Commit | Line | Data |
---|---|---|
78849248 ILT |
1 | @c Copyright 1991, 1992, 1993, 1994, 1995, 1997, 1999, 2000, 2001, |
2 | @c 2002, 2003, 2004 | |
f7e42eb4 | 3 | @c Free Software Foundation, Inc. |
252b5132 RH |
4 | @c This is part of the GAS manual. |
5 | @c For copying conditions, see the file as.texinfo. | |
6 | @ifset GENERIC | |
7 | @page | |
8 | @node MIPS-Dependent | |
9 | @chapter MIPS Dependent Features | |
10 | @end ifset | |
11 | @ifclear GENERIC | |
12 | @node Machine Dependencies | |
13 | @chapter MIPS Dependent Features | |
14 | @end ifclear | |
15 | ||
16 | @cindex MIPS processor | |
17 | @sc{gnu} @code{@value{AS}} for @sc{mips} architectures supports several | |
84ea6cf2 | 18 | different @sc{mips} processors, and MIPS ISA levels I through V, MIPS32, |
584da044 NC |
19 | and MIPS64. For information about the @sc{mips} instruction set, see |
20 | @cite{MIPS RISC Architecture}, by Kane and Heindrich (Prentice-Hall). | |
21 | For an overview of @sc{mips} assembly conventions, see ``Appendix D: | |
22 | Assembly Language Programming'' in the same work. | |
252b5132 RH |
23 | |
24 | @menu | |
25 | * MIPS Opts:: Assembler options | |
26 | * MIPS Object:: ECOFF object code | |
27 | * MIPS Stabs:: Directives for debugging information | |
28 | * MIPS ISA:: Directives to override the ISA level | |
aed1a261 | 29 | * MIPS symbol sizes:: Directives to override the size of symbols |
252b5132 RH |
30 | * MIPS autoextend:: Directives for extending MIPS 16 bit instructions |
31 | * MIPS insn:: Directive to mark data as an instruction | |
32 | * MIPS option stack:: Directives to save and restore options | |
0eb7102d AJ |
33 | * MIPS ASE instruction generation overrides:: Directives to control |
34 | generation of MIPS ASE instructions | |
252b5132 RH |
35 | @end menu |
36 | ||
37 | @node MIPS Opts | |
38 | @section Assembler options | |
39 | ||
40 | The @sc{mips} configurations of @sc{gnu} @code{@value{AS}} support these | |
41 | special options: | |
42 | ||
43 | @table @code | |
44 | @cindex @code{-G} option (MIPS) | |
45 | @item -G @var{num} | |
46 | This option sets the largest size of an object that can be referenced | |
47 | implicitly with the @code{gp} register. It is only accepted for targets | |
48 | that use @sc{ecoff} format. The default value is 8. | |
49 | ||
50 | @cindex @code{-EB} option (MIPS) | |
51 | @cindex @code{-EL} option (MIPS) | |
52 | @cindex MIPS big-endian output | |
53 | @cindex MIPS little-endian output | |
54 | @cindex big-endian output, MIPS | |
55 | @cindex little-endian output, MIPS | |
56 | @item -EB | |
57 | @itemx -EL | |
58 | Any @sc{mips} configuration of @code{@value{AS}} can select big-endian or | |
59 | little-endian output at run time (unlike the other @sc{gnu} development | |
60 | tools, which must be configured for one or the other). Use @samp{-EB} | |
61 | to select big-endian output, and @samp{-EL} for little-endian. | |
62 | ||
63 | @cindex MIPS architecture options | |
64 | @item -mips1 | |
65 | @itemx -mips2 | |
66 | @itemx -mips3 | |
67 | @itemx -mips4 | |
84ea6cf2 | 68 | @itemx -mips5 |
e7af610e | 69 | @itemx -mips32 |
af7ee8bf | 70 | @itemx -mips32r2 |
84ea6cf2 | 71 | @itemx -mips64 |
5f74bc13 | 72 | @itemx -mips64r2 |
252b5132 RH |
73 | Generate code for a particular MIPS Instruction Set Architecture level. |
74 | @samp{-mips1} corresponds to the @sc{r2000} and @sc{r3000} processors, | |
75 | @samp{-mips2} to the @sc{r6000} processor, @samp{-mips3} to the | |
84ea6cf2 | 76 | @sc{r4000} processor, and @samp{-mips4} to the @sc{r8000} and |
5f74bc13 CD |
77 | @sc{r10000} processors. @samp{-mips5}, @samp{-mips32}, @samp{-mips32r2}, |
78 | @samp{-mips64}, and @samp{-mips64r2} | |
79 | correspond to generic | |
80 | @sc{MIPS V}, @sc{MIPS32}, @sc{MIPS32 Release 2}, @sc{MIPS64}, | |
81 | and @sc{MIPS64 Release 2} | |
82 | ISA processors, respectively. You can also switch | |
584da044 | 83 | instruction sets during the assembly; see @ref{MIPS ISA, Directives to |
ec68c924 | 84 | override the ISA level}. |
252b5132 | 85 | |
6349b5f4 | 86 | @item -mgp32 |
ca4e0257 RS |
87 | @itemx -mfp32 |
88 | Some macros have different expansions for 32-bit and 64-bit registers. | |
89 | The register sizes are normally inferred from the ISA and ABI, but these | |
90 | flags force a certain group of registers to be treated as 32 bits wide at | |
91 | all times. @samp{-mgp32} controls the size of general-purpose registers | |
92 | and @samp{-mfp32} controls the size of floating-point registers. | |
93 | ||
94 | On some MIPS variants there is a 32-bit mode flag; when this flag is | |
95 | set, 64-bit instructions generate a trap. Also, some 32-bit OSes only | |
96 | save the 32-bit registers on a context switch, so it is essential never | |
97 | to use the 64-bit registers. | |
6349b5f4 AH |
98 | |
99 | @item -mgp64 | |
100 | Assume that 64-bit general purpose registers are available. This is | |
101 | provided in the interests of symmetry with -gp32. | |
102 | ||
252b5132 RH |
103 | @item -mips16 |
104 | @itemx -no-mips16 | |
105 | Generate code for the MIPS 16 processor. This is equivalent to putting | |
106 | @samp{.set mips16} at the start of the assembly file. @samp{-no-mips16} | |
107 | turns off this option. | |
108 | ||
1f25f5d3 CD |
109 | @item -mips3d |
110 | @itemx -no-mips3d | |
111 | Generate code for the MIPS-3D Application Specific Extension. | |
112 | This tells the assembler to accept MIPS-3D instructions. | |
113 | @samp{-no-mips3d} turns off this option. | |
114 | ||
deec1734 CD |
115 | @item -mdmx |
116 | @itemx -no-mdmx | |
117 | Generate code for the MDMX Application Specific Extension. | |
118 | This tells the assembler to accept MDMX instructions. | |
119 | @samp{-no-mdmx} turns off this option. | |
120 | ||
ef2e4d86 CF |
121 | @item -mmt |
122 | @itemx -mno-mt | |
123 | Generate code for the MT Application Specific Extension. | |
124 | This tells the assembler to accept MT instructions. | |
125 | @samp{-mno-mt} turns off this option. | |
126 | ||
6b76fefe | 127 | @item -mfix7000 |
9ee72ff1 | 128 | @itemx -mno-fix7000 |
6b76fefe CM |
129 | Cause nops to be inserted if the read of the destination register |
130 | of an mfhi or mflo instruction occurs in the following two instructions. | |
131 | ||
d766e8ec RS |
132 | @item -mfix-vr4120 |
133 | @itemx -no-mfix-vr4120 | |
134 | Insert nops to work around certain VR4120 errata. This option is | |
135 | intended to be used on GCC-generated code: it is not designed to catch | |
136 | all problems in hand-written assembler code. | |
60b63b72 | 137 | |
11db99f8 RS |
138 | @item -mfix-vr4130 |
139 | @itemx -no-mfix-vr4130 | |
140 | Insert nops to work around the VR4130 @samp{mflo}/@samp{mfhi} errata. | |
141 | ||
252b5132 RH |
142 | @item -m4010 |
143 | @itemx -no-m4010 | |
144 | Generate code for the LSI @sc{r4010} chip. This tells the assembler to | |
145 | accept the @sc{r4010} specific instructions (@samp{addciu}, @samp{ffc}, | |
146 | etc.), and to not schedule @samp{nop} instructions around accesses to | |
147 | the @samp{HI} and @samp{LO} registers. @samp{-no-m4010} turns off this | |
148 | option. | |
149 | ||
150 | @item -m4650 | |
151 | @itemx -no-m4650 | |
152 | Generate code for the MIPS @sc{r4650} chip. This tells the assembler to accept | |
153 | the @samp{mad} and @samp{madu} instruction, and to not schedule @samp{nop} | |
154 | instructions around accesses to the @samp{HI} and @samp{LO} registers. | |
155 | @samp{-no-m4650} turns off this option. | |
156 | ||
157 | @itemx -m3900 | |
158 | @itemx -no-m3900 | |
159 | @itemx -m4100 | |
160 | @itemx -no-m4100 | |
161 | For each option @samp{-m@var{nnnn}}, generate code for the MIPS | |
162 | @sc{r@var{nnnn}} chip. This tells the assembler to accept instructions | |
163 | specific to that chip, and to schedule for that chip's hazards. | |
164 | ||
ec68c924 | 165 | @item -march=@var{cpu} |
252b5132 RH |
166 | Generate code for a particular MIPS cpu. It is exactly equivalent to |
167 | @samp{-m@var{cpu}}, except that there are more value of @var{cpu} | |
168 | understood. Valid @var{cpu} value are: | |
169 | ||
170 | @quotation | |
171 | 2000, | |
172 | 3000, | |
173 | 3900, | |
174 | 4000, | |
175 | 4010, | |
176 | 4100, | |
177 | 4111, | |
60b63b72 RS |
178 | vr4120, |
179 | vr4130, | |
180 | vr4181, | |
252b5132 RH |
181 | 4300, |
182 | 4400, | |
183 | 4600, | |
184 | 4650, | |
185 | 5000, | |
b946ec34 NC |
186 | rm5200, |
187 | rm5230, | |
188 | rm5231, | |
189 | rm5261, | |
190 | rm5721, | |
60b63b72 RS |
191 | vr5400, |
192 | vr5500, | |
252b5132 | 193 | 6000, |
b946ec34 | 194 | rm7000, |
252b5132 | 195 | 8000, |
963ac363 | 196 | rm9000, |
e7af610e | 197 | 10000, |
18ae5d72 | 198 | 12000, |
c6c98b38 NC |
199 | mips32-4k, |
200 | sb1 | |
252b5132 RH |
201 | @end quotation |
202 | ||
ec68c924 EC |
203 | @item -mtune=@var{cpu} |
204 | Schedule and tune for a particular MIPS cpu. Valid @var{cpu} values are | |
205 | identical to @samp{-march=@var{cpu}}. | |
206 | ||
316f5878 RS |
207 | @item -mabi=@var{abi} |
208 | Record which ABI the source code uses. The recognized arguments | |
209 | are: @samp{32}, @samp{n32}, @samp{o64}, @samp{64} and @samp{eabi}. | |
252b5132 | 210 | |
aed1a261 RS |
211 | @item -msym32 |
212 | @itemx -mno-sym32 | |
213 | @cindex -msym32 | |
214 | @cindex -mno-sym32 | |
215 | Equivalent to adding @code{.set sym32} or @code{.set nosym32} to | |
216 | the beginning of the assembler input. @xref{MIPS symbol sizes}. | |
217 | ||
252b5132 RH |
218 | @cindex @code{-nocpp} ignored (MIPS) |
219 | @item -nocpp | |
220 | This option is ignored. It is accepted for command-line compatibility with | |
221 | other assemblers, which use it to turn off C style preprocessing. With | |
222 | @sc{gnu} @code{@value{AS}}, there is no need for @samp{-nocpp}, because the | |
223 | @sc{gnu} assembler itself never runs the C preprocessor. | |
224 | ||
119d663a NC |
225 | @item --construct-floats |
226 | @itemx --no-construct-floats | |
227 | @cindex --construct-floats | |
228 | @cindex --no-construct-floats | |
229 | The @code{--no-construct-floats} option disables the construction of | |
230 | double width floating point constants by loading the two halves of the | |
231 | value into the two single width floating point registers that make up | |
232 | the double width register. This feature is useful if the processor | |
233 | support the FR bit in its status register, and this bit is known (by | |
234 | the programmer) to be set. This bit prevents the aliasing of the double | |
235 | width register by the single width registers. | |
236 | ||
63bf5651 | 237 | By default @code{--construct-floats} is selected, allowing construction |
119d663a NC |
238 | of these floating point constants. |
239 | ||
252b5132 RH |
240 | @item --trap |
241 | @itemx --no-break | |
242 | @c FIXME! (1) reflect these options (next item too) in option summaries; | |
243 | @c (2) stop teasing, say _which_ instructions expanded _how_. | |
244 | @code{@value{AS}} automatically macro expands certain division and | |
245 | multiplication instructions to check for overflow and division by zero. This | |
246 | option causes @code{@value{AS}} to generate code to take a trap exception | |
247 | rather than a break exception when an error is detected. The trap instructions | |
248 | are only supported at Instruction Set Architecture level 2 and higher. | |
249 | ||
250 | @item --break | |
251 | @itemx --no-trap | |
252 | Generate code to take a break exception rather than a trap exception when an | |
253 | error is detected. This is the default. | |
63486801 | 254 | |
dcd410fe RO |
255 | @item -mpdr |
256 | @itemx -mno-pdr | |
257 | Control generation of @code{.pdr} sections. Off by default on IRIX, on | |
258 | elsewhere. | |
aa6975fb ILT |
259 | |
260 | @item -mshared | |
261 | @itemx -mno-shared | |
262 | When generating code using the Unix calling conventions (selected by | |
263 | @samp{-KPIC} or @samp{-mcall_shared}), gas will normally generate code | |
264 | which can go into a shared library. The @samp{-mno-shared} option | |
265 | tells gas to generate code which uses the calling convention, but can | |
266 | not go into a shared library. The resulting code is slightly more | |
267 | efficient. This option only affects the handling of the | |
268 | @samp{.cpload} and @samp{.cpsetup} pseudo-ops. | |
252b5132 RH |
269 | @end table |
270 | ||
271 | @node MIPS Object | |
272 | @section MIPS ECOFF object code | |
273 | ||
274 | @cindex ECOFF sections | |
275 | @cindex MIPS ECOFF sections | |
276 | Assembling for a @sc{mips} @sc{ecoff} target supports some additional sections | |
277 | besides the usual @code{.text}, @code{.data} and @code{.bss}. The | |
278 | additional sections are @code{.rdata}, used for read-only data, | |
279 | @code{.sdata}, used for small data, and @code{.sbss}, used for small | |
280 | common objects. | |
281 | ||
282 | @cindex small objects, MIPS ECOFF | |
283 | @cindex @code{gp} register, MIPS | |
284 | When assembling for @sc{ecoff}, the assembler uses the @code{$gp} (@code{$28}) | |
285 | register to form the address of a ``small object''. Any object in the | |
286 | @code{.sdata} or @code{.sbss} sections is considered ``small'' in this sense. | |
287 | For external objects, or for objects in the @code{.bss} section, you can use | |
288 | the @code{@value{GCC}} @samp{-G} option to control the size of objects addressed via | |
289 | @code{$gp}; the default value is 8, meaning that a reference to any object | |
290 | eight bytes or smaller uses @code{$gp}. Passing @samp{-G 0} to | |
291 | @code{@value{AS}} prevents it from using the @code{$gp} register on the basis | |
292 | of object size (but the assembler uses @code{$gp} for objects in @code{.sdata} | |
293 | or @code{sbss} in any case). The size of an object in the @code{.bss} section | |
294 | is set by the @code{.comm} or @code{.lcomm} directive that defines it. The | |
295 | size of an external object may be set with the @code{.extern} directive. For | |
296 | example, @samp{.extern sym,4} declares that the object at @code{sym} is 4 bytes | |
297 | in length, whie leaving @code{sym} otherwise undefined. | |
298 | ||
299 | Using small @sc{ecoff} objects requires linker support, and assumes that the | |
300 | @code{$gp} register is correctly initialized (normally done automatically by | |
301 | the startup code). @sc{mips} @sc{ecoff} assembly code must not modify the | |
302 | @code{$gp} register. | |
303 | ||
304 | @node MIPS Stabs | |
305 | @section Directives for debugging information | |
306 | ||
307 | @cindex MIPS debugging directives | |
308 | @sc{mips} @sc{ecoff} @code{@value{AS}} supports several directives used for | |
309 | generating debugging information which are not support by traditional @sc{mips} | |
310 | assemblers. These are @code{.def}, @code{.endef}, @code{.dim}, @code{.file}, | |
311 | @code{.scl}, @code{.size}, @code{.tag}, @code{.type}, @code{.val}, | |
312 | @code{.stabd}, @code{.stabn}, and @code{.stabs}. The debugging information | |
313 | generated by the three @code{.stab} directives can only be read by @sc{gdb}, | |
314 | not by traditional @sc{mips} debuggers (this enhancement is required to fully | |
315 | support C++ debugging). These directives are primarily used by compilers, not | |
316 | assembly language programmers! | |
317 | ||
aed1a261 RS |
318 | @node MIPS symbol sizes |
319 | @section Directives to override the size of symbols | |
320 | ||
321 | @cindex @code{.set sym32} | |
322 | @cindex @code{.set nosym32} | |
323 | The n64 ABI allows symbols to have any 64-bit value. Although this | |
324 | provides a great deal of flexibility, it means that some macros have | |
325 | much longer expansions than their 32-bit counterparts. For example, | |
326 | the non-PIC expansion of @samp{dla $4,sym} is usually: | |
327 | ||
328 | @smallexample | |
329 | lui $4,%highest(sym) | |
330 | lui $1,%hi(sym) | |
331 | daddