodefaultlibs} | |
4298 | @cindex @option{-nodefaultlibs} and unresolved references | |
4299 | @cindex unresolved references and @option{-nodefaultlibs} | |
4300 | One of the standard libraries bypassed by @option{-nostdlib} and | |
4301 | @option{-nodefaultlibs} is @file{libgcc.a}, a library of internal subroutines | |
4302 | that GCC uses to overcome shortcomings of particular machines, or special | |
4303 | needs for some languages. | |
4304 | (@xref{Interface,,Interfacing to GCC Output,gccint,GNU Compiler | |
4305 | Collection (GCC) Internals}, | |
4306 | for more discussion of @file{libgcc.a}.) | |
4307 | In most cases, you need @file{libgcc.a} even when you want to avoid | |
4308 | other standard libraries. In other words, when you specify @option{-nostdlib} | |
4309 | or @option{-nodefaultlibs} you should usually specify @option{-lgcc} as well. | |
4310 | This ensures that you have no unresolved references to internal GCC | |
4311 | library subroutines. (For example, @samp{__main}, used to ensure C++ | |
4312 | constructors will be called; @pxref{Collect2,,@code{collect2}, gccint, | |
4313 | GNU Compiler Collection (GCC) Internals}.) | |
4314 | ||
4315 | @item -s | |
4316 | @opindex s | |
4317 | Remove all symbol table and relocation information from the executable. | |
4318 | ||
4319 | @item -static | |
4320 | @opindex static | |
4321 | On systems that support dynamic linking, this prevents linking with the shared | |
4322 | libraries. On other systems, this option has no effect. | |
4323 | ||
4324 | @item -shared | |
4325 | @opindex shared | |
4326 | Produce a shared object which can then be linked with other objects to | |
4327 | form an executable. Not all systems support this option. For predictable | |
4328 | results, you must also specify the same set of options that were used to | |
4329 | generate code (@option{-fpic}, @option{-fPIC}, or model suboptions) | |
4330 | when you specify this option.@footnote{On some systems, @samp{gcc -shared} | |
4331 | needs to build supplementary stub code for constructors to work. On | |
4332 | multi-libbed systems, @samp{gcc -shared} must select the correct support | |
4333 | libraries to link against. Failing to supply the correct flags may lead | |
4334 | to subtle defects. Supplying them in cases where they are not necessary | |
4335 | is innocuous.} | |
4336 | ||
4337 | @item -shared-libgcc | |
4338 | @itemx -static-libgcc | |
4339 | @opindex shared-libgcc | |
4340 | @opindex static-libgcc | |
4341 | On systems that provide @file{libgcc} as a shared library, these options | |
4342 | force the use of either the shared or static version respectively. | |
4343 | If no shared version of @file{libgcc} was built when the compiler was | |
4344 | configured, these options have no effect. | |
4345 | ||
4346 | There are several situations in which an application should use the | |
4347 | shared @file{libgcc} instead of the static version. The most common | |
4348 | of these is when the application wishes to throw and catch exceptions | |
4349 | across different shared libraries. In that case, each of the libraries | |
4350 | as well as the application itself should use the shared @file{libgcc}. | |
4351 | ||
4352 | Therefore, the G++ and GCJ drivers automatically add | |
4353 | @option{-shared-libgcc} whenever you build a shared library or a main | |
4354 | executable, because C++ and Java programs typically use exceptions, so | |
4355 | this is the right thing to do. | |
4356 | ||
4357 | If, instead, you use the GCC driver to create shared libraries, you may | |
4358 | find that they will not always be linked with the shared @file{libgcc}. | |
4359 | If GCC finds, at its configuration time, that you have a GNU linker that | |
4360 | does not support option @option{--eh-frame-hdr}, it will link the shared | |
4361 | version of @file{libgcc} into shared libraries by default. Otherwise, | |
4362 | it will take advantage of the linker and optimize away the linking with | |
4363 | the shared version of @file{libgcc}, linking with the static version of | |
4364 | libgcc by default. This allows exceptions to propagate through such | |
4365 | shared libraries, without incurring relocation costs at library load | |
4366 | time. | |
4367 | ||
4368 | However, if a library or main executable is supposed to throw or catch | |
4369 | exceptions, you must link it using the G++ or GCJ driver, as appropriate | |
4370 | for the languages used in the program, or using the option | |
4371 | @option{-shared-libgcc}, such that it is linked with the shared | |
4372 | @file{libgcc}. | |
4373 | ||
4374 | @item -symbolic | |
4375 | @opindex symbolic | |
4376 | Bind references to global symbols when building a shared object. Warn | |
4377 | about any unresolved references (unless overridden by the link editor | |
4378 | option @samp{-Xlinker -z -Xlinker defs}). Only a few systems support | |
4379 | this option. | |
4380 | ||
4381 | @item -Xlinker @var{option} | |
4382 | @opindex Xlinker | |
4383 | Pass @var{option} as an option to the linker. You can use this to | |
4384 | supply system-specific linker options which GCC does not know how to | |
4385 | recognize. | |
4386 | ||
4387 | If you want to pass an option that takes an argument, you must use | |
4388 | @option{-Xlinker} twice, once for the option and once for the argument. | |
4389 | For example, to pass @option{-assert definitions}, you must write | |
4390 | @samp{-Xlinker -assert -Xlinker definitions}. It does not work to write | |
4391 | @option{-Xlinker "-assert definitions"}, because this passes the entire | |
4392 | string as a single argument, which is not what the linker expects. | |
4393 | ||
4394 | @item -Wl,@var{option} | |
4395 | @opindex Wl | |
4396 | Pass @var{option} as an option to the linker. If @var{option} contains | |
4397 | commas, it is split into multiple options at the commas. | |
4398 | ||
4399 | @item -u @var{symbol} | |
4400 | @opindex u | |
4401 | Pretend the symbol @var{symbol} is undefined, to force linking of | |
4402 | library modules to define it. You can use @option{-u} multiple times with | |
4403 | different symbols to force loading of additional library modules. | |
4404 | @end table | |
4405 | ||
4406 | @node Directory Options | |
4407 | @section Options for Directory Search | |
4408 | @cindex directory options | |
4409 | @cindex options, directory search | |
4410 | @cindex search path | |
4411 | ||
4412 | These options specify directories to search for header files, for | |
4413 | libraries and for parts of the compiler: | |
4414 | ||
4415 | @table @gcctabopt | |
4416 | @item -I@var{dir} | |
4417 | @opindex I | |
4418 | Add the directory @var{dir} to the head of the list of directories to be | |
4419 | searched for header files. This can be used to override a system header | |
4420 | file, substituting your own version, since these directories are | |
4421 | searched before the system header file directories. However, you should | |
4422 | not use this option to add directories that contain vendor-supplied | |
4423 | system header files (use @option{-isystem} for that). If you use more than | |
4424 | one @option{-I} option, the directories are scanned in left-to-right | |
4425 | order; the standard system directories come after. | |
4426 | ||
4427 | If a standard system include directory, or a directory specified with | |
4428 | @option{-isystem}, is also specified with @option{-I}, it will be | |
4429 | searched only in the position requested by @option{-I}. Also, it will | |
4430 | not be considered a system include directory. If that directory really | |
4431 | does contain system headers, there is a good chance that they will | |
4432 | break. For instance, if GCC's installation procedure edited the headers | |
4433 | in @file{/usr/include} to fix bugs, @samp{-I/usr/include} will cause the | |
4434 | original, buggy headers to be found instead of the corrected ones. GCC | |
4435 | will issue a warning when a system include directory is hidden in this | |
4436 | way. | |
4437 | ||
4438 | @item -I- | |
4439 | @opindex I- | |
4440 | Any directories you specify with @option{-I} options before the @option{-I-} | |
4441 | option are searched only for the case of @samp{#include "@var{file}"}; | |
4442 | they are not searched for @samp{#include <@var{file}>}. | |
4443 | ||
4444 | If additional directories are specified with @option{-I} options after | |
4445 | the @option{-I-}, these directories are searched for all @samp{#include} | |
4446 | directives. (Ordinarily @emph{all} @option{-I} directories are used | |
4447 | this way.) | |
4448 | ||
4449 | In addition, the @option{-I-} option inhibits the use of the current | |
4450 | directory (where the current input file came from) as the first search | |
4451 | directory for @samp{#include "@var{file}"}. There is no way to | |
4452 | override this effect of @option{-I-}. With @option{-I.} you can specify | |
4453 | searching the directory which was current when the compiler was | |
4454 | invoked. That is not exactly the same as what the preprocessor does | |
4455 | by default, but it is often satisfactory. | |
4456 | ||
4457 | @option{-I-} does not inhibit the use of the standard system directories | |
4458 | for header files. Thus, @option{-I-} and @option{-nostdinc} are | |
4459 | independent. | |
4460 | ||
4461 | @item -L@var{dir} | |
4462 | @opindex L | |
4463 | Add directory @var{dir} to the list of directories to be searched | |
4464 | for @option{-l}. | |
4465 | ||
4466 | @item -B@var{prefix} | |
4467 | @opindex B | |
4468 | This option specifies where to find the executables, libraries, | |
4469 | include files, and data files of the compiler itself. | |
4470 | ||
4471 | The compiler driver program runs one or more of the subprograms | |
4472 | @file{cpp}, @file{cc1}, @file{as} and @file{ld}. It tries | |
4473 | @var{prefix} as a prefix for each program it tries to run, both with and | |
4474 | without @samp{@var{machine}/@var{version}/} (@pxref{Target Options}). | |
4475 | ||
4476 | For each subprogram to be run, the compiler driver first tries the | |
4477 | @option{-B} prefix, if any. If that name is not found, or if @option{-B} | |
4478 | was not specified, the driver tries two standard prefixes, which are | |
4479 | @file{/usr/lib/gcc/} and @file{/usr/local/lib/gcc-lib/}. If neither of | |
4480 | those results in a file name that is found, the unmodified program | |
4481 | name is searched for using the directories specified in your | |
4482 | @env{PATH} environment variable. | |
4483 | ||
4484 | The compiler will check to see if the path provided by the @option{-B} | |
4485 | refers to a directory, and if necessary it will add a directory | |
4486 | separator character at the end of the path. | |
4487 | ||
4488 | @option{-B} prefixes that effectively specify directory names also apply | |
4489 | to libraries in the linker, because the compiler translates these | |
4490 | options into @option{-L} options for the linker. They also apply to | |
4491 | includes files in the preprocessor, because the compiler translates these | |
4492 | options into @option{-isystem} options for the preprocessor. In this case, | |
4493 | the compiler appends @samp{include} to the prefix. | |
4494 | ||
4495 | The run-time support file @file{libgcc.a} can also be searched for using | |
4496 | the @option{-B} prefix, if needed. If it is not found there, the two | |
4497 | standard prefixes above are tried, and that is all. The file is left | |
4498 | out of the link if it is not found by those means. | |
4499 | ||
4500 | Another way to specify a prefix much like the @option{-B} prefix is to use | |
4501 | the environment variable @env{GCC_EXEC_PREFIX}. @xref{Environment | |
4502 | Variables}. | |
4503 | ||
4504 | As a special kludge, if the path provided by @option{-B} is | |
4505 | @file{[dir/]stage@var{N}/}, where @var{N} is a number in the range 0 to | |
4506 | 9, then it will be replaced by @file{[dir/]include}. This is to help | |
4507 | with boot-strapping the compiler. | |
4508 | ||
4509 | @item -specs=@var{file} | |
4510 | @opindex specs | |
4511 | Process @var{file} after the compiler reads in the standard @file{specs} | |
4512 | file, in order to override the defaults that the @file{gcc} driver | |
4513 | program uses when determining what switches to pass to @file{cc1}, | |
4514 | @file{cc1plus}, @file{as}, @file{ld}, etc. More than one | |
4515 | @option{-specs=@var{file}} can be specified on the command line, and they | |
4516 | are processed in order, from left to right. | |
4517 | @end table | |
4518 | ||
4519 | @c man end | |
4520 | ||
4521 | @node Spec Files | |
4522 | @section Specifying subprocesses and the switches to pass to them | |
4523 | @cindex Spec Files | |
4524 | @command{gcc} is a driver program. It performs its job by invoking a | |
4525 | sequence of other programs to do the work of compiling, assembling and | |
4526 | linking. GCC interprets its command-line parameters and uses these to | |
4527 | deduce which programs it should invoke, and which command-line options | |
4528 | it ought to place on their command lines. This behavior is controlled | |
4529 | by @dfn{spec strings}. In most cases there is one spec string for each | |
4530 | program that GCC can invoke, but a few programs have multiple spec | |
4531 | strings to control their behavior. The spec strings built into GCC can | |
4532 | be overridden by using the @option{-specs=} command-line switch to specify | |
4533 | a spec file. | |
4534 | ||
4535 | @dfn{Spec files} are plaintext files that are used to construct spec | |
4536 | strings. They consist of a sequence of directives separated by blank | |
4537 | lines. The type of directive is determined by the first non-whitespace | |
4538 | character on the line and it can be one of the following: | |
4539 | ||
4540 | @table @code | |
4541 | @item %@var{command} | |
4542 | Issues a @var{command} to the spec file processor. The commands that can | |
4543 | appear here are: | |
4544 | ||
4545 | @table @code | |
4546 | @item %include <@var{file}> | |
4547 | @cindex %include | |
4548 | Search for @var{file} and insert its text at the current point in the | |
4549 | specs file. | |
4550 | ||
4551 | @item %include_noerr <@var{file}> | |
4552 | @cindex %include_noerr | |
4553 | Just like @samp{%include}, but do not generate an error message if the include | |
4554 | file cannot be found. | |
4555 | ||
4556 | @item %rename @var{old_name} @var{new_name} | |
4557 | @cindex %rename | |
4558 | Rename the spec string @var{old_name} to @var{new_name}. | |
4559 | ||
4560 | @end table | |
4561 | ||
4562 | @item *[@var{spec_name}]: | |
4563 | This tells the compiler to create, override or delete the named spec | |
4564 | string. All lines after this directive up to the next directive or | |
4565 | blank line are considered to be the text for the spec string. If this | |
4566 | results in an empty string then the spec will be deleted. (Or, if the | |
4567 | spec did not exist, then nothing will happened.) Otherwise, if the spec | |
4568 | does not currently exist a new spec will be created. If the spec does | |
4569 | exist then its contents will be overridden by the text of this | |
4570 | directive, unless the first character of that text is the @samp{+} | |
4571 | character, in which case the text will be appended to the spec. | |
4572 | ||
4573 | @item [@var{suffix}]: | |
4574 | Creates a new @samp{[@var{suffix}] spec} pair. All lines after this directive | |
4575 | and up to the next directive or blank line are considered to make up the | |
4576 | spec string for the indicated suffix. When the compiler encounters an | |
4577 | input file with the named suffix, it will processes the spec string in | |
4578 | order to work out how to compile that file. For example: | |
4579 | ||
4580 | @smallexample | |
4581 | .ZZ: | |
4582 | z-compile -input %i | |
4583 | @end smallexample | |
4584 | ||
4585 | This says that any input file whose name ends in @samp{.ZZ} should be | |
4586 | passed to the program @samp{z-compile}, which should be invoked with the | |
4587 | command-line switch @option{-input} and with the result of performing the | |
4588 | @samp{%i} substitution. (See below.) | |
4589 | ||
4590 | As an alternative to providing a spec string, the text that follows a | |
4591 | suffix directive can be one of the following: | |
4592 | ||
4593 | @table @code | |
4594 | @item @@@var{language} | |
4595 | This says that the suffix is an alias for a known @var{language}. This is | |
4596 | similar to using the @option{-x} command-line switch to GCC to specify a | |
4597 | language explicitly. For example: | |
4598 | ||
4599 | @smallexample | |
4600 | .ZZ: | |
4601 | @@c++ | |
4602 | @end smallexample | |
4603 | ||
4604 | Says that .ZZ files are, in fact, C++ source files. | |
4605 | ||
4606 | @item #@var{name} | |
4607 | This causes an error messages saying: | |
4608 | ||
4609 | @smallexample | |
4610 | @var{name} compiler not installed on this system. | |
4611 | @end smallexample | |
4612 | @end table | |
4613 | ||
4614 | GCC already has an extensive list of suffixes built into it. | |
4615 | This directive will add an entry to the end of the list of suffixes, but | |
4616 | since the list is searched from the end backwards, it is effectively | |
4617 | possible to override earlier entries using this technique. | |
4618 | ||
4619 | @end table | |
4620 | ||
4621 | GCC has the following spec strings built into it. Spec files can | |
4622 | override these strings or create their own. Note that individual | |
4623 | targets can also add their own spec strings to this list. | |
4624 | ||
4625 | @smallexample | |
4626 | asm Options to pass to the assembler | |
4627 | asm_final Options to pass to the assembler post-processor | |
4628 | cpp Options to pass to the C preprocessor | |
4629 | cc1 Options to pass to the C compiler | |
4630 | cc1plus Options to pass to the C++ compiler | |
4631 | endfile Object files to include at the end of the link | |
4632 | link Options to pass to the linker | |
4633 | lib Libraries to include on the command line to the linker | |
4634 | libgcc Decides which GCC support library to pass to the linker | |
4635 | linker Sets the name of the linker | |
4636 | predefines Defines to be passed to the C preprocessor | |
4637 | signed_char Defines to pass to CPP to say whether @code{char} is signed | |
4638 | by default | |
4639 | startfile Object files to include at the start of the link | |
4640 | @end smallexample | |
4641 | ||
4642 | Here is a small example of a spec file: | |
4643 | ||
4644 | @smallexample | |
4645 | %rename lib old_lib | |
4646 | ||
4647 | *lib: | |
4648 | --start-group -lgcc -lc -leval1 --end-group %(old_lib) | |
4649 | @end smallexample | |
4650 | ||
4651 | This example renames the spec called @samp{lib} to @samp{old_lib} and | |
4652 | then overrides the previous definition of @samp{lib} with a new one. | |
4653 | The new definition adds in some extra command-line options before | |
4654 | including the text of the old definition. | |
4655 | ||
4656 | @dfn{Spec strings} are a list of command-line options to be passed to their | |
4657 | corresponding program. In addition, the spec strings can contain | |
4658 | @samp{%}-prefixed sequences to substitute variable text or to | |
4659 | conditionally insert text into the command line. Using these constructs | |
4660 | it is possible to generate quite complex command lines. | |
4661 | ||
4662 | Here is a table of all defined @samp{%}-sequences for spec | |
4663 | strings. Note that spaces are not generated automatically around the | |
4664 | results of expanding these sequences. Therefore you can concatenate them | |
4665 | together or combine them with constant text in a single argument. | |
4666 | ||
4667 | @table @code | |
4668 | @item %% | |
4669 | Substitute one @samp{%} into the program name or argument. | |
4670 | ||
4671 | @item %i | |
4672 | Substitute the name of the input file being processed. | |
4673 | ||
4674 | @item %b | |
4675 | Substitute the basename of the input file being processed. | |
4676 | This is the substring up to (and not including) the last period | |
4677 | and not including the directory. | |
4678 | ||
4679 | @item %B | |
4680 | This is the same as @samp{%b}, but include the file suffix (text after | |
4681 | the last period). | |
4682 | ||
4683 | @item %d | |
4684 | Marks the argument containing or following the @samp{%d} as a | |
4685 | temporary file name, so that that file will be deleted if GCC exits | |
4686 | successfully. Unlike @samp{%g}, this contributes no text to the | |
4687 | argument. | |
4688 | ||
4689 | @item %g@var{suffix} | |
4690 | Substitute a file name that has suffix @var{suffix} and is chosen | |
4691 | once per compilation, and mark the argument in the same way as | |
4692 | @samp{%d}. To reduce exposure to denial-of-service attacks, the file | |
4693 | name is now chosen in a way that is hard to predict even when previously | |
4694 | chosen file names are known. For example, @samp{%g.s @dots{} %g.o @dots{} %g.s} | |
4695 | might turn into @samp{ccUVUUAU.s ccXYAXZ12.o ccUVUUAU.s}. @var{suffix} matches | |
4696 | the regexp @samp{[.A-Za-z]*} or the special string @samp{%O}, which is | |
4697 | treated exactly as if @samp{%O} had been preprocessed. Previously, @samp{%g} | |
4698 | was simply substituted with a file name chosen once per compilation, | |
4699 | without regard to any appended suffix (which was therefore treated | |
4700 | just like ordinary text), making such attacks more likely to succeed. | |
4701 | ||
4702 | @item %u@var{suffix} | |
4703 | Like @samp{%g}, but generates a new temporary file name even if | |
4704 | @samp{%u@var{suffix}} was already seen. | |
4705 | ||
4706 | @item %U@var{suffix} | |
4707 | Substitutes the last file name generated with @samp{%u@var{suffix}}, generating a | |
4708 | new one if there is no such last file name. In the absence of any | |
4709 | @samp{%u@var{suffix}}, this is just like @samp{%g@var{suffix}}, except they don't share | |
4710 | the same suffix @emph{space}, so @samp{%g.s @dots{} %U.s @dots{} %g.s @dots{} %U.s} | |
4711 | would involve the generation of two distinct file names, one | |
4712 | for each @samp{%g.s} and another for each @samp{%U.s}. Previously, @samp{%U} was | |
4713 | simply substituted with a file name chosen for the previous @samp{%u}, | |
4714 | without regard to any appended suffix. | |
4715 | ||
4716 | @item %j@var{SUFFIX} | |
4717 | Substitutes the name of the @code{HOST_BIT_BUCKET}, if any, and if it is | |
4718 | writable, and if save-temps is off; otherwise, substitute the name | |
4719 | of a temporary file, just like @samp{%u}. This temporary file is not | |
4720 | meant for communication between processes, but rather as a junk | |
4721 | disposal mechanism. | |
4722 | ||
4723 | @item %.@var{SUFFIX} | |
4724 | Substitutes @var{.SUFFIX} for the suffixes of a matched switch's args | |
4725 | when it is subsequently output with @samp{%*}. @var{SUFFIX} is | |
4726 | terminated by the next space or %. | |
4727 | ||
4728 | @item %w | |
4729 | Marks the argument containing or following the @samp{%w} as the | |
4730 | designated output file of this compilation. This puts the argument | |
4731 | into the sequence of arguments that @samp{%o} will substitute later. | |
4732 | ||
4733 | @item %o | |
4734 | Substitutes the names of all the output files, with spaces | |
4735 | automatically placed around them. You should write spaces | |
4736 | around the @samp{%o} as well or the results are undefined. | |
4737 | @samp{%o} is for use in the specs for running the linker. | |
4738 | Input files whose names have no recognized suffix are not compiled | |
4739 | at all, but they are included among the output files, so they will | |
4740 | be linked. | |
4741 | ||
4742 | @item %O | |
4743 | Substitutes the suffix for object files. Note that this is | |
4744 | handled specially when it immediately follows @samp{%g, %u, or %U}, | |
4745 | because of the need for those to form complete file names. The | |
4746 | handling is such that @samp{%O} is treated exactly as if it had already | |
4747 | been substituted, except that @samp{%g, %u, and %U} do not currently | |
4748 | support additional @var{suffix} characters following @samp{%O} as they would | |
4749 | following, for example, @samp{.o}. | |
4750 | ||
4751 | @item %p | |
4752 | Substitutes the standard macro predefinitions for the | |
4753 | current target machine. Use this when running @code{cpp}. | |
4754 | ||
4755 | @item %P | |
4756 | Like @samp{%p}, but puts @samp{__} before and after the name of each | |
4757 | predefined macro, except for macros that start with @samp{__} or with | |
4758 | @samp{_@var{L}}, where @var{L} is an uppercase letter. This is for ISO | |
4759 | C@. | |
4760 | ||
4761 | @item %I | |
4762 | Substitute a @option{-iprefix} option made from @env{GCC_EXEC_PREFIX}. | |
4763 | ||
4764 | @item %s | |
4765 | Current argument is the name of a library or startup file of some sort. | |
4766 | Search for that file in a standard list of directories and substitute | |
4767 | the full name found. | |
4768 | ||
4769 | @item %e@var{str} | |
4770 | Print @var{str} as an error message. @var{str} is terminated by a newline. | |
4771 | Use this when inconsistent options are detected. | |
4772 | ||
4773 | @item %| | |
4774 | Output @samp{-} if the input for the current command is coming from a pipe. | |
4775 | ||
4776 | @item %(@var{name}) | |
4777 | Substitute the contents of spec string @var{name} at this point. | |
4778 | ||
4779 | @item %[@var{name}] | |
4780 | Like @samp{%(@dots{})} but put @samp{__} around @option{-D} arguments. | |
4781 | ||
4782 | @item %x@{@var{option}@} | |
4783 | Accumulate an option for @samp{%X}. | |
4784 | ||
4785 | @item %X | |
4786 | Output the accumulated linker options specified by @option{-Wl} or a @samp{%x} | |
4787 | spec string. | |
4788 | ||
4789 | @item %Y | |
4790 | Output the accumulated assembler options specified by @option{-Wa}. | |
4791 | ||
4792 | @item %Z | |
4793 | Output the accumulated preprocessor options specified by @option{-Wp}. | |
4794 | ||
4795 | @item %v1 | |
4796 | Substitute the major version number of GCC@. | |
4797 | (For version 2.9.5, this is 2.) | |
4798 | ||
4799 | @item %v2 | |
4800 | Substitute the minor version number of GCC@. | |
4801 | (For version 2.9.5, this is 9.) | |
4802 | ||
4803 | @item %v3 | |
4804 | Substitute the patch level number of GCC@. | |
4805 | (For version 2.9.5, this is 5.) | |
4806 | ||
4807 | @item %a | |
4808 | Process the @code{asm} spec. This is used to compute the | |
4809 | switches to be passed to the assembler. | |
4810 | ||
4811 | @item %A | |
4812 | Process the @code{asm_final} spec. This is a spec string for | |
4813 | passing switches to an assembler post-processor, if such a program is | |
4814 | needed. | |
4815 | ||
4816 | @item %l | |
4817 | Process the @code{link} spec. This is the spec for computing the | |
4818 | command line passed to the linker. Typically it will make use of the | |
4819 | @samp{%L %G %S %D and %E} sequences. | |
4820 | ||
4821 | @item %D | |
4822 | Dump out a @option{-L} option for each directory that GCC believes might | |
4823 | contain startup files. If the target supports multilibs then the | |
4824 | current multilib directory will be prepended to each of these paths. | |
4825 | ||
4826 | @item %M | |
4827 | Output the multilib directory with directory separators replaced with | |
4828 | @samp{_}. If multilib directories are not set, or the multilib directory is | |
4829 | @file{.} then this option emits nothing. | |
4830 | ||
4831 | @item %L | |
4832 | Process the @code{lib} spec. This is a spec string for deciding which | |
4833 | libraries should be included on the command line to the linker. | |
4834 | ||
4835 | @item %G | |
4836 | Process the @code{libgcc} spec. This is a spec string for deciding | |
4837 | which GCC support library should be included on the command line to the linker. | |
4838 | ||
4839 | @item %S | |
4840 | Process the @code{startfile} spec. This is a spec for deciding which | |
4841 | object files should be the first ones passed to the linker. Typically | |
4842 | this might be a file named @file{crt0.o}. | |
4843 | ||
4844 | @item %E | |
4845 | Process the @code{endfile} spec. This is a spec string that specifies | |
4846 | the last object files that will be passed to the linker. | |
4847 | ||
4848 | @item %C | |
4849 | Process the @code{cpp} spec. This is used to construct the arguments | |
4850 | to be passed to the C preprocessor. | |
4851 | ||
4852 | @item %c | |
4853 | Process the @code{signed_char} spec. This is intended to be used | |
4854 | to tell cpp whether a char is signed. It typically has the definition: | |
4855 | @smallexample | |
4856 | %@{funsigned-char:-D__CHAR_UNSIGNED__@} | |
4857 | @end smallexample | |
4858 | ||
4859 | @item %1 | |
4860 | Process the @code{cc1} spec. This is used to construct the options to be | |
4861 | passed to the actual C compiler (@samp{cc1}). | |
4862 | ||
4863 | @item %2 | |
4864 | Process the @code{cc1plus} spec. This is used to construct the options to be | |
4865 | passed to the actual C++ compiler (@samp{cc1plus}). | |
4866 | ||
4867 | @item %* | |
4868 | Substitute the variable part of a matched option. See below. | |
4869 | Note that each comma in the substituted string is replaced by | |
4870 | a single space. | |
4871 | ||
4872 | @item %@{@code{S}@} | |
4873 | Substitutes the @code{-S} switch, if that switch was given to GCC@. | |
4874 | If that switch was not specified, this substitutes nothing. Note that | |
4875 | the leading dash is omitted when specifying this option, and it is | |
4876 | automatically inserted if the substitution is performed. Thus the spec | |
4877 | string @samp{%@{foo@}} would match the command-line option @option{-foo} | |
4878 | and would output the command line option @option{-foo}. | |
4879 | ||
4880 | @item %W@{@code{S}@} | |
4881 | Like %@{@code{S}@} but mark last argument supplied within as a file to be | |
4882 | deleted on failure. | |
4883 | ||
4884 | @item %@{@code{S}*@} | |
4885 | Substitutes all the switches specified to GCC whose names start | |
4886 | with @code{-S}, but which also take an argument. This is used for | |
4887 | switches like @option{-o}, @option{-D}, @option{-I}, etc. | |
4888 | GCC considers @option{-o foo} as being | |
4889 | one switch whose names starts with @samp{o}. %@{o*@} would substitute this | |
4890 | text, including the space. Thus two arguments would be generated. | |
4891 | ||
4892 | @item %@{^@code{S}*@} | |
4893 | Like %@{@code{S}*@}, but don't put a blank between a switch and its | |
4894 | argument. Thus %@{^o*@} would only generate one argument, not two. | |
4895 | ||
4896 | @item %@{@code{S}*&@code{T}*@} | |
4897 | Like %@{@code{S}*@}, but preserve order of @code{S} and @code{T} options | |
4898 | (the order of @code{S} and @code{T} in the spec is not significant). | |
4899 | There can be any number of ampersand-separated variables; for each the | |
4900 | wild card is optional. Useful for CPP as @samp{%@{D*&U*&A*@}}. | |
4901 | ||
4902 | @item %@{<@code{S}@} | |
4903 | Remove all occurrences of @code{-S} from the command line. Note---this | |
4904 | command is position dependent. @samp{%} commands in the spec string | |
4905 | before this option will see @code{-S}, @samp{%} commands in the spec | |
4906 | string after this option will not. | |
4907 | ||
4908 | @item %@{@code{S}*:@code{X}@} | |
4909 | Substitutes @code{X} if one or more switches whose names start with | |
4910 | @code{-S} are specified to GCC@. Note that the tail part of the | |
4911 | @code{-S} option (i.e.@: the part matched by the @samp{*}) will be substituted | |
4912 | for each occurrence of @samp{%*} within @code{X}. | |
4913 | ||
4914 | @item %@{@code{S}:@code{X}@} | |
4915 | Substitutes @code{X}, but only if the @samp{-S} switch was given to GCC@. | |
4916 | ||
4917 | @item %@{!@code{S}:@code{X}@} | |
4918 | Substitutes @code{X}, but only if the @samp{-S} switch was @emph{not} given to GCC@. | |
4919 | ||
4920 | @item %@{|@code{S}:@code{X}@} | |
4921 | Like %@{@code{S}:@code{X}@}, but if no @code{S} switch, substitute @samp{-}. | |
4922 | ||
4923 | @item %@{|!@code{S}:@code{X}@} | |
4924 | Like %@{!@code{S}:@code{X}@}, but if there is an @code{S} switch, substitute @samp{-}. | |
4925 | ||
4926 | @item %@{.@code{S}:@code{X}@} | |
4927 | Substitutes @code{X}, but only if processing a file with suffix @code{S}. | |
4928 | ||
4929 | @item %@{!.@code{S}:@code{X}@} | |
4930 | Substitutes @code{X}, but only if @emph{not} processing a file with suffix @code{S}. | |
4931 | ||
4932 | @item %@{@code{S}|@code{P}:@code{X}@} | |
4933 | Substitutes @code{X} if either @code{-S} or @code{-P} was given to GCC@. This may be | |
4934 | combined with @samp{!} and @samp{.} sequences as well, although they | |
4935 | have a stronger binding than the @samp{|}. For example a spec string | |
4936 | like this: | |
4937 | ||
4938 | @smallexample | |
4939 | %@{.c:-foo@} %@{!.c:-bar@} %@{.c|d:-baz@} %@{!.c|d:-boggle@} | |
4940 | @end smallexample | |
4941 | ||
4942 | will output the following command-line options from the following input | |
4943 | command-line options: | |
4944 | ||
4945 | @smallexample | |
4946 | fred.c -foo -baz | |
4947 | jim.d -bar -boggle | |
4948 | -d fred.c -foo -baz -boggle | |
4949 | -d jim.d -bar -baz -boggle | |
4950 | @end smallexample | |
4951 | ||
4952 | @end table | |
4953 | ||
4954 | The conditional text @code{X} in a %@{@code{S}:@code{X}@} or | |
4955 | %@{!@code{S}:@code{X}@} construct may contain other nested @samp{%} constructs | |
4956 | or spaces, or even newlines. They are processed as usual, as described | |
4957 | above. | |
4958 | ||
4959 | The @option{-O}, @option{-f}, @option{-m}, and @option{-W} | |
4960 | switches are handled specifically in these | |
4961 | constructs. If another value of @option{-O} or the negated form of a @option{-f}, @option{-m}, or | |
4962 | @option{-W} switch is found later in the command line, the earlier switch | |
4963 | value is ignored, except with @{@code{S}*@} where @code{S} is just one | |
4964 | letter, which passes all matching options. | |
4965 | ||
4966 | The character @samp{|} at the beginning of the predicate text is used to indicate | |
4967 | that a command should be piped to the following command, but only if @option{-pipe} | |
4968 | is specified. | |
4969 | ||
4970 | It is built into GCC which switches take arguments and which do not. | |
4971 | (You might think it would be useful to generalize this to allow each | |
4972 | compiler's spec to say which switches take arguments. But this cannot | |
4973 | be done in a consistent fashion. GCC cannot even decide which input | |
4974 | files have been specified without knowing which switches take arguments, | |
4975 | and it must know which input files to compile in order to tell which | |
4976 | compilers to run). | |
4977 | ||
4978 | GCC also knows implicitly that arguments starting in @option{-l} are to be | |
4979 | treated as compiler output files, and passed to the linker in their | |
4980 | proper position among the other output files. | |
4981 | ||
4982 | @c man begin OPTIONS | |
4983 | ||
4984 | @node Target Options | |
4985 | @section Specifying Target Machine and Compiler Version | |
4986 | @cindex target options | |
4987 | @cindex cross compiling | |
4988 | @cindex specifying machine version | |
4989 | @cindex specifying compiler version and target machine | |
4990 | @cindex compiler version, specifying | |
4991 | @cindex target machine, specifying | |
4992 | ||
4993 | By default, GCC compiles code for the same type of machine that you | |
4994 | are using. However, it can also be installed as a cross-compiler, to | |
4995 | compile for some other type of machine. In fact, several different | |
4996 | configurations of GCC, for different target machines, can be | |
4997 | installed side by side. Then you specify which one to use with the | |
4998 | @option{-b} option. | |
4999 | ||
5000 | In addition, older and newer versions of GCC can be installed side | |
5001 | by side. One of them (probably the newest) will be the default, but | |
5002 | you may sometimes wish to use another. | |
5003 | ||
5004 | @table @gcctabopt | |
5005 | @item -b @var{machine} | |
5006 | @opindex b | |
5007 | The argument @var{machine} specifies the target machine for compilation. | |
5008 | This is useful when you have installed GCC as a cross-compiler. | |
5009 | ||
5010 | The value to use for @var{machine} is the same as was specified as the | |
5011 | machine type when configuring GCC as a cross-compiler. For | |
5012 | example, if a cross-compiler was configured with @samp{configure | |
5013 | i386v}, meaning to compile for an 80386 running System V, then you | |
5014 | would specify @option{-b i386v} to run that cross compiler. | |
5015 | ||
5016 | When you do not specify @option{-b}, it normally means to compile for | |
5017 | the same type of machine that you are using. | |
5018 | ||
5019 | @item -V @var{version} | |
5020 | @opindex V | |
5021 | The argument @var{version} specifies which version of GCC to run. | |
5022 | This is useful when multiple versions are installed. For example, | |
5023 | @var{version} might be @samp{2.0}, meaning to run GCC version 2.0. | |
5024 | ||
5025 | The default version, when you do not specify @option{-V}, is the last | |
5026 | version of GCC that you installed. | |
5027 | @end table | |
5028 | ||
5029 | The @option{-b} and @option{-V} options actually work by controlling part of | |
5030 | the file name used for the executable files and libraries used for | |
5031 | compilation. A given version of GCC, for a given target machine, is | |
5032 | normally kept in the directory @file{/usr/local/lib/gcc-lib/@var{machine}/@var{version}}. | |
5033 | ||
5034 | Thus, sites can customize the effect of @option{-b} or @option{-V} either by | |
5035 | changing the names of these directories or adding alternate names (or | |
5036 | symbolic links). If in directory @file{/usr/local/lib/gcc-lib/} the | |
5037 | file @file{80386} is a link to the file @file{i386v}, then @option{-b | |
5038 | 80386} becomes an alias for @option{-b i386v}. | |
5039 | ||
5040 | In one respect, the @option{-b} or @option{-V} do not completely change | |
5041 | to a different compiler: the top-level driver program @command{gcc} | |
5042 | that you originally invoked continues to run and invoke the other | |
5043 | executables (preprocessor, compiler per se, assembler and linker) | |
5044 | that do the real work. However, since no real work is done in the | |
5045 | driver program, it usually does not matter that the driver program | |
5046 | in use is not the one for the specified target. It is common for the | |
5047 | interface to the other executables to change incompatibly between | |
5048 | compiler versions, so unless the version specified is very close to that | |
5049 | of the driver (for example, @option{-V 3.0} with a driver program from GCC | |
5050 | version 3.0.1), use of @option{-V} may not work; for example, using | |
5051 | @option{-V 2.95.2} will not work with a driver program from GCC 3.0. | |
5052 | ||
5053 | The only way that the driver program depends on the target machine is | |
5054 | in the parsing and handling of special machine-specific options. | |
5055 | However, this is controlled by a file which is found, along with the | |
5056 | other executables, in the directory for the specified version and | |
5057 | target machine. As a result, a single installed driver program adapts | |
5058 | to any specified target machine, and sufficiently similar compiler | |
5059 | versions. | |
5060 | ||
5061 | The driver program executable does control one significant thing, | |
5062 | however: the default version and target machine. Therefore, you can | |
5063 | install different instances of the driver program, compiled for | |
5064 | different targets or versions, under different names. | |
5065 | ||
5066 | For example, if the driver for version 2.0 is installed as @command{ogcc} | |
5067 | and that for version 2.1 is installed as @command{gcc}, then the command | |
5068 | @command{gcc} will use version 2.1 by default, while @command{ogcc} will use | |
5069 | 2.0 by default. However, you can choose either version with either | |
5070 | command with the @option{-V} option. | |
5071 | ||
5072 | @node Submodel Options | |
5073 | @section Hardware Models and Configurations | |
5074 | @cindex submodel options | |
5075 | @cindex specifying hardware config | |
5076 | @cindex hardware models and configurations, specifying | |
5077 | @cindex machine dependent options | |
5078 | ||
5079 | Earlier we discussed the standard option @option{-b} which chooses among | |
5080 | different installed compilers for completely different target | |
5081 | machines, such as VAX vs.@: 68000 vs.@: 80386. | |
5082 | ||
5083 | In addition, each of these target machine types can have its own | |
5084 | special options, starting with @samp{-m}, to choose among various | |
5085 | hardware models or configurations---for example, 68010 vs 68020, | |
5086 | floating coprocessor or none. A single installed version of the | |
5087 | compiler can compile for any model or configuration, according to the | |
5088 | options specified. | |
5089 | ||
5090 | Some configurations of the compiler also support additional special | |
5091 | options, usually for compatibility with other compilers on the same | |
5092 | platform. | |
5093 | ||
5094 | These options are defined by the macro @code{TARGET_SWITCHES} in the | |
5095 | machine description. The default for the options is also defined by | |
5096 | that macro, which enables you to change the defaults. | |
5097 | ||
5098 | @menu | |
5099 | * M680x0 Options:: | |
5100 | * M68hc1x Options:: | |
5101 | * VAX Options:: | |
5102 | * SPARC Options:: | |
5103 | * Convex Options:: | |
5104 | * AMD29K Options:: | |
5105 | * ARM Options:: | |
5106 | * MN10200 Options:: | |
5107 | * MN10300 Options:: | |
5108 | * M32R/D Options:: | |
5109 | * M88K Options:: | |
5110 | * RS/6000 and PowerPC Options:: | |
5111 | * RT Options:: | |
5112 | * MIPS Options:: | |
5113 | * i386 and x86-64 Options:: | |
5114 | * HPPA Options:: | |
5115 | * Intel 960 Options:: | |
5116 | * DEC Alpha Options:: | |
5117 | * DEC Alpha/VMS Options:: | |
5118 | * Clipper Options:: | |
5119 | * H8/300 Options:: | |
5120 | * SH Options:: | |
5121 | * System V Options:: | |
5122 | * TMS320C3x/C4x Options:: | |
5123 | * V850 Options:: | |
5124 | * ARC Options:: | |
5125 | * NS32K Options:: | |
5126 | * AVR Options:: | |
5127 | * MCore Options:: | |
5128 | * IA-64 Options:: | |
5129 | * D30V Options:: | |
5130 | * S/390 and zSeries Options:: | |
5131 | * CRIS Options:: | |
5132 | * MMIX Options:: | |
5133 | * PDP-11 Options:: | |
5134 | * Xstormy16 Options:: | |
5135 | * Xtensa Options:: | |
5136 | @end menu | |
5137 | ||
5138 | @node M680x0 Options | |
5139 | @subsection M680x0 Options | |
5140 | @cindex M680x0 options | |
5141 | ||
5142 | These are the @samp{-m} options defined for the 68000 series. The default | |
5143 | values for these options depends on which style of 68000 was selected when | |
5144 | the compiler was configured; the defaults for the most common choices are | |
5145 | given below. | |
5146 | ||
5147 | @table @gcctabopt | |
5148 | @item -m68000 | |
5149 | @itemx -mc68000 | |
5150 | @opindex m68000 | |
5151 | @opindex mc68000 | |
5152 | Generate output for a 68000. This is the default | |
5153 | when the compiler is configured for 68000-based systems. | |
5154 | ||
5155 | Use this option for microcontrollers with a 68000 or EC000 core, | |
5156 | including the 68008, 68302, 68306, 68307, 68322, 68328 and 68356. | |
5157 | ||
5158 | @item -m68020 | |
5159 | @itemx -mc68020 | |
5160 | @opindex m68020 | |
5161 | @opindex mc68020 | |
5162 | Generate output for a 68020. This is the default | |
5163 | when the compiler is configured for 68020-based systems. | |
5164 | ||
5165 | @item -m68881 | |
5166 | @opindex m68881 | |
5167 | Generate output containing 68881 instructions for floating point. | |
5168 | This is the default for most 68020 systems unless @option{--nfp} was | |
5169 | specified when the compiler was configured. | |
5170 | ||
5171 | @item -m68030 | |
5172 | @opindex m68030 | |
5173 | Generate output for a 68030. This is the default when the compiler is | |
5174 | configured for 68030-based systems. | |
5175 | ||
5176 | @item -m68040 | |
5177 | @opindex m68040 | |
5178 | Generate output for a 68040. This is the default when the compiler is | |
5179 | configured for 68040-based systems. | |
5180 | ||
5181 | This option inhibits the use of 68881/68882 instructions that have to be | |
5182 | emulated by software on the 68040. Use this option if your 68040 does not | |
5183 | have code to emulate those instructions. | |
5184 | ||
5185 | @item -m68060 | |
5186 | @opindex m68060 | |
5187 | Generate output for a 68060. This is the default when the compiler is | |
5188 | configured for 68060-based systems. | |
5189 | ||
5190 | This option inhibits the use of 68020 and 68881/68882 instructions that | |
5191 | have to be emulated by software on the 68060. Use this option if your 68060 | |
5192 | does not have code to emulate those instructions. | |
5193 | ||
5194 | @item -mcpu32 | |
5195 | @opindex mcpu32 | |
5196 | Generate output for a CPU32. This is the default | |
5197 | when the compiler is configured for CPU32-based systems. | |
5198 | ||
5199 | Use this option for microcontrollers with a | |
5200 | CPU32 or CPU32+ core, including the 68330, 68331, 68332, 68333, 68334, | |
5201 | 68336, 68340, 68341, 68349 and 68360. | |
5202 | ||
5203 | @item -m5200 | |
5204 | @opindex m5200 | |
5205 | Generate output for a 520X ``coldfire'' family cpu. This is the default | |
5206 | when the compiler is configured for 520X-based systems. | |
5207 | ||
5208 | Use this option for microcontroller with a 5200 core, including | |
5209 | the MCF5202, MCF5203, MCF5204 and MCF5202. | |
5210 | ||
5211 | ||
5212 | @item -m68020-40 | |
5213 | @opindex m68020-40 | |
5214 | Generate output for a 68040, without using any of the new instructions. | |
5215 | This results in code which can run relatively efficiently on either a | |
5216 | 68020/68881 or a 68030 or a 68040. The generated code does use the | |
5217 | 68881 instructions that are emulated on the 68040. | |
5218 | ||
5219 | @item -m68020-60 | |
5220 | @opindex m68020-60 | |
5221 | Generate output for a 68060, without using any of the new instructions. | |
5222 | This results in code which can run relatively efficiently on either a | |
5223 | 68020/68881 or a 68030 or a 68040. The generated code does use the | |
5224 | 68881 instructions that are emulated on the 68060. | |
5225 | ||
5226 | @item -mfpa | |
5227 | @opindex mfpa | |
5228 | Generate output containing Sun FPA instructions for floating point. | |
5229 | ||
5230 | @item -msoft-float | |
5231 | @opindex msoft-float | |
5232 | Generate output containing library calls for floating point. | |
5233 | @strong{Warning:} the requisite libraries are not available for all m68k | |
5234 | targets. Normally the facilities of the machine's usual C compiler are | |
5235 | used, but this can't be done directly in cross-compilation. You must | |
5236 | make your own arrangements to provide suitable library functions for | |
5237 | cross-compilation. The embedded targets @samp{m68k-*-aout} and | |
5238 | @samp{m68k-*-coff} do provide software floating point support. | |
5239 | ||
5240 | @item -mshort | |
5241 | @opindex mshort | |
5242 | Consider type @code{int} to be 16 bits wide, like @code{short int}. | |
5243 | ||
5244 | @item -mnobitfield | |
5245 | @opindex mnobitfield | |
5246 | Do not use the bit-field instructions. The @option{-m68000}, @option{-mcpu32} | |
5247 | and @option{-m5200} options imply @w{@option{-mnobitfield}}. | |
5248 | ||
5249 | @item -mbitfield | |
5250 | @opindex mbitfield | |
5251 | Do use the bit-field instructions. The @option{-m68020} option implies | |
5252 | @option{-mbitfield}. This is the default if you use a configuration | |
5253 | designed for a 68020. | |
5254 | ||
5255 | @item -mrtd | |
5256 | @opindex mrtd | |
5257 | Use a different function-calling convention, in which functions | |
5258 | that take a fixed number of arguments return with the @code{rtd} | |
5259 | instruction, which pops their arguments while returning. This | |
5260 | saves one instruction in the caller since there is no need to pop | |
5261 | the arguments there. | |
5262 | ||
5263 | This calling convention is incompatible with the one normally | |
5264 | used on Unix, so you cannot use it if you need to call libraries | |
5265 | compiled with the Unix compiler. | |
5266 | ||
5267 | Also, you must provide function prototypes for all functions that | |
5268 | take variable numbers of arguments (including @code{printf}); | |
5269 | otherwise incorrect code will be generated for calls to those | |
5270 | functions. | |
5271 | ||
5272 | In addition, seriously incorrect code will result if you call a | |
5273 | function with too many arguments. (Normally, extra arguments are | |
5274 | harmlessly ignored.) | |
5275 | ||
5276 | The @code{rtd} instruction is supported by the 68010, 68020, 68030, | |
5277 | 68040, 68060 and CPU32 processors, but not by the 68000 or 5200. | |
5278 | ||
5279 | @item -malign-int | |
5280 | @itemx -mno-align-int | |
5281 | @opindex malign-int | |
5282 | @opindex mno-align-int | |
5283 | Control whether GCC aligns @code{int}, @code{long}, @code{long long}, | |
5284 | @code{float}, @code{double}, and @code{long double} variables on a 32-bit | |
5285 | boundary (@option{-malign-int}) or a 16-bit boundary (@option{-mno-align-int}). | |
5286 | Aligning variables on 32-bit boundaries produces code that runs somewhat | |
5287 | faster on processors with 32-bit busses at the expense of more memory. | |
5288 | ||
5289 | @strong{Warning:} if you use the @option{-malign-int} switch, GCC will | |
5290 | align structures containing the above types differently than | |
5291 | most published application binary interface specifications for the m68k. | |
5292 | ||
5293 | @item -mpcrel | |
5294 | @opindex mpcrel | |
5295 | Use the pc-relative addressing mode of the 68000 directly, instead of | |
5296 | using a global offset table. At present, this option implies @option{-fpic}, | |
5297 | allowing at most a 16-bit offset for pc-relative addressing. @option{-fPIC} is | |
5298 | not presently supported with @option{-mpcrel}, though this could be supported for | |
5299 | 68020 and higher processors. | |
5300 | ||
5301 | @item -mno-strict-align | |
5302 | @itemx -mstrict-align | |
5303 | @opindex mno-strict-align | |
5304 | @opindex mstrict-align | |
5305 | Do not (do) assume that unaligned memory references will be handled by | |
5306 | the system. | |
5307 | ||
5308 | @end table | |
5309 | ||
5310 | @node M68hc1x Options | |
5311 | @subsection M68hc1x Options | |
5312 | @cindex M68hc1x options | |
5313 | ||
5314 | These are the @samp{-m} options defined for the 68hc11 and 68hc12 | |
5315 | microcontrollers. The default values for these options depends on | |
5316 | which style of microcontroller was selected when the compiler was configured; | |
5317 | the defaults for the most common choices are given below. | |
5318 | ||
5319 | @table @gcctabopt | |
5320 | @item -m6811 | |
5321 | @itemx -m68hc11 | |
5322 | @opindex m6811 | |
5323 | @opindex m68hc11 | |
5324 | Generate output for a 68HC11. This is the default | |
5325 | when the compiler is configured for 68HC11-based systems. | |
5326 | ||
5327 | @item -m6812 | |
5328 | @itemx -m68hc12 | |
5329 | @opindex m6812 | |
5330 | @opindex m68hc12 | |
5331 | Generate output for a 68HC12. This is the default | |
5332 | when the compiler is configured for 68HC12-based systems. | |
5333 | ||
5334 | @item -mauto-incdec | |
5335 | @opindex mauto-incdec | |
5336 | Enable the use of 68HC12 pre and post auto-increment and auto-decrement | |
5337 | addressing modes. | |
5338 | ||
5339 | @item -mshort | |
5340 | @opindex mshort | |
5341 | Consider type @code{int} to be 16 bits wide, like @code{short int}. | |
5342 | ||
5343 | @item -msoft-reg-count=@var{count} | |
5344 | @opindex msoft-reg-count | |
5345 | Specify the number of pseudo-soft registers which are used for the | |
5346 | code generation. The maximum number is 32. Using more pseudo-soft | |
5347 | register may or may not result in better code depending on the program. | |
5348 | The default is 4 for 68HC11 and 2 for 68HC12. | |
5349 | ||
5350 | @end table | |
5351 | ||
5352 | @node VAX Options | |
5353 | @subsection VAX Options | |
5354 | @cindex VAX options | |
5355 | ||
5356 | These @samp{-m} options are defined for the VAX: | |
5357 | ||
5358 | @table @gcctabopt | |
5359 | @item -munix | |
5360 | @opindex munix | |
5361 | Do not output certain jump instructions (@code{aobleq} and so on) | |
5362 | that the Unix assembler for the VAX cannot handle across long | |
5363 | ranges. | |
5364 | ||
5365 | @item -mgnu | |
5366 | @opindex mgnu | |
5367 | Do output those jump instructions, on the assumption that you | |
5368 | will assemble with the GNU assembler. | |
5369 | ||
5370 | @item -mg | |
5371 | @opindex mg | |
5372 | Output code for g-format floating point numbers instead of d-format. | |
5373 | @end table | |
5374 | ||
5375 | @node SPARC Options | |
5376 | @subsection SPARC Options | |
5377 | @cindex SPARC options | |
5378 | ||
5379 | These @samp{-m} switches are supported on the SPARC: | |
5380 | ||
5381 | @table @gcctabopt | |
5382 | @item -mno-app-regs | |
5383 | @itemx -mapp-regs | |
5384 | @opindex mno-app-regs | |
5385 | @opindex mapp-regs | |
5386 | Specify @option{-mapp-regs} to generate output using the global registers | |
5387 | 2 through 4, which the SPARC SVR4 ABI reserves for applications. This | |
5388 | is the default. | |
5389 | ||
5390 | To be fully SVR4 ABI compliant at the cost of some performance loss, | |
5391 | specify @option{-mno-app-regs}. You should compile libraries and system | |
5392 | software with this option. | |
5393 | ||
5394 | @item -mfpu | |
5395 | @itemx -mhard-float | |
5396 | @opindex mfpu | |
5397 | @opindex mhard-float | |
5398 | Generate output containing floating point instructions. This is the | |
5399 | default. | |
5400 | ||
5401 | @item -mno-fpu | |
5402 | @itemx -msoft-float | |
5403 | @opindex mno-fpu | |
5404 | @opindex msoft-float | |
5405 | Generate output containing library calls for floating point. | |
5406 | @strong{Warning:} the requisite libraries are not available for all SPARC | |
5407 | targets. Normally the facilities of the machine's usual C compiler are | |
5408 | used, but this cannot be done directly in cross-compilation. You must make | |
5409 | your own arrangements to provide suitable library functions for | |
5410 | cross-compilation. The embedded targets @samp{sparc-*-aout} and | |
5411 | @samp{sparclite-*-*} do provide software floating point support. | |
5412 | ||
5413 | @option{-msoft-float} changes the calling convention in the output file; | |
5414 | therefore, it is only useful if you compile @emph{all} of a program with | |
5415 | this option. In particular, you need to compile @file{libgcc.a}, the | |
5416 | library that comes with GCC, with @option{-msoft-float} in order for | |
5417 | this to work. | |
5418 | ||
5419 | @item -mhard-quad-float | |
5420 | @opindex mhard-quad-float | |
5421 | Generate output containing quad-word (long double) floating point | |
5422 | instructions. | |
5423 | ||
5424 | @item -msoft-quad-float | |
5425 | @opindex msoft-quad-float | |
5426 | Generate output containing library calls for quad-word (long double) | |
5427 | floating point instructions. The functions called are those specified | |
5428 | in the SPARC ABI@. This is the default. | |
5429 | ||
5430 | As of this writing, there are no sparc implementations that have hardware | |
5431 | support for the quad-word floating point instructions. They all invoke | |
5432 | a trap handler for one of these instructions, and then the trap handler | |
5433 | emulates the effect of the instruction. Because of the trap handler overhead, | |
5434 | this is much slower than calling the ABI library routines. Thus the | |
5435 | @option{-msoft-quad-float} option is the default. | |
5436 | ||
5437 | @item -mno-epilogue | |
5438 | @itemx -mepilogue | |
5439 | @opindex mno-epilogue | |
5440 | @opindex mepilogue | |
5441 | With @option{-mepilogue} (the default), the compiler always emits code for | |
5442 | function exit at the end of each function. Any function exit in | |
5443 | the middle of the function (such as a return statement in C) will | |
5444 | generate a jump to the exit code at the end of the function. | |
5445 | ||
5446 | With @option{-mno-epilogue}, the compiler tries to emit exit code inline | |
5447 | at every function exit. | |
5448 | ||
5449 | @item -mno-flat | |
5450 | @itemx -mflat | |
5451 | @opindex mno-flat | |
5452 | @opindex mflat | |
5453 | With @option{-mflat}, the compiler does not generate save/restore instructions | |
5454 | and will use a ``flat'' or single register window calling convention. | |
5455 | This model uses %i7 as the frame pointer and is compatible with the normal | |
5456 | register window model. Code from either may be intermixed. | |
5457 | The local registers and the input registers (0--5) are still treated as | |
5458 | ``call saved'' registers and will be saved on the stack as necessary. | |
5459 | ||
5460 | With @option{-mno-flat} (the default), the compiler emits save/restore | |
5461 | instructions (except for leaf functions) and is the normal mode of operation. | |
5462 | ||
5463 | @item -mno-unaligned-doubles | |
5464 | @itemx -munaligned-doubles | |
5465 | @opindex mno-unaligned-doubles | |
5466 | @opindex munaligned-doubles | |
5467 | Assume that doubles have 8 byte alignment. This is the default. | |
5468 | ||
5469 | With @option{-munaligned-doubles}, GCC assumes that doubles have 8 byte | |
5470 | alignment only if they are contained in another type, or if they have an | |
5471 | absolute address. Otherwise, it assumes they have 4 byte alignment. | |
5472 | Specifying this option avoids some rare compatibility problems with code | |
5473 | generated by other compilers. It is not the default because it results | |
5474 | in a performance loss, especially for floating point code. | |
5475 | ||
5476 | @item -mno-faster-structs | |
5477 | @itemx -mfaster-structs | |
5478 | @opindex mno-faster-structs | |
5479 | @opindex mfaster-structs | |
5480 | With @option{-mfaster-structs}, the compiler assumes that structures | |
5481 | should have 8 byte alignment. This enables the use of pairs of | |
5482 | @code{ldd} and @code{std} instructions for copies in structure | |
5483 | assignment, in place of twice as many @code{ld} and @code{st} pairs. | |
5484 | However, the use of this changed alignment directly violates the Sparc | |
5485 | ABI@. Thus, it's intended only for use on targets where the developer | |
5486 | acknowledges that their resulting code will not be directly in line with | |
5487 | the rules of the ABI@. | |
5488 | ||
5489 | @item -mv8 | |
5490 | @itemx -msparclite | |
5491 | @opindex mv8 | |
5492 | @opindex msparclite | |
5493 | These two options select variations on the SPARC architecture. | |
5494 | ||
5495 | By default (unless specifically configured for the Fujitsu SPARClite), | |
5496 | GCC generates code for the v7 variant of the SPARC architecture. | |
5497 | ||
5498 | @option{-mv8} will give you SPARC v8 code. The only difference from v7 | |
5499 | code is that the compiler emits the integer multiply and integer | |
5500 | divide instructions which exist in SPARC v8 but not in SPARC v7. | |
5501 | ||
5502 | @option{-msparclite} will give you SPARClite code. This adds the integer | |
5503 | multiply, integer divide step and scan (@code{ffs}) instructions which | |
5504 | exist in SPARClite but not in SPARC v7. | |
5505 | ||
5506 | These options are deprecated and will be deleted in a future GCC release. | |
5507 | They have been replaced with @option{-mcpu=xxx}. | |
5508 | ||
5509 | @item -mcypress | |
5510 | @itemx -msupersparc | |
5511 | @opindex mcypress | |
5512 | @opindex msupersparc | |
5513 | These two options select the processor for which the code is optimized. | |
5514 | ||
5515 | With @option{-mcypress} (the default), the compiler optimizes code for the | |
5516 | Cypress CY7C602 chip, as used in the SparcStation/SparcServer 3xx series. | |
5517 | This is also appropriate for the older SparcStation 1, 2, IPX etc. | |
5518 | ||
5519 | With @option{-msupersparc} the compiler optimizes code for the SuperSparc cpu, as | |
5520 | used in the SparcStation 10, 1000 and 2000 series. This flag also enables use | |
5521 | of the full SPARC v8 instruction set. | |
5522 | ||
5523 | These options are deprecated and will be deleted in a future GCC release. | |
5524 | They have been replaced with @option{-mcpu=xxx}. | |
5525 | ||
5526 | @item -mcpu=@var{cpu_type} | |
5527 | @opindex mcpu | |
5528 | Set the instruction set, register set, and instruction scheduling parameters | |
5529 | for machine type @var{cpu_type}. Supported values for @var{cpu_type} are | |
5530 | @samp{v7}, @samp{cypress}, @samp{v8}, @samp{supersparc}, @samp{sparclite}, | |
5531 | @samp{hypersparc}, @samp{sparclite86x}, @samp{f930}, @samp{f934}, | |
5532 | @samp{sparclet}, @samp{tsc701}, @samp{v9}, and @samp{ultrasparc}. | |
5533 | ||
5534 | Default instruction scheduling parameters are used for values that select | |
5535 | an architecture and not an implementation. These are @samp{v7}, @samp{v8}, | |
5536 | @samp{sparclite}, @samp{sparclet}, @samp{v9}. | |
5537 | ||
5538 | Here is a list of each supported architecture and their supported | |
5539 | implementations. | |
5540 | ||
5541 | @smallexample | |
5542 | v7: cypress | |
5543 | v8: supersparc, hypersparc | |
5544 | sparclite: f930, f934, sparclite86x | |
5545 | sparclet: tsc701 | |
5546 | v9: ultrasparc | |
5547 | @end smallexample | |
5548 | ||
5549 | @item -mtune=@var{cpu_type} | |
5550 | @opindex mtune | |
5551 | Set the instruction scheduling parameters for machine type | |
5552 | @var{cpu_type}, but do not set the instruction set or register set that the | |
5553 | option @option{-mcpu=@var{cpu_type}} would. | |
5554 | ||
5555 | The same values for @option{-mcpu=@var{cpu_type}} can be used for | |
5556 | @option{-mtune=@var{cpu_type}}, but the only useful values are those | |
5557 | that select a particular cpu implementation. Those are @samp{cypress}, | |
5558 | @samp{supersparc}, @samp{hypersparc}, @samp{f930}, @samp{f934}, | |
5559 | @samp{sparclite86x}, @samp{tsc701}, and @samp{ultrasparc}. | |
5560 | ||
5561 | @end table | |
5562 | ||
5563 | These @samp{-m} switches are supported in addition to the above | |
5564 | on the SPARCLET processor. | |
5565 | ||
5566 | @table @gcctabopt | |
5567 | @item -mlittle-endian | |
5568 | @opindex mlittle-endian | |
5569 | Generate code for a processor running in little-endian mode. | |
5570 | ||
5571 | @item -mlive-g0 | |
5572 | @opindex mlive-g0 | |
5573 | Treat register @code{%g0} as a normal register. | |
5574 | GCC will continue to clobber it as necessary but will not assume | |
5575 | it always reads as 0. | |
5576 | ||
5577 | @item -mbroken-saverestore | |
5578 | @opindex mbroken-saverestore | |
5579 | Generate code that does not use non-trivial forms of the @code{save} and | |
5580 | @code{restore} instructions. Early versions of the SPARCLET processor do | |
5581 | not correctly handle @code{save} and @code{restore} instructions used with | |
5582 | arguments. They correctly handle them used without arguments. A @code{save} | |
5583 | instruction used without arguments increments the current window pointer | |
5584 | but does not allocate a new stack frame. It is assumed that the window | |
5585 | overflow trap handler will properly handle this case as will interrupt | |
5586 | handlers. | |
5587 | @end table | |
5588 | ||
5589 | These @samp{-m} switches are supported in addition to the above | |
5590 | on SPARC V9 processors in 64-bit environments. | |
5591 | ||
5592 | @table @gcctabopt | |
5593 | @item -mlittle-endian | |
5594 | @opindex mlittle-endian | |
5595 | Generate code for a processor running in little-endian mode. | |
5596 | ||
5597 | @item -m32 | |
5598 | @itemx -m64 | |
5599 | @opindex m32 | |
5600 | @opindex m64 | |
5601 | Generate code for a 32-bit or 64-bit environment. | |
5602 | The 32-bit environment sets int, long and pointer to 32 bits. | |
5603 | The 64-bit environment sets int to 32 bits and long and pointer | |
5604 | to 64 bits. | |
5605 | ||
5606 | @item -mcmodel=medlow | |
5607 | @opindex mcmodel=medlow | |
5608 | Generate code for the Medium/Low code model: the program must be linked | |
5609 | in the low 32 bits of the address space. Pointers are 64 bits. | |
5610 | Programs can be statically or dynamically linked. | |
5611 | ||
5612 | @item -mcmodel=medmid | |
5613 | @opindex mcmodel=medmid | |
5614 | Generate code for the Medium/Middle code model: the program must be linked | |
5615 | in the low 44 bits of the address space, the text segment must be less than | |
5616 | 2G bytes, and data segment must be within 2G of the text segment. | |
5617 | Pointers are 64 bits. | |
5618 | ||
5619 | @item -mcmodel=medany | |
5620 | @opindex mcmodel=medany | |
5621 | Generate code for the Medium/Anywhere code model: the program may be linked | |
5622 | anywhere in the address space, the text segment must be less than | |
5623 | 2G bytes, and data segment must be within 2G of the text segment. | |
5624 | Pointers are 64 bits. | |
5625 | ||
5626 | @item -mcmodel=embmedany | |
5627 | @opindex mcmodel=embmedany | |
5628 | Generate code for the Medium/Anywhere code model for embedded systems: | |
5629 | assume a 32-bit text and a 32-bit data segment, both starting anywhere | |
5630 | (determined at link time). Register %g4 points to the base of the | |
5631 | data segment. Pointers are still 64 bits. | |
5632 | Programs are statically linked, PIC is not supported. | |
5633 | ||
5634 | @item -mstack-bias | |
5635 | @itemx -mno-stack-bias | |
5636 | @opindex mstack-bias | |
5637 | @opindex mno-stack-bias | |
5638 | With @option{-mstack-bias}, GCC assumes that the stack pointer, and | |
5639 | frame pointer if present, are offset by @minus{}2047 which must be added back | |
5640 | when making stack frame references. | |
5641 | Otherwise, assume no such offset is present. | |
5642 | @end table | |
5643 | ||
5644 | @node Convex Options | |
5645 | @subsection Convex Options | |
5646 | @cindex Convex options | |
5647 | ||
5648 | These @samp{-m} options are defined for Convex: | |
5649 | ||
5650 | @table @gcctabopt | |
5651 | @item -mc1 | |
5652 | @opindex mc1 | |
5653 | Generate output for C1. The code will run on any Convex machine. | |
5654 | The preprocessor symbol @code{__convex__c1__} is defined. | |
5655 | ||
5656 | @item -mc2 | |
5657 | @opindex mc2 | |
5658 | Generate output for C2. Uses instructions not available on C1. | |
5659 | Scheduling and other optimizations are chosen for max performance on C2. | |
5660 | The preprocessor symbol @code{__convex_c2__} is defined. | |
5661 | ||
5662 | @item -mc32 | |
5663 | @opindex mc32 | |
5664 | Generate output for C32xx. Uses instructions not available on C1. | |
5665 | Scheduling and other optimizations are chosen for max performance on C32. | |
5666 | The preprocessor symbol @code{__convex_c32__} is defined. | |
5667 | ||
5668 | @item -mc34 | |
5669 | @opindex mc34 | |
5670 | Generate output for C34xx. Uses instructions not available on C1. | |
5671 | Scheduling and other optimizations are chosen for max performance on C34. | |
5672 | The preprocessor symbol @code{__convex_c34__} is defined. | |
5673 | ||
5674 | @item -mc38 | |
5675 | @opindex mc38 | |
5676 | Generate output for C38xx. Uses instructions not available on C1. | |
5677 | Scheduling and other optimizations are chosen for max performance on C38. | |
5678 | The preprocessor symbol @code{__convex_c38__} is defined. | |
5679 | ||
5680 | @item -margcount | |
5681 | @opindex margcount | |
5682 | Generate code which puts an argument count in the word preceding each | |
5683 | argument list. This is compatible with regular CC, and a few programs | |
5684 | may need the argument count word. GDB and other source-level debuggers | |
5685 | do not need it; this info is in the symbol table. | |
5686 | ||
5687 | @item -mnoargcount | |
5688 | @opindex mnoargcount | |
5689 | Omit the argument count word. This is the default. | |
5690 | ||
5691 | @item -mvolatile-cache | |
5692 | @opindex mvolatile-cache | |
5693 | Allow volatile references to be cached. This is the default. | |
5694 | ||
5695 | @item -mvolatile-nocache | |
5696 | @opindex mvolatile-nocache | |
5697 | Volatile references bypass the data cache, going all the way to memory. | |
5698 | This is only needed for multi-processor code that does not use standard | |
5699 | synchronization instructions. Making non-volatile references to volatile | |
5700 | locations will not necessarily work. | |
5701 | ||
5702 | @item -mlong32 | |
5703 | @opindex mlong32 | |
5704 | Type long is 32 bits, the same as type int. This is the default. | |
5705 | ||
5706 | @item -mlong64 | |
5707 | @opindex mlong64 | |
5708 | Type long is 64 bits, the same as type long long. This option is useless, | |
5709 | because no library support exists for it. | |
5710 | @end table | |
5711 | ||
5712 | @node AMD29K Options | |
5713 | @subsection AMD29K Options | |
5714 | @cindex AMD29K options | |
5715 | ||
5716 | These @samp{-m} options are defined for the AMD Am29000: | |
5717 | ||
5718 | @table @gcctabopt | |
5719 | @item -mdw | |
5720 | @opindex mdw | |
5721 | @cindex DW bit (29k) | |
5722 | Generate code that assumes the @code{DW} bit is set, i.e., that byte and | |
5723 | halfword operations are directly supported by the hardware. This is the | |
5724 | default. | |
5725 | ||
5726 | @item -mndw | |
5727 | @opindex mndw | |
5728 | Generate code that assumes the @code{DW} bit is not set. | |
5729 | ||
5730 | @item -mbw | |
5731 | @opindex mbw | |
5732 | @cindex byte writes (29k) | |
5733 | Generate code that assumes the system supports byte and halfword write | |
5734 | operations. This is the default. | |
5735 | ||
5736 | @item -mnbw | |
5737 | @opindex mnbw | |
5738 | Generate code that assumes the systems does not support byte and | |
5739 | halfword write operations. @option{-mnbw} implies @option{-mndw}. | |
5740 | ||
5741 | @item -msmall | |
5742 | @opindex msmall | |
5743 | @cindex memory model (29k) | |
5744 | Use a small memory model that assumes that all function addresses are | |
5745 | either within a single 256 KB segment or at an absolute address of less | |
5746 | than 256k. This allows the @code{call} instruction to be used instead | |
5747 | of a @code{const}, @code{consth}, @code{calli} sequence. | |
5748 | ||
5749 | @item -mnormal | |
5750 | @opindex mnormal | |
5751 | Use the normal memory model: Generate @code{call} instructions only when | |
5752 | calling functions in the same file and @code{calli} instructions | |
5753 | otherwise. This works if each file occupies less than 256 KB but allows | |
5754 | the entire executable to be larger than 256 KB@. This is the default. | |
5755 | ||
5756 | @item -mlarge | |
5757 | @opindex mlarge | |
5758 | Always use @code{calli} instructions. Specify this option if you expect | |
5759 | a single file to compile into more than 256 KB of code. | |
5760 | ||
5761 | @item -m29050 | |
5762 | @opindex m29050 | |
5763 | @cindex processor selection (29k) | |
5764 | Generate code for the Am29050. | |
5765 | ||
5766 | @item -m29000 | |
5767 | @opindex m29000 | |
5768 | Generate code for the Am29000. This is the default. | |
5769 | ||
5770 | @item -mkernel-registers | |
5771 | @opindex mkernel-registers | |
5772 | @cindex kernel and user registers (29k) | |
5773 | Generate references to registers @code{gr64-gr95} instead of to | |
5774 | registers @code{gr96-gr127}. This option can be used when compiling | |
5775 | kernel code that wants a set of global registers disjoint from that used | |
5776 | by user-mode code. | |
5777 | ||
5778 | Note that when this option is used, register names in @samp{-f} flags | |
5779 | must use the normal, user-mode, names. | |
5780 | ||
5781 | @item -muser-registers | |
5782 | @opindex muser-registers | |
5783 | Use the normal set of global registers, @code{gr96-gr127}. This is the | |
5784 | default. | |
5785 | ||
5786 | @item -mstack-check | |
5787 | @itemx -mno-stack-check | |
5788 | @opindex mstack-check | |
5789 | @opindex mno-stack-check | |
5790 | @cindex stack checks (29k) | |
5791 | Insert (or do not insert) a call to @code{__msp_check} after each stack | |
5792 | adjustment. This is often used for kernel code. | |
5793 | ||
5794 | @item -mstorem-bug | |
5795 | @itemx -mno-storem-bug | |
5796 | @opindex mstorem-bug | |
5797 | @opindex mno-storem-bug | |
5798 | @cindex storem bug (29k) | |
5799 | @option{-mstorem-bug} handles 29k processors which cannot handle the | |
5800 | separation of a mtsrim insn and a storem instruction (most 29000 chips | |
5801 | to date, but not the 29050). | |
5802 | ||
5803 | @item -mno-reuse-arg-regs | |
5804 | @itemx -mreuse-arg-regs | |
5805 | @opindex mno-reuse-arg-regs | |
5806 | @opindex mreuse-arg-regs | |
5807 | @option{-mno-reuse-arg-regs} tells the compiler to only use incoming argument | |
5808 | registers for copying out arguments. This helps detect calling a function | |
5809 | with fewer arguments than it was declared with. | |
5810 | ||
5811 | @item -mno-impure-text | |
5812 | @itemx -mimpure-text | |
5813 | @opindex mno-impure-text | |
5814 | @opindex mimpure-text | |
5815 | @option{-mimpure-text}, used in addition to @option{-shared}, tells the compiler to | |
5816 | not pass @option{-assert pure-text} to the linker when linking a shared object. | |
5817 | ||
5818 | @item -msoft-float | |
5819 | @opindex msoft-float | |
5820 | Generate output containing library calls for floating point. | |
5821 | @strong{Warning:} the requisite libraries are not part of GCC@. | |
5822 | Normally the facilities of the machine's usual C compiler are used, but | |
5823 | this can't be done directly in cross-compilation. You must make your | |
5824 | own arrangements to provide suitable library functions for | |
5825 | cross-compilation. | |
5826 | ||
5827 | @item -mno-multm | |
5828 | @opindex mno-multm | |
5829 | Do not generate multm or multmu instructions. This is useful for some embedded | |
5830 | systems which do not have trap handlers for these instructions. | |
5831 | @end table | |
5832 | ||
5833 | @node ARM Options | |
5834 | @subsection ARM Options | |
5835 | @cindex ARM options | |
5836 | ||
5837 | These @samp{-m} options are defined for Advanced RISC Machines (ARM) | |
5838 | architectures: | |
5839 | ||
5840 | @table @gcctabopt | |
5841 | @item -mapcs-frame | |
5842 | @opindex mapcs-frame | |
5843 | Generate a stack frame that is compliant with the ARM Procedure Call | |
5844 | Standard for all functions, even if this is not strictly necessary for | |
5845 | correct execution of the code. Specifying @option{-fomit-frame-pointer} | |
5846 | with this option will cause the stack frames not to be generated for | |
5847 | leaf functions. The default is @option{-mno-apcs-frame}. | |
5848 | ||
5849 | @item -mapcs | |
5850 | @opindex mapcs | |
5851 | This is a synonym for @option{-mapcs-frame}. | |
5852 | ||
5853 | @item -mapcs-26 | |
5854 | @opindex mapcs-26 | |
5855 | Generate code for a processor running with a 26-bit program counter, | |
5856 | and conforming to the function calling standards for the APCS 26-bit | |
5857 | option. This option replaces the @option{-m2} and @option{-m3} options | |
5858 | of previous releases of the compiler. | |
5859 | ||
5860 | @item -mapcs-32 | |
5861 | @opindex mapcs-32 | |
5862 | Generate code for a processor running with a 32-bit program counter, | |
5863 | and conforming to the function calling standards for the APCS 32-bit | |
5864 | option. This option replaces the @option{-m6} option of previous releases | |
5865 | of the compiler. | |
5866 | ||
5867 | @ignore | |
5868 | @c not currently implemented | |
5869 | @item -mapcs-stack-check | |
5870 | @opindex mapcs-stack-check | |
5871 | Generate code to check the amount of stack space available upon entry to | |
5872 | every function (that actually uses some stack space). If there is | |
5873 | insufficient space available then either the function | |
5874 | @samp{__rt_stkovf_split_small} or @samp{__rt_stkovf_split_big} will be | |
5875 | called, depending upon the amount of stack space required. The run time | |
5876 | system is required to provide these functions. The default is | |
5877 | @option{-mno-apcs-stack-check}, since this produces smaller code. | |
5878 | ||
5879 | @c not currently implemented | |
5880 | @item -mapcs-float | |
5881 | @opindex mapcs-float | |
5882 | Pass floating point arguments using the float point registers. This is | |
5883 | one of the variants of the APCS@. This option is recommended if the | |
5884 | target hardware has a floating point unit or if a lot of floating point | |
5885 | arithmetic is going to be performed by the code. The default is | |
5886 | @option{-mno-apcs-float}, since integer only code is slightly increased in | |
5887 | size if @option{-mapcs-float} is used. | |
5888 | ||
5889 | @c not currently implemented | |
5890 | @item -mapcs-reentrant | |
5891 | @opindex mapcs-reentrant | |
5892 | Generate reentrant, position independent code. The default is | |
5893 | @option{-mno-apcs-reentrant}. | |
5894 | @end ignore | |
5895 | ||
5896 | @item -mthumb-interwork | |
5897 | @opindex mthumb-interwork | |
5898 | Generate code which supports calling between the ARM and Thumb | |
5899 | instruction sets. Without this option the two instruction sets cannot | |
5900 | be reliably used inside one program. The default is | |
5901 | @option{-mno-thumb-interwork}, since slightly larger code is generated | |
5902 | when @option{-mthumb-interwork} is specified. | |
5903 | ||
5904 | @item -mno-sched-prolog | |
5905 | @opindex mno-sched-prolog | |
5906 | Prevent the reordering of instructions in the function prolog, or the | |
5907 | merging of those instruction with the instructions in the function's | |
5908 | body. This means that all functions will start with a recognizable set | |
5909 | of instructions (or in fact one of a choice from a small set of | |
5910 | different function prologues), and this information can be used to | |
5911 | locate the start if functions inside an executable piece of code. The | |
5912 | default is @option{-msched-prolog}. | |
5913 | ||
5914 | @item -mhard-float | |
5915 | @opindex mhard-float | |
5916 | Generate output containing floating point instructions. This is the | |
5917 | default. | |
5918 | ||
5919 | @item -msoft-float | |
5920 | @opindex msoft-float | |
5921 | Generate output containing library calls for floating point. | |
5922 | @strong{Warning:} the requisite libraries are not available for all ARM | |
5923 | targets. Normally the facilities of the machine's usual C compiler are | |
5924 | used, but this cannot be done directly in cross-compilation. You must make | |
5925 | your own arrangements to provide suitable library functions for | |
5926 | cross-compilation. | |
5927 | ||
5928 | @option{-msoft-float} changes the calling convention in the output file; | |
5929 | therefore, it is only useful if you compile @emph{all} of a program with | |
5930 | this option. In particular, you need to compile @file{libgcc.a}, the | |
5931 | library that comes with GCC, with @option{-msoft-float} in order for | |
5932 | this to work. | |
5933 | ||
5934 | @item -mlittle-endian | |
5935 | @opindex mlittle-endian | |
5936 | Generate code for a processor running in little-endian mode. This is | |
5937 | the default for all standard configurations. | |
5938 | ||
5939 | @item -mbig-endian | |
5940 | @opindex mbig-endian | |
5941 | Generate code for a processor running in big-endian mode; the default is | |
5942 | to compile code for a little-endian processor. | |
5943 | ||
5944 | @item -mwords-little-endian | |
5945 | @opindex mwords-little-endian | |
5946 | This option only applies when generating code for big-endian processors. | |
5947 | Generate code for a little-endian word order but a big-endian byte | |
5948 | order. That is, a byte order of the form @samp{32107654}. Note: this | |
5949 | option should only be used if you require compatibility with code for | |
5950 | big-endian ARM processors generated by versions of the compiler prior to | |
5951 | 2.8. | |
5952 | ||
5953 | @item -malignment-traps | |
5954 | @opindex malignment-traps | |
5955 | Generate code that will not trap if the MMU has alignment traps enabled. | |
5956 | On ARM architectures prior to ARMv4, there were no instructions to | |
5957 | access half-word objects stored in memory. However, when reading from | |
5958 | memory a feature of the ARM architecture allows a word load to be used, | |
5959 | even if the address is unaligned, and the processor core will rotate the | |
5960 | data as it is being loaded. This option tells the compiler that such | |
5961 | misaligned accesses will cause a MMU trap and that it should instead | |
5962 | synthesise the access as a series of byte accesses. The compiler can | |
5963 | still use word accesses to load half-word data if it knows that the | |
5964 | address is aligned to a word boundary. | |
5965 | ||
5966 | This option is ignored when compiling for ARM architecture 4 or later, | |
5967 | since these processors have instructions to directly access half-word | |
5968 | objects in memory. | |
5969 | ||
5970 | @item -mno-alignment-traps | |
5971 | @opindex mno-alignment-traps | |
5972 | Generate code that assumes that the MMU will not trap unaligned | |
5973 | accesses. This produces better code when the target instruction set | |
5974 | does not have half-word memory operations (i.e.@: implementations prior to | |
5975 | ARMv4). | |
5976 | ||
5977 | Note that you cannot use this option to access unaligned word objects, | |
5978 | since the processor will only fetch one 32-bit aligned object from | |
5979 | memory. | |
5980 | ||
5981 | The default setting for most targets is @option{-mno-alignment-traps}, since | |
5982 | this produces better code when there are no half-word memory | |
5983 | instructions available. | |
5984 | ||
5985 | @item -mshort-load-bytes | |
5986 | @itemx -mno-short-load-words | |
5987 | @opindex mshort-load-bytes | |
5988 | @opindex mno-short-load-words | |
5989 | These are deprecated aliases for @option{-malignment-traps}. | |
5990 | ||
5991 | @item -mno-short-load-bytes | |
5992 | @itemx -mshort-load-words | |
5993 | @opindex mno-short-load-bytes | |
5994 | @opindex mshort-load-words | |
5995 | This are deprecated aliases for @option{-mno-alignment-traps}. | |
5996 | ||
5997 | @item -mbsd | |
5998 | @opindex mbsd | |
5999 | This option only applies to RISC iX@. Emulate the native BSD-mode | |
6000 | compiler. This is the default if @option{-ansi} is not specified. | |
6001 | ||
6002 | @item -mxopen | |
6003 | @opindex mxopen | |
6004 | This option only applies to RISC iX@. Emulate the native X/Open-mode | |
6005 | compiler. | |
6006 | ||
6007 | @item -mno-symrename | |
6008 | @opindex mno-symrename | |
6009 | This option only applies to RISC iX@. Do not run the assembler | |
6010 | post-processor, @samp{symrename}, after code has been assembled. | |
6011 | Normally it is necessary to modify some of the standard symbols in | |
6012 | preparation for linking with the RISC iX C library; this option | |
6013 | suppresses this pass. The post-processor is never run when the | |
6014 | compiler is built for cross-compilation. | |
6015 | ||
6016 | @item -mcpu=@var{name} | |
6017 | @opindex mcpu | |
6018 | This specifies the name of the target ARM processor. GCC uses this name | |
6019 | to determine what kind of instructions it can emit when generating | |
6020 | assembly code. Permissible names are: @samp{arm2}, @samp{arm250}, | |
6021 | @samp{arm3}, @samp{arm6}, @samp{arm60}, @samp{arm600}, @samp{arm610}, | |
6022 | @samp{arm620}, @samp{arm7}, @samp{arm7m}, @samp{arm7d}, @samp{arm7dm}, | |
6023 | @samp{arm7di}, @samp{arm7dmi}, @samp{arm70}, @samp{arm700}, | |
6024 | @samp{arm700i}, @samp{arm710}, @samp{arm710c}, @samp{arm7100}, | |
6025 | @samp{arm7500}, @samp{arm7500fe}, @samp{arm7tdmi}, @samp{arm8}, | |
6026 | @samp{strongarm}, @samp{strongarm110}, @samp{strongarm1100}, | |
6027 | @samp{arm8}, @samp{arm810}, @samp{arm9}, @samp{arm9e}, @samp{arm920}, | |
6028 | @samp{arm920t}, @samp{arm940t}, @samp{arm9tdmi}, @samp{arm10tdmi}, | |
6029 | @samp{arm1020t}, @samp{xscale}. | |
6030 | ||
6031 | @itemx -mtune=@var{name} | |
6032 | @opindex mtune | |
6033 | This option is very similar to the @option{-mcpu=} option, except that | |
6034 | instead of specifying the actual target processor type, and hence | |
6035 | restricting which instructions can be used, it specifies that GCC should | |
6036 | tune the performance of the code as if the target were of the type | |
6037 | specified in this option, but still choosing the instructions that it | |
6038 | will generate based on the cpu specified by a @option{-mcpu=} option. | |
6039 | For some ARM implementations better performance can be obtained by using | |
6040 | this option. | |
6041 | ||
6042 | @item -march=@var{name} | |
6043 | @opindex march | |
6044 | This specifies the name of the target ARM architecture. GCC uses this | |
6045 | name to determine what kind of instructions it can emit when generating | |
6046 | assembly code. This option can be used in conjunction with or instead | |
6047 | of the @option{-mcpu=} option. Permissible names are: @samp{armv2}, | |
6048 | @samp{armv2a}, @samp{armv3}, @samp{armv3m}, @samp{armv4}, @samp{armv4t}, | |
6049 | @samp{armv5}, @samp{armv5t}, @samp{armv5te}. | |
6050 | ||
6051 | @item -mfpe=@var{number} | |
6052 | @itemx -mfp=@var{number} | |
6053 | @opindex mfpe | |
6054 | @opindex mfp | |
6055 | This specifies the version of the floating point emulation available on | |
6056 | the target. Permissible values are 2 and 3. @option{-mfp=} is a synonym | |
6057 | for @option{-mfpe=}, for compatibility with older versions of GCC@. | |
6058 | ||
6059 | @item -mstructure-size-boundary=@var{n} | |
6060 | @opindex mstructure-size-boundary | |
6061 | The size of all structures and unions will be rounded up to a multiple | |
6062 | of the number of bits set by this option. Permissible values are 8 and | |
6063 | 32. The default value varies for different toolchains. For the COFF | |
6064 | targeted toolchain the default value is 8. Specifying the larger number | |
6065 | can produce faster, more efficient code, but can also increase the size | |
6066 | of the program. The two values are potentially incompatible. Code | |
6067 | compiled with one value cannot necessarily expect to work with code or | |
6068 | libraries compiled with the other value, if they exchange information | |
6069 | using structures or unions. | |
6070 | ||
6071 | @item -mabort-on-noreturn | |
6072 | @opindex mabort-on-noreturn | |
6073 | Generate a call to the function @code{abort} at the end of a | |
6074 | @code{noreturn} function. It will be executed if the function tries to | |
6075 | return. | |
6076 | ||
6077 | @item -mlong-calls | |
6078 | @itemx -mno-long-calls | |
6079 | @opindex mlong-calls | |
6080 | @opindex mno-long-calls | |
6081 | Tells the compiler to perform function calls by first loading the | |
6082 | address of the function into a register and then performing a subroutine | |
6083 | call on this register. This switch is needed if the target function | |
6084 | will lie outside of the 64 megabyte addressing range of the offset based | |
6085 | version of subroutine call instruction. | |
6086 | ||
6087 | Even if this switch is enabled, not all function calls will be turned | |
6088 | into long calls. The heuristic is that static functions, functions | |
6089 | which have the @samp{short-call} attribute, functions that are inside | |
6090 | the scope of a @samp{#pragma no_long_calls} directive and functions whose | |
6091 | definitions have already been compiled within the current compilation | |
6092 | unit, will not be turned into long calls. The exception to this rule is | |
6093 | that weak function definitions, functions with the @samp{long-call} | |
6094 | attribute or the @samp{section} attribute, and functions that are within | |
6095 | the scope of a @samp{#pragma long_calls} directive, will always be | |
6096 | turned into long calls. | |
6097 | ||
6098 | This feature is not enabled by default. Specifying | |
6099 | @option{-mno-long-calls} will restore the default behavior, as will | |
6100 | placing the function calls within the scope of a @samp{#pragma | |
6101 | long_calls_off} directive. Note these switches have no effect on how | |
6102 | the compiler generates code to handle function calls via function | |
6103 | pointers. | |
6104 | ||
6105 | @item -mnop-fun-dllimport | |
6106 | @opindex mnop-fun-dllimport | |
6107 | Disable support for the @code{dllimport} attribute. | |
6108 | ||
6109 | @item -msingle-pic-base | |
6110 | @opindex msingle-pic-base | |
6111 | Treat the register used for PIC addressing as read-only, rather than | |
6112 | loading it in the prologue for each function. The run-time system is | |
6113 | responsible for initializing this register with an appropriate value | |
6114 | before execution begins. | |
6115 | ||
6116 | @item -mpic-register=@var{reg} | |
6117 | @opindex mpic-register | |
6118 | Specify the register to be used for PIC addressing. The default is R10 | |
6119 | unless stack-checking is enabled, when R9 is used. | |
6120 | ||
6121 | @item -mpoke-function-name | |
6122 | @opindex mpoke-function-name | |
6123 | Write the name of each function into the text section, directly | |
6124 | preceding the function prologue. The generated code is similar to this: | |
6125 | ||
6126 | @smallexample | |
6127 | t0 | |
6128 | .ascii "arm_poke_function_name", 0 | |
6129 | .align | |
6130 | t1 | |
6131 | .word 0xff000000 + (t1 - t0) | |
6132 | arm_poke_function_name | |
6133 | mov ip, sp | |
6134 | stmfd sp!, @{fp, ip, lr, pc@} | |
6135 | sub fp, ip, #4 | |
6136 | @end smallexample | |
6137 | ||
6138 | When performing a stack backtrace, code can inspect the value of | |
6139 | @code{pc} stored at @code{fp + 0}. If the trace function then looks at | |
6140 | location @code{pc - 12} and the top 8 bits are set, then we know that | |
6141 | there is a function name embedded immediately preceding this location | |
6142 | and has length @code{((pc[-3]) & 0xff000000)}. | |
6143 | ||
6144 | @item -mthumb | |
6145 | @opindex mthumb | |
6146 | Generate code for the 16-bit Thumb instruction set. The default is to | |
6147 | use the 32-bit ARM instruction set. | |
6148 | ||
6149 | @item -mtpcs-frame | |
6150 | @opindex mtpcs-frame | |
6151 | Generate a stack frame that is compliant with the Thumb Procedure Call | |
6152 | Standard for all non-leaf functions. (A leaf function is one that does | |
6153 | not call any other functions.) The default is @option{-mno-tpcs-frame}. | |
6154 | ||
6155 | @item -mtpcs-leaf-frame | |
6156 | @opindex mtpcs-leaf-frame | |
6157 | Generate a stack frame that is compliant with the Thumb Procedure Call | |
6158 | Standard for all leaf functions. (A leaf function is one that does | |
6159 | not call any other functions.) The default is @option{-mno-apcs-leaf-frame}. | |
6160 | ||
6161 | @item -mcallee-super-interworking | |
6162 | @opindex mcallee-super-interworking | |
6163 | Gives all externally visible functions in the file being compiled an ARM | |
6164 | instruction set header which switches to Thumb mode before executing the | |
6165 | rest of the function. This allows these functions to be called from | |
6166 | non-interworking code. | |
6167 | ||
6168 | @item -mcaller-super-interworking | |
6169 | @opindex mcaller-super-interworking | |
6170 | Allows calls via function pointers (including virtual functions) to | |
6171 | execute correctly regardless of whether the target code has been | |
6172 | compiled for interworking or not. There is a small overhead in the cost | |
6173 | of executing a function pointer if this option is enabled. | |
6174 | ||
6175 | @end table | |
6176 | ||
6177 | @node MN10200 Options | |
6178 | @subsection MN10200 Options | |
6179 | @cindex MN10200 options | |
6180 | These @option{-m} options are defined for Matsushita MN10200 architectures: | |
6181 | @table @gcctabopt | |
6182 | ||
6183 | @item -mrelax | |
6184 | @opindex mrelax | |
6185 | Indicate to the linker that it should perform a relaxation optimization pass | |
6186 | to shorten branches, calls and absolute memory addresses. This option only | |
6187 | has an effect when used on the command line for the final link step. | |
6188 | ||
6189 | This option makes symbolic debugging impossible. | |
6190 | @end table | |
6191 | ||
6192 | @node MN10300 Options | |
6193 | @subsection MN10300 Options | |
6194 | @cindex MN10300 options | |
6195 | These @option{-m} options are defined for Matsushita MN10300 architectures: | |
6196 | ||
6197 | @table @gcctabopt | |
6198 | @item -mmult-bug | |
6199 | @opindex mmult-bug | |
6200 | Generate code to avoid bugs in the multiply instructions for the MN10300 | |
6201 | processors. This is the default. | |
6202 | ||
6203 | @item -mno-mult-bug | |
6204 | @opindex mno-mult-bug | |
6205 | Do not generate code to avoid bugs in the multiply instructions for the | |
6206 | MN10300 processors. | |
6207 | ||
6208 | @item -mam33 | |
6209 | @opindex mam33 | |
6210 | Generate code which uses features specific to the AM33 processor. | |
6211 | ||
6212 | @item -mno-am33 | |
6213 | @opindex mno-am33 | |
6214 | Do not generate code which uses features specific to the AM33 processor. This | |
6215 | is the default. | |
6216 | ||
6217 | @item -mno-crt0 | |
6218 | @opindex mno-crt0 | |
6219 | Do not link in the C run-time initialization object file. | |
6220 | ||
6221 | @item -mrelax | |
6222 | @opindex mrelax | |
6223 | Indicate to the linker that it should perform a relaxation optimization pass | |
6224 | to shorten branches, calls and absolute memory addresses. This option only | |
6225 | has an effect when used on the command line for the final link step. | |
6226 | ||
6227 | This option makes symbolic debugging impossible. | |
6228 | @end table | |
6229 | ||
6230 | ||
6231 | @node M32R/D Options | |
6232 | @subsection M32R/D Options | |
6233 | @cindex M32R/D options | |
6234 | ||
6235 | These @option{-m} options are defined for Mitsubishi M32R/D architectures: | |
6236 | ||
6237 | @table @gcctabopt | |
6238 | @item -m32rx | |
6239 | @opindex m32rx | |
6240 | Generate code for the M32R/X@. | |
6241 | ||
6242 | @item -m32r | |
6243 | @opindex m32r | |
6244 | Generate code for the M32R@. This is the default. | |
6245 | ||
6246 | @item -mcode-model=small | |
6247 | @opindex mcode-model=small | |
6248 | Assume all objects live in the lower 16MB of memory (so that their addresses | |
6249 | can be loaded with the @code{ld24} instruction), and assume all subroutines | |
6250 | are reachable with the @code{bl} instruction. | |
6251 | This is the default. | |
6252 | ||
6253 | The addressability of a particular object can be set with the | |
6254 | @code{model} attribute. | |
6255 | ||
6256 | @item -mcode-model=medium | |
6257 | @opindex mcode-model=medium | |
6258 | Assume objects may be anywhere in the 32-bit address space (the compiler | |
6259 | will generate @code{seth/add3} instructions to load their addresses), and | |
6260 | assume all subroutines are reachable with the @code{bl} instruction. | |
6261 | ||
6262 | @item -mcode-model=large | |
6263 | @opindex mcode-model=large | |
6264 | Assume objects may be anywhere in the 32-bit address space (the compiler | |
6265 | will generate @code{seth/add3} instructions to load their addresses), and | |
6266 | assume subroutines may not be reachable with the @code{bl} instruction | |
6267 | (the compiler will generate the much slower @code{seth/add3/jl} | |
6268 | instruction sequence). | |
6269 | ||
6270 | @item -msdata=none | |
6271 | @opindex msdata=none | |
6272 | Disable use of the small data area. Variables will be put into | |
6273 | one of @samp{.data}, @samp{bss}, or @samp{.rodata} (unless the | |
6274 | @code{section} attribute has been specified). | |
6275 | This is the default. | |
6276 | ||
6277 | The small data area consists of sections @samp{.sdata} and @samp{.sbss}. | |
6278 | Objects may be explicitly put in the small data area with the | |
6279 | @code{section} attribute using one of these sections. | |
6280 | ||
6281 | @item -msdata=sdata | |
6282 | @opindex msdata=sdata | |
6283 | Put small global and static data in the small data area, but do not | |
6284 | generate special code to reference them. | |
6285 | ||
6286 | @item -msdata=use | |
6287 | @opindex msdata=use | |
6288 | Put small global and static data in the small data area, and generate | |
6289 | special instructions to reference them. | |
6290 | ||
6291 | @item -G @var{num} | |
6292 | @opindex G | |
6293 | @cindex smaller data references | |
6294 | Put global and static objects less than or equal to @var{num} bytes | |
6295 | into the small data or bss sections instead of the normal data or bss | |
6296 | sections. The default value of @var{num} is 8. | |
6297 | The @option{-msdata} option must be set to one of @samp{sdata} or @samp{use} | |
6298 | for this option to have any effect. | |
6299 | ||
6300 | All modules should be compiled with the same @option{-G @var{num}} value. | |
6301 | Compiling with different values of @var{num} may or may not work; if it | |
6302 | doesn't the linker will give an error message---incorrect code will not be | |
6303 | generated. | |
6304 | ||
6305 | @end table | |
6306 | ||
6307 | @node M88K Options | |
6308 | @subsection M88K Options | |
6309 | @cindex M88k options | |
6310 | ||
6311 | These @samp{-m} options are defined for Motorola 88k architectures: | |
6312 | ||
6313 | @table @gcctabopt | |
6314 | @item -m88000 | |
6315 | @opindex m88000 | |
6316 | Generate code that works well on both the m88100 and the | |
6317 | m88110. | |
6318 | ||
6319 | @item -m88100 | |
6320 | @opindex m88100 | |
6321 | Generate code that works best for the m88100, but that also | |
6322 | runs on the m88110. | |
6323 | ||
6324 | @item -m88110 | |
6325 | @opindex m88110 | |
6326 | Generate code that works best for the m88110, and may not run | |
6327 | on the m88100. | |
6328 | ||
6329 | @item -mbig-pic | |
6330 | @opindex mbig-pic | |
6331 | Obsolete option to be removed from the next revision. | |
6332 | Use @option{-fPIC}. | |
6333 | ||
6334 | @item -midentify-revision | |
6335 | @opindex midentify-revision | |
6336 | @cindex identifying source, compiler (88k) | |
6337 | Include an @code{ident} directive in the assembler output recording the | |
6338 | source file name, compiler name and version, timestamp, and compilation | |
6339 | flags used. | |
6340 | ||
6341 | @item -mno-underscores | |
6342 | @opindex mno-underscores | |
6343 | @cindex underscores, avoiding (88k) | |
6344 | In assembler output, emit symbol names without adding an underscore | |
6345 | character at the beginning of each name. The default is to use an | |
6346 | underscore as prefix on each name. | |
6347 | ||
6348 | @item -mocs-debug-info | |
6349 | @itemx -mno-ocs-debug-info | |
6350 | @opindex mocs-debug-info | |
6351 | @opindex mno-ocs-debug-info | |
6352 | @cindex OCS (88k) | |
6353 | @cindex debugging, 88k OCS | |
6354 | Include (or omit) additional debugging information (about registers used | |
6355 | in each stack frame) as specified in the 88open Object Compatibility | |
6356 | Standard, ``OCS''@. This extra information allows debugging of code that | |
6357 | has had the frame pointer eliminated. The default for DG/UX, SVr4, and | |
6358 | Delta 88 SVr3.2 is to include this information; other 88k configurations | |
6359 | omit this information by default. | |
6360 | ||
6361 | @item -mocs-frame-position | |
6362 | @opindex mocs-frame-position | |
6363 | @cindex register positions in frame (88k) | |
6364 | When emitting COFF debugging information for automatic variables and | |
6365 | parameters stored on the stack, use the offset from the canonical frame | |
6366 | address, which is the stack pointer (register 31) on entry to the | |
6367 | function. The DG/UX, SVr4, Delta88 SVr3.2, and BCS configurations use | |
6368 | @option{-mocs-frame-position}; other 88k configurations have the default | |
6369 | @option{-mno-ocs-frame-position}. | |
6370 | ||
6371 | @item -mno-ocs-frame-position | |
6372 | @opindex mno-ocs-frame-position | |
6373 | @cindex register positions in frame (88k) | |
6374 | When emitting COFF debugging information for automatic variables and | |
6375 | parameters stored on the stack, use the offset from the frame pointer | |
6376 | register (register 30). When this option is in effect, the frame | |
6377 | pointer is not eliminated when debugging information is selected by the | |
6378 | -g switch. | |
6379 | ||
6380 | @item -moptimize-arg-area | |
6381 | @opindex moptimize-arg-area | |
6382 | @cindex arguments in frame (88k) | |
6383 | Save space by reorganizing the stack frame. This option generates code | |
6384 | that does not agree with the 88open specifications, but uses less | |
6385 | memory. | |
6386 | ||
6387 | @itemx -mno-optimize-arg-area | |
6388 | @opindex mno-optimize-arg-area | |
6389 | Do not reorganize the stack frame to save space. This is the default. | |
6390 | The generated conforms to the specification, but uses more memory. | |
6391 | ||
6392 | @item -mshort-data-@var{num} | |
6393 | @opindex mshort-data | |
6394 | @cindex smaller data references (88k) | |
6395 | @cindex r0-relative references (88k) | |
6396 | Generate smaller data references by making them relative to @code{r0}, | |
6397 | which allows loading a value using a single instruction (rather than the | |
6398 | usual two). You control which data references are affected by | |
6399 | specifying @var{num} with this option. For example, if you specify | |
6400 | @option{-mshort-data-512}, then the data references affected are those | |
6401 | involving displacements of less than 512 bytes. | |
6402 | @option{-mshort-data-@var{num}} is not effective for @var{num} greater | |
6403 | than 64k. | |
6404 | ||
6405 | @item -mserialize-volatile | |
6406 | @opindex mserialize-volatile | |
6407 | @itemx -mno-serialize-volatile | |
6408 | @opindex mno-serialize-volatile | |
6409 | @cindex sequential consistency on 88k | |
6410 | Do, or don't, generate code to guarantee sequential consistency | |
6411 | of volatile memory references. By default, consistency is | |
6412 | guaranteed. | |
6413 | ||
6414 | The order of memory references made by the MC88110 processor does | |
6415 | not always match the order of the instructions requesting those | |
6416 | references. In particular, a load instruction may execute before | |
6417 | a preceding store instruction. Such reordering violates | |
6418 | sequential consistency of volatile memory references, when there | |
6419 | are multiple processors. When consistency must be guaranteed, | |
6420 | GCC generates special instructions, as needed, to force | |
6421 | execution in the proper order. | |
6422 | ||
6423 | The MC88100 processor does not reorder memory references and so | |
6424 | always provides sequential consistency. However, by default, GCC | |
6425 | generates the special instructions to guarantee consistency | |
6426 | even when you use @option{-m88100}, so that the code may be run on an | |
6427 | MC88110 processor. If you intend to run your code only on the | |
6428 | MC88100 processor, you may use @option{-mno-serialize-volatile}. | |
6429 | ||
6430 | The extra code generated to guarantee consistency may affect the | |
6431 | performance of your application. If you know that you can safely | |
6432 | forgo this guarantee, you may use @option{-mno-serialize-volatile}. | |
6433 | ||
6434 | @item -msvr4 | |
6435 | @itemx -msvr3 | |
6436 | @opindex msvr4 | |
6437 | @opindex msvr3 | |
6438 | @cindex assembler syntax, 88k | |
6439 | @cindex SVr4 | |
6440 | Turn on (@option{-msvr4}) or off (@option{-msvr3}) compiler extensions | |
6441 | related to System V release 4 (SVr4). This controls the following: | |
6442 | ||
6443 | @enumerate | |
6444 | @item | |
6445 | Which variant of the assembler syntax to emit. | |
6446 | @item | |
6447 | @option{-msvr4} makes the C preprocessor recognize @samp{#pragma weak} | |
6448 | that is used on System V release 4. | |
6449 | @item | |
6450 | @option{-msvr4} makes GCC issue additional declaration directives used in | |
6451 | SVr4. | |
6452 | @end enumerate | |
6453 | ||
6454 | @option{-msvr4} is the default for the m88k-motorola-sysv4 and | |
6455 | m88k-dg-dgux m88k configurations. @option{-msvr3} is the default for all | |
6456 | other m88k configurations. | |
6457 | ||
6458 | @item -mversion-03.00 | |
6459 | @opindex mversion-03.00 | |
6460 | This option is obsolete, and is ignored. | |
6461 | @c ??? which asm syntax better for GAS? option there too? | |
6462 | ||
6463 | @item -mno-check-zero-division | |
6464 | @itemx -mcheck-zero-division | |
6465 | @opindex mno-check-zero-division | |
6466 | @opindex mcheck-zero-division | |
6467 | @cindex zero division on 88k | |
6468 | Do, or don't, generate code to guarantee that integer division by | |
6469 | zero will be detected. By default, detection is guaranteed. | |
6470 | ||
6471 | Some models of the MC88100 processor fail to trap upon integer | |
6472 | division by zero under certain conditions. By default, when | |
6473 | compiling code that might be run on such a processor, GCC | |
6474 | generates code that explicitly checks for zero-valued divisors | |
6475 | and traps with exception number 503 when one is detected. Use of | |
6476 | @option{-mno-check-zero-division} suppresses such checking for code | |
6477 | generated to run on an MC88100 processor. | |
6478 | ||
6479 | GCC assumes that the MC88110 processor correctly detects all instances | |
6480 | of integer division by zero. When @option{-m88110} is specified, no | |
6481 | explicit checks for zero-valued divisors are generated, and both | |
6482 | @option{-mcheck-zero-division} and @option{-mno-check-zero-division} are | |
6483 | ignored. | |
6484 | ||
6485 | @item -muse-div-instruction | |
6486 | @opindex muse-div-instruction | |
6487 | @cindex divide instruction, 88k | |
6488 | Use the div instruction for signed integer division on the | |
6489 | MC88100 processor. By default, the div instruction is not used. | |
6490 | ||
6491 | On the MC88100 processor the signed integer division instruction | |
6492 | div) traps to the operating system on a negative operand. The | |
6493 | operating system transparently completes the operation, but at a | |
6494 | large cost in execution time. By default, when compiling code | |
6495 | that might be run on an MC88100 processor, GCC emulates signed | |
6496 | integer division using the unsigned integer division instruction | |
6497 | divu), thereby avoiding the large penalty of a trap to the | |
6498 | operating system. Such emulation has its own, smaller, execution | |
6499 | cost in both time and space. To the extent that your code's | |
6500 | important signed integer division operations are performed on two | |
6501 | nonnegative operands, it may be desirable to use the div | |
6502 | instruction directly. | |
6503 | ||
6504 | On the MC88110 processor the div instruction (also known as the | |
6505 | divs instruction) processes negative operands without trapping to | |
6506 | the operating system. When @option{-m88110} is specified, | |
6507 | @option{-muse-div-instruction} is ignored, and the div instruction is used | |
6508 | for signed integer division. | |
6509 | ||
6510 | Note that the result of dividing @code{INT_MIN} by @minus{}1 is undefined. In | |
6511 | particular, the behavior of such a division with and without | |
6512 | @option{-muse-div-instruction} may differ. | |
6513 | ||
6514 | @item -mtrap-large-shift | |
6515 | @itemx -mhandle-large-shift | |
6516 | @opindex mtrap-large-shift | |
6517 | @opindex mhandle-large-shift | |
6518 | @cindex bit shift overflow (88k) | |
6519 | @cindex large bit shifts (88k) | |
6520 | Include code to detect bit-shifts of more than 31 bits; respectively, | |
6521 | trap such shifts or emit code to handle them properly. By default GCC | |
6522 | makes no special provision for large bit shifts. | |
6523 | ||
6524 | @item -mwarn-passed-structs | |
6525 | @opindex mwarn-passed-structs | |
6526 | @cindex structure passing (88k) | |
6527 | Warn when a function passes a struct as an argument or result. | |
6528 | Structure-passing conventions have changed during the evolution of the C | |
6529 | language, and are often the source of portability problems. By default, | |
6530 | GCC issues no such warning. | |
6531 | @end table | |
6532 | ||
6533 | @c break page here to avoid unsightly interparagraph stretch. | |
6534 | @c -zw, 2001-8-17 | |
6535 | @page | |
6536 | ||
6537 | @node RS/6000 and PowerPC Options | |
6538 | @subsection IBM RS/6000 and PowerPC Options | |
6539 | @cindex RS/6000 and PowerPC Options | |
6540 | @cindex IBM RS/6000 and PowerPC Options | |
6541 | ||
6542 | These @samp{-m} options are defined for the IBM RS/6000 and PowerPC: | |
6543 | @table @gcctabopt | |
6544 | @item -mpower | |
6545 | @itemx -mno-power | |
6546 | @itemx -mpower2 | |
6547 | @itemx -mno-power2 | |
6548 | @itemx -mpowerpc | |
6549 | @itemx -mno-powerpc | |
6550 | @itemx -mpowerpc-gpopt | |
6551 | @itemx -mno-powerpc-gpopt | |
6552 | @itemx -mpowerpc-gfxopt | |
6553 | @itemx -mno-powerpc-gfxopt | |
6554 | @itemx -mpowerpc64 | |
6555 | @itemx -mno-powerpc64 | |
6556 | @opindex mpower | |
6557 | @opindex mno-power | |
6558 | @opindex mpower2 | |
6559 | @opindex mno-power2 | |
6560 | @opindex mpowerpc | |
6561 | @opindex mno-powerpc | |
6562 | @opindex mpowerpc-gpopt | |
6563 | @opindex mno-powerpc-gpopt | |
6564 | @opindex mpowerpc-gfxopt | |
6565 | @opindex mno-powerpc-gfxopt | |
6566 | @opindex mpowerpc64 | |
6567 | @opindex mno-powerpc64 | |
6568 | GCC supports two related instruction set architectures for the | |
6569 | RS/6000 and PowerPC@. The @dfn{POWER} instruction set are those | |
6570 | instructions supported by the @samp{rios} chip set used in the original | |
6571 | RS/6000 systems and the @dfn{PowerPC} instruction set is the | |
6572 | architecture of the Motorola MPC5xx, MPC6xx, MPC8xx microprocessors, and | |
6573 | the IBM 4xx microprocessors. | |
6574 | ||
6575 | Neither architecture is a subset of the other. However there is a | |
6576 | large common subset of instructions supported by both. An MQ | |
6577 | register is included in processors supporting the POWER architecture. | |
6578 | ||
6579 | You use these options to specify which instructions are available on the | |
6580 | processor you are using. The default value of these options is | |
6581 | determined when configuring GCC@. Specifying the | |
6582 | @option{-mcpu=@var{cpu_type}} overrides the specification of these | |
6583 | options. We recommend you use the @option{-mcpu=@var{cpu_type}} option | |
6584 | rather than the options listed above. | |
6585 | ||
6586 | The @option{-mpower} option allows GCC to generate instructions that | |
6587 | are found only in the POWER architecture and to use the MQ register. | |
6588 | Specifying @option{-mpower2} implies @option{-power} and also allows GCC | |
6589 | to generate instructions that are present in the POWER2 architecture but | |
6590 | not the original POWER architecture. | |
6591 | ||
6592 | The @option{-mpowerpc} option allows GCC to generate instructions that | |
6593 | are found only in the 32-bit subset of the PowerPC architecture. | |
6594 | Specifying @option{-mpowerpc-gpopt} implies @option{-mpowerpc} and also allows | |
6595 | GCC to use the optional PowerPC architecture instructions in the | |
6596 | General Purpose group, including floating-point square root. Specifying | |
6597 | @option{-mpowerpc-gfxopt} implies @option{-mpowerpc} and also allows GCC to | |
6598 | use the optional PowerPC architecture instructions in the Graphics | |
6599 | group, including floating-point select. | |
6600 | ||
6601 | The @option{-mpowerpc64} option allows GCC to generate the additional | |
6602 | 64-bit instructions that are found in the full PowerPC64 architecture | |
6603 | and to treat GPRs as 64-bit, doubleword quantities. GCC defaults to | |
6604 | @option{-mno-powerpc64}. | |
6605 | ||
6606 | If you specify both @option{-mno-power} and @option{-mno-powerpc}, GCC | |
6607 | will use only the instructions in the common subset of both | |
6608 | architectures plus some special AIX common-mode calls, and will not use | |
6609 | the MQ register. Specifying both @option{-mpower} and @option{-mpowerpc} | |
6610 | permits GCC to use any instruction from either architecture and to | |
6611 | allow use of the MQ register; specify this for the Motorola MPC601. | |
6612 | ||
6613 | @item -mnew-mnemonics | |
6614 | @itemx -mold-mnemonics | |
6615 | @opindex mnew-mnemonics | |
6616 | @opindex mold-mnemonics | |
6617 | Select which mnemonics to use in the generated assembler code. With | |
6618 | @option{-mnew-mnemonics}, GCC uses the assembler mnemonics defined for | |
6619 | the PowerPC architecture. With @option{-mold-mnemonics} it uses the | |
6620 | assembler mnemonics defined for the POWER architecture. Instructions | |
6621 | defined in only one architecture have only one mnemonic; GCC uses that | |
6622 | mnemonic irrespective of which of these options is specified. | |
6623 | ||
6624 | GCC defaults to the mnemonics appropriate for the architecture in | |
6625 | use. Specifying @option{-mcpu=@var{cpu_type}} sometimes overrides the | |
6626 | value of these option. Unless you are building a cross-compiler, you | |
6627 | should normally not specify either @option{-mnew-mnemonics} or | |
6628 | @option{-mold-mnemonics}, but should instead accept the default. | |
6629 | ||
6630 | @item -mcpu=@var{cpu_type} | |
6631 | @opindex mcpu | |
6632 | Set architecture type, register usage, choice of mnemonics, and | |
6633 | instruction scheduling parameters for machine type @var{cpu_type}. | |
6634 | Supported values for @var{cpu_type} are @samp{rios}, @samp{rios1}, | |
6635 | @samp{rsc}, @samp{rios2}, @samp{rs64a}, @samp{601}, @samp{602}, | |
6636 | @samp{603}, @samp{603e}, @samp{604}, @samp{604e}, @samp{620}, | |
6637 | @samp{630}, @samp{740}, @samp{7400}, @samp{7450}, @samp{750}, | |
6638 | @samp{power}, @samp{power2}, @samp{powerpc}, @samp{403}, @samp{505}, | |
6639 | @samp{801}, @samp{821}, @samp{823}, and @samp{860} and @samp{common}. | |
6640 | ||
6641 | @option{-mcpu=common} selects a completely generic processor. Code | |
6642 | generated under this option will run on any POWER or PowerPC processor. | |
6643 | GCC will use only the instructions in the common subset of both | |
6644 | architectures, and will not use the MQ register. GCC assumes a generic | |
6645 | processor model for scheduling purposes. | |
6646 | ||
6647 | @option{-mcpu=power}, @option{-mcpu=power2}, @option{-mcpu=powerpc}, and | |
6648 | @option{-mcpu=powerpc64} specify generic POWER, POWER2, pure 32-bit | |
6649 | PowerPC (i.e., not MPC601), and 64-bit PowerPC architecture machine | |
6650 | types, with an appropriate, generic processor model assumed for | |
6651 | scheduling purposes. | |
6652 | ||
6653 | The other options specify a specific processor. Code generated under | |
6654 | those options will run best on that processor, and may not run at all on | |
6655 | others. | |
6656 | ||
6657 | The @option{-mcpu} options automatically enable or disable other | |
6658 | @option{-m} options as follows: | |
6659 | ||
6660 | @table @samp | |
6661 | @item common | |
6662 | @option{-mno-power}, @option{-mno-powerc} | |
6663 | ||
6664 | @item power | |
6665 | @itemx power2 | |
6666 | @itemx rios1 | |
6667 | @itemx rios2 | |
6668 | @itemx rsc | |
6669 | @option{-mpower}, @option{-mno-powerpc}, @option{-mno-new-mnemonics} | |
6670 | ||
6671 | @item powerpc | |
6672 | @itemx rs64a | |
6673 | @itemx 602 | |
6674 | @itemx 603 | |
6675 | @itemx 603e | |
6676 | @itemx 604 | |
6677 | @itemx 620 | |
6678 | @itemx 630 | |
6679 | @itemx 740 | |
6680 | @itemx 7400 | |
6681 | @itemx 7450 | |
6682 | @itemx 750 | |
6683 | @itemx 505 | |
6684 | @option{-mno-power}, @option{-mpowerpc}, @option{-mnew-mnemonics} | |
6685 | ||
6686 | @item 601 | |
6687 | @option{-mpower}, @option{-mpowerpc}, @option{-mnew-mnemonics} | |
6688 | ||
6689 | @item 403 | |
6690 | @itemx 821 | |
6691 | @itemx 860 | |
6692 | @option{-mno-power}, @option{-mpowerpc}, @option{-mnew-mnemonics}, @option{-msoft-float} | |
6693 | @end table | |
6694 | ||
6695 | @item -mtune=@var{cpu_type} | |
6696 | @opindex mtune | |
6697 | Set the instruction scheduling parameters for machine type | |
6698 | @var{cpu_type}, but do not set the architecture type, register usage, or | |
6699 | choice of mnemonics, as @option{-mcpu=@var{cpu_type}} would. The same | |
6700 | values for @var{cpu_type} are used for @option{-mtune} as for | |
6701 | @option{-mcpu}. If both are specified, the code generated will use the | |
6702 | architecture, registers, and mnemonics set by @option{-mcpu}, but the | |
6703 | scheduling parameters set by @option{-mtune}. | |
6704 | ||
6705 | @item -maltivec | |
6706 | @itemx -mno-altivec | |
6707 | @opindex maltivec | |
6708 | @opindex mno-altivec | |
6709 | These switches enable or disable the use of built-in functions that | |
6710 | allow access to the AltiVec instruction set. You may also need to set | |
6711 | @option{-mabi=altivec} to adjust the current ABI with AltiVec ABI | |
6712 | enhancements. | |
6713 | ||
6714 | @item -mfull-toc | |
6715 | @itemx -mno-fp-in-toc | |
6716 | @itemx -mno-sum-in-toc | |
6717 | @itemx -mminimal-toc | |
6718 | @opindex mfull-toc | |
6719 | @opindex mno-fp-in-toc | |
6720 | @opindex mno-sum-in-toc | |
6721 | @opindex mminimal-toc | |
6722 | Modify generation of the TOC (Table Of Contents), which is created for | |
6723 | every executable file. The @option{-mfull-toc} option is selected by | |
6724 | default. In that case, GCC will allocate at least one TOC entry for | |
6725 | each unique non-automatic variable reference in your program. GCC | |
6726 | will also place floating-point constants in the TOC@. However, only | |
6727 | 16,384 entries are available in the TOC@. | |
6728 | ||
6729 | If you receive a linker error message that saying you have overflowed | |
6730 | the available TOC space, you can reduce the amount of TOC space used | |
6731 | with the @option{-mno-fp-in-toc} and @option{-mno-sum-in-toc} options. | |
6732 | @option{-mno-fp-in-toc} prevents GCC from putting floating-point | |
6733 | constants in the TOC and @option{-mno-sum-in-toc} forces GCC to | |
6734 | generate code to calculate the sum of an address and a constant at | |
6735 | run-time instead of putting that sum into the TOC@. You may specify one | |
6736 | or both of these options. Each causes GCC to produce very slightly | |
6737 | slower and larger code at the expense of conserving TOC space. | |
6738 | ||
6739 | If you still run out of space in the TOC even when you specify both of | |
6740 | these options, specify @option{-mminimal-toc} instead. This option causes | |
6741 | GCC to make only one TOC entry for every file. When you specify this | |
6742 | option, GCC will produce code that is slower and larger but which | |
6743 | uses extremely little TOC space. You may wish to use this option | |
6744 | only on files that contain less frequently executed code. | |
6745 | ||
6746 | @item -maix64 | |
6747 | @itemx -maix32 | |
6748 | @opindex maix64 | |
6749 | @opindex maix32 | |
6750 | Enable 64-bit AIX ABI and calling convention: 64-bit pointers, 64-bit | |
6751 | @code{long} type, and the infrastructure needed to support them. | |
6752 | Specifying @option{-maix64} implies @option{-mpowerpc64} and | |
6753 | @option{-mpowerpc}, while @option{-maix32} disables the 64-bit ABI and | |
6754 | implies @option{-mno-powerpc64}. GCC defaults to @option{-maix32}. | |
6755 | ||
6756 | @item -mxl-call | |
6757 | @itemx -mno-xl-call | |
6758 | @opindex mxl-call | |
6759 | @opindex mno-xl-call | |
6760 | On AIX, pass floating-point arguments to prototyped functions beyond the | |
6761 | register save area (RSA) on the stack in addition to argument FPRs. The | |
6762 | AIX calling convention was extended but not initially documented to | |
6763 | handle an obscure K&R C case of calling a function that takes the | |
6764 | address of its arguments with fewer arguments than declared. AIX XL | |
6765 | compilers access floating point arguments which do not fit in the | |
6766 | RSA from the stack when a subroutine is compiled without | |
6767 | optimization. Because always storing floating-point arguments on the | |
6768 | stack is inefficient and rarely needed, this option is not enabled by | |
6769 | default and only is necessary when calling subroutines compiled by AIX | |
6770 | XL compilers without optimization. | |
6771 | ||
6772 | @item -mpe | |
6773 | @opindex mpe | |
6774 | Support @dfn{IBM RS/6000 SP} @dfn{Parallel Environment} (PE)@. Link an | |
6775 | application written to use message passing with special startup code to | |
6776 | enable the application to run. The system must have PE installed in the | |
6777 | standard location (@file{/usr/lpp/ppe.poe/}), or the @file{specs} file | |
6778 | must be overridden with the @option{-specs=} option to specify the | |
6779 | appropriate directory location. The Parallel Environment does not | |
6780 | support threads, so the @option{-mpe} option and the @option{-pthread} | |
6781 | option are incompatible. | |
6782 | ||
6783 | @item -msoft-float | |
6784 | @itemx -mhard-float | |
6785 | @opindex msoft-float | |
6786 | @opindex mhard-float | |
6787 | Generate code that does not use (uses) the floating-point register set. | |
6788 | Software floating point emulation is provided if you use the | |
6789 | @option{-msoft-float} option, and pass the option to GCC when linking. | |
6790 | ||
6791 | @item -mmultiple | |
6792 | @itemx -mno-multiple | |
6793 | @opindex mmultiple | |
6794 | @opindex mno-multiple | |
6795 | Generate code that uses (does not use) the load multiple word | |
6796 | instructions and the store multiple word instructions. These | |
6797 | instructions are generated by default on POWER systems, and not | |
6798 | generated on PowerPC systems. Do not use @option{-mmultiple} on little | |
6799 | endian PowerPC systems, since those instructions do not work when the | |
6800 | processor is in little endian mode. The exceptions are PPC740 and | |
6801 | PPC750 which permit the instructions usage in little endian mode. | |
6802 | ||
6803 | @item -mstring | |
6804 | @itemx -mno-string | |
6805 | @opindex mstring | |
6806 | @opindex mno-string | |
6807 | Generate code that uses (does not use) the load string instructions | |
6808 | and the store string word instructions to save multiple registers and | |
6809 | do small block moves. These instructions are generated by default on | |
6810 | POWER systems, and not generated on PowerPC systems. Do not use | |
6811 | @option{-mstring} on little endian PowerPC systems, since those | |
6812 | instructions do not work when the processor is in little endian mode. | |
6813 | The exceptions are PPC740 and PPC750 which permit the instructions | |
6814 | usage in little endian mode. | |
6815 | ||
6816 | @item -mupdate | |
6817 | @itemx -mno-update | |
6818 | @opindex mupdate | |
6819 | @opindex mno-update | |
6820 | Generate code that uses (does not use) the load or store instructions | |
6821 | that update the base register to the address of the calculated memory | |
6822 | location. These instructions are generated by default. If you use | |
6823 | @option{-mno-update}, there is a small window between the time that the | |
6824 | stack pointer is updated and the address of the previous frame is | |
6825 | stored, which means code that walks the stack frame across interrupts or | |
6826 | signals may get corrupted data. | |
6827 | ||
6828 | @item -mfused-madd | |
6829 | @itemx -mno-fused-madd | |
6830 | @opindex mfused-madd | |
6831 | @opindex mno-fused-madd | |
6832 | Generate code that uses (does not use) the floating point multiply and | |
6833 | accumulate instructions. These instructions are generated by default if | |
6834 | hardware floating is used. | |
6835 | ||
6836 | @item -mno-bit-align | |
6837 | @itemx -mbit-align | |
6838 | @opindex mno-bit-align | |
6839 | @opindex mbit-align | |
6840 | On System V.4 and embedded PowerPC systems do not (do) force structures | |
6841 | and unions that contain bit-fields to be aligned to the base type of the | |
6842 | bit-field. | |
6843 | ||
6844 | For example, by default a structure containing nothing but 8 | |
6845 | @code{unsigned} bit-fields of length 1 would be aligned to a 4 byte | |
6846 | boundary and have a size of 4 bytes. By using @option{-mno-bit-align}, | |
6847 | the structure would be aligned to a 1 byte boundary and be one byte in | |
6848 | size. | |
6849 | ||
6850 | @item -mno-strict-align | |
6851 | @itemx -mstrict-align | |
6852 | @opindex mno-strict-align | |
6853 | @opindex mstrict-align | |
6854 | On System V.4 and embedded PowerPC systems do not (do) assume that | |
6855 | unaligned memory references will be handled by the system. | |
6856 | ||
6857 | @item -mrelocatable | |
6858 | @itemx -mno-relocatable | |
6859 | @opindex mrelocatable | |
6860 | @opindex mno-relocatable | |
6861 | On embedded PowerPC systems generate code that allows (does not allow) | |
6862 | the program to be relocated to a different address at runtime. If you | |
6863 | use @option{-mrelocatable} on any module, all objects linked together must | |
6864 | be compiled with @option{-mrelocatable} or @option{-mrelocatable-lib}. | |
6865 | ||
6866 | @item -mrelocatable-lib | |
6867 | @itemx -mno-relocatable-lib | |
6868 | @opindex mrelocatable-lib | |
6869 | @opindex mno-relocatable-lib | |
6870 | On embedded PowerPC systems generate code that allows (does not allow) | |
6871 | the program to be relocated to a different address at runtime. Modules | |
6872 | compiled with @option{-mrelocatable-lib} can be linked with either modules | |
6873 | compiled without @option{-mrelocatable} and @option{-mrelocatable-lib} or | |
6874 | with modules compiled with the @option{-mrelocatable} options. | |
6875 | ||
6876 | @item -mno-toc | |
6877 | @itemx -mtoc | |
6878 | @opindex mno-toc | |
6879 | @opindex mtoc | |
6880 | On System V.4 and embedded PowerPC systems do not (do) assume that | |
6881 | register 2 contains a pointer to a global area pointing to the addresses | |
6882 | used in the program. | |
6883 | ||
6884 | @item -mlittle | |
6885 | @itemx -mlittle-endian | |
6886 | @opindex mlittle | |
6887 | @opindex mlittle-endian | |
6888 | On System V.4 and embedded PowerPC systems compile code for the | |
6889 | processor in little endian mode. The @option{-mlittle-endian} option is | |
6890 | the same as @option{-mlittle}. | |
6891 | ||
6892 | @item -mbig | |
6893 | @itemx -mbig-endian | |
6894 | @opindex mbig | |
6895 | @opindex mbig-endian | |
6896 | On System V.4 and embedded PowerPC systems compile code for the | |
6897 | processor in big endian mode. The @option{-mbig-endian} option is | |
6898 | the same as @option{-mbig}. | |
6899 | ||
6900 | @item -mcall-sysv | |
6901 | @opindex mcall-sysv | |
6902 | On System V.4 and embedded PowerPC systems compile code using calling | |
6903 | conventions that adheres to the March 1995 draft of the System V | |
6904 | Application Binary Interface, PowerPC processor supplement. This is the | |
6905 | default unless you configured GCC using @samp{powerpc-*-eabiaix}. | |
6906 | ||
6907 | @item -mcall-sysv-eabi | |
6908 | @opindex mcall-sysv-eabi | |
6909 | Specify both @option{-mcall-sysv} and @option{-meabi} options. | |
6910 | ||
6911 | @item -mcall-sysv-noeabi | |
6912 | @opindex mcall-sysv-noeabi | |
6913 | Specify both @option{-mcall-sysv} and @option{-mno-eabi} options. | |
6914 | ||
6915 | @item -mcall-aix | |
6916 | @opindex mcall-aix | |
6917 | On System V.4 and embedded PowerPC systems compile code using calling | |
6918 | conventions that are similar to those used on AIX@. This is the | |
6919 | default if you configured GCC using @samp{powerpc-*-eabiaix}. | |
6920 | ||
6921 | @item -mcall-solaris | |
6922 | @opindex mcall-solaris | |
6923 | On System V.4 and embedded PowerPC systems compile code for the Solaris | |
6924 | operating system. | |
6925 | ||
6926 | @item -mcall-linux | |
6927 | @opindex mcall-linux | |
6928 | On System V.4 and embedded PowerPC systems compile code for the | |
6929 | Linux-based GNU system. | |
6930 | ||
6931 | @item -mcall-gnu | |
6932 | @opindex mcall-gnu | |
6933 | On System V.4 and embedded PowerPC systems compile code for the | |
6934 | Hurd-based GNU system. | |
6935 | ||
6936 | @item -mcall-netbsd | |
6937 | @opindex mcall-netbsd | |
6938 | On System V.4 and embedded PowerPC systems compile code for the | |
6939 | NetBSD operating system. | |
6940 | ||
6941 | @item -maix-struct-return | |
6942 | @opindex maix-struct-return | |
6943 | Return all structures in memory (as specified by the AIX ABI)@. | |
6944 | ||
6945 | @item -msvr4-struct-return | |
6946 | @opindex msvr4-struct-return | |
6947 | Return structures smaller than 8 bytes in registers (as specified by the | |
6948 | SVR4 ABI)@. | |
6949 | ||
6950 | @item -mabi=altivec | |
6951 | @opindex mabi=altivec | |
6952 | Extend the current ABI with AltiVec ABI extensions. This does not | |
6953 | change the default ABI, instead it adds the AltiVec ABI extensions to | |
6954 | the current ABI@. | |
6955 | ||
6956 | @item -mprototype | |
6957 | @itemx -mno-prototype | |
6958 | @opindex mprototype | |
6959 | @opindex mno-prototype | |
6960 | On System V.4 and embedded PowerPC systems assume that all calls to | |
6961 | variable argument functions are properly prototyped. Otherwise, the | |
6962 | compiler must insert an instruction before every non prototyped call to | |
6963 | set or clear bit 6 of the condition code register (@var{CR}) to | |
6964 | indicate whether floating point values were passed in the floating point | |
6965 | registers in case the function takes a variable arguments. With | |
6966 | @option{-mprototype}, only calls to prototyped variable argument functions | |
6967 | will set or clear the bit. | |
6968 | ||
6969 | @item -msim | |
6970 | @opindex msim | |
6971 | On embedded PowerPC systems, assume that the startup module is called | |
6972 | @file{sim-crt0.o} and that the standard C libraries are @file{libsim.a} and | |
6973 | @file{libc.a}. This is the default for @samp{powerpc-*-eabisim}. | |
6974 | configurations. | |
6975 | ||
6976 | @item -mmvme | |
6977 | @opindex mmvme | |
6978 | On embedded PowerPC systems, assume that the startup module is called | |
6979 | @file{crt0.o} and the standard C libraries are @file{libmvme.a} and | |
6980 | @file{libc.a}. | |
6981 | ||
6982 | @item -mads | |
6983 | @opindex mads | |
6984 | On embedded PowerPC systems, assume that the startup module is called | |
6985 | @file{crt0.o} and the standard C libraries are @file{libads.a} and | |
6986 | @file{libc.a}. | |
6987 | ||
6988 | @item -myellowknife | |
6989 | @opindex myellowknife | |
6990 | On embedded PowerPC systems, assume that the startup module is called | |
6991 | @file{crt0.o} and the standard C libraries are @file{libyk.a} and | |
6992 | @file{libc.a}. | |
6993 | ||
6994 | @item -mvxworks | |
6995 | @opindex mvxworks | |
6996 | On System V.4 and embedded PowerPC systems, specify that you are | |
6997 | compiling for a VxWorks system. | |
6998 | ||
6999 | @item -memb | |
7000 | @opindex memb | |
7001 | On embedded PowerPC systems, set the @var{PPC_EMB} bit in the ELF flags | |
7002 | header to indicate that @samp{eabi} extended relocations are used. | |
7003 | ||
7004 | @item -meabi | |
7005 | @itemx -mno-eabi | |
7006 | @opindex meabi | |
7007 | @opindex mno-eabi | |
7008 | On System V.4 and embedded PowerPC systems do (do not) adhere to the | |
7009 | Embedded Applications Binary Interface (eabi) which is a set of | |
7010 | modifications to the System V.4 specifications. Selecting @option{-meabi} | |
7011 | means that the stack is aligned to an 8 byte boundary, a function | |
7012 | @code{__eabi} is called to from @code{main} to set up the eabi | |
7013 | environment, and the @option{-msdata} option can use both @code{r2} and | |
7014 | @code{r13} to point to two separate small data areas. Selecting | |
7015 | @option{-mno-eabi} means that the stack is aligned to a 16 byte boundary, | |
7016 | do not call an initialization function from @code{main}, and the | |
7017 | @option{-msdata} option will only use @code{r13} to point to a single | |
7018 | small data area. The @option{-meabi} option is on by default if you | |
7019 | configured GCC using one of the @samp{powerpc*-*-eabi*} options. | |
7020 | ||
7021 | @item -msdata=eabi | |
7022 | @opindex msdata=eabi | |
7023 | On System V.4 and embedded PowerPC systems, put small initialized | |
7024 | @code{const} global and static data in the @samp{.sdata2} section, which | |
7025 | is pointed to by register @code{r2}. Put small initialized | |
7026 | non-@code{const} global and static data in the @samp{.sdata} section, | |
7027 | which is pointed to by register @code{r13}. Put small uninitialized | |
7028 | global and static data in the @samp{.sbss} section, which is adjacent to | |
7029 | the @samp{.sdata} section. The @option{-msdata=eabi} option is | |
7030 | incompatible with the @option{-mrelocatable} option. The | |
7031 | @option{-msdata=eabi} option also sets the @option{-memb} option. | |
7032 | ||
7033 | @item -msdata=sysv | |
7034 | @opindex msdata=sysv | |
7035 | On System V.4 and embedded PowerPC systems, put small global and static | |
7036 | data in the @samp{.sdata} section, which is pointed to by register | |
7037 | @code{r13}. Put small uninitialized global and static data in the | |
7038 | @samp{.sbss} section, which is adjacent to the @samp{.sdata} section. | |
7039 | The @option{-msdata=sysv} option is incompatible with the | |
7040 | @option{-mrelocatable} option. | |
7041 | ||
7042 | @item -msdata=default | |
7043 | @itemx -msdata | |
7044 | @opindex msdata=default | |
7045 | @opindex msdata | |
7046 | On System V.4 and embedded PowerPC systems, if @option{-meabi} is used, | |
7047 | compile code the same as @option{-msdata=eabi}, otherwise compile code the | |
7048 | same as @option{-msdata=sysv}. | |
7049 | ||
7050 | @item -msdata-data | |
7051 | @opindex msdata-data | |
7052 | On System V.4 and embedded PowerPC systems, put small global and static | |
7053 | data in the @samp{.sdata} section. Put small uninitialized global and | |
7054 | static data in the @samp{.sbss} section. Do not use register @code{r13} | |
7055 | to address small data however. This is the default behavior unless | |
7056 | other @option{-msdata} options are used. | |
7057 | ||
7058 | @item -msdata=none | |
7059 | @itemx -mno-sdata | |
7060 | @opindex msdata=none | |
7061 | @opindex mno-sdata | |
7062 | On embedded PowerPC systems, put all initialized global and static data | |
7063 | in the @samp{.data} section, and all uninitialized data in the | |
7064 | @samp{.bss} section. | |
7065 | ||
7066 | @item -G @var{num} | |
7067 | @opindex G | |
7068 | @cindex smaller data references (PowerPC) | |
7069 | @cindex .sdata/.sdata2 references (PowerPC) | |
7070 | On embedded PowerPC systems, put global and static items less than or | |
7071 | equal to @var{num} bytes into the small data or bss sections instead of | |
7072 | the normal data or bss section. By default, @var{num} is 8. The | |
7073 | @option{-G @var{num}} switch is also passed to the linker. | |
7074 | All modules should be compiled with the same @option{-G @var{num}} value. | |
7075 | ||
7076 | @item -mregnames | |
7077 | @itemx -mno-regnames | |
7078 | @opindex mregnames | |
7079 | @opindex mno-regnames | |
7080 | On System V.4 and embedded PowerPC systems do (do not) emit register | |
7081 | names in the assembly language output using symbolic forms. | |
7082 | ||
7083 | @item -pthread | |
7084 | @opindex pthread | |
7085 | Adds support for multithreading with the @dfn{pthreads} library. | |
7086 | This option sets flags for both the preprocessor and linker. | |
7087 | ||
7088 | @end table | |
7089 | ||
7090 | @node RT Options | |
7091 | @subsection IBM RT Options | |
7092 | @cindex RT options | |
7093 | @cindex IBM RT options | |
7094 | ||
7095 | These @samp{-m} options are defined for the IBM RT PC: | |
7096 | ||
7097 | @table @gcctabopt | |
7098 | @item -min-line-mul | |
7099 | @opindex min-line-mul | |
7100 | Use an in-line code sequence for integer multiplies. This is the | |
7101 | default. | |
7102 | ||
7103 | @item -mcall-lib-mul | |
7104 | @opindex mcall-lib-mul | |
7105 | Call @code{lmul$$} for integer multiples. | |
7106 | ||
7107 | @item -mfull-fp-blocks | |
7108 | @opindex mfull-fp-blocks | |
7109 | Generate full-size floating point data blocks, including the minimum | |
7110 | amount of scratch space recommended by IBM@. This is the default. | |
7111 | ||
7112 | @item -mminimum-fp-blocks | |
7113 | @opindex mminimum-fp-blocks | |
7114 | Do not include extra scratch space in floating point data blocks. This | |
7115 | results in smaller code, but slower execution, since scratch space must | |
7116 | be allocated dynamically. | |
7117 | ||
7118 | @cindex @file{varargs.h} and RT PC | |
7119 | @cindex @file{stdarg.h} and RT PC | |
7120 | @item -mfp-arg-in-fpregs | |
7121 | @opindex mfp-arg-in-fpregs | |
7122 | Use a calling sequence incompatible with the IBM calling convention in | |
7123 | which floating point arguments are passed in floating point registers. | |
7124 | Note that @code{varargs.h} and @code{stdarg.h} will not work with | |
7125 | floating point operands if this option is specified. | |
7126 | ||
7127 | @item -mfp-arg-in-gregs | |
7128 | @opindex mfp-arg-in-gregs | |
7129 | Use the normal calling convention for floating point arguments. This is | |
7130 | the default. | |
7131 | ||
7132 | @item -mhc-struct-return | |
7133 | @opindex mhc-struct-return | |
7134 | Return structures of more than one word in memory, rather than in a | |
7135 | register. This provides compatibility with the MetaWare HighC (hc) | |
7136 | compiler. Use the option @option{-fpcc-struct-return} for compatibility | |
7137 | with the Portable C Compiler (pcc). | |
7138 | ||
7139 | @item -mnohc-struct-return | |
7140 | @opindex mnohc-struct-return | |
7141 | Return some structures of more than one word in registers, when | |
7142 | convenient. This is the default. For compatibility with the | |
7143 | IBM-supplied compilers, use the option @option{-fpcc-struct-return} or the | |
7144 | option @option{-mhc-struct-return}. | |
7145 | @end table | |
7146 | ||
7147 | @node MIPS Options | |
7148 | @subsection MIPS Options | |
7149 | @cindex MIPS options | |
7150 | ||
7151 | These @samp{-m} options are defined for the MIPS family of computers: | |
7152 | ||
7153 | @table @gcctabopt | |
7154 | ||
7155 | @item -march=@var{cpu-type} | |
7156 | @opindex march | |
7157 | Assume the defaults for the machine type @var{cpu-type} when generating | |
7158 | instructions. The choices for @var{cpu-type} are @samp{r2000}, @samp{r3000}, | |
7159 | @samp{r3900}, @samp{r4000}, @samp{r4100}, @samp{r4300}, @samp{r4400}, | |
7160 | @samp{r4600}, @samp{r4650}, @samp{r5000}, @samp{r6000}, @samp{r8000}, | |
7161 | and @samp{orion}. Additionally, the @samp{r2000}, @samp{r3000}, | |
7162 | @samp{r4000}, @samp{r5000}, and @samp{r6000} can be abbreviated as | |
7163 | @samp{r2k} (or @samp{r2K}), @samp{r3k}, etc. | |
7164 | ||
7165 | @item -mtune=@var{cpu-type} | |
7166 | @opindex mtune | |
7167 | Assume the defaults for the machine type @var{cpu-type} when scheduling | |
7168 | instructions. The choices for @var{cpu-type} are @samp{r2000}, @samp{r3000}, | |
7169 | @samp{r3900}, @samp{r4000}, @samp{r4100}, @samp{r4300}, @samp{r4400}, | |
7170 | @samp{r4600}, @samp{r4650}, @samp{r5000}, @samp{r6000}, @samp{r8000}, | |
7171 | and @samp{orion}. Additionally, the @samp{r2000}, @samp{r3000}, | |
7172 | @samp{r4000}, @samp{r5000}, and @samp{r6000} can be abbreviated as | |
7173 | @samp{r2k} (or @samp{r2K}), @samp{r3k}, etc. While picking a specific | |
7174 | @var{cpu-type} will schedule things appropriately for that particular | |
7175 | chip, the compiler will not generate any code that does not meet level 1 | |
7176 | of the MIPS ISA (instruction set architecture) without a @option{-mipsX} | |
7177 | or @option{-mabi} switch being used. | |
7178 | ||
7179 | @item -mcpu=@var{cpu-type} | |
7180 | @opindex mcpu | |
7181 | This is identical to specifying both @option{-march} and @option{-mtune}. | |
7182 | ||
7183 | @item -mips1 | |
7184 | @opindex mips1 | |
7185 | Issue instructions from level 1 of the MIPS ISA@. This is the default. | |
7186 | @samp{r3000} is the default @var{cpu-type} at this ISA level. | |
7187 | ||
7188 | @item -mips2 | |
7189 | @opindex mips2 | |
7190 | Issue instructions from level 2 of the MIPS ISA (branch likely, square | |
7191 | root instructions). @samp{r6000} is the default @var{cpu-type} at this | |
7192 | ISA level. | |
7193 | ||
7194 | @item -mips3 | |
7195 | @opindex mips3 | |
7196 | Issue instructions from level 3 of the MIPS ISA (64-bit instructions). | |
7197 | @samp{r4000} is the default @var{cpu-type} at this ISA level. | |
7198 | ||
7199 | @item -mips4 | |
7200 | @opindex mips4 | |
7201 | Issue instructions from level 4 of the MIPS ISA (conditional move, | |
7202 | prefetch, enhanced FPU instructions). @samp{r8000} is the default | |
7203 | @var{cpu-type} at this ISA level. | |
7204 | ||
7205 | @item -mfp32 | |
7206 | @opindex mfp32 | |
7207 | Assume that 32 32-bit floating point registers are available. This is | |
7208 | the default. | |
7209 | ||
7210 | @item -mfp64 | |
7211 | @opindex mfp64 | |
7212 | Assume that 32 64-bit floating point registers are available. This is | |
7213 | the default when the @option{-mips3} option is used. | |
7214 | ||
7215 | @item -mfused-madd | |
7216 | @itemx -mno-fused-madd | |
7217 | @opindex mfused-madd | |
7218 | @opindex mno-fused-madd | |
7219 | Generate code that uses (does not use) the floating point multiply and | |
7220 | accumulate instructions, when they are available. These instructions | |
7221 | are generated by default if they are available, but this may be | |
7222 | undesirable if the extra precision causes problems or on certain chips | |
7223 | in the mode where denormals are rounded to zero where denormals | |
7224 | generated by multiply and accumulate instructions cause exceptions | |
7225 | anyway. | |
7226 | ||
7227 | @item -mgp32 | |
7228 | @opindex mgp32 | |
7229 | Assume that 32 32-bit general purpose registers are available. This is | |
7230 | the default. | |
7231 | ||
7232 | @item -mgp64 | |
7233 | @opindex mgp64 | |
7234 | Assume that 32 64-bit general purpose registers are available. This is | |
7235 | the default when the @option{-mips3} option is used. | |
7236 | ||
7237 | @item -mint64 | |
7238 | @opindex mint64 | |
7239 | Force int and long types to be 64 bits wide. See @option{-mlong32} for an | |
7240 | explanation of the default, and the width of pointers. | |
7241 | ||
7242 | @item -mlong64 | |
7243 | @opindex mlong64 | |
7244 | Force long types to be 64 bits wide. See @option{-mlong32} for an | |
7245 | explanation of the default, and the width of pointers. | |
7246 | ||
7247 | @item -mlong32 | |
7248 | @opindex mlong32 | |
7249 | Force long, int, and pointer types to be 32 bits wide. | |
7250 | ||
7251 | If none of @option{-mlong32}, @option{-mlong64}, or @option{-mint64} are set, | |
7252 | the size of ints, longs, and pointers depends on the ABI and ISA chosen. | |
7253 | For @option{-mabi=32}, and @option{-mabi=n32}, ints and longs are 32 bits | |
7254 | wide. For @option{-mabi=64}, ints are 32 bits, and longs are 64 bits wide. | |
7255 | For @option{-mabi=eabi} and either @option{-mips1} or @option{-mips2}, ints | |
7256 | and longs are 32 bits wide. For @option{-mabi=eabi} and higher ISAs, ints | |
7257 | are 32 bits, and longs are 64 bits wide. The width of pointer types is | |
7258 | the smaller of the width of longs or the width of general purpose | |
7259 | registers (which in turn depends on the ISA)@. | |
7260 | ||
7261 | @item -mabi=32 | |
7262 | @itemx -mabi=o64 | |
7263 | @itemx -mabi=n32 | |
7264 | @itemx -mabi=64 | |
7265 | @itemx -mabi=eabi | |
7266 | @opindex mabi=32 | |
7267 | @opindex mabi=o64 | |
7268 | @opindex mabi=n32 | |
7269 | @opindex mabi=64 | |
7270 | @opindex mabi=eabi | |
7271 | Generate code for the indicated ABI@. The default instruction level is | |
7272 | @option{-mips1} for @samp{32}, @option{-mips3} for @samp{n32}, and | |
7273 | @option{-mips4} otherwise. Conversely, with @option{-mips1} or | |
7274 | @option{-mips2}, the default ABI is @samp{32}; otherwise, the default ABI | |
7275 | is @samp{64}. | |
7276 | ||
7277 | @item -mmips-as | |
7278 | @opindex mmips-as | |
7279 | Generate code for the MIPS assembler, and invoke @file{mips-tfile} to | |
7280 | add normal debug information. This is the default for all | |
7281 | platforms except for the OSF/1 reference platform, using the OSF/rose | |
7282 | object format. If the either of the @option{-gstabs} or @option{-gstabs+} | |
7283 | switches are used, the @file{mips-tfile} program will encapsulate the | |
7284 | stabs within MIPS ECOFF@. | |
7285 | ||
7286 | @item -mgas | |
7287 | @opindex mgas | |
7288 | Generate code for the GNU assembler. This is the default on the OSF/1 | |
7289 | reference platform, using the OSF/rose object format. Also, this is | |
7290 | the default if the configure option @option{--with-gnu-as} is used. | |
7291 | ||
7292 | @item -msplit-addresses | |
7293 | @itemx -mno-split-addresses | |
7294 | @opindex msplit-addresses | |
7295 | @opindex mno-split-addresses | |
7296 | Generate code to load the high and low parts of address constants separately. | |
7297 | This allows GCC to optimize away redundant loads of the high order | |
7298 | bits of addresses. This optimization requires GNU as and GNU ld. | |
7299 | This optimization is enabled by default for some embedded targets where | |
7300 | GNU as and GNU ld are standard. | |
7301 | ||
7302 | @item -mrnames | |
7303 | @itemx -mno-rnames | |
7304 | @opindex mrnames | |
7305 | @opindex mno-rnames | |
7306 | The @option{-mrnames} switch says to output code using the MIPS software | |
7307 | names for the registers, instead of the hardware names (ie, @var{a0} | |
7308 | instead of @var{$4}). The only known assembler that supports this option | |
7309 | is the Algorithmics assembler. | |
7310 | ||
7311 | @item -mgpopt | |
7312 | @itemx -mno-gpopt | |
7313 | @opindex mgpopt | |
7314 | @opindex mno-gpopt | |
7315 | The @option{-mgpopt} switch says to write all of the data declarations | |
7316 | before the instructions in the text section, this allows the MIPS | |
7317 | assembler to generate one word memory references instead of using two | |
7318 | words for short global or static data items. This is on by default if | |
7319 | optimization is selected. | |
7320 | ||
7321 | @item -mstats | |
7322 | @itemx -mno-stats | |
7323 | @opindex mstats | |
7324 | @opindex mno-stats | |
7325 | For each non-inline function processed, the @option{-mstats} switch | |
7326 | causes the compiler to emit one line to the standard error file to | |
7327 | print statistics about the program (number of registers saved, stack | |
7328 | size, etc.). | |
7329 | ||
7330 | @item -mmemcpy | |
7331 | @itemx -mno-memcpy | |
7332 | @opindex mmemcpy | |
7333 | @opindex mno-memcpy | |
7334 | The @option{-mmemcpy} switch makes all block moves call the appropriate | |
7335 | string function (@samp{memcpy} or @samp{bcopy}) instead of possibly | |
7336 | generating inline code. | |
7337 | ||
7338 | @item -mmips-tfile | |
7339 | @itemx -mno-mips-tfile | |
7340 | @opindex mmips-tfile | |
7341 | @opindex mno-mips-tfile | |
7342 | The @option{-mno-mips-tfile} switch causes the compiler not | |
7343 | postprocess the object file with the @file{mips-tfile} program, | |
7344 | after the MIPS assembler has generated it to add debug support. If | |
7345 | @file{mips-tfile} is not run, then no local variables will be | |
7346 | available to the debugger. In addition, @file{stage2} and | |
7347 | @file{stage3} objects will have the temporary file names passed to the | |
7348 | assembler embedded in the object file, which means the objects will | |
7349 | not compare the same. The @option{-mno-mips-tfile} switch should only | |
7350 | be used when there are bugs in the @file{mips-tfile} program that | |
7351 | prevents compilation. | |
7352 | ||
7353 | @item -msoft-float | |
7354 | @opindex msoft-float | |
7355 | Generate output containing library calls for floating point. | |
7356 | @strong{Warning:} the requisite libraries are not part of GCC@. | |
7357 | Normally the facilities of the machine's usual C compiler are used, but | |
7358 | this can't be done directly in cross-compilation. You must make your | |
7359 | own arrangements to provide suitable library functions for | |
7360 | cross-compilation. | |
7361 | ||
7362 | @item -mhard-float | |
7363 | @opindex mhard-float | |
7364 | Generate output containing floating point instructions. This is the | |
7365 | default if you use the unmodified sources. | |
7366 | ||
7367 | @item -mabicalls | |
7368 | @itemx -mno-abicalls | |
7369 | @opindex mabicalls | |
7370 | @opindex mno-abicalls | |
7371 | Emit (or do not emit) the pseudo operations @samp{.abicalls}, | |
7372 | @samp{.cpload}, and @samp{.cprestore} that some System V.4 ports use for | |
7373 | position independent code. | |
7374 | ||
7375 | @item -mlong-calls | |
7376 | @itemx -mno-long-calls | |
7377 | @opindex mlong-calls | |
7378 | @opindex mno-long-calls | |
7379 | Do all calls with the @samp{JALR} instruction, which requires | |
7380 | loading up a function's address into a register before the call. | |
7381 | You need to use this switch, if you call outside of the current | |
7382 | 512 megabyte segment to functions that are not through pointers. | |
7383 | ||
7384 | @item -mhalf-pic | |
7385 | @itemx -mno-half-pic | |
7386 | @opindex mhalf-pic | |
7387 | @opindex mno-half-pic | |
7388 | Put pointers to extern references into the data section and load them | |
7389 | up, rather than put the references in the text section. | |
7390 | ||
7391 | @item -membedded-pic | |
7392 | @itemx -mno-embedded-pic | |
7393 | @opindex membedded-pic | |
7394 | @opindex mno-embedded-pic | |
7395 | Generate PIC code suitable for some embedded systems. All calls are | |
7396 | made using PC relative address, and all data is addressed using the $gp | |
7397 | register. No more than 65536 bytes of global data may be used. This | |
7398 | requires GNU as and GNU ld which do most of the work. This currently | |
7399 | only works on targets which use ECOFF; it does not work with ELF@. | |
7400 | ||
7401 | @item -membedded-data | |
7402 | @itemx -mno-embedded-data | |
7403 | @opindex membedded-data | |
7404 | @opindex mno-embedded-data | |
7405 | Allocate variables to the read-only data section first if possible, then | |
7406 | next in the small data section if possible, otherwise in data. This gives | |
7407 | slightly slower code than the default, but reduces the amount of RAM required | |
7408 | when executing, and thus may be preferred for some embedded systems. | |
7409 | ||
7410 | @item -muninit-const-in-rodata | |
7411 | @itemx -mno-uninit-const-in-rodata | |
7412 | @opindex muninit-const-in-rodata | |
7413 | @opindex mno-uninit-const-in-rodata | |
7414 | When used together with @option{-membedded-data}, it will always store uninitialized | |
7415 | const variables in the read-only data section. | |
7416 | ||
7417 | @item -msingle-float | |
7418 | @itemx -mdouble-float | |
7419 | @opindex msingle-float | |
7420 | @opindex mdouble-float | |
7421 | The @option{-msingle-float} switch tells gcc to assume that the floating | |
7422 | point coprocessor only supports single precision operations, as on the | |
7423 | @samp{r4650} chip. The @option{-mdouble-float} switch permits gcc to use | |
7424 | double precision operations. This is the default. | |
7425 | ||
7426 | @item -mmad | |
7427 | @itemx -mno-mad | |
7428 | @opindex mmad | |
7429 | @opindex mno-mad | |
7430 | Permit use of the @samp{mad}, @samp{madu} and @samp{mul} instructions, | |
7431 | as on the @samp{r4650} chip. | |
7432 | ||
7433 | @item -m4650 | |
7434 | @opindex m4650 | |
7435 | Turns on @option{-msingle-float}, @option{-mmad}, and, at least for now, | |
7436 | @option{-mcpu=r4650}. | |
7437 | ||
7438 | @item -mips16 | |
7439 | @itemx -mno-mips16 | |
7440 | @opindex mips16 | |
7441 | @opindex mno-mips16 | |
7442 | Enable 16-bit instructions. | |
7443 | ||
7444 | @item -mentry | |
7445 | @opindex mentry | |
7446 | Use the entry and exit pseudo ops. This option can only be used with | |
7447 | @option{-mips16}. | |
7448 | ||
7449 | @item -EL | |
7450 | @opindex EL | |
7451 | Compile code for the processor in little endian mode. | |
7452 | The requisite libraries are assumed to exist. | |
7453 | ||
7454 | @item -EB | |
7455 | @opindex EB | |
7456 | Compile code for the processor in big endian mode. | |
7457 | The requisite libraries are assumed to exist. | |
7458 | ||
7459 | @item -G @var{num} | |
7460 | @opindex G | |
7461 | @cindex smaller data references (MIPS) | |
7462 | @cindex gp-relative references (MIPS) | |
7463 | Put global and static items less than or equal to @var{num} bytes into | |
7464 | the small data or bss sections instead of the normal data or bss | |
7465 | section. This allows the assembler to emit one word memory reference | |
7466 | instructions based on the global pointer (@var{gp} or @var{$28}), | |
7467 | instead of the normal two words used. By default, @var{num} is 8 when | |
7468 | the MIPS assembler is used, and 0 when the GNU assembler is used. The | |
7469 | @option{-G @var{num}} switch is also passed to the assembler and linker. | |
7470 | All modules should be compiled with the same @option{-G @var{num}} | |
7471 | value. | |
7472 | ||
7473 | @item -nocpp | |
7474 | @opindex nocpp | |
7475 | Tell the MIPS assembler to not run its preprocessor over user | |
7476 | assembler files (with a @samp{.s} suffix) when assembling them. | |
7477 | ||
7478 | @item -mfix7000 | |
7479 | @opindex mfix7000 | |
7480 | Pass an option to gas which will cause nops to be inserted if | |
7481 | the read of the destination register of an mfhi or mflo instruction | |
7482 | occurs in the following two instructions. | |
7483 | ||
7484 | @item -no-crt0 | |
7485 | @opindex no-crt0 | |
7486 | Do not include the default crt0. | |
7487 | ||
7488 | @item -mflush-func=@var{func} | |
7489 | @itemx -mno-flush-func | |
7490 | @opindex mflush-func | |
7491 | Specifies the function to call to flush the I and D caches, or to not | |
7492 | call any such function. If called, the function must take the same | |
7493 | arguments as the common @code{_flush_func()}, that is, the address of the | |
7494 | memory range for which the cache is being flushed, the size of the | |
7495 | memory range, and the number 3 (to flush both caches). The default | |
7496 | depends on the target gcc was configured for, but commonly is either | |
7497 | @samp{_flush_func} or @samp{__cpu_flush}. | |
7498 | @end table | |
7499 | ||
7500 | These options are defined by the macro | |
7501 | @code{TARGET_SWITCHES} in the machine description. The default for the | |
7502 | options is also defined by that macro, which enables you to change the | |
7503 | defaults. | |
7504 | ||
7505 | @node i386 and x86-64 Options | |
7506 | @subsection Intel 386 and AMD x86-64 Options | |
7507 | @cindex i386 Options | |
7508 | @cindex x86-64 Options | |
7509 | @cindex Intel 386 Options | |
7510 | @cindex AMD x86-64 Options | |
7511 | ||
7512 | These @samp{-m} options are defined for the i386 and x86-64 family of | |
7513 | computers: | |
7514 | ||
7515 | @table @gcctabopt | |
7516 | @item -mcpu=@var{cpu-type} | |
7517 | @opindex mcpu | |
7518 | Tune to @var{cpu-type} everything applicable about the generated code, except | |
7519 | for the ABI and the set of available instructions. The choices for | |
7520 | @var{cpu-type} are @samp{i386}, @samp{i486}, @samp{i586}, @samp{i686}, | |
7521 | @samp{pentium}, @samp{pentium-mmx}, @samp{pentiumpro}, @samp{pentium2}, | |
7522 | @samp{pentium3}, @samp{pentium4}, @samp{k6}, @samp{k6-2}, @samp{k6-3}, | |
7523 | @samp{athlon}, @samp{athlon-tbird}, @samp{athlon-4}, @samp{athlon-xp} | |
7524 | and @samp{athlon-mp}. | |
7525 | ||
7526 | While picking a specific @var{cpu-type} will schedule things appropriately | |
7527 | for that particular chip, the compiler will not generate any code that | |
7528 | does not run on the i386 without the @option{-march=@var{cpu-type}} option | |
7529 | being used. @samp{i586} is equivalent to @samp{pentium} and @samp{i686} | |
7530 | is equivalent to @samp{pentiumpro}. @samp{k6} and @samp{athlon} are the | |
7531 | AMD chips as opposed to the Intel ones. | |
7532 | ||
7533 | @item -march=@var{cpu-type} | |
7534 | @opindex march | |
7535 | Generate instructions for the machine type @var{cpu-type}. The choices | |
7536 | for @var{cpu-type} are the same as for @option{-mcpu}. Moreover, | |
7537 | specifying @option{-march=@var{cpu-type}} implies @option{-mcpu=@var{cpu-type}}. | |
7538 | ||
7539 | @item -m386 | |
7540 | @itemx -m486 | |
7541 | @itemx -mpentium | |
7542 | @itemx -mpentiumpro | |
7543 | @opindex m386 | |
7544 | @opindex m486 | |
7545 | @opindex mpentium | |
7546 | @opindex mpentiumpro | |
7547 | These options are synonyms for @option{-mcpu=i386}, @option{-mcpu=i486}, | |
7548 | @option{-mcpu=pentium}, and @option{-mcpu=pentiumpro} respectively. | |
7549 | These synonyms are deprecated. | |
7550 | ||
7551 | @item -mfpmath=@var{unit} | |
7552 | @opindex march | |
7553 | generate floating point arithmetics for selected unit @var{unit}. the choices | |
7554 | for @var{unit} are: | |
7555 | ||
7556 | @table @samp | |
7557 | @item 387 | |
7558 | Use the standard 387 floating point coprocessor present majority of chips and | |
7559 | emulated otherwise. Code compiled with this option will run almost everywhere. | |
7560 | The temporary results are computed in 80bit precesion instead of precision | |
7561 | specified by the type resulting in slightly different results compared to most | |
7562 | of other chips. See @option{-ffloat-store} for more detailed description. | |
7563 | ||
7564 | This is the default choice for i386 compiler. | |
7565 | ||
7566 | @item sse | |
7567 | Use scalar floating point instructions present in the SSE instruction set. | |
7568 | This instruction set is supported by Pentium3 and newer chips, in the AMD line | |
7569 | by Athlon-4, Athlon-xp and Athlon-mp chips. The earlier version of SSE | |
7570 | instruction set supports only single precision arithmetics, thus the double and | |
7571 | extended precision arithmetics is still done using 387. Later version, present | |
7572 | only in Pentium4 and the future AMD x86-64 chips supports double precision | |
7573 | arithmetics too. | |
7574 | ||
7575 | For i387 you need to use @option{-march=@var{cpu-type}}, @option{-msse} or | |
7576 | @option{-msse2} switches to enable SSE extensions and make this option | |
7577 | effective. For x86-64 compiler, these extensions are enabled by default. | |
7578 | ||
7579 | The resulting code should be considerably faster in majority of cases and avoid | |
7580 | the numerical instability problems of 387 code, but may break some existing | |
7581 | code that expects temporaries to be 80bit. | |
7582 | ||
7583 | This is the default choice for x86-64 compiler. | |
7584 | ||
7585 | @item sse,387 | |
7586 | Attempt to utilize both instruction sets at once. This effectivly double the | |
7587 | amount of available registers and on chips with separate execution units for | |
7588 | 387 and SSE the execution resources too. Use this option with care, as it is | |
7589 | still experimental, because gcc register allocator does not model separate | |
7590 | functional units well resulting in instable performance. | |
7591 | @end table | |
7592 | ||
7593 | @item -masm=@var{dialect} | |
7594 | @opindex masm=@var{dialect} | |
7595 | Output asm instructions using selected @var{dialect}. Supported choices are | |
7596 | @samp{intel} or @samp{att} (the default one). | |
7597 | ||
7598 | @item -mieee-fp | |
7599 | @itemx -mno-ieee-fp | |
7600 | @opindex mieee-fp | |
7601 | @opindex mno-ieee-fp | |
7602 | Control whether or not the compiler uses IEEE floating point | |
7603 | comparisons. These handle correctly the case where the result of a | |
7604 | comparison is unordered. | |
7605 | ||
7606 | @item -msoft-float | |
7607 | @opindex msoft-float | |
7608 | Generate output containing library calls for floating point. | |
7609 | @strong{Warning:} the requisite libraries are not part of GCC@. | |
7610 | Normally the facilities of the machine's usual C compiler are used, but | |
7611 | this can't be done directly in cross-compilation. You must make your | |
7612 | own arrangements to provide suitable library functions for | |
7613 | cross-compilation. | |
7614 | ||
7615 | On machines where a function returns floating point results in the 80387 | |
7616 | register stack, some floating point opcodes may be emitted even if | |
7617 | @option{-msoft-float} is used. | |
7618 | ||
7619 | @item -mno-fp-ret-in-387 | |
7620 | @opindex mno-fp-ret-in-387 | |
7621 | Do not use the FPU registers for return values of functions. | |
7622 | ||
7623 | The usual calling convention has functions return values of types | |
7624 | @code{float} and @code{double} in an FPU register, even if there | |
7625 | is no FPU@. The idea is that the operating system should emulate | |
7626 | an FPU@. | |
7627 | ||
7628 | The option @option{-mno-fp-ret-in-387} causes such values to be returned | |
7629 | in ordinary CPU registers instead. | |
7630 | ||
7631 | @item -mno-fancy-math-387 | |
7632 | @opindex mno-fancy-math-387 | |
7633 | Some 387 emulators do not support the @code{sin}, @code{cos} and | |
7634 | @code{sqrt} instructions for the 387. Specify this option to avoid | |
7635 | generating those instructions. This option is the default on FreeBSD@. | |
7636 | As of revision 2.6.1, these instructions are not generated unless you | |
7637 | also use the @option{-funsafe-math-optimizations} switch. | |
7638 | ||
7639 | @item -malign-double | |
7640 | @itemx -mno-align-double | |
7641 | @opindex malign-double | |
7642 | @opindex mno-align-double | |
7643 | Control whether GCC aligns @code{double}, @code{long double}, and | |
7644 | @code{long long} variables on a two word boundary or a one word | |
7645 | boundary. Aligning @code{double} variables on a two word boundary will | |
7646 | produce code that runs somewhat faster on a @samp{Pentium} at the | |
7647 | expense of more memory. | |
7648 | ||
7649 | @item -m128bit-long-double | |
7650 | @opindex m128bit-long-double | |
7651 | Control the size of @code{long double} type. i386 application binary interface | |
7652 | specify the size to be 12 bytes, while modern architectures (Pentium and newer) | |
7653 | prefer @code{long double} aligned to 8 or 16 byte boundary. This is | |
7654 | impossible to reach with 12 byte long doubles in the array accesses. | |
7655 | ||
7656 | @strong{Warning:} if you use the @option{-m128bit-long-double} switch, the | |
7657 | structures and arrays containing @code{long double} will change their size as | |
7658 | well as function calling convention for function taking @code{long double} | |
7659 | will be modified. | |
7660 | ||
7661 | @item -m96bit-long-double | |
7662 | @opindex m96bit-long-double | |
7663 | Set the size of @code{long double} to 96 bits as required by the i386 | |
7664 | application binary interface. This is the default. | |
7665 | ||
7666 | @item -msvr3-shlib | |
7667 | @itemx -mno-svr3-shlib | |
7668 | @opindex msvr3-shlib | |
7669 | @opindex mno-svr3-shlib | |
7670 | Control whether GCC places uninitialized local variables into the | |
7671 | @code{bss} or @code{data} segments. @option{-msvr3-shlib} places them | |
7672 | into @code{bss}. These options are meaningful only on System V Release 3. | |
7673 | ||
7674 | @item -mrtd | |
7675 | @opindex mrtd | |
7676 | Use a different function-calling convention, in which functions that | |
7677 | take a fixed number of arguments return with the @code{ret} @var{num} | |
7678 | instruction, which pops their arguments while returning. This saves one | |
7679 | instruction in the caller since there is no need to pop the arguments | |
7680 | there. | |
7681 | ||
7682 | You can specify that an individual function is called with this calling | |
7683 | sequence with the function attribute @samp{stdcall}. You can also | |
7684 | override the @option{-mrtd} option by using the function attribute | |
7685 | @samp{cdecl}. @xref{Function Attributes}. | |
7686 | ||
7687 | @strong{Warning:} this calling convention is incompatible with the one | |
7688 | normally used on Unix, so you cannot use it if you need to call | |
7689 | libraries compiled with the Unix compiler. | |
7690 | ||
7691 | Also, you must provide function prototypes for all functions that | |
7692 | take variable numbers of arguments (including @code{printf}); | |
7693 | otherwise incorrect code will be generated for calls to those | |
7694 | functions. | |
7695 | ||
7696 | In addition, seriously incorrect code will result if you call a | |
7697 | function with too many arguments. (Normally, extra arguments are | |
7698 | harmlessly ignored.) | |
7699 | ||
7700 | @item -mregparm=@var{num} | |
7701 | @opindex mregparm | |
7702 | Control how many registers are used to pass integer arguments. By | |
7703 | default, no registers are used to pass arguments, and at most 3 | |
7704 | registers can be used. You can control this behavior for a specific | |
7705 | function by using the function attribute @samp{regparm}. | |
7706 | @xref{Function Attributes}. | |
7707 | ||
7708 | @strong{Warning:} if you use this switch, and | |
7709 | @var{num} is nonzero, then you must build all modules with the same | |
7710 | value, including any libraries. This includes the system libraries and | |
7711 | startup modules. | |
7712 | ||
7713 | @item -mpreferred-stack-boundary=@var{num} | |
7714 | @opindex mpreferred-stack-boundary | |
7715 | Attempt to keep the stack boundary aligned to a 2 raised to @var{num} | |
7716 | byte boundary. If @option{-mpreferred-stack-boundary} is not specified, | |
7717 | the default is 4 (16 bytes or 128 bits), except when optimizing for code | |
7718 | size (@option{-Os}), in which case the default is the minimum correct | |
7719 | alignment (4 bytes for x86, and 8 bytes for x86-64). | |
7720 | ||
7721 | On Pentium and PentiumPro, @code{double} and @code{long double} values | |
7722 | should be aligned to an 8 byte boundary (see @option{-malign-double}) or | |
7723 | suffer significant run time performance penalties. On Pentium III, the | |
7724 | Streaming SIMD Extension (SSE) data type @code{__m128} suffers similar | |
7725 | penalties if it is not 16 byte aligned. | |
7726 | ||
7727 | To ensure proper alignment of this values on the stack, the stack boundary | |
7728 | must be as aligned as that required by any value stored on the stack. | |
7729 | Further, every function must be generated such that it keeps the stack | |
7730 | aligned. Thus calling a function compiled with a higher preferred | |
7731 | stack boundary from a function compiled with a lower preferred stack | |
7732 | boundary will most likely misalign the stack. It is recommended that | |
7733 | libraries that use callbacks always use the default setting. | |
7734 | ||
7735 | This extra alignment does consume extra stack space, and generally | |
7736 | increases code size. Code that is sensitive to stack space usage, such | |
7737 | as embedded systems and operating system kernels, may want to reduce the | |
7738 | preferred alignment to @option{-mpreferred-stack-boundary=2}. | |
7739 | ||
7740 | @item -mmmx | |
7741 | @itemx -mno-mmx | |
7742 | @item -msse | |
7743 | @itemx -mno-sse | |
7744 | @item -msse2 | |
7745 | @itemx -mno-sse2 | |
7746 | @item -m3dnow | |
7747 | @itemx -mno-3dnow | |
7748 | @opindex mmmx | |
7749 | @opindex mno-mmx | |
7750 | @opindex msse | |
7751 | @opindex mno-sse | |
7752 | @opindex m3dnow | |
7753 | @opindex mno-3dnow | |
7754 | These switches enable or disable the use of built-in functions that allow | |
7755 | direct access to the MMX, SSE and 3Dnow extensions of the instruction set. | |
7756 | ||
7757 | @xref{X86 Built-in Functions}, for details of the functions enabled | |
7758 | and disabled by these switches. | |
7759 | ||
7760 | @item -mpush-args | |
7761 | @itemx -mno-push-args | |
7762 | @opindex mpush-args | |
7763 | @opindex mno-push-args | |
7764 | Use PUSH operations to store outgoing parameters. This method is shorter | |
7765 | and usually equally fast as method using SUB/MOV operations and is enabled | |
7766 | by default. In some cases disabling it may improve performance because of | |
7767 | improved scheduling and reduced dependencies. | |
7768 | ||
7769 | @item -maccumulate-outgoing-args | |
7770 | @opindex maccumulate-outgoing-args | |
7771 | If enabled, the maximum amount of space required for outgoing arguments will be | |
7772 | computed in the function prologue. This is faster on most modern CPUs | |
7773 | because of reduced dependencies, improved scheduling and reduced stack usage | |
7774 | when preferred stack boundary is not equal to 2. The drawback is a notable | |
7775 | increase in code size. This switch implies @option{-mno-push-args}. | |
7776 | ||
7777 | @item -mthreads | |
7778 | @opindex mthreads | |
7779 | Support thread-safe exception handling on @samp{Mingw32}. Code that relies | |
7780 | on thread-safe exception handling must compile and link all code with the | |
7781 | @option{-mthreads} option. When compiling, @option{-mthreads} defines | |
7782 | @option{-D_MT}; when linking, it links in a special thread helper library | |
7783 | @option{-lmingwthrd} which cleans up per thread exception handling data. | |
7784 | ||
7785 | @item -mno-align-stringops | |
7786 | @opindex mno-align-stringops | |
7787 | Do not align destination of inlined string operations. This switch reduces | |
7788 | code size and improves performance in case the destination is already aligned, | |
7789 | but gcc don't know about it. | |
7790 | ||
7791 | @item -minline-all-stringops | |
7792 | @opindex minline-all-stringops | |
7793 | By default GCC inlines string operations only when destination is known to be | |
7794 | aligned at least to 4 byte boundary. This enables more inlining, increase code | |
7795 | size, but may improve performance of code that depends on fast memcpy, strlen | |
7796 | and memset for short lengths. | |
7797 | ||
7798 | @item -momit-leaf-frame-pointer | |
7799 | @opindex momit-leaf-frame-pointer | |
7800 | Don't keep the frame pointer in a register for leaf functions. This | |
7801 | avoids the instructions to save, set up and restore frame pointers and | |
7802 | makes an extra register available in leaf functions. The option | |
7803 | @option{-fomit-frame-pointer} removes the frame pointer for all functions | |
7804 | which might make debugging harder. | |
7805 | @end table | |
7806 | ||
7807 | These @samp{-m} switches are supported in addition to the above | |
7808 | on AMD x86-64 processors in 64-bit environments. | |
7809 | ||
7810 | @table @gcctabopt | |
7811 | @item -m32 | |
7812 | @itemx -m64 | |
7813 | @opindex m32 | |
7814 | @opindex m64 | |
7815 | Generate code for a 32-bit or 64-bit environment. | |
7816 | The 32-bit environment sets int, long and pointer to 32 bits and | |
7817 | generates code that runs on any i386 system. | |
7818 | The 64-bit environment sets int to 32 bits and long and pointer | |
7819 | to 64 bits and generates code for AMD's x86-64 architecture. | |
7820 | ||
7821 | @item -mno-red-zone | |
7822 | @opindex no-red-zone | |
7823 | Do not use a so called red zone for x86-64 code. The red zone is mandated | |
7824 | by the x86-64 ABI, it is a 128-byte area beyond the location of the | |
7825 | stack pointer that will not be modified by signal or interrupt handlers | |
7826 | and therefore can be used for temporary data without adjusting the stack | |
7827 | pointer. The flag @option{-mno-red-zone} disables this red zone. | |
7828 | @end table | |
7829 | ||
7830 | @node HPPA Options | |
7831 | @subsection HPPA Options | |
7832 | @cindex HPPA Options | |
7833 | ||
7834 | These @samp{-m} options are defined for the HPPA family of computers: | |
7835 | ||
7836 | @table @gcctabopt | |
7837 | @item -march=@var{architecture-type} | |
7838 | @opindex march | |
7839 | Generate code for the specified architecture. The choices for | |
7840 | @var{architecture-type} are @samp{1.0} for PA 1.0, @samp{1.1} for PA | |
7841 | 1.1, and @samp{2.0} for PA 2.0 processors. Refer to | |
7842 | @file{/usr/lib/sched.models} on an HP-UX system to determine the proper | |
7843 | architecture option for your machine. Code compiled for lower numbered | |
7844 | architectures will run on higher numbered architectures, but not the | |
7845 | other way around. | |
7846 | ||
7847 | PA 2.0 support currently requires gas snapshot 19990413 or later. The | |
7848 | next release of binutils (current is 2.9.1) will probably contain PA 2.0 | |
7849 | support. | |
7850 | ||
7851 | @item -mpa-risc-1-0 | |
7852 | @itemx -mpa-risc-1-1 | |
7853 | @itemx -mpa-risc-2-0 | |
7854 | @opindex mpa-risc-1-0 | |
7855 | @opindex mpa-risc-1-1 | |
7856 | @opindex mpa-risc-2-0 | |
7857 | Synonyms for @option{-march=1.0}, @option{-march=1.1}, and @option{-march=2.0} respectively. | |
7858 | ||
7859 | @item -mbig-switch | |
7860 | @opindex mbig-switch | |
7861 | Generate code suitable for big switch tables. Use this option only if | |
7862 | the assembler/linker complain about out of range branches within a switch | |
7863 | table. | |
7864 | ||
7865 | @item -mjump-in-delay | |
7866 | @opindex mjump-in-delay | |
7867 | Fill delay slots of function calls with unconditional jump instructions | |
7868 | by modifying the return pointer for the function call to be the target | |
7869 | of the conditional jump. | |
7870 | ||
7871 | @item -mdisable-fpregs | |
7872 | @opindex mdisable-fpregs | |
7873 | Prevent floating point registers from being used in any manner. This is | |
7874 | necessary for compiling kernels which perform lazy context switching of | |
7875 | floating point registers. If you use this option and attempt to perform | |
7876 | floating point operations, the compiler will abort. | |
7877 | ||
7878 | @item -mdisable-indexing | |
7879 | @opindex mdisable-indexing | |
7880 | Prevent the compiler from using indexing address modes. This avoids some | |
7881 | rather obscure problems when compiling MIG generated code under MACH@. | |
7882 | ||
7883 | @item -mno-space-regs | |
7884 | @opindex mno-space-regs | |
7885 | Generate code that assumes the target has no space registers. This allows | |
7886 | GCC to generate faster indirect calls and use unscaled index address modes. | |
7887 | ||
7888 | Such code is suitable for level 0 PA systems and kernels. | |
7889 | ||
7890 | @item -mfast-indirect-calls | |
7891 | @opindex mfast-indirect-calls | |
7892 | Generate code that assumes calls never cross space boundaries. This | |
7893 | allows GCC to emit code which performs faster indirect calls. | |
7894 | ||
7895 | This option will not work in the presence of shared libraries or nested | |
7896 | functions. | |
7897 | ||
7898 | @item -mlong-load-store | |
7899 | @opindex mlong-load-store | |
7900 | Generate 3-instruction load and store sequences as sometimes required by | |
7901 | the HP-UX 10 linker. This is equivalent to the @samp{+k} option to | |
7902 | the HP compilers. | |
7903 | ||
7904 | @item -mportable-runtime | |
7905 | @opindex mportable-runtime | |
7906 | Use the portable calling conventions proposed by HP for ELF systems. | |
7907 | ||
7908 | @item -mgas | |
7909 | @opindex mgas | |
7910 | Enable the use of assembler directives only GAS understands. | |
7911 | ||
7912 | @item -mschedule=@var{cpu-type} | |
7913 | @opindex mschedule | |
7914 | Schedule code according to the constraints for the machine type | |
7915 | @var{cpu-type}. The choices for @var{cpu-type} are @samp{700} | |
7916 | @samp{7100}, @samp{7100LC}, @samp{7200}, and @samp{8000}. Refer to | |
7917 | @file{/usr/lib/sched.models} on an HP-UX system to determine the | |
7918 | proper scheduling option for your machine. | |
7919 | ||
7920 | @item -mlinker-opt | |
7921 | @opindex mlinker-opt | |
7922 | Enable the optimization pass in the HPUX linker. Note this makes symbolic | |
7923 | debugging impossible. It also triggers a bug in the HPUX 8 and HPUX 9 linkers | |
7924 | in which they give bogus error messages when linking some programs. | |
7925 | ||
7926 | @item -msoft-float | |
7927 | @opindex msoft-float | |
7928 | Generate output containing library calls for floating point. | |
7929 | @strong{Warning:} the requisite libraries are not available for all HPPA | |
7930 | targets. Normally the facilities of the machine's usual C compiler are | |
7931 | used, but this cannot be done directly in cross-compilation. You must make | |
7932 | your own arrangements to provide suitable library functions for | |
7933 | cross-compilation. The embedded target @samp{hppa1.1-*-pro} | |
7934 | does provide software floating point support. | |
7935 | ||
7936 | @option{-msoft-float} changes the calling convention in the output file; | |
7937 | therefore, it is only useful if you compile @emph{all} of a program with | |
7938 | this option. In particular, you need to compile @file{libgcc.a}, the | |
7939 | library that comes with GCC, with @option{-msoft-float} in order for | |
7940 | this to work. | |
7941 | @end table | |
7942 | ||
7943 | @node Intel 960 Options | |
7944 | @subsection Intel 960 Options | |
7945 | ||
7946 | These @samp{-m} options are defined for the Intel 960 implementations: | |
7947 | ||
7948 | @table @gcctabopt | |
7949 | @item -m@var{cpu-type} | |
7950 | @opindex mka | |
7951 | @opindex mkb | |
7952 | @opindex mmc | |
7953 | @opindex mca | |
7954 | @opindex mcf | |
7955 | @opindex msa | |
7956 | @opindex msb | |
7957 | Assume the defaults for the machine type @var{cpu-type} for some of | |
7958 | the other options, including instruction scheduling, floating point | |
7959 | support, and addressing modes. The choices for @var{cpu-type} are | |
7960 | @samp{ka}, @samp{kb}, @samp{mc}, @samp{ca}, @samp{cf}, | |
7961 | @samp{sa}, and @samp{sb}. | |
7962 | The default is | |
7963 | @samp{kb}. | |
7964 | ||
7965 | @item -mnumerics | |
7966 | @itemx -msoft-float | |
7967 | @opindex mnumerics | |
7968 | @opindex msoft-float | |
7969 | The @option{-mnumerics} option indicates that the processor does support | |
7970 | floating-point instructions. The @option{-msoft-float} option indicates | |
7971 | that floating-point support should not be assumed. | |
7972 | ||
7973 | @item -mleaf-procedures | |
7974 | @itemx -mno-leaf-procedures | |
7975 | @opindex mleaf-procedures | |
7976 | @opindex mno-leaf-procedures | |
7977 | Do (or do not) attempt to alter leaf procedures to be callable with the | |
7978 | @code{bal} instruction as well as @code{call}. This will result in more | |
7979 | efficient code for explicit calls when the @code{bal} instruction can be | |
7980 | substituted by the assembler or linker, but less efficient code in other | |
7981 | cases, such as calls via function pointers, or using a linker that doesn't | |
7982 | support this optimization. | |
7983 | ||
7984 | @item -mtail-call | |
7985 | @itemx -mno-tail-call | |
7986 | @opindex mtail-call | |
7987 | @opindex mno-tail-call | |
7988 | Do (or do not) make additional attempts (beyond those of the | |
7989 | machine-independent portions of the compiler) to optimize tail-recursive | |
7990 | calls into branches. You may not want to do this because the detection of | |
7991 | cases where this is not valid is not totally complete. The default is | |
7992 | @option{-mno-tail-call}. | |
7993 | ||
7994 | @item -mcomplex-addr | |
7995 | @itemx -mno-complex-addr | |
7996 | @opindex mcomplex-addr | |
7997 | @opindex mno-complex-addr | |
7998 | Assume (or do not assume) that the use of a complex addressing mode is a | |
7999 | win on this implementation of the i960. Complex addressing modes may not | |
8000 | be worthwhile on the K-series, but they definitely are on the C-series. | |
8001 | The default is currently @option{-mcomplex-addr} for all processors except | |
8002 | the CB and CC@. | |
8003 | ||
8004 | @item -mcode-align | |
8005 | @itemx -mno-code-align | |
8006 | @opindex mcode-align | |
8007 | @opindex mno-code-align | |
8008 | Align code to 8-byte boundaries for faster fetching (or don't bother). | |
8009 | Currently turned on by default for C-series implementations only. | |
8010 | ||
8011 | @ignore | |
8012 | @item -mclean-linkage | |
8013 | @itemx -mno-clean-linkage | |
8014 | @opindex mclean-linkage | |
8015 | @opindex mno-clean-linkage | |
8016 | These options are not fully implemented. | |
8017 | @end ignore | |
8018 | ||
8019 | @item -mic-compat | |
8020 | @itemx -mic2.0-compat | |
8021 | @itemx -mic3.0-compat | |
8022 | @opindex mic-compat | |
8023 | @opindex mic2.0-compat | |
8024 | @opindex mic3.0-compat | |
8025 | Enable compatibility with iC960 v2.0 or v3.0. | |
8026 | ||
8027 | @item -masm-compat | |
8028 | @itemx -mintel-asm | |
8029 | @opindex masm-compat | |
8030 | @opindex mintel-asm | |
8031 | Enable compatibility with the iC960 assembler. | |
8032 | ||
8033 | @item -mstrict-align | |
8034 | @itemx -mno-strict-align | |
8035 | @opindex mstrict-align | |
8036 | @opindex mno-strict-align | |
8037 | Do not permit (do permit) unaligned accesses. | |
8038 | ||
8039 | @item -mold-align | |
8040 | @opindex mold-align | |
8041 | Enable structure-alignment compatibility with Intel's gcc release version | |
8042 | 1.3 (based on gcc 1.37). This option implies @option{-mstrict-align}. | |
8043 | ||
8044 | @item -mlong-double-64 | |
8045 | @opindex mlong-double-64 | |
8046 | Implement type @samp{long double} as 64-bit floating point numbers. | |
8047 | Without the option @samp{long double} is implemented by 80-bit | |
8048 | floating point numbers. The only reason we have it because there is | |
8049 | no 128-bit @samp{long double} support in @samp{fp-bit.c} yet. So it | |
8050 | is only useful for people using soft-float targets. Otherwise, we | |
8051 | should recommend against use of it. | |
8052 | ||
8053 | @end table | |
8054 | ||
8055 | @node DEC Alpha Options | |
8056 | @subsection DEC Alpha Options | |
8057 | ||
8058 | These @samp{-m} options are defined for the DEC Alpha implementations: | |
8059 | ||
8060 | @table @gcctabopt | |
8061 | @item -mno-soft-float | |
8062 | @itemx -msoft-float | |
8063 | @opindex mno-soft-float | |
8064 | @opindex msoft-float | |
8065 | Use (do not use) the hardware floating-point instructions for | |
8066 | floating-point operations. When @option{-msoft-float} is specified, | |
8067 | functions in @file{libgcc.a} will be used to perform floating-point | |
8068 | operations. Unless they are replaced by routines that emulate the | |
8069 | floating-point operations, or compiled in such a way as to call such | |
8070 | emulations routines, these routines will issue floating-point | |
8071 | operations. If you are compiling for an Alpha without floating-point | |
8072 | operations, you must ensure that the library is built so as not to call | |
8073 | them. | |
8074 | ||
8075 | Note that Alpha implementations without floating-point operations are | |
8076 | required to have floating-point registers. | |
8077 | ||
8078 | @item -mfp-reg | |
8079 | @itemx -mno-fp-regs | |
8080 | @opindex mfp-reg | |
8081 | @opindex mno-fp-regs | |
8082 | Generate code that uses (does not use) the floating-point register set. | |
8083 | @option{-mno-fp-regs} implies @option{-msoft-float}. If the floating-point | |
8084 | register set is not used, floating point operands are passed in integer | |
8085 | registers as if they were integers and floating-point results are passed | |
8086 | in @code{$0} instead of @code{$f0}. This is a non-standard calling sequence, | |
8087 | so any function with a floating-point argument or return value called by code | |
8088 | compiled with @option{-mno-fp-regs} must also be compiled with that | |
8089 | option. | |
8090 | ||
8091 | A typical use of this option is building a kernel that does not use, | |
8092 | and hence need not save and restore, any floating-point registers. | |
8093 | ||
8094 | @item -mieee | |
8095 | @opindex mieee | |
8096 | The Alpha architecture implements floating-point hardware optimized for | |
8097 | maximum performance. It is mostly compliant with the IEEE floating | |
8098 | point standard. However, for full compliance, software assistance is | |
8099 | required. This option generates code fully IEEE compliant code | |
8100 | @emph{except} that the @var{inexact-flag} is not maintained (see below). | |
8101 | If this option is turned on, the preprocessor macro @code{_IEEE_FP} is | |
8102 | defined during compilation. The resulting code is less efficient but is | |
8103 | able to correctly support denormalized numbers and exceptional IEEE | |
8104 | values such as not-a-number and plus/minus infinity. Other Alpha | |
8105 | compilers call this option @option{-ieee_with_no_inexact}. | |
8106 | ||
8107 | @item -mieee-with-inexact | |
8108 | @opindex mieee-with-inexact | |
8109 | This is like @option{-mieee} except the generated code also maintains | |
8110 | the IEEE @var{inexact-flag}. Turning on this option causes the | |
8111 | generated code to implement fully-compliant IEEE math. In addition to | |
8112 | @code{_IEEE_FP}, @code{_IEEE_FP_EXACT} is defined as a preprocessor | |
8113 | macro. On some Alpha implementations the resulting code may execute | |
8114 | significantly slower than the code generated by default. Since there is | |
8115 | very little code that depends on the @var{inexact-flag}, you should | |
8116 | normally not specify this option. Other Alpha compilers call this | |
8117 | option @option{-ieee_with_inexact}. | |
8118 | ||
8119 | @item -mfp-trap-mode=@var{trap-mode} | |
8120 | @opindex mfp-trap-mode | |
8121 | This option controls what floating-point related traps are enabled. | |
8122 | Other Alpha compilers call this option @option{-fptm @var{trap-mode}}. | |
8123 | The trap mode can be set to one of four values: | |
8124 | ||
8125 | @table @samp | |
8126 | @item n | |
8127 | This is the default (normal) setting. The only traps that are enabled | |
8128 | are the ones that cannot be disabled in software (e.g., division by zero | |
8129 | trap). | |
8130 | ||
8131 | @item u | |
8132 | In addition to the traps enabled by @samp{n}, underflow traps are enabled | |
8133 | as well. | |
8134 | ||
8135 | @item su | |
8136 | Like @samp{su}, but the instructions are marked to be safe for software | |
8137 | completion (see Alpha architecture manual for details). | |
8138 | ||
8139 | @item sui | |
8140 | Like @samp{su}, but inexact traps are enabled as well. | |
8141 | @end table | |
8142 | ||
8143 | @item -mfp-rounding-mode=@var{rounding-mode} | |
8144 | @opindex mfp-rounding-mode | |
8145 | Selects the IEEE rounding mode. Other Alpha compilers call this option | |
8146 | @option{-fprm @var{rounding-mode}}. The @var{rounding-mode} can be one | |
8147 | of: | |
8148 | ||
8149 | @table @samp | |
8150 | @item n | |
8151 | Normal IEEE rounding mode. Floating point numbers are rounded towards | |
8152 | the nearest machine number or towards the even machine number in case | |
8153 | of a tie. | |
8154 | ||
8155 | @item m | |
8156 | Round towards minus infinity. | |
8157 | ||
8158 | @item c | |
8159 | Chopped rounding mode. Floating point numbers are rounded towards zero. | |
8160 | ||
8161 | @item d | |
8162 | Dynamic rounding mode. A field in the floating point control register | |
8163 | (@var{fpcr}, see Alpha architecture reference manual) controls the | |
8164 | rounding mode in effect. The C library initializes this register for | |
8165 | rounding towards plus infinity. Thus, unless your program modifies the | |
8166 | @var{fpcr}, @samp{d} corresponds to round towards plus infinity. | |
8167 | @end table | |
8168 | ||
8169 | @item -mtrap-precision=@var{trap-precision} | |
8170 | @opindex mtrap-precision | |
8171 | In the Alpha architecture, floating point traps are imprecise. This | |
8172 | means without software assistance it is impossible to recover from a | |
8173 | floating trap and program execution normally needs to be terminated. | |
8174 | GCC can generate code that can assist operating system trap handlers | |
8175 | in determining the exact location that caused a floating point trap. | |
8176 | Depending on the requirements of an application, different levels of | |
8177 | precisions can be selected: | |
8178 | ||
8179 | @table @samp | |
8180 | @item p | |
8181 | Program precision. This option is the default and means a trap handler | |
8182 | can only identify which program caused a floating point exception. | |
8183 | ||
8184 | @item f | |
8185 | Function precision. The trap handler can determine the function that | |
8186 | caused a floating point exception. | |
8187 | ||
8188 | @item i | |
8189 | Instruction precision. The trap handler can determine the exact | |
8190 | instruction that caused a floating point exception. | |
8191 | @end table | |
8192 | ||
8193 | Other Alpha compilers provide the equivalent options called | |
8194 | @option{-scope_safe} and @option{-resumption_safe}. | |
8195 | ||
8196 | @item -mieee-conformant | |
8197 | @opindex mieee-conformant | |
8198 | This option marks the generated code as IEEE conformant. You must not | |
8199 | use this option unless you also specify @option{-mtrap-precision=i} and either | |
8200 | @option{-mfp-trap-mode=su} or @option{-mfp-trap-mode=sui}. Its only effect | |
8201 | is to emit the line @samp{.eflag 48} in the function prologue of the | |
8202 | generated assembly file. Under DEC Unix, this has the effect that | |
8203 | IEEE-conformant math library routines will be linked in. | |
8204 | ||
8205 | @item -mbuild-constants | |
8206 | @opindex mbuild-constants | |
8207 | Normally GCC examines a 32- or 64-bit integer constant to | |
8208 | see if it can construct it from smaller constants in two or three | |
8209 | instructions. If it cannot, it will output the constant as a literal and | |
8210 | generate code to load it from the data segment at runtime. | |
8211 | ||
8212 | Use this option to require GCC to construct @emph{all} integer constants | |
8213 | using code, even if it takes more instructions (the maximum is six). | |
8214 | ||
8215 | You would typically use this option to build a shared library dynamic | |
8216 | loader. Itself a shared library, it must relocate itself in memory | |
8217 | before it can find the variables and constants in its own data segment. | |
8218 | ||
8219 | @item -malpha-as | |
8220 | @itemx -mgas | |
8221 | @opindex malpha-as | |
8222 | @opindex mgas | |
8223 | Select whether to generate code to be assembled by the vendor-supplied | |
8224 | assembler (@option{-malpha-as}) or by the GNU assembler @option{-mgas}. | |
8225 | ||
8226 | @item -mbwx | |
8227 | @itemx -mno-bwx | |
8228 | @itemx -mcix | |
8229 | @itemx -mno-cix | |
8230 | @itemx -mfix | |
8231 | @itemx -mno-fix | |
8232 | @itemx -mmax | |
8233 | @itemx -mno-max | |
8234 | @opindex mbwx | |
8235 | @opindex mno-bwx | |
8236 | @opindex mcix | |
8237 | @opindex mno-cix | |
8238 | @opindex mfix | |
8239 | @opindex mno-fix | |
8240 | @opindex mmax | |
8241 | @opindex mno-max | |
8242 | Indicate whether GCC should generate code to use the optional BWX, | |
8243 | CIX, FIX and MAX instruction sets. The default is to use the instruction | |
8244 | sets supported by the CPU type specified via @option{-mcpu=} option or that | |
8245 | of the CPU on which GCC was built if none was specified. | |
8246 | ||
8247 | @item -mfloat-vax | |
8248 | @itemx -mfloat-ieee | |
8249 | @opindex mfloat-vax | |
8250 | @opindex mfloat-ieee | |
8251 | Generate code that uses (does not use) VAX F and G floating point | |
8252 | arithmetic instead of IEEE single and double precision. | |
8253 | ||
8254 | @item -mexplicit-relocs | |
8255 | @itemx -mno-explicit-relocs | |
8256 | @opindex mexplicit-relocs | |
8257 | @opindex mno-explicit-relocs | |
8258 | Older Alpha assemblers provided no way to generate symbol relocations | |
8259 | except via assembler macros. Use of these macros does not allow | |
8260 | optimial instruction scheduling. GNU binutils as of version 2.12 | |
8261 | supports a new syntax that allows the compiler to explicitly mark | |
8262 | which relocations should apply to which instructions. This option | |
8263 | is mostly useful for debugging, as GCC detects the capabilities of | |
8264 | the assembler when it is built and sets the default accordingly. | |
8265 | ||
8266 | @item -msmall-data | |
8267 | @itemx -mlarge-data | |
8268 | @opindex msmall-data | |
8269 | @opindex mlarge-data | |
8270 | When @option{-mexplicit-relocs} is in effect, static data is | |
8271 | accessed via @dfn{gp-relative} relocations. When @option{-msmall-data} | |
8272 | is used, objects 8 bytes long or smaller are placed in a @dfn{small data area} | |
8273 | (the @code{.sdata} and @code{.sbss} sections) and are accessed via | |
8274 | 16-bit relocations off of the @code{$gp} register. This limits the | |
8275 | size of the small data area to 64KB, but allows the variables to be | |
8276 | directly accessed via a single instruction. | |
8277 | ||
8278 | The default is @option{-mlarge-data}. With this option the data area | |
8279 | is limited to just below 2GB. Programs that require more than 2GB of | |
8280 | data must use @code{malloc} or @code{mmap} to allocate the data in the | |
8281 | heap instead of in the program's data segment. | |
8282 | ||
8283 | When generating code for shared libraries, @option{-fpic} implies | |
8284 | @option{-msmall-data} and @option{-fPIC} implies @option{-mlarge-data}. | |
8285 | ||
8286 | @item -mcpu=@var{cpu_type} | |
8287 | @opindex mcpu | |
8288 | Set the instruction set and instruction scheduling parameters for | |
8289 | machine type @var{cpu_type}. You can specify either the @samp{EV} | |
8290 | style name or the corresponding chip number. GCC supports scheduling | |
8291 | parameters for the EV4, EV5 and EV6 family of processors and will | |
8292 | choose the default values for the instruction set from the processor | |
8293 | you specify. If you do not specify a processor type, GCC will default | |
8294 | to the processor on which the compiler was built. | |
8295 | ||
8296 | Supported values for @var{cpu_type} are | |
8297 | ||
8298 | @table @samp | |
8299 | @item ev4 | |
8300 | @item ev45 | |
8301 | @itemx 21064 | |
8302 | Schedules as an EV4 and has no instruction set extensions. | |
8303 | ||
8304 | @item ev5 | |
8305 | @itemx 21164 | |
8306 | Schedules as an EV5 and has no instruction set extensions. | |
8307 | ||
8308 | @item ev56 | |
8309 | @itemx 21164a | |
8310 | Schedules as an EV5 and supports the BWX extension. | |
8311 | ||
8312 | @item pca56 | |
8313 | @itemx 21164pc | |
8314 | @itemx 21164PC | |
8315 | Schedules as an EV5 and supports the BWX and MAX extensions. | |
8316 | ||
8317 | @item ev6 | |
8318 | @itemx 21264 | |
8319 | Schedules as an EV6 and supports the BWX, FIX, and MAX extensions. | |
8320 | ||
8321 | @item ev67 | |
8322 | @item 21264a | |
8323 | Schedules as an EV6 and supports the BWX, CIX, FIX, and MAX extensions. | |
8324 | @end table | |
8325 | ||
8326 | @item -mtune=@var{cpu_type} | |
8327 | @opindex mtune | |
8328 | Set only the instruction scheduling parameters for machine type | |
8329 | @var{cpu_type}. The instruction set is not changed. | |
8330 | ||
8331 | @item -mmemory-latency=@var{time} | |
8332 | @opindex mmemory-latency | |
8333 | Sets the latency the scheduler should assume for typical memory | |
8334 | references as seen by the application. This number is highly | |
8335 | dependent on the memory access patterns used by the application | |
8336 | and the size of the external cache on the machine. | |
8337 | ||
8338 | Valid options for @var{time} are | |
8339 | ||
8340 | @table @samp | |
8341 | @item @var{number} | |
8342 | A decimal number representing clock cycles. | |
8343 | ||
8344 | @item L1 | |
8345 | @itemx L2 | |
8346 | @itemx L3 | |
8347 | @itemx main | |
8348 | The compiler contains estimates of the number of clock cycles for | |
8349 | ``typical'' EV4 & EV5 hardware for the Level 1, 2 & 3 caches | |
8350 | (also called Dcache, Scache, and Bcache), as well as to main memory. | |
8351 | Note that L3 is only valid for EV5. | |
8352 | ||
8353 | @end table | |
8354 | @end table | |
8355 | ||
8356 | @node DEC Alpha/VMS Options | |
8357 | @subsection DEC Alpha/VMS Options | |
8358 | ||
8359 | These @samp{-m} options are defined for the DEC Alpha/VMS implementations: | |
8360 | ||
8361 | @table @gcctabopt | |
8362 | @item -mvms-return-codes | |
8363 | @opindex mvms-return-codes | |
8364 | Return VMS condition codes from main. The default is to return POSIX | |
8365 | style condition (e.g.@ error) codes. | |
8366 | @end table | |
8367 | ||
8368 | @node Clipper Options | |
8369 | @subsection Clipper Options | |
8370 | ||
8371 | These @samp{-m} options are defined for the Clipper implementations: | |
8372 | ||
8373 | @table @gcctabopt | |
8374 | @item -mc300 | |
8375 | @opindex mc300 | |
8376 | Produce code for a C300 Clipper processor. This is the default. | |
8377 | ||
8378 | @item -mc400 | |
8379 | @opindex mc400 | |
8380 | Produce code for a C400 Clipper processor, i.e.@: use floating point | |
8381 | registers f8--f15. | |
8382 | @end table | |
8383 | ||
8384 | @node H8/300 Options | |
8385 | @subsection H8/300 Options | |
8386 | ||
8387 | These @samp{-m} options are defined for the H8/300 implementations: | |
8388 | ||
8389 | @table @gcctabopt | |
8390 | @item -mrelax | |
8391 | @opindex mrelax | |
8392 | Shorten some address references at link time, when possible; uses the | |
8393 | linker option @option{-relax}. @xref{H8/300,, @code{ld} and the H8/300, | |
8394 | ld.info, Using ld}, for a fuller description. | |
8395 | ||
8396 | @item -mh | |
8397 | @opindex mh | |
8398 | Generate code for the H8/300H@. | |
8399 | ||
8400 | @item -ms | |
8401 | @opindex ms | |
8402 | Generate code for the H8/S@. | |
8403 | ||
8404 | @item -ms2600 | |
8405 | @opindex ms2600 | |
8406 | Generate code for the H8/S2600. This switch must be used with @option{-ms}. | |
8407 | ||
8408 | @item -mint32 | |
8409 | @opindex mint32 | |
8410 | Make @code{int} data 32 bits by default. | |
8411 | ||
8412 | @item -malign-300 | |
8413 | @opindex malign-300 | |
8414 | On the H8/300H and H8/S, use the same alignment rules as for the H8/300. | |
8415 | The default for the H8/300H and H8/S is to align longs and floats on 4 | |
8416 | byte boundaries. | |
8417 | @option{-malign-300} causes them to be aligned on 2 byte boundaries. | |
8418 | This option has no effect on the H8/300. | |
8419 | @end table | |
8420 | ||
8421 | @node SH Options | |
8422 | @subsection SH Options | |
8423 | ||
8424 | These @samp{-m} options are defined for the SH implementations: | |
8425 | ||
8426 | @table @gcctabopt | |
8427 | @item -m1 | |
8428 | @opindex m1 | |
8429 | Generate code for the SH1. | |
8430 | ||
8431 | @item -m2 | |
8432 | @opindex m2 | |
8433 | Generate code for the SH2. | |
8434 | ||
8435 | @item -m3 | |
8436 | @opindex m3 | |
8437 | Generate code for the SH3. | |
8438 | ||
8439 | @item -m3e | |
8440 | @opindex m3e | |
8441 | Generate code for the SH3e. | |
8442 | ||
8443 | @item -m4-nofpu | |
8444 | @opindex m4-nofpu | |
8445 | Generate code for the SH4 without a floating-point unit. | |
8446 | ||
8447 | @item -m4-single-only | |
8448 | @opindex m4-single-only | |
8449 | Generate code for the SH4 with a floating-point unit that only | |
8450 | supports single-precision arithmetic. | |
8451 | ||
8452 | @item -m4-single | |
8453 | @opindex m4-single | |
8454 | Generate code for the SH4 assuming the floating-point unit is in | |
8455 | single-precision mode by default. | |
8456 | ||
8457 | @item -m4 | |
8458 | @opindex m4 | |
8459 | Generate code for the SH4. | |
8460 | ||
8461 | @item -mb | |
8462 | @opindex mb | |
8463 | Compile code for the processor in big endian mode. | |
8464 | ||
8465 | @item -ml | |
8466 | @opindex ml | |
8467 | Compile code for the processor in little endian mode. | |
8468 | ||
8469 | @item -mdalign | |
8470 | @opindex mdalign | |
8471 | Align doubles at 64-bit boundaries. Note that this changes the calling | |
8472 | conventions, and thus some functions from the standard C library will | |
8473 | not work unless you recompile it first with @option{-mdalign}. | |
8474 | ||
8475 | @item -mrelax | |
8476 | @opindex mrelax | |
8477 | Shorten some address references at link time, when possible; uses the | |
8478 | linker option @option{-relax}. | |
8479 | ||
8480 | @item -mbigtable | |
8481 | @opindex mbigtable | |
8482 | Use 32-bit offsets in @code{switch} tables. The default is to use | |
8483 | 16-bit offsets. | |
8484 | ||
8485 | @item -mfmovd | |
8486 | @opindex mfmovd | |
8487 | Enable the use of the instruction @code{fmovd}. | |
8488 | ||
8489 | @item -mhitachi | |
8490 | @opindex mhitachi | |
8491 | Comply with the calling conventions defined by Hitachi. | |
8492 | ||
8493 | @item -mnomacsave | |
8494 | @opindex mnomacsave | |
8495 | Mark the @code{MAC} register as call-clobbered, even if | |
8496 | @option{-mhitachi} is given. | |
8497 | ||
8498 | @item -mieee | |
8499 | @opindex mieee | |
8500 | Increase IEEE-compliance of floating-point code. | |
8501 | ||
8502 | @item -misize | |
8503 | @opindex misize | |
8504 | Dump instruction size and location in the assembly code. | |
8505 | ||
8506 | @item -mpadstruct | |
8507 | @opindex mpadstruct | |
8508 | This option is deprecated. It pads structures to multiple of 4 bytes, | |
8509 | which is incompatible with the SH ABI@. | |
8510 | ||
8511 | @item -mspace | |
8512 | @opindex mspace | |
8513 | Optimize for space instead of speed. Implied by @option{-Os}. | |
8514 | ||
8515 | @item -mprefergot | |
8516 | @opindex mprefergot | |
8517 | When generating position-independent code, emit function calls using | |
8518 | the Global Offset Table instead of the Procedure Linkage Table. | |
8519 | ||
8520 | @item -musermode | |
8521 | @opindex musermode | |
8522 | Generate a library function call to invalidate instruction cache | |
8523 | entries, after fixing up a trampoline. This library function call | |
8524 | doesn't assume it can write to the whole memory address space. This | |
8525 | is the default when the target is @code{sh-*-linux*}. | |
8526 | @end table | |
8527 | ||
8528 | @node System V Options | |
8529 | @subsection Options for System V | |
8530 | ||
8531 | These additional options are available on System V Release 4 for | |
8532 | compatibility with other compilers on those systems: | |
8533 | ||
8534 | @table @gcctabopt | |
8535 | @item -G | |
8536 | @opindex G | |
8537 | Create a shared object. | |
8538 | It is recommended that @option{-symbolic} or @option{-shared} be used instead. | |
8539 | ||
8540 | @item -Qy | |
8541 | @opindex Qy | |
8542 | Identify the versions of each tool used by the compiler, in a | |
8543 | @code{.ident} assembler directive in the output. | |
8544 | ||
8545 | @item -Qn | |
8546 | @opindex Qn | |
8547 | Refrain from adding @code{.ident} directives to the output file (this is | |
8548 | the default). | |
8549 | ||
8550 | @item -YP,@var{dirs} | |
8551 | @opindex YP | |
8552 | Search the directories @var{dirs}, and no others, for libraries | |
8553 | specified with @option{-l}. | |
8554 | ||
8555 | @item -Ym,@var{dir} | |
8556 | @opindex Ym | |
8557 | Look in the directory @var{dir} to find the M4 preprocessor. | |
8558 | The assembler uses this option. | |
8559 | @c This is supposed to go with a -Yd for predefined M4 macro files, but | |
8560 | @c the generic assembler that comes with Solaris takes just -Ym. | |
8561 | @end table | |
8562 | ||
8563 | @node TMS320C3x/C4x Options | |
8564 | @subsection TMS320C3x/C4x Options | |
8565 | @cindex TMS320C3x/C4x Options | |
8566 | ||
8567 | These @samp{-m} options are defined for TMS320C3x/C4x implementations: | |
8568 | ||
8569 | @table @gcctabopt | |
8570 | ||
8571 | @item -mcpu=@var{cpu_type} | |
8572 | @opindex mcpu | |
8573 | Set the instruction set, register set, and instruction scheduling | |
8574 | parameters for machine type @var{cpu_type}. Supported values for | |
8575 | @var{cpu_type} are @samp{c30}, @samp{c31}, @samp{c32}, @samp{c40}, and | |
8576 | @samp{c44}. The default is @samp{c40} to generate code for the | |
8577 | TMS320C40. | |
8578 | ||
8579 | @item -mbig-memory | |
8580 | @item -mbig | |
8581 | @itemx -msmall-memory | |
8582 | @itemx -msmall | |
8583 | @opindex mbig-memory | |
8584 | @opindex mbig | |
8585 | @opindex msmall-memory | |
8586 | @opindex msmall | |
8587 | Generates code for the big or small memory model. The small memory | |
8588 | model assumed that all data fits into one 64K word page. At run-time | |
8589 | the data page (DP) register must be set to point to the 64K page | |
8590 | containing the .bss and .data program sections. The big memory model is | |
8591 | the default and requires reloading of the DP register for every direct | |
8592 | memory access. | |
8593 | ||
8594 | @item -mbk | |
8595 | @itemx -mno-bk | |
8596 | @opindex mbk | |
8597 | @opindex mno-bk | |
8598 | Allow (disallow) allocation of general integer operands into the block | |
8599 | count register BK@. | |
8600 | ||
8601 | @item -mdb | |
8602 | @itemx -mno-db | |
8603 | @opindex mdb | |
8604 | @opindex mno-db | |
8605 | Enable (disable) generation of code using decrement and branch, | |
8606 | DBcond(D), instructions. This is enabled by default for the C4x. To be | |
8607 | on the safe side, this is disabled for the C3x, since the maximum | |
8608 | iteration count on the C3x is @math{2^23 + 1} (but who iterates loops more than | |
8609 | @math{2^23} times on the C3x?). Note that GCC will try to reverse a loop so | |
8610 | that it can utilise the decrement and branch instruction, but will give | |
8611 | up if there is more than one memory reference in the loop. Thus a loop | |
8612 | where the loop counter is decremented can generate slightly more | |
8613 | efficient code, in cases where the RPTB instruction cannot be utilised. | |
8614 | ||
8615 | @item -mdp-isr-reload | |
8616 | @itemx -mparanoid | |
8617 | @opindex mdp-isr-reload | |
8618 | @opindex mparanoid | |
8619 | Force the DP register to be saved on entry to an interrupt service | |
8620 | routine (ISR), reloaded to point to the data section, and restored on | |
8621 | exit from the ISR@. This should not be required unless someone has | |
8622 | violated the small memory model by modifying the DP register, say within | |
8623 | an object library. | |
8624 | ||
8625 | @item -mmpyi | |
8626 | @itemx -mno-mpyi | |
8627 | @opindex mmpyi | |
8628 | @opindex mno-mpyi | |
8629 | For the C3x use the 24-bit MPYI instruction for integer multiplies | |
8630 | instead of a library call to guarantee 32-bit results. Note that if one | |
8631 | of the operands is a constant, then the multiplication will be performed | |
8632 | using shifts and adds. If the @option{-mmpyi} option is not specified for the C3x, | |
8633 | then squaring operations are performed inline instead of a library call. | |
8634 | ||
8635 | @item -mfast-fix | |
8636 | @itemx -mno-fast-fix | |
8637 | @opindex mfast-fix | |
8638 | @opindex mno-fast-fix | |
8639 | The C3x/C4x FIX instruction to convert a floating point value to an | |
8640 | integer value chooses the nearest integer less than or equal to the | |
8641 | floating point value rather than to the nearest integer. Thus if the | |
8642 | floating point number is negative, the result will be incorrectly | |
8643 | truncated an additional code is necessary to detect and correct this | |
8644 | case. This option can be used to disable generation of the additional | |
8645 | code required to correct the result. | |
8646 | ||
8647 | @item -mrptb | |
8648 | @itemx -mno-rptb | |
8649 | @opindex mrptb | |
8650 | @opindex mno-rptb | |
8651 | Enable (disable) generation of repeat block sequences using the RPTB | |
8652 | instruction for zero overhead looping. The RPTB construct is only used | |
8653 | for innermost loops that do not call functions or jump across the loop | |
8654 | boundaries. There is no advantage having nested RPTB loops due to the | |
8655 | overhead required to save and restore the RC, RS, and RE registers. | |
8656 | This is enabled by default with @option{-O2}. | |
8657 | ||
8658 | @item -mrpts=@var{count} | |
8659 | @itemx -mno-rpts | |
8660 | @opindex mrpts | |
8661 | @opindex mno-rpts | |
8662 | Enable (disable) the use of the single instruction repeat instruction | |
8663 | RPTS@. If a repeat block contains a single instruction, and the loop | |
8664 | count can be guaranteed to be less than the value @var{count}, GCC will | |
8665 | emit a RPTS instruction instead of a RPTB@. If no value is specified, | |
8666 | then a RPTS will be emitted even if the loop count cannot be determined | |
8667 | at compile time. Note that the repeated instruction following RPTS does | |
8668 | not have to be reloaded from memory each iteration, thus freeing up the | |
8669 | CPU buses for operands. However, since interrupts are blocked by this | |
8670 | instruction, it is disabled by default. | |
8671 | ||
8672 | @item -mloop-unsigned | |
8673 | @itemx -mno-loop-unsigned | |
8674 | @opindex mloop-unsigned | |
8675 | @opindex mno-loop-unsigned | |
8676 | The maximum iteration count when using RPTS and RPTB (and DB on the C40) | |
8677 | is @math{2^31 + 1} since these instructions test if the iteration count is | |
8678 | negative to terminate the loop. If the iteration count is unsigned | |
8679 | there is a possibility than the @math{2^31 + 1} maximum iteration count may be | |
8680 | exceeded. This switch allows an unsigned iteration count. | |
8681 | ||
8682 | @item -mti | |
8683 | @opindex mti | |
8684 | Try to emit an assembler syntax that the TI assembler (asm30) is happy | |
8685 | with. This also enforces compatibility with the API employed by the TI | |
8686 | C3x C compiler. For example, long doubles are passed as structures | |
8687 | rather than in floating point registers. | |
8688 | ||
8689 | @item -mregparm | |
8690 | @itemx -mmemparm | |
8691 | @opindex mregparm | |
8692 | @opindex mmemparm | |
8693 | Generate code that uses registers (stack) for passing arguments to functions. | |
8694 | By default, arguments are passed in registers where possible rather | |
8695 | than by pushing arguments on to the stack. | |
8696 | ||
8697 | @item -mparallel-insns | |
8698 | @itemx -mno-parallel-insns | |
8699 | @opindex mparallel-insns | |
8700 | @opindex mno-parallel-insns | |
8701 | Allow the generation of parallel instructions. This is enabled by | |
8702 | default with @option{-O2}. | |
8703 | ||
8704 | @item -mparallel-mpy | |
8705 | @itemx -mno-parallel-mpy | |
8706 | @opindex mparallel-mpy | |
8707 | @opindex mno-parallel-mpy | |
8708 | Allow the generation of MPY||ADD and MPY||SUB parallel instructions, | |
8709 | provided @option{-mparallel-insns} is also specified. These instructions have | |
8710 | tight register constraints which can pessimize the code generation | |
8711 | of large functions. | |
8712 | ||
8713 | @end table | |
8714 | ||
8715 | @node V850 Options | |
8716 | @subsection V850 Options | |
8717 | @cindex V850 Options | |
8718 | ||
8719 | These @samp{-m} options are defined for V850 implementations: | |
8720 | ||
8721 | @table @gcctabopt | |
8722 | @item -mlong-calls | |
8723 | @itemx -mno-long-calls | |
8724 | @opindex mlong-calls | |
8725 | @opindex mno-long-calls | |
8726 | Treat all calls as being far away (near). If calls are assumed to be | |
8727 | far away, the compiler will always load the functions address up into a | |
8728 | register, and call indirect through the pointer. | |
8729 | ||
8730 | @item -mno-ep | |
8731 | @itemx -mep | |
8732 | @opindex mno-ep | |
8733 | @opindex mep | |
8734 | Do not optimize (do optimize) basic blocks that use the same index | |
8735 | pointer 4 or more times to copy pointer into the @code{ep} register, and | |
8736 | use the shorter @code{sld} and @code{sst} instructions. The @option{-mep} | |
8737 | option is on by default if you optimize. | |
8738 | ||
8739 | @item -mno-prolog-function | |
8740 | @itemx -mprolog-function | |
8741 | @opindex mno-prolog-function | |
8742 | @opindex mprolog-function | |
8743 | Do not use (do use) external functions to save and restore registers at | |
8744 | the prolog and epilog of a function. The external functions are slower, | |
8745 | but use less code space if more than one function saves the same number | |
8746 | of registers. The @option{-mprolog-function} option is on by default if | |
8747 | you optimize. | |
8748 | ||
8749 | @item -mspace | |
8750 | @opindex mspace | |
8751 | Try to make the code as small as possible. At present, this just turns | |
8752 | on the @option{-mep} and @option{-mprolog-function} options. | |
8753 | ||
8754 | @item -mtda=@var{n} | |
8755 | @opindex mtda | |
8756 | Put static or global variables whose size is @var{n} bytes or less into | |
8757 | the tiny data area that register @code{ep} points to. The tiny data | |
8758 | area can hold up to 256 bytes in total (128 bytes for byte references). | |
8759 | ||
8760 | @item -msda=@var{n} | |
8761 | @opindex msda | |
8762 | Put static or global variables whose size is @var{n} bytes or less into | |
8763 | the small data area that register @code{gp} points to. The small data | |
8764 | area can hold up to 64 kilobytes. | |
8765 | ||
8766 | @item -mzda=@var{n} | |
8767 | @opindex mzda | |
8768 | Put static or global variables whose size is @var{n} bytes or less into | |
8769 | the first 32 kilobytes of memory. | |
8770 | ||
8771 | @item -mv850 | |
8772 | @opindex mv850 | |
8773 | Specify that the target processor is the V850. | |
8774 | ||
8775 | @item -mbig-switch | |
8776 | @opindex mbig-switch | |
8777 | Generate code suitable for big switch tables. Use this option only if | |
8778 | the assembler/linker complain about out of range branches within a switch | |
8779 | table. | |
8780 | @end table | |
8781 | ||
8782 | @node ARC Options | |
8783 | @subsection ARC Options | |
8784 | @cindex ARC Options | |
8785 | ||
8786 | These options are defined for ARC implementations: | |
8787 | ||
8788 | @table @gcctabopt | |
8789 | @item -EL | |
8790 | @opindex EL | |
8791 | Compile code for little endian mode. This is the default. | |
8792 | ||
8793 | @item -EB | |
8794 | @opindex EB | |
8795 | Compile code for big endian mode. | |
8796 | ||
8797 | @item -mmangle-cpu | |
8798 | @opindex mmangle-cpu | |
8799 | Prepend the name of the cpu to all public symbol names. | |
8800 | In multiple-processor systems, there are many ARC variants with different | |
8801 | instruction and register set characteristics. This flag prevents code | |
8802 | compiled for one cpu to be linked with code compiled for another. | |
8803 | No facility exists for handling variants that are ``almost identical''. | |
8804 | This is an all or nothing option. | |
8805 | ||
8806 | @item -mcpu=@var{cpu} | |
8807 | @opindex mcpu | |
8808 | Compile code for ARC variant @var{cpu}. | |
8809 | Which variants are supported depend on the configuration. | |
8810 | All variants support @option{-mcpu=base}, this is the default. | |
8811 | ||
8812 | @item -mtext=@var{text-section} | |
8813 | @itemx -mdata=@var{data-section} | |
8814 | @itemx -mrodata=@var{readonly-data-section} | |
8815 | @opindex mtext | |
8816 | @opindex mdata | |
8817 | @opindex mrodata | |
8818 | Put functions, data, and readonly data in @var{text-section}, | |
8819 | @var{data-section}, and @var{readonly-data-section} respectively | |
8820 | by default. This can be overridden with the @code{section} attribute. | |
8821 | @xref{Variable Attributes}. | |
8822 | ||
8823 | @end table | |
8824 | ||
8825 | @node NS32K Options | |
8826 | @subsection NS32K Options | |
8827 | @cindex NS32K options | |
8828 | ||
8829 | These are the @samp{-m} options defined for the 32000 series. The default | |
8830 | values for these options depends on which style of 32000 was selected when | |
8831 | the compiler was configured; the defaults for the most common choices are | |
8832 | given below. | |
8833 | ||
8834 | @table @gcctabopt | |
8835 | @item -m32032 | |
8836 | @itemx -m32032 | |
8837 | @opindex m32032 | |
8838 | @opindex m32032 | |
8839 | Generate output for a 32032. This is the default | |
8840 | when the compiler is configured for 32032 and 32016 based systems. | |
8841 | ||
8842 | @item -m32332 | |
8843 | @itemx -m32332 | |
8844 | @opindex m32332 | |
8845 | @opindex m32332 | |
8846 | Generate output for a 32332. This is the default | |
8847 | when the compiler is configured for 32332-based systems. | |
8848 | ||
8849 | @item -m32532 | |
8850 | @itemx -m32532 | |
8851 | @opindex m32532 | |
8852 | @opindex m32532 | |
8853 | Generate output for a 32532. This is the default | |
8854 | when the compiler is configured for 32532-based systems. | |
8855 | ||
8856 | @item -m32081 | |
8857 | @opindex m32081 | |
8858 | Generate output containing 32081 instructions for floating point. | |
8859 | This is the default for all systems. | |
8860 | ||
8861 | @item -m32381 | |
8862 | @opindex m32381 | |
8863 | Generate output containing 32381 instructions for floating point. This | |
8864 | also implies @option{-m32081}. The 32381 is only compatible with the 32332 | |
8865 | and 32532 cpus. This is the default for the pc532-netbsd configuration. | |
8866 | ||
8867 | @item -mmulti-add | |
8868 | @opindex mmulti-add | |
8869 | Try and generate multiply-add floating point instructions @code{polyF} | |
8870 | and @code{dotF}. This option is only available if the @option{-m32381} | |
8871 | option is in effect. Using these instructions requires changes to | |
8872 | register allocation which generally has a negative impact on | |
8873 | performance. This option should only be enabled when compiling code | |
8874 | particularly likely to make heavy use of multiply-add instructions. | |
8875 | ||
8876 | @item -mnomulti-add | |
8877 | @opindex mnomulti-add | |
8878 | Do not try and generate multiply-add floating point instructions | |
8879 | @code{polyF} and @code{dotF}. This is the default on all platforms. | |
8880 | ||
8881 | @item -msoft-float | |
8882 | @opindex msoft-float | |
8883 | Generate output containing library calls for floating point. | |
8884 | @strong{Warning:} the requisite libraries may not be available. | |
8885 | ||
8886 | @item -mnobitfield | |
8887 | @opindex mnobitfield | |
8888 | Do not use the bit-field instructions. On some machines it is faster to | |
8889 | use shifting and masking operations. This is the default for the pc532. | |
8890 | ||
8891 | @item -mbitfield | |
8892 | @opindex mbitfield | |
8893 | Do use the bit-field instructions. This is the default for all platforms | |
8894 | except the pc532. | |
8895 | ||
8896 | @item -mrtd | |
8897 | @opindex mrtd | |
8898 | Use a different function-calling convention, in which functions | |
8899 | that take a fixed number of arguments return pop their | |
8900 | arguments on return with the @code{ret} instruction. | |
8901 | ||
8902 | This calling convention is incompatible with the one normally | |
8903 | used on Unix, so you cannot use it if you need to call libraries | |
8904 | compiled with the Unix compiler. | |
8905 | ||
8906 | Also, you must provide function prototypes for all functions that | |
8907 | take variable numbers of arguments (including @code{printf}); | |
8908 | otherwise incorrect code will be generated for calls to those | |
8909 | functions. | |
8910 | ||
8911 | In addition, seriously incorrect code will result if you call a | |
8912 | function with too many arguments. (Normally, extra arguments are | |
8913 | harmlessly ignored.) | |
8914 | ||
8915 | This option takes its name from the 680x0 @code{rtd} instruction. | |
8916 | ||
8917 | ||
8918 | @item -mregparam | |
8919 | @opindex mregparam | |
8920 | Use a different function-calling convention where the first two arguments | |
8921 | are passed in registers. | |
8922 | ||
8923 | This calling convention is incompatible with the one normally | |
8924 | used on Unix, so you cannot use it if you need to call libraries | |
8925 | compiled with the Unix compiler. | |
8926 | ||
8927 | @item -mnoregparam | |
8928 | @opindex mnoregparam | |
8929 | Do not pass any arguments in registers. This is the default for all | |
8930 | targets. | |
8931 | ||
8932 | @item -msb | |
8933 | @opindex msb | |
8934 | It is OK to use the sb as an index register which is always loaded with | |
8935 | zero. This is the default for the pc532-netbsd target. | |
8936 | ||
8937 | @item -mnosb | |
8938 | @opindex mnosb | |
8939 | The sb register is not available for use or has not been initialized to | |
8940 | zero by the run time system. This is the default for all targets except | |
8941 | the pc532-netbsd. It is also implied whenever @option{-mhimem} or | |
8942 | @option{-fpic} is set. | |
8943 | ||
8944 | @item -mhimem | |
8945 | @opindex mhimem | |
8946 | Many ns32000 series addressing modes use displacements of up to 512MB@. | |
8947 | If an address is above 512MB then displacements from zero can not be used. | |
8948 | This option causes code to be generated which can be loaded above 512MB@. | |
8949 | This may be useful for operating systems or ROM code. | |
8950 | ||
8951 | @item -mnohimem | |
8952 | @opindex mnohimem | |
8953 | Assume code will be loaded in the first 512MB of virtual address space. | |
8954 | This is the default for all platforms. | |
8955 | ||
8956 | ||
8957 | @end table | |
8958 | ||
8959 | @node AVR Options | |
8960 | @subsection AVR Options | |
8961 | @cindex AVR Options | |
8962 | ||
8963 | These options are defined for AVR implementations: | |
8964 | ||
8965 | @table @gcctabopt | |
8966 | @item -mmcu=@var{mcu} | |
8967 | @opindex mmcu | |
8968 | Specify ATMEL AVR instruction set or MCU type. | |
8969 | ||
8970 | Instruction set avr1 is for the minimal AVR core, not supported by the C | |
8971 | compiler, only for assembler programs (MCU types: at90s1200, attiny10, | |
8972 | attiny11, attiny12, attiny15, attiny28). | |
8973 | ||
8974 | Instruction set avr2 (default) is for the classic AVR core with up to | |
8975 | 8K program memory space (MCU types: at90s2313, at90s2323, attiny22, | |
8976 | at90s2333, at90s2343, at90s4414, at90s4433, at90s4434, at90s8515, | |
8977 | at90c8534, at90s8535). | |
8978 | ||
8979 | Instruction set avr3 is for the classic AVR core with up to 128K program | |
8980 | memory space (MCU types: atmega103, atmega603, at43usb320, at76c711). | |
8981 | ||
8982 | Instruction set avr4 is for the enhanced AVR core with up to 8K program | |
8983 | memory space (MCU types: atmega8, atmega83, atmega85). | |
8984 | ||
8985 | Instruction set avr5 is for the enhanced AVR core with up to 128K program | |
8986 | memory space (MCU types: atmega16, atmega161, atmega163, atmega32, atmega323, | |
8987 | atmega64, atmega128, at43usb355, at94k). | |
8988 | ||
8989 | @item -msize | |
8990 | @opindex msize | |
8991 | Output instruction sizes to the asm file. | |
8992 | ||
8993 | @item -minit-stack=@var{N} | |
8994 | @opindex minit-stack | |
8995 | Specify the initial stack address, which may be a symbol or numeric value, | |
8996 | @samp{__stack} is the default. | |
8997 | ||
8998 | @item -mno-interrupts | |
8999 | @opindex mno-interrupts | |
9000 | Generated code is not compatible with hardware interrupts. | |
9001 | Code size will be smaller. | |
9002 | ||
9003 | @item -mcall-prologues | |
9004 | @opindex mcall-prologues | |
9005 | Functions prologues/epilogues expanded as call to appropriate | |
9006 | subroutines. Code size will be smaller. | |
9007 | ||
9008 | @item -mno-tablejump | |
9009 | @opindex mno-tablejump | |
9010 | Do not generate tablejump insns which sometimes increase code size. | |
9011 | ||
9012 | @item -mtiny-stack | |
9013 | @opindex mtiny-stack | |
9014 | Change only the low 8 bits of the stack pointer. | |
9015 | @end table | |
9016 | ||
9017 | @node MCore Options | |
9018 | @subsection MCore Options | |
9019 | @cindex MCore options | |
9020 | ||
9021 | These are the @samp{-m} options defined for the Motorola M*Core | |
9022 | processors. | |
9023 | ||
9024 | @table @gcctabopt | |
9025 | ||
9026 | @item -mhardlit | |
9027 | @itemx -mhardlit | |
9028 | @itemx -mno-hardlit | |
9029 | @opindex mhardlit | |
9030 | @opindex mhardlit | |
9031 | @opindex mno-hardlit | |
9032 | Inline constants into the code stream if it can be done in two | |
9033 | instructions or less. | |
9034 | ||
9035 | @item -mdiv | |
9036 | @itemx -mdiv | |
9037 | @itemx -mno-div | |
9038 | @opindex mdiv | |
9039 | @opindex mdiv | |
9040 | @opindex mno-div | |
9041 | Use the divide instruction. (Enabled by default). | |
9042 | ||
9043 | @item -mrelax-immediate | |
9044 | @itemx -mrelax-immediate | |
9045 | @itemx -mno-relax-immediate | |
9046 | @opindex mrelax-immediate | |
9047 | @opindex mrelax-immediate | |
9048 | @opindex mno-relax-immediate | |
9049 | Allow arbitrary sized immediates in bit operations. | |
9050 | ||
9051 | @item -mwide-bitfields | |
9052 | @itemx -mwide-bitfields | |
9053 | @itemx -mno-wide-bitfields | |
9054 | @opindex mwide-bitfields | |
9055 | @opindex mwide-bitfields | |
9056 | @opindex mno-wide-bitfields | |
9057 | Always treat bit-fields as int-sized. | |
9058 | ||
9059 | @item -m4byte-functions | |
9060 | @itemx -m4byte-functions | |
9061 | @itemx -mno-4byte-functions | |
9062 | @opindex m4byte-functions | |
9063 | @opindex m4byte-functions | |
9064 | @opindex mno-4byte-functions | |
9065 | Force all functions to be aligned to a four byte boundary. | |
9066 | ||
9067 | @item -mcallgraph-data | |
9068 | @itemx -mcallgraph-data | |
9069 | @itemx -mno-callgraph-data | |
9070 | @opindex mcallgraph-data | |
9071 | @opindex mcallgraph-data | |
9072 | @opindex mno-callgraph-data | |
9073 | Emit callgraph information. | |
9074 | ||
9075 | @item -mslow-bytes | |
9076 | @itemx -mslow-bytes | |
9077 | @itemx -mno-slow-bytes | |
9078 | @opindex mslow-bytes | |
9079 | @opindex mslow-bytes | |
9080 | @opindex mno-slow-bytes | |
9081 | Prefer word access when reading byte quantities. | |
9082 | ||
9083 | @item -mlittle-endian | |
9084 | @itemx -mlittle-endian | |
9085 | @itemx -mbig-endian | |
9086 | @opindex mlittle-endian | |
9087 | @opindex mlittle-endian | |
9088 | @opindex mbig-endian | |
9089 | Generate code for a little endian target. | |
9090 | ||
9091 | @item -m210 | |
9092 | @itemx -m210 | |
9093 | @itemx -m340 | |
9094 | @opindex m210 | |
9095 | @opindex m210 | |
9096 | @opindex m340 | |
9097 | Generate code for the 210 processor. | |
9098 | @end table | |
9099 | ||
9100 | @node IA-64 Options | |
9101 | @subsection IA-64 Options | |
9102 | @cindex IA-64 Options | |
9103 | ||
9104 | These are the @samp{-m} options defined for the Intel IA-64 architecture. | |
9105 | ||
9106 | @table @gcctabopt | |
9107 | @item -mbig-endian | |
9108 | @opindex mbig-endian | |
9109 | Generate code for a big endian target. This is the default for HPUX@. | |
9110 | ||
9111 | @item -mlittle-endian | |
9112 | @opindex mlittle-endian | |
9113 | Generate code for a little endian target. This is the default for AIX5 | |
9114 | and Linux. | |
9115 | ||
9116 | @item -mgnu-as | |
9117 | @itemx -mno-gnu-as | |
9118 | @opindex mgnu-as | |
9119 | @opindex mno-gnu-as | |
9120 | Generate (or don't) code for the GNU assembler. This is the default. | |
9121 | @c Also, this is the default if the configure option @option{--with-gnu-as} | |
9122 | @c is used. | |
9123 | ||
9124 | @item -mgnu-ld | |
9125 | @itemx -mno-gnu-ld | |
9126 | @opindex mgnu-ld | |
9127 | @opindex mno-gnu-ld | |
9128 | Generate (or don't) code for the GNU linker. This is the default. | |
9129 | @c Also, this is the default if the configure option @option{--with-gnu-ld} | |
9130 | @c is used. | |
9131 | ||
9132 | @item -mno-pic | |
9133 | @opindex mno-pic | |
9134 | Generate code that does not use a global pointer register. The result | |
9135 | is not position independent code, and violates the IA-64 ABI@. | |
9136 | ||
9137 | @item -mvolatile-asm-stop | |
9138 | @itemx -mno-volatile-asm-stop | |
9139 | @opindex mvolatile-asm-stop | |
9140 | @opindex mno-volatile-asm-stop | |
9141 | Generate (or don't) a stop bit immediately before and after volatile asm | |
9142 | statements. | |
9143 | ||
9144 | @item -mb-step | |
9145 | @opindex mb-step | |
9146 | Generate code that works around Itanium B step errata. | |
9147 | ||
9148 | @item -mregister-names | |
9149 | @itemx -mno-register-names | |
9150 | @opindex mregister-names | |
9151 | @opindex mno-register-names | |
9152 | Generate (or don't) @samp{in}, @samp{loc}, and @samp{out} register names for | |
9153 | the stacked registers. This may make assembler output more readable. | |
9154 | ||
9155 | @item -mno-sdata | |
9156 | @itemx -msdata | |
9157 | @opindex mno-sdata | |
9158 | @opindex msdata | |
9159 | Disable (or enable) optimizations that use the small data section. This may | |
9160 | be useful for working around optimizer bugs. | |
9161 | ||
9162 | @item -mconstant-gp | |
9163 | @opindex mconstant-gp | |
9164 | Generate code that uses a single constant global pointer value. This is | |
9165 | useful when compiling kernel code. | |
9166 | ||
9167 | @item -mauto-pic | |
9168 | @opindex mauto-pic | |
9169 | Generate code that is self-relocatable. This implies @option{-mconstant-gp}. | |
9170 | This is useful when compiling firmware code. | |
9171 | ||
9172 | @item -minline-divide-min-latency | |
9173 | @opindex minline-divide-min-latency | |
9174 | Generate code for inline divides using the minimum latency algorithm. | |
9175 | ||
9176 | @item -minline-divide-max-throughput | |
9177 | @opindex minline-divide-max-throughput | |
9178 | Generate code for inline divides using the maximum throughput algorithm. | |
9179 | ||
9180 | @item -mno-dwarf2-asm | |
9181 | @itemx -mdwarf2-asm | |
9182 | @opindex mno-dwarf2-asm | |
9183 | @opindex mdwarf2-asm | |
9184 | Don't (or do) generate assembler code for the DWARF2 line number debugging | |
9185 | info. This may be useful when not using the GNU assembler. | |
9186 | ||
9187 | @item -mfixed-range=@var{register-range} | |
9188 | @opindex mfixed-range | |
9189 | Generate code treating the given register range as fixed registers. | |
9190 | A fixed register is one that the register allocator can not use. This is | |
9191 | useful when compiling kernel code. A register range is specified as | |
9192 | two registers separated by a dash. Multiple register ranges can be | |
9193 | specified separated by a comma. | |
9194 | @end table | |
9195 | ||
9196 | @node D30V Options | |
9197 | @subsection D30V Options | |
9198 | @cindex D30V Options | |
9199 | ||
9200 | These @samp{-m} options are defined for D30V implementations: | |
9201 | ||
9202 | @table @gcctabopt | |
9203 | @item -mextmem | |
9204 | @opindex mextmem | |
9205 | Link the @samp{.text}, @samp{.data}, @samp{.bss}, @samp{.strings}, | |
9206 | @samp{.rodata}, @samp{.rodata1}, @samp{.data1} sections into external | |
9207 | memory, which starts at location @code{0x80000000}. | |
9208 | ||
9209 | @item -mextmemory | |
9210 | @opindex mextmemory | |
9211 | Same as the @option{-mextmem} switch. | |
9212 | ||
9213 | @item -monchip | |
9214 | @opindex monchip | |
9215 | Link the @samp{.text} section into onchip text memory, which starts at | |
9216 | location @code{0x0}. Also link @samp{.data}, @samp{.bss}, | |
9217 | @samp{.strings}, @samp{.rodata}, @samp{.rodata1}, @samp{.data1} sections | |
9218 | into onchip data memory, which starts at location @code{0x20000000}. | |
9219 | ||
9220 | @item -mno-asm-optimize | |
9221 | @itemx -masm-optimize | |
9222 | @opindex mno-asm-optimize | |
9223 | @opindex masm-optimize | |
9224 | Disable (enable) passing @option{-O} to the assembler when optimizing. | |
9225 | The assembler uses the @option{-O} option to automatically parallelize | |
9226 | adjacent short instructions where possible. | |
9227 | ||
9228 | @item -mbranch-cost=@var{n} | |
9229 | @opindex mbranch-cost | |
9230 | Increase the internal costs of branches to @var{n}. Higher costs means | |
9231 | that the compiler will issue more instructions to avoid doing a branch. | |
9232 | The default is 2. | |
9233 | ||
9234 | @item -mcond-exec=@var{n} | |
9235 | @opindex mcond-exec | |
9236 | Specify the maximum number of conditionally executed instructions that | |
9237 | replace a branch. The default is 4. | |
9238 | @end table | |
9239 | ||
9240 | @node S/390 and zSeries Options | |
9241 | @subsection S/390 and zSeries Options | |
9242 | @cindex S/390 and zSeries Options | |
9243 | ||
9244 | These are the @samp{-m} options defined for the S/390 and zSeries architecture. | |
9245 | ||
9246 | @table @gcctabopt | |
9247 | @item -mhard-float | |
9248 | @itemx -msoft-float | |
9249 | @opindex mhard-float | |
9250 | @opindex msoft-float | |
9251 | Use (do not use) the hardware floating-point instructions and registers | |
9252 | for floating-point operations. When @option{-msoft-float} is specified, | |
9253 | functions in @file{libgcc.a} will be used to perform floating-point | |
9254 | operations. When @option{-mhard-float} is specified, the compiler | |
9255 | generates IEEE floating-point instructions. This is the default. | |
9256 | ||
9257 | @item -mbackchain | |
9258 | @itemx -mno-backchain | |
9259 | @opindex mbackchain | |
9260 | @opindex mno-backchain | |
9261 | Generate (or do not generate) code which maintains an explicit | |
9262 | backchain within the stack frame that points to the caller's frame. | |
9263 | This is currently needed to allow debugging. The default is to | |
9264 | generate the backchain. | |
9265 | ||
9266 | @item -msmall-exec | |
9267 | @itemx -mno-small-exec | |
9268 | @opindex msmall-exec | |
9269 | @opindex mno-small-exec | |
9270 | Generate (or do not generate) code using the @code{bras} instruction | |
9271 | to do subroutine calls. | |
9272 | This only works reliably if the total executable size does not | |
9273 | exceed 64k. The default is to use the @code{basr} instruction instead, | |
9274 | which does not have this limitation. | |
9275 | ||
9276 | @item -m64 | |
9277 | @itemx -m31 | |
9278 | @opindex m64 | |
9279 | @opindex m31 | |
9280 | When @option{-m31} is specified, generate code compliant to the | |
9281 | Linux for S/390 ABI@. When @option{-m64} is specified, generate | |
9282 | code compliant to the Linux for zSeries ABI@. This allows GCC in | |
9283 | particular to generate 64-bit instructions. For the @samp{s390} | |
9284 | targets, the default is @option{-m31}, while the @samp{s390x} | |
9285 | targets default to @option{-m64}. | |
9286 | ||
9287 | @item -mmvcle | |
9288 | @itemx -mno-mvcle | |
9289 | @opindex mmvcle | |
9290 | @opindex mno-mvcle | |
9291 | Generate (or do not generate) code using the @code{mvcle} instruction | |
9292 | to perform block moves. When @option{-mno-mvcle} is specifed, | |
9293 | use a @code{mvc} loop instead. This is the default. | |
9294 | ||
9295 | @item -mdebug | |
9296 | @itemx -mno-debug | |
9297 | @opindex mdebug | |
9298 | @opindex mno-debug | |
9299 | Print (or do not print) additional debug information when compiling. | |
9300 | The default is to not print debug information. | |
9301 | ||
9302 | @end table | |
9303 | ||
9304 | @node CRIS Options | |
9305 | @subsection CRIS Options | |
9306 | @cindex CRIS Options | |
9307 | ||
9308 | These options are defined specifically for the CRIS ports. | |
9309 | ||
9310 | @table @gcctabopt | |
9311 | @item -march=@var{architecture-type} | |
9312 | @itemx -mcpu=@var{architecture-type} | |
9313 | @opindex march | |
9314 | @opindex mcpu | |
9315 | Generate code for the specified architecture. The choices for | |
9316 | @var{architecture-type} are @samp{v3}, @samp{v8} and @samp{v10} for | |
9317 | respectively ETRAX@w{ }4, ETRAX@w{ }100, and ETRAX@w{ }100@w{ }LX. | |
9318 | Default is @samp{v0} except for cris-axis-linux-gnu, where the default is | |
9319 | @samp{v10}. | |
9320 | ||
9321 | @item -mtune=@var{architecture-type} | |
9322 | @opindex mtune | |
9323 | Tune to @var{architecture-type} everything applicable about the generated | |
9324 | code, except for the ABI and the set of available instructions. The | |
9325 | choices for @var{architecture-type} are the same as for | |
9326 | @option{-march=@var{architecture-type}}. | |
9327 | ||
9328 | @item -mmax-stack-frame=@var{n} | |
9329 | @opindex mmax-stack-frame | |
9330 | Warn when the stack frame of a function exceeds @var{n} bytes. | |
9331 | ||
9332 | @item -melinux-stacksize=@var{n} | |
9333 | @opindex melinux-stacksize | |
9334 | Only available with the @samp{cris-axis-aout} target. Arranges for | |
9335 | indications in the program to the kernel loader that the stack of the | |
9336 | program should be set to @var{n} bytes. | |
9337 | ||
9338 | @item -metrax4 | |
9339 | @itemx -metrax100 | |
9340 | @opindex metrax4 | |
9341 | @opindex metrax100 | |
9342 | The options @option{-metrax4} and @option{-metrax100} are synonyms for | |
9343 | @option{-march=v3} and @option{-march=v8} respectively. | |
9344 | ||
9345 | @item -mpdebug | |
9346 | @opindex mpdebug | |
9347 | Enable CRIS-specific verbose debug-related information in the assembly | |
9348 | code. This option also has the effect to turn off the @samp{#NO_APP} | |
9349 | formatted-code indicator to the assembler at the beginning of the | |
9350 | assembly file. | |
9351 | ||
9352 | @item -mcc-init | |
9353 | @opindex mcc-init | |
9354 | Do not use condition-code results from previous instruction; always emit | |
9355 | compare and test instructions before use of condition codes. | |
9356 | ||
9357 | @item -mno-side-effects | |
9358 | @opindex mno-side-effects | |
9359 | Do not emit instructions with side-effects in addressing modes other than | |
9360 | post-increment. | |
9361 | ||
9362 | @item -mstack-align | |
9363 | @itemx -mno-stack-align | |
9364 | @itemx -mdata-align | |
9365 | @itemx -mno-data-align | |
9366 | @itemx -mconst-align | |
9367 | @itemx -mno-const-align | |
9368 | @opindex mstack-align | |
9369 | @opindex mno-stack-align | |
9370 | @opindex mdata-align | |
9371 | @opindex mno-data-align | |
9372 | @opindex mconst-align | |
9373 | @opindex mno-const-align | |
9374 | These options (no-options) arranges (eliminate arrangements) for the | |
9375 | stack-frame, individual data and constants to be aligned for the maximum | |
9376 | single data access size for the chosen CPU model. The default is to | |
9377 | arrange for 32-bit alignment. ABI details such as structure layout are | |
9378 | not affected by these options. | |
9379 | ||
9380 | @item -m32-bit | |
9381 | @itemx -m16-bit | |
9382 | @itemx -m8-bit | |
9383 | @opindex m32-bit | |
9384 | @opindex m16-bit | |
9385 | @opindex m8-bit | |
9386 | Similar to the stack- data- and const-align options above, these options | |
9387 | arrange for stack-frame, writable data and constants to all be 32-bit, | |
9388 | 16-bit or 8-bit aligned. The default is 32-bit alignment. | |
9389 | ||
9390 | @item -mno-prologue-epilogue | |
9391 | @itemx -mprologue-epilogue | |
9392 | @opindex mno-prologue-epilogue | |
9393 | @opindex mprologue-epilogue | |
9394 | With @option{-mno-prologue-epilogue}, the normal function prologue and | |
9395 | epilogue that sets up the stack-frame are omitted and no return | |
9396 | instructions or return sequences are generated in the code. Use this | |
9397 | option only together with visual inspection of the compiled code: no | |
9398 | warnings or errors are generated when call-saved registers must be saved, | |
9399 | or storage for local variable needs to be allocated. | |
9400 | ||
9401 | @item -mno-gotplt | |
9402 | @itemx -mgotplt | |
9403 | @opindex mno-gotplt | |
9404 | @opindex mgotplt | |
9405 | With @option{-fpic} and @option{-fPIC}, don't generate (do generate) | |
9406 | instruction sequences that load addresses for functions from the PLT part | |
9407 | of the GOT rather than (traditional on other architectures) calls to the | |
9408 | PLT. The default is @option{-mgotplt}. | |
9409 | ||
9410 | @item -maout | |
9411 | @opindex maout | |
9412 | Legacy no-op option only recognized with the cris-axis-aout target. | |
9413 | ||
9414 | @item -melf | |
9415 | @opindex melf | |
9416 | Legacy no-op option only recognized with the cris-axis-elf and | |
9417 | cris-axis-linux-gnu targets. | |
9418 | ||
9419 | @item -melinux | |
9420 | @opindex melinux | |
9421 | Only recognized with the cris-axis-aout target, where it selects a | |
9422 | GNU/linux-like multilib, include files and instruction set for | |
9423 | @option{-march=v8}. | |
9424 | ||
9425 | @item -mlinux | |
9426 | @opindex mlinux | |
9427 | Legacy no-op option only recognized with the cris-axis-linux-gnu target. | |
9428 | ||
9429 | @item -sim | |
9430 | @opindex sim | |
9431 | This option, recognized for the cris-axis-aout and cris-axis-elf arranges | |
9432 | to link with input-output functions from a simulator library. Code, | |
9433 | initialized data and zero-initialized data are allocated consecutively. | |
9434 | ||
9435 | @item -sim2 | |
9436 | @opindex sim2 | |
9437 | Like @option{-sim}, but pass linker options to locate initialized data at | |
9438 | 0x40000000 and zero-initialized data at 0x80000000. | |
9439 | @end table | |
9440 | ||
9441 | @node MMIX Options | |
9442 | @subsection MMIX Options | |
9443 | @cindex MMIX Options | |
9444 | ||
9445 | These options are defined for the MMIX: | |
9446 | ||
9447 | @table @gcctabopt | |
9448 | @item -mlibfuncs | |
9449 | @itemx -mno-libfuncs | |
9450 | @opindex mlibfuncs | |
9451 | @opindex mno-libfuncs | |
9452 | Specify that intrinsic library functions are being compiled, passing all | |
9453 | values in registers, no matter the size. | |
9454 | ||
9455 | @item -mepsilon | |
9456 | @itemx -mno-epsilon | |
9457 | @opindex mepsilon | |
9458 | @opindex mno-epsilon | |
9459 | Generate floating-point comparison instructions that compare with respect | |
9460 | to the @code{rE} epsilon register. | |
9461 | ||
9462 | @item -mabi=mmixware | |
9463 | @itemx -mabi=gnu | |
9464 | @opindex mabi-mmixware | |
9465 | @opindex mabi=gnu | |
9466 | Generate code that passes function parameters and return values that (in | |
9467 | the called function) are seen as registers @code{$0} and up, as opposed to | |
9468 | the GNU ABI which uses global registers @code{$231} and up. | |
9469 | ||
9470 | @item -mzero-extend | |
9471 | @itemx -mno-zero-extend | |
9472 | @opindex mzero-extend | |
9473 | @opindex mno-zero-extend | |
9474 | When reading data from memory in sizes shorter than 64 bits, use (do not | |
9475 | use) zero-extending load instructions by default, rather than | |
9476 | sign-extending ones. | |
9477 | ||
9478 | @item -mknuthdiv | |
9479 | @itemx -mno-knuthdiv | |
9480 | @opindex mknuthdiv | |
9481 | @opindex mno-knuthdiv | |
9482 | Make the result of a division yielding a remainder have the same sign as | |
9483 | the divisor. With the default, @option{-mno-knuthdiv}, the sign of the | |
9484 | remainder follows the sign of the dividend. Both methods are | |
9485 | arithmetically valid, the latter being almost exclusively used. | |
9486 | ||
9487 | @item -mtoplevel-symbols | |
9488 | @itemx -mno-toplevel-symbols | |
9489 | @opindex mtoplevel-symbols | |
9490 | @opindex mno-toplevel-symbols | |
9491 | Prepend (do not prepend) a @samp{:} to all global symbols, so the assembly | |
9492 | code can be used with the @code{PREFIX} assembly directive. | |
9493 | ||
9494 | @item -melf | |
9495 | @opindex melf | |
9496 | Generate an executable in the ELF format, rather than the default | |
9497 | @samp{mmo} format used by the @command{mmix} simulator. | |
9498 | ||
9499 | @item -mbranch-predict | |
9500 | @itemx -mno-branch-predict | |
9501 | @opindex mbranch-predict | |
9502 | @opindex mno-branch-predict | |
9503 | Use (do not use) the probable-branch instructions, when static branch | |
9504 | prediction indicates a probable branch. | |
9505 | ||
9506 | @item -mbase-addresses | |
9507 | @itemx -mno-base-addresses | |
9508 | @opindex mbase-addresses | |
9509 | @opindex mno-base-addresses | |
9510 | Generate (do not generate) code that uses @emph{base addresses}. Using a | |
9511 | base address automatically generates a request (handled by the assembler | |
9512 | and the linker) for a constant to be set up in a global register. The | |
9513 | register is used for one or more base address requests within the range 0 | |
9514 | to 255 from the value held in the register. The generally leads to short | |
9515 | and fast code, but the number of different data items that can be | |
9516 | addressed is limited. This means that a program that uses lots of static | |
9517 | data may require @option{-mno-base-addresses}. | |
9518 | @end table | |
9519 | ||
9520 | @node PDP-11 Options | |
9521 | @subsection PDP-11 Options | |
9522 | @cindex PDP-11 Options | |
9523 | ||
9524 | These options are defined for the PDP-11: | |
9525 | ||
9526 | @table @gcctabopt | |
9527 | @item -mfpu | |
9528 | @opindex mfpu | |
9529 | Use hardware FPP floating point. This is the default. (FIS floating | |
9530 | point on the PDP-11/40 is not supported.) | |
9531 | ||
9532 | @item -msoft-float | |
9533 | @opindex msoft-float | |
9534 | Do not use hardware floating point. | |
9535 | ||
9536 | @item -mac0 | |
9537 | @opindex mac0 | |
9538 | Return floating-point results in ac0 (fr0 in Unix assembler syntax). | |
9539 | ||
9540 | @item -mno-ac0 | |
9541 | @opindex mno-ac0 | |
9542 | Return floating-point results in memory. This is the default. | |
9543 | ||
9544 | @item -m40 | |
9545 | @opindex m40 | |
9546 | Generate code for a PDP-11/40. | |
9547 | ||
9548 | @item -m45 | |
9549 | @opindex m45 | |
9550 | Generate code for a PDP-11/45. This is the default. | |
9551 | ||
9552 | @item -m10 | |
9553 | @opindex m10 | |
9554 | Generate code for a PDP-11/10. | |
9555 | ||
9556 | @item -mbcopy-builtin | |
9557 | @opindex bcopy-builtin | |
9558 | Use inline @code{movstrhi} patterns for copying memory. This is the | |
9559 | default. | |
9560 | ||
9561 | @item -mbcopy | |
9562 | @opindex mbcopy | |
9563 | Do not use inline @code{movstrhi} patterns for copying memory. | |
9564 | ||
9565 | @item -mint16 | |
9566 | @itemx -mno-int32 | |
9567 | @opindex mint16 | |
9568 | @opindex mno-int32 | |
9569 | Use 16-bit @code{int}. This is the default. | |
9570 | ||
9571 | @item -mint32 | |
9572 | @itemx -mno-int16 | |
9573 | @opindex mint32 | |
9574 | @opindex mno-int16 | |
9575 | Use 32-bit @code{int}. | |
9576 | ||
9577 | @item -mfloat64 | |
9578 | @itemx -mno-float32 | |
9579 | @opindex mfloat64 | |
9580 | @opindex mno-float32 | |
9581 | Use 64-bit @code{float}. This is the default. | |
9582 | ||
9583 | @item -mfloat32 | |
9584 | @item -mno-float64 | |
9585 | @opindex mfloat32 | |
9586 | @opindex mno-float64 | |
9587 | Use 32-bit @code{float}. | |
9588 | ||
9589 | @item -mabshi | |
9590 | @opindex mabshi | |
9591 | Use @code{abshi2} pattern. This is the default. | |
9592 | ||
9593 | @item -mno-abshi | |
9594 | @opindex mno-abshi | |
9595 | Do not use @code{abshi2} pattern. | |
9596 | ||
9597 | @item -mbranch-expensive | |
9598 | @opindex mbranch-expensive | |
9599 | Pretend that branches are expensive. This is for experimenting with | |
9600 | code generation only. | |
9601 | ||
9602 | @item -mbranch-cheap | |
9603 | @opindex mbranch-cheap | |
9604 | Do not pretend that branches are expensive. This is the default. | |
9605 | ||
9606 | @item -msplit | |
9607 | @opindex msplit | |
9608 | Generate code for a system with split I&D. | |
9609 | ||
9610 | @item -mno-split | |
9611 | @opindex mno-split | |
9612 | Generate code for a system without split I&D. This is the default. | |
9613 | ||
9614 | @item -munix-asm | |
9615 | @opindex munix-asm | |
9616 | Use Unix assembler syntax. This is the default when configured for | |
9617 | @samp{pdp11-*-bsd}. | |
9618 | ||
9619 | @item -mdec-asm | |
9620 | @opindex mdec-asm | |
9621 | Use DEC assembler syntax. This is the default when configured for any | |
9622 | PDP-11 target other than @samp{pdp11-*-bsd}. | |
9623 | @end table | |
9624 | ||
9625 | @node Xstormy16 Options | |
9626 | @subsection Xstormy16 Options | |
9627 | @cindex Xstormy16 Options | |
9628 | ||
9629 | These options are defined for Xstormy16: | |
9630 | ||
9631 | @table @gcctabopt | |
9632 | @item -msim | |
9633 | @opindex msim | |
9634 | Choose startup files and linker script suitable for the simulator. | |
9635 | @end table | |
9636 | ||
9637 | @node Xtensa Options | |
9638 | @subsection Xtensa Options | |
9639 | @cindex Xtensa Options | |
9640 | ||
9641 | The Xtensa architecture is designed to support many different | |
9642 | configurations. The compiler's default options can be set to match a | |
9643 | particular Xtensa configuration by copying a configuration file into the | |
9644 | GCC sources when building GCC@. The options below may be used to | |
9645 | override the default options. | |
9646 | ||
9647 | @table @gcctabopt | |
9648 | @item -mbig-endian | |
9649 | @itemx -mlittle-endian | |
9650 | @opindex mbig-endian | |
9651 | @opindex mlittle-endian | |
9652 | Specify big-endian or little-endian byte ordering for the target Xtensa | |
9653 | processor. | |
9654 | ||
9655 | @item -mdensity | |
9656 | @itemx -mno-density | |
9657 | @opindex mdensity | |
9658 | @opindex mno-density | |
9659 | Enable or disable use of the optional Xtensa code density instructions. | |
9660 | ||
9661 | @item -mmac16 | |
9662 | @itemx -mno-mac16 | |
9663 | @opindex mmac16 | |
9664 | @opindex mno-mac16 | |
9665 | Enable or disable use of the Xtensa MAC16 option. When enabled, GCC | |
9666 | will generate MAC16 instructions from standard C code, with the | |
9667 | limitation that it will use neither the MR register file nor any | |
9668 | instruction that operates on the MR registers. When this option is | |
9669 | disabled, GCC will translate 16-bit multiply/accumulate operations to a | |
9670 | combination of core instructions and library calls, depending on whether | |
9671 | any other multiplier options are enabled. | |
9672 | ||
9673 | @item -mmul16 | |
9674 | @itemx -mno-mul16 | |
9675 | @opindex mmul16 | |
9676 | @opindex mno-mul16 | |
9677 | Enable or disable use of the 16-bit integer multiplier option. When | |
9678 | enabled, the compiler will generate 16-bit multiply instructions for | |
9679 | multiplications of 16 bits or smaller in standard C code. When this | |
9680 | option is disabled, the compiler will either use 32-bit multiply or | |
9681 | MAC16 instructions if they are available or generate library calls to | |
9682 | perform the multiply operations using shifts and adds. | |
9683 | ||
9684 | @item -mmul32 | |
9685 | @itemx -mno-mul32 | |
9686 | @opindex mmul32 | |
9687 | @opindex mno-mul32 | |
9688 | Enable or disable use of the 32-bit integer multiplier option. When | |
9689 | enabled, the compiler will generate 32-bit multiply instructions for | |
9690 | multiplications of 32 bits or smaller in standard C code. When this | |
9691 | option is disabled, the compiler will generate library calls to perform | |
9692 | the multiply operations using either shifts and adds or 16-bit multiply | |
9693 | instructions if they are available. | |
9694 | ||
9695 | @item -mnsa | |
9696 | @itemx -mno-nsa | |
9697 | @opindex mnsa | |
9698 | @opindex mno-nsa | |
9699 | Enable or disable use of the optional normalization shift amount | |
9700 | (@code{NSA}) instructions to implement the built-in @code{ffs} function. | |
9701 | ||
9702 | @item -mminmax | |
9703 | @itemx -mno-minmax | |
9704 | @opindex mminmax | |
9705 | @opindex mno-minmax | |
9706 | Enable or disable use of the optional minimum and maximum value | |
9707 | instructions. | |
9708 | ||
9709 | @item -msext | |
9710 | @itemx -mno-sext | |
9711 | @opindex msext | |
9712 | @opindex mno-sext | |
9713 | Enable or disable use of the optional sign extend (@code{SEXT}) | |
9714 | instruction. | |
9715 | ||
9716 | @item -mbooleans | |
9717 | @itemx -mno-booleans | |
9718 | @opindex mbooleans | |
9719 | @opindex mno-booleans | |
9720 | Enable or disable support for the boolean register file used by Xtensa | |
9721 | coprocessors. This is not typically useful by itself but may be | |
9722 | required for other options that make use of the boolean registers (e.g., | |
9723 | the floating-point option). | |
9724 | ||
9725 | @item -mhard-float | |
9726 | @itemx -msoft-float | |
9727 | @opindex mhard-float | |
9728 | @opindex msoft-float | |
9729 | Enable or disable use of the floating-point option. When enabled, GCC | |
9730 | generates floating-point instructions for 32-bit @code{float} | |
9731 | operations. When this option is disabled, GCC generates library calls | |
9732 | to emulate 32-bit floating-point operations using integer instructions. | |
9733 | Regardless of this option, 64-bit @code{double} operations are always | |
9734 | emulated with calls to library functions. | |
9735 | ||
9736 | @item -mfused-madd | |
9737 | @itemx -mno-fused-madd | |
9738 | @opindex mfused-madd | |
9739 | @opindex mno-fused-madd | |
9740 | Enable or disable use of fused multiply/add and multiply/subtract | |
9741 | instructions in the floating-point option. This has no effect if the | |
9742 | floating-point option is not also enabled. Disabling fused multiply/add | |
9743 | and multiply/subtract instructions forces the compiler to use separate | |
9744 | instructions for the multiply and add/subtract operations. This may be | |
9745 | desirable in some cases where strict IEEE 754-compliant results are | |
9746 | required: the fused multiply add/subtract instructions do not round the | |
9747 | intermediate result, thereby producing results with @emph{more} bits of | |
9748 | precision than specified by the IEEE standard. Disabling fused multiply | |
9749 | add/subtract instructions also ensures that the program output is not | |
9750 | sensitive to the compiler's ability to combine multiply and add/subtract | |
9751 | operations. | |
9752 | ||
9753 | @item -mserialize-volatile | |
9754 | @itemx -mno-serialize-volatile | |
9755 | @opindex mserialize-volatile | |
9756 | @opindex mno-serialize-volatile | |
9757 | When this option is enabled, GCC inserts @code{MEMW} instructions before | |
9758 | @code{volatile} memory references to guarantee sequential consistency. | |
9759 | The default is @option{-mserialize-volatile}. Use | |
9760 | @option{-mno-serialize-volatile} to omit the @code{MEMW} instructions. | |
9761 | ||
9762 | @item -mtext-section-literals | |
9763 | @itemx -mno-text-section-literals | |
9764 | @opindex mtext-section-literals | |
9765 | @opindex mno-text-section-literals | |
9766 | Control the treatment of literal pools. The default is | |
9767 | @option{-mno-text-section-literals}, which places literals in a separate | |
9768 | section in the output file. This allows the literal pool to be placed | |
9769 | in a data RAM/ROM, and it also allows the linker to combine literal | |
9770 | pools from separate object files to remove redundant literals and | |
9771 | improve code size. With @option{-mtext-section-literals}, the literals | |
9772 | are interspersed in the text section in order to keep them as close as | |
9773 | possible to their references. This may be necessary for large assembly | |
9774 | files. | |
9775 | ||
9776 | @item -mtarget-align | |
9777 | @itemx -mno-target-align | |
9778 | @opindex mtarget-align | |
9779 | @opindex mno-target-align | |
9780 | When this option is enabled, GCC instructs the assembler to | |
9781 | automatically align instructions to reduce branch penalties at the | |
9782 | expense of some code density. The assembler attempts to widen density | |
9783 | instructions to align branch targets and the instructions following call | |
9784 | instructions. If there are not enough preceding safe density | |
9785 | instructions to align a target, no widening will be performed. The | |
9786 | default is @option{-mtarget-align}. These options do not affect the | |
9787 | treatment of auto-aligned instructions like @code{LOOP}, which the | |
9788 | assembler will always align, either by widening density instructions or | |
9789 | by inserting no-op instructions. | |
9790 | ||
9791 | @item -mlongcalls | |
9792 | @itemx -mno-longcalls | |
9793 | @opindex mlongcalls | |
9794 | @opindex mno-longcalls | |
9795 | When this option is enabled, GCC instructs the assembler to translate | |
9796 | direct calls to indirect calls unless it can determine that the target | |
9797 | of a direct call is in the range allowed by the call instruction. This | |
9798 | translation typically occurs for calls to functions in other source | |
9799 | files. Specifically, the assembler translates a direct @code{CALL} | |
9800 | instruction into an @code{L32R} followed by a @code{CALLX} instruction. | |
9801 | The default is @option{-mno-longcalls}. This option should be used in | |
9802 | programs where the call target can potentially be out of range. This | |
9803 | option is implemented in the assembler, not the compiler, so the | |
9804 | assembly code generated by GCC will still show direct call | |
9805 | instructions---look at the disassembled object code to see the actual | |
9806 | instructions. Note that the assembler will use an indirect call for | |
9807 | every cross-file call, not just those that really will be out of range. | |
9808 | @end table | |
9809 | ||
9810 | @node Code Gen Options | |
9811 | @section Options for Code Generation Conventions | |
9812 | @cindex code generation conventions | |
9813 | @cindex options, code generation | |
9814 | @cindex run-time options | |
9815 | ||
9816 | These machine-independent options control the interface conventions | |
9817 | used in code generation. | |
9818 | ||
9819 | Most of them have both positive and negative forms; the negative form | |
9820 | of @option{-ffoo} would be @option{-fno-foo}. In the table below, only | |
9821 | one of the forms is listed---the one which is not the default. You | |
9822 | can figure out the other form by either removing @samp{no-} or adding | |
9823 | it. | |
9824 | ||
9825 | @table @gcctabopt | |
9826 | @item -fexceptions | |
9827 | @opindex fexceptions | |
9828 | Enable exception handling. Generates extra code needed to propagate | |
9829 | exceptions. For some targets, this implies GCC will generate frame | |
9830 | unwind information for all functions, which can produce significant data | |
9831 | size overhead, although it does not affect execution. If you do not | |
9832 | specify this option, GCC will enable it by default for languages like | |
9833 | C++ which normally require exception handling, and disable it for | |
9834 | languages like C that do not normally require it. However, you may need | |
9835 | to enable this option when compiling C code that needs to interoperate | |
9836 | properly with exception handlers written in C++. You may also wish to | |
9837 | disable this option if you are compiling older C++ programs that don't | |
9838 | use exception handling. | |
9839 | ||
9840 | @item -fnon-call-exceptions | |
9841 | @opindex fnon-call-exceptions | |
9842 | Generate code that allows trapping instructions to throw exceptions. | |
9843 | Note that this requires platform-specific runtime support that does | |
9844 | not exist everywhere. Moreover, it only allows @emph{trapping} | |
9845 | instructions to throw exceptions, i.e.@: memory references or floating | |
9846 | point instructions. It does not allow exceptions to be thrown from | |
9847 | arbitrary signal handlers such as @code{SIGALRM}. | |
9848 | ||
9849 | @item -funwind-tables | |
9850 | @opindex funwind-tables | |
9851 | Similar to @option{-fexceptions}, except that it will just generate any needed | |
9852 | static data, but will not affect the generated code in any other way. | |
9853 | You will normally not enable this option; instead, a language processor | |
9854 | that needs this handling would enable it on your behalf. | |
9855 | ||
9856 | @item -fasynchronous-unwind-tables | |
9857 | @opindex funwind-tables | |
9858 | Generate unwind table in dwarf2 format, if supported by target machine. The | |
9859 | table is exact at each instruction boundary, so it can be used for stack | |
9860 | unwinding from asynchronous events (such as debugger or garbage collector). | |
9861 | ||
9862 | @item -fpcc-struct-return | |
9863 | @opindex fpcc-struct-return | |
9864 | Return ``short'' @code{struct} and @code{union} values in memory like | |
9865 | longer ones, rather than in registers. This convention is less | |
9866 | efficient, but it has the advantage of allowing intercallability between | |
9867 | GCC-compiled files and files compiled with other compilers. | |
9868 | ||
9869 | The precise convention for returning structures in memory depends | |
9870 | on the target configuration macros. | |
9871 | ||
9872 | Short structures and unions are those whose size and alignment match | |
9873 | that of some integer type. | |
9874 | ||
9875 | @item -freg-struct-return | |
9876 | @opindex freg-struct-return | |
9877 | Return @code{struct} and @code{union} values in registers when possible. | |
9878 | This is more efficient for small structures than | |
9879 | @option{-fpcc-struct-return}. | |
9880 | ||
9881 | If you specify neither @option{-fpcc-struct-return} nor | |
9882 | @option{-freg-struct-return}, GCC defaults to whichever convention is | |
9883 | standard for the target. If there is no standard convention, GCC | |
9884 | defaults to @option{-fpcc-struct-return}, except on targets where GCC is | |
9885 | the principal compiler. In those cases, we can choose the standard, and | |
9886 | we chose the more efficient register return alternative. | |
9887 | ||
9888 | @item -fshort-enums | |
9889 | @opindex fshort-enums | |
9890 | Allocate to an @code{enum} type only as many bytes as it needs for the | |
9891 | declared range of possible values. Specifically, the @code{enum} type | |
9892 | will be equivalent to the smallest integer type which has enough room. | |
9893 | ||
9894 | @item -fshort-double | |
9895 | @opindex fshort-double | |
9896 | Use the same size for @code{double} as for @code{float}. | |
9897 | ||
9898 | @item -fshared-data | |
9899 | @opindex fshared-data | |
9900 | Requests that the data and non-@code{const} variables of this | |
9901 | compilation be shared data rather than private data. The distinction | |
9902 | makes sense only on certain operating systems, where shared data is | |
9903 | shared between processes running the same program, while private data | |
9904 | exists in one copy per process. | |
9905 | ||
9906 | @item -fno-common | |
9907 | @opindex fno-common | |
9908 | In C, allocate even uninitialized global variables in the data section of the | |
9909 | object file, rather than generating them as common blocks. This has the | |
9910 | effect that if the same variable is declared (without @code{extern}) in | |
9911 | two different compilations, you will get an error when you link them. | |
9912 | The only reason this might be useful is if you wish to verify that the | |
9913 | program will work on other systems which always work this way. | |
9914 | ||
9915 | @item -fno-ident | |
9916 | @opindex fno-ident | |
9917 | Ignore the @samp{#ident} directive. | |
9918 | ||
9919 | @item -fno-gnu-linker | |
9920 | @opindex fno-gnu-linker | |
9921 | Do not output global initializations (such as C++ constructors and | |
9922 | destructors) in the form used by the GNU linker (on systems where the GNU | |
9923 | linker is the standard method of handling them). Use this option when | |
9924 | you want to use a non-GNU linker, which also requires using the | |
9925 | @command{collect2} program to make sure the system linker includes | |
9926 | constructors and destructors. (@command{collect2} is included in the GCC | |
9927 | distribution.) For systems which @emph{must} use @command{collect2}, the | |
9928 | compiler driver @command{gcc} is configured to do this automatically. | |
9929 | ||
9930 | @item -finhibit-size-directive | |
9931 | @opindex finhibit-size-directive | |
9932 | Don't output a @code{.size} assembler directive, or anything else that | |
9933 | would cause trouble if the function is split in the middle, and the | |
9934 | two halves are placed at locations far apart in memory. This option is | |
9935 | used when compiling @file{crtstuff.c}; you should not need to use it | |
9936 | for anything else. | |
9937 | ||
9938 | @item -fverbose-asm | |
9939 | @opindex fverbose-asm | |
9940 | Put extra commentary information in the generated assembly code to | |
9941 | make it more readable. This option is generally only of use to those | |
9942 | who actually need to read the generated assembly code (perhaps while | |
9943 | debugging the compiler itself). | |
9944 | ||
9945 | @option{-fno-verbose-asm}, the default, causes the | |
9946 | extra information to be omitted and is useful when comparing two assembler | |
9947 | files. | |
9948 | ||
9949 | @item -fvolatile | |
9950 | @opindex fvolatile | |
9951 | Consider all memory references through pointers to be volatile. | |
9952 | ||
9953 | @item -fvolatile-global | |
9954 | @opindex fvolatile-global | |
9955 | Consider all memory references to extern and global data items to | |
9956 | be volatile. GCC does not consider static data items to be volatile | |
9957 | because of this switch. | |
9958 | ||
9959 | @item -fvolatile-static | |
9960 | @opindex fvolatile-static | |
9961 | Consider all memory references to static data to be volatile. | |
9962 | ||
9963 | @item -fpic | |
9964 | @opindex fpic | |
9965 | @cindex global offset table | |
9966 | @cindex PIC | |
9967 | Generate position-independent code (PIC) suitable for use in a shared | |
9968 | library, if supported for the target machine. Such code accesses all | |
9969 | constant addresses through a global offset table (GOT)@. The dynamic | |
9970 | loader resolves the GOT entries when the program starts (the dynamic | |
9971 | loader is not part of GCC; it is part of the operating system). If | |
9972 | the GOT size for the linked executable exceeds a machine-specific | |
9973 | maximum size, you get an error message from the linker indicating that | |
9974 | @option{-fpic} does not work; in that case, recompile with @option{-fPIC} | |
9975 | instead. (These maximums are 16k on the m88k, 8k on the Sparc, and 32k | |
9976 | on the m68k and RS/6000. The 386 has no such limit.) | |
9977 | ||
9978 | Position-independent code requires special support, and therefore works | |
9979 | only on certain machines. For the 386, GCC supports PIC for System V | |
9980 | but not for the Sun 386i. Code generated for the IBM RS/6000 is always | |
9981 | position-independent. | |
9982 | ||
9983 | @item -fPIC | |
9984 | @opindex fPIC | |
9985 | If supported for the target machine, emit position-independent code, | |
9986 | suitable for dynamic linking and avoiding any limit on the size of the | |
9987 | global offset table. This option makes a difference on the m68k, m88k, | |
9988 | and the Sparc. | |
9989 | ||
9990 | Position-independent code requires special support, and therefore works | |
9991 | only on certain machines. | |
9992 | ||
9993 | @item -ffixed-@var{reg} | |
9994 | @opindex ffixed | |
9995 | Treat the register named @var{reg} as a fixed register; generated code | |
9996 | should never refer to it (except perhaps as a stack pointer, frame | |
9997 | pointer or in some other fixed role). | |
9998 | ||
9999 | @var{reg} must be the name of a register. The register names accepted | |
10000 | are machine-specific and are defined in the @code{REGISTER_NAMES} | |
10001 | macro in the machine description macro file. | |
10002 | ||
10003 | This flag does not have a negative form, because it specifies a | |
10004 | three-way choice. | |
10005 | ||
10006 | @item -fcall-used-@var{reg} | |
10007 | @opindex fcall-used | |
10008 | Treat the register named @var{reg} as an allocable register that is | |
10009 | clobbered by function calls. It may be allocated for temporaries or | |
10010 | variables that do not live across a call. Functions compiled this way | |
10011 | will not save and restore the register @var{reg}. | |
10012 | ||
10013 | It is an error to used this flag with the frame pointer or stack pointer. | |
10014 | Use of this flag for other registers that have fixed pervasive roles in | |
10015 | the machine's execution model will produce disastrous results. | |
10016 | ||
10017 | This flag does not have a negative form, because it specifies a | |
10018 | three-way choice. | |
10019 | ||
10020 | @item -fcall-saved-@var{reg} | |
10021 | @opindex fcall-saved | |
10022 | Treat the register named @var{reg} as an allocable register saved by | |
10023 | functions. It may be allocated even for temporaries or variables that | |
10024 | live across a call. Functions compiled this way will save and restore | |
10025 | the register @var{reg} if they use it. | |
10026 | ||
10027 | It is an error to used this flag with the frame pointer or stack pointer. | |
10028 | Use of this flag for other registers that have fixed pervasive roles in | |
10029 | the machine's execution model will produce disastrous results. | |
10030 | ||
10031 | A different sort of disaster will result from the use of this flag for | |
10032 | a register in which function values may be returned. | |
10033 | ||
10034 | This flag does not have a negative form, because it specifies a | |
10035 | three-way choice. | |
10036 | ||
10037 | @item -fpack-struct | |
10038 | @opindex fpack-struct | |
10039 | Pack all structure members together without holes. Usually you would | |
10040 | not want to use this option, since it makes the code suboptimal, and | |
10041 | the offsets of structure members won't agree with system libraries. | |
10042 | ||
10043 | @item -finstrument-functions | |
10044 | @opindex finstrument-functions | |
10045 | Generate instrumentation calls for entry and exit to functions. Just | |
10046 | after function entry and just before function exit, the following | |
10047 | profiling functions will be called with the address of the current | |
10048 | function and its call site. (On some platforms, | |
10049 | @code{__builtin_return_address} does not work beyond the current | |
10050 | function, so the call site information may not be available to the | |
10051 | profiling functions otherwise.) | |
10052 | ||
10053 | @example | |
10054 | void __cyg_profile_func_enter (void *this_fn, | |
10055 | void *call_site); | |
10056 | void __cyg_profile_func_exit (void *this_fn, | |
10057 | void *call_site); | |
10058 | @end example | |
10059 | ||
10060 | The first argument is the address of the start of the current function, | |
10061 | which may be looked up exactly in the symbol table. | |
10062 | ||
10063 | This instrumentation is also done for functions expanded inline in other | |
10064 | functions. The profiling calls will indicate where, conceptually, the | |
10065 | inline function is entered and exited. This means that addressable | |
10066 | versions of such functions must be available. If all your uses of a | |
10067 | function are expanded inline, this may mean an additional expansion of | |
10068 | code size. If you use @samp{extern inline} in your C code, an | |
10069 | addressable version of such functions must be provided. (This is | |
10070 | normally the case anyways, but if you get lucky and the optimizer always | |
10071 | expands the functions inline, you might have gotten away without | |
10072 | providing static copies.) | |
10073 | ||
10074 | A function may be given the attribute @code{no_instrument_function}, in | |
10075 | which case this instrumentation will not be done. This can be used, for | |
10076 | example, for the profiling functions listed above, high-priority | |
10077 | interrupt routines, and any functions from which the profiling functions | |
10078 | cannot safely be called (perhaps signal handlers, if the profiling | |
10079 | routines generate output or allocate memory). | |
10080 | ||
10081 | @item -fstack-check | |
10082 | @opindex fstack-check | |
10083 | Generate code to verify that you do not go beyond the boundary of the | |
10084 | stack. You should specify this flag if you are running in an | |
10085 | environment with multiple threads, but only rarely need to specify it in | |
10086 | a single-threaded environment since stack overflow is automatically | |
10087 | detected on nearly all systems if there is only one stack. | |
10088 | ||
10089 | Note that this switch does not actually cause checking to be done; the | |
10090 | operating system must do that. The switch causes generation of code | |
10091 | to ensure that the operating system sees the stack being extended. | |
10092 | ||
10093 | @item -fstack-limit-register=@var{reg} | |
10094 | @itemx -fstack-limit-symbol=@var{sym} | |
10095 | @itemx -fno-stack-limit | |
10096 | @opindex fstack-limit-register | |
10097 | @opindex fstack-limit-symbol | |
10098 | @opindex fno-stack-limit | |
10099 | Generate code to ensure that the stack does not grow beyond a certain value, | |
10100 | either the value of a register or the address of a symbol. If the stack | |
10101 | would grow beyond the value, a signal is raised. For most targets, | |
10102 | the signal is raised before the stack overruns the boundary, so | |
10103 | it is possible to catch the signal without taking special precautions. | |
10104 | ||
10105 | For instance, if the stack starts at absolute address @samp{0x80000000} | |
10106 | and grows downwards, you can use the flags | |
10107 | @option{-fstack-limit-symbol=__stack_limit} and | |
10108 | @option{-Wl,--defsym,__stack_limit=0x7ffe0000} to enforce a stack limit | |
10109 | of 128KB@. Note that this may only work with the GNU linker. | |
10110 | ||
10111 | @cindex aliasing of parameters | |
10112 | @cindex parameters, aliased | |
10113 | @item -fargument-alias | |
10114 | @itemx -fargument-noalias | |
10115 | @itemx -fargument-noalias-global | |
10116 | @opindex fargument-alias | |
10117 | @opindex fargument-noalias | |
10118 | @opindex fargument-noalias-global | |
10119 | Specify the possible relationships among parameters and between | |
10120 | parameters and global data. | |
10121 | ||
10122 | @option{-fargument-alias} specifies that arguments (parameters) may | |
10123 | alias each other and may alias global storage.@* | |
10124 | @option{-fargument-noalias} specifies that arguments do not alias | |
10125 | each other, but may alias global storage.@* | |
10126 | @option{-fargument-noalias-global} specifies that arguments do not | |
10127 | alias each other and do not alias global storage. | |
10128 | ||
10129 | Each language will automatically use whatever option is required by | |
10130 | the language standard. You should not need to use these options yourself. | |
10131 | ||
10132 | @item -fleading-underscore | |
10133 | @opindex fleading-underscore | |
10134 | This option and its counterpart, @option{-fno-leading-underscore}, forcibly | |
10135 | change the way C symbols are represented in the object file. One use | |
10136 | is to help link with legacy assembly code. | |
10137 | ||
10138 | Be warned that you should know what you are doing when invoking this | |
10139 | option, and that not all targets provide complete support for it. | |
10140 | @end table | |
10141 | ||
10142 | @c man end | |
10143 | ||
10144 | @node Environment Variables | |
10145 | @section Environment Variables Affecting GCC | |
10146 | @cindex environment variables | |
10147 | ||
10148 | @c man begin ENVIRONMENT | |
10149 | ||
10150 | This section describes several environment variables that affect how GCC | |
10151 | operates. Some of them work by specifying directories or prefixes to use | |
10152 | when searching for various kinds of files. Some are used to specify other | |
10153 | aspects of the compilation environment. | |
10154 | ||
10155 | Note that you can also specify places to search using options such as | |
10156 | @option{-B}, @option{-I} and @option{-L} (@pxref{Directory Options}). These | |
10157 | take precedence over places specified using environment variables, which | |
10158 | in turn take precedence over those specified by the configuration of GCC@. | |
10159 | @xref{Driver,, Controlling the Compilation Driver @file{gcc}, gccint, | |
10160 | GNU Compiler Collection (GCC) Internals}. | |
10161 | ||
10162 | @table @env | |
10163 | @item LANG | |
10164 | @itemx LC_CTYPE | |
10165 | @c @itemx LC_COLLATE | |
10166 | @itemx LC_MESSAGES | |
10167 | @c @itemx LC_MONETARY | |
10168 | @c @itemx LC_NUMERIC | |
10169 | @c @itemx LC_TIME | |
10170 | @itemx LC_ALL | |
10171 | @findex LANG | |
10172 | @findex LC_CTYPE | |
10173 | @c @findex LC_COLLATE | |
10174 | @findex LC_MESSAGES | |
10175 | @c @findex LC_MONETARY | |
10176 | @c @findex LC_NUMERIC | |
10177 | @c @findex LC_TIME | |
10178 | @findex LC_ALL | |
10179 | @cindex locale | |
10180 | These environment variables control the way that GCC uses | |
10181 | localization information that allow GCC to work with different | |
10182 | national conventions. GCC inspects the locale categories | |
10183 | @env{LC_CTYPE} and @env{LC_MESSAGES} if it has been configured to do | |
10184 | so. These locale categories can be set to any value supported by your | |
10185 | installation. A typical value is @samp{en_UK} for English in the United | |
10186 | Kingdom. | |
10187 | ||
10188 | The @env{LC_CTYPE} environment variable specifies character | |
10189 | classification. GCC uses it to determine the character boundaries in | |
10190 | a string; this is needed for some multibyte encodings that contain quote | |
10191 | and escape characters that would otherwise be interpreted as a string | |
10192 | end or escape. | |
10193 | ||
10194 | The @env{LC_MESSAGES} environment variable specifies the language to | |
10195 | use in diagnostic messages. | |
10196 | ||
10197 | If the @env{LC_ALL} environment variable is set, it overrides the value | |
10198 | of @env{LC_CTYPE} and @env{LC_MESSAGES}; otherwise, @env{LC_CTYPE} | |
10199 | and @env{LC_MESSAGES} default to the value of the @env{LANG} | |
10200 | environment variable. If none of these variables are set, GCC | |
10201 | defaults to traditional C English behavior. | |
10202 | ||
10203 | @item TMPDIR | |
10204 | @findex TMPDIR | |
10205 | If @env{TMPDIR} is set, it specifies the directory to use for temporary | |
10206 | files. GCC uses temporary files to hold the output of one stage of | |
10207 | compilation which is to be used as input to the next stage: for example, | |
10208 | the output of the preprocessor, which is the input to the compiler | |
10209 | proper. | |
10210 | ||
10211 | @item GCC_EXEC_PREFIX | |
10212 | @findex GCC_EXEC_PREFIX | |
10213 | If @env{GCC_EXEC_PREFIX} is set, it specifies a prefix to use in the | |
10214 | names of the subprograms executed by the compiler. No slash is added | |
10215 | when this prefix is combined with the name of a subprogram, but you can | |
10216 | specify a prefix that ends with a slash if you wish. | |
10217 | ||
10218 | If @env{GCC_EXEC_PREFIX} is not set, GCC will attempt to figure out | |
10219 | an appropriate prefix to use based on the pathname it was invoked with. | |
10220 | ||
10221 | If GCC cannot find the subprogram using the specified prefix, it | |
10222 | tries looking in the usual places for the subprogram. | |
10223 | ||
10224 | The default value of @env{GCC_EXEC_PREFIX} is | |
10225 | @file{@var{prefix}/lib/gcc-lib/} where @var{prefix} is the value | |
10226 | of @code{prefix} when you ran the @file{configure} script. | |
10227 | ||
10228 | Other prefixes specified with @option{-B} take precedence over this prefix. | |
10229 | ||
10230 | This prefix is also used for finding files such as @file{crt0.o} that are | |
10231 | used for linking. | |
10232 | ||
10233 | In addition, the prefix is used in an unusual way in finding the | |
10234 | directories to search for header files. For each of the standard | |
10235 | directories whose name normally begins with @samp{/usr/local/lib/gcc-lib} | |
10236 | (more precisely, with the value of @env{GCC_INCLUDE_DIR}), GCC tries | |
10237 | replacing that beginning with the specified prefix to produce an | |
10238 | alternate directory name. Thus, with @option{-Bfoo/}, GCC will search | |
10239 | @file{foo/bar} where it would normally search @file{/usr/local/lib/bar}. | |
10240 | These alternate directories are searched first; the standard directories | |
10241 | come next. | |
10242 | ||
10243 | @item COMPILER_PATH | |
10244 | @findex COMPILER_PATH | |
10245 | The value of @env{COMPILER_PATH} is a colon-separated list of | |
10246 | directories, much like @env{PATH}. GCC tries the directories thus | |
10247 | specified when searching for subprograms, if it can't find the | |
10248 | subprograms using @env{GCC_EXEC_PREFIX}. | |
10249 | ||
10250 | @item LIBRARY_PATH | |
10251 | @findex LIBRARY_PATH | |
10252 | The value of @env{LIBRARY_PATH} is a colon-separated list of | |
10253 | directories, much like @env{PATH}. When configured as a native compiler, | |
10254 | GCC tries the directories thus specified when searching for special | |
10255 | linker files, if it can't find them using @env{GCC_EXEC_PREFIX}. Linking | |
10256 | using GCC also uses these directories when searching for ordinary | |
10257 | libraries for the @option{-l} option (but directories specified with | |
10258 | @option{-L} come first). | |
10259 | ||
10260 | @item C_INCLUDE_PATH | |
10261 | @itemx CPLUS_INCLUDE_PATH | |
10262 | @itemx OBJC_INCLUDE_PATH | |
10263 | @findex C_INCLUDE_PATH | |
10264 | @findex CPLUS_INCLUDE_PATH | |
10265 | @findex OBJC_INCLUDE_PATH | |
10266 | @c @itemx OBJCPLUS_INCLUDE_PATH | |
10267 | These environment variables pertain to particular languages. Each | |
10268 | variable's value is a colon-separated list of directories, much like | |
10269 | @env{PATH}. When GCC searches for header files, it tries the | |
10270 | directories listed in the variable for the language you are using, after | |
10271 | the directories specified with @option{-I} but before the standard header | |
10272 | file directories. | |
10273 | ||
10274 | @item DEPENDENCIES_OUTPUT | |
10275 | @findex DEPENDENCIES_OUTPUT | |
10276 | @cindex dependencies for make as output | |
10277 | If this variable is set, its value specifies how to output dependencies | |
10278 | for Make based on the header files processed by the compiler. This | |
10279 | output looks much like the output from the @option{-M} option | |
10280 | (@pxref{Preprocessor Options}), but it goes to a separate file, and is | |
10281 | in addition to the usual results of compilation. | |
10282 | ||
10283 | The value of @env{DEPENDENCIES_OUTPUT} can be just a file name, in | |
10284 | which case the Make rules are written to that file, guessing the target | |
10285 | name from the source file name. Or the value can have the form | |
10286 | @samp{@var{file} @var{target}}, in which case the rules are written to | |
10287 | file @var{file} using @var{target} as the target name. | |
10288 | ||
10289 | @item LANG | |
10290 | @findex LANG | |
10291 | @cindex locale definition | |
10292 | This variable is used to pass locale information to the compiler. One way in | |
10293 | which this information is used is to determine the character set to be used | |
10294 | when character literals, string literals and comments are parsed in C and C++. | |
10295 | When the compiler is configured to allow multibyte characters, | |
10296 | the following values for @env{LANG} are recognized: | |
10297 | ||
10298 | @table @samp | |
10299 | @item C-JIS | |
10300 | Recognize JIS characters. | |
10301 | @item C-SJIS | |
10302 | Recognize SJIS characters. | |
10303 | @item C-EUCJP | |
10304 | Recognize EUCJP characters. | |
10305 | @end table | |
10306 | ||
10307 | If @env{LANG} is not defined, or if it has some other value, then the | |
10308 | compiler will use mblen and mbtowc as defined by the default locale to | |
10309 | recognize and translate multibyte characters. | |
10310 | @end table | |
10311 | ||
10312 | @c man end | |
10313 | ||
10314 | @node Running Protoize | |
10315 | @section Running Protoize | |
10316 | ||
10317 | The program @code{protoize} is an optional part of GCC@. You can use | |
10318 | it to add prototypes to a program, thus converting the program to ISO | |
10319 | C in one respect. The companion program @code{unprotoize} does the | |
10320 | reverse: it removes argument types from any prototypes that are found. | |
10321 | ||
10322 | When you run these programs, you must specify a set of source files as | |
10323 | command line arguments. The conversion programs start out by compiling | |
10324 | these files to see what functions they define. The information gathered | |
10325 | about a file @var{foo} is saved in a file named @file{@var{foo}.X}. | |
10326 | ||
10327 | After scanning comes actual conversion. The specified files are all | |
10328 | eligible to be converted; any files they include (whether sources or | |
10329 | just headers) are eligible as well. | |
10330 | ||
10331 | But not all the eligible files are converted. By default, | |
10332 | @code{protoize} and @code{unprotoize} convert only source and header | |
10333 | files in the current directory. You can specify additional directories | |
10334 | whose files should be converted with the @option{-d @var{directory}} | |
10335 | option. You can also specify particular files to exclude with the | |
10336 | @option{-x @var{file}} option. A file is converted if it is eligible, its | |
10337 | directory name matches one of the specified directory names, and its | |
10338 | name within the directory has not been excluded. | |
10339 | ||
10340 | Basic conversion with @code{protoize} consists of rewriting most | |
10341 | function definitions and function declarations to specify the types of | |
10342 | the arguments. The only ones not rewritten are those for varargs | |
10343 | functions. | |
10344 | ||
10345 | @code{protoize} optionally inserts prototype declarations at the | |
10346 | beginning of the source file, to make them available for any calls that | |
10347 | precede the function's definition. Or it can insert prototype | |
10348 | declarations with block scope in the blocks where undeclared functions | |
10349 | are called. | |
10350 | ||
10351 | Basic conversion with @code{unprotoize} consists of rewriting most | |
10352 | function declarations to remove any argument types, and rewriting | |
10353 | function definitions to the old-style pre-ISO form. | |
10354 | ||
10355 | Both conversion programs print a warning for any function declaration or | |
10356 | definition that they can't convert. You can suppress these warnings | |
10357 | with @option{-q}. | |
10358 | ||
10359 | The output from @code{protoize} or @code{unprotoize} replaces the | |
10360 | original source file. The original file is renamed to a name ending | |
10361 | with @samp{.save} (for DOS, the saved filename ends in @samp{.sav} | |
10362 | without the original @samp{.c} suffix). If the @samp{.save} (@samp{.sav} | |
10363 | for DOS) file already exists, then the source file is simply discarded. | |
10364 | ||
10365 | @code{protoize} and @code{unprotoize} both depend on GCC itself to | |
10366 | scan the program and collect information about the functions it uses. | |
10367 | So neither of these programs will work until GCC is installed. | |
10368 | ||
10369 | Here is a table of the options you can use with @code{protoize} and | |
10370 | @code{unprotoize}. Each option works with both programs unless | |
10371 | otherwise stated. | |
10372 | ||
10373 | @table @code | |
10374 | @item -B @var{directory} | |
10375 | Look for the file @file{SYSCALLS.c.X} in @var{directory}, instead of the | |
10376 | usual directory (normally @file{/usr/local/lib}). This file contains | |
10377 | prototype information about standard system functions. This option | |
10378 | applies only to @code{protoize}. | |
10379 | ||
10380 | @item -c @var{compilation-options} | |
10381 | Use @var{compilation-options} as the options when running @code{gcc} to | |
10382 | produce the @samp{.X} files. The special option @option{-aux-info} is | |
10383 | always passed in addition, to tell @code{gcc} to write a @samp{.X} file. | |
10384 | ||
10385 | Note that the compilation options must be given as a single argument to | |
10386 | @code{protoize} or @code{unprotoize}. If you want to specify several | |
10387 | @code{gcc} options, you must quote the entire set of compilation options | |
10388 | to make them a single word in the shell. | |
10389 | ||
10390 | There are certain @code{gcc} arguments that you cannot use, because they | |
10391 | would produce the wrong kind of output. These include @option{-g}, | |
10392 | @option{-O}, @option{-c}, @option{-S}, and @option{-o} If you include these in | |
10393 | the @var{compilation-options}, they are ignored. | |
10394 | ||
10395 | @item -C | |
10396 | Rename files to end in @samp{.C} (@samp{.cc} for DOS-based file | |
10397 | systems) instead of @samp{.c}. This is convenient if you are converting | |
10398 | a C program to C++. This option applies only to @code{protoize}. | |
10399 | ||
10400 | @item -g | |
10401 | Add explicit global declarations. This means inserting explicit | |
10402 | declarations at the beginning of each source file for each function | |
10403 | that is called in the file and was not declared. These declarations | |
10404 | precede the first function definition that contains a call to an | |
10405 | undeclared function. This option applies only to @code{protoize}. | |
10406 | ||
10407 | @item -i @var{string} | |
10408 | Indent old-style parameter declarations with the string @var{string}. | |
10409 | This option applies only to @code{protoize}. | |
10410 | ||
10411 | @code{unprotoize} converts prototyped function definitions to old-style | |
10412 | function definitions, where the arguments are declared between the | |
10413 | argument list and the initial @samp{@{}. By default, @code{unprotoize} | |
10414 | uses five spaces as the indentation. If you want to indent with just | |
10415 | one space instead, use @option{-i " "}. | |
10416 | ||
10417 | @item -k | |
10418 | Keep the @samp{.X} files. Normally, they are deleted after conversion | |
10419 | is finished. | |
10420 | ||
10421 | @item -l | |
10422 | Add explicit local declarations. @code{protoize} with @option{-l} inserts | |
10423 | a prototype declaration for each function in each block which calls the | |
10424 | function without any declaration. This option applies only to | |
10425 | @code{protoize}. | |
10426 | ||
10427 | @item -n | |
10428 | Make no real changes. This mode just prints information about the conversions | |
10429 | that would have been done without @option{-n}. | |
10430 | ||
10431 | @item -N | |
10432 | Make no @samp{.save} files. The original files are simply deleted. | |
10433 | Use this option with caution. | |
10434 | ||
10435 | @item -p @var{program} | |
10436 | Use the program @var{program} as the compiler. Normally, the name | |
10437 | @file{gcc} is used. | |
10438 | ||
10439 | @item -q | |
10440 | Work quietly. Most warnings are suppressed. | |
10441 | ||
10442 | @item -v | |
10443 | Print the version number, just like @option{-v} for @code{gcc}. | |
10444 | @end table | |
10445 | ||
10446 | If you need special compiler options to compile one of your program's | |
10447 | source files, then you should generate that file's @samp{.X} file | |
10448 | specially, by running @code{gcc} on that source file with the | |
10449 | appropriate options and the option @option{-aux-info}. Then run | |
10450 | @code{protoize} on the entire set of files. @code{protoize} will use | |
10451 | the existing @samp{.X} file because it is newer than the source file. | |
10452 | For example: | |
10453 | ||
10454 | @example | |
10455 | gcc -Dfoo=bar file1.c -aux-info file1.X | |
10456 | protoize *.c | |
10457 | @end example | |
10458 | ||
10459 | @noindent | |
10460 | You need to include the special files along with the rest in the | |
10461 | @code{protoize} command, even though their @samp{.X} files already | |
10462 | exist, because otherwise they won't get converted. | |
10463 | ||
10464 | @xref{Protoize Caveats}, for more information on how to use | |
10465 | @code{protoize} successfully. |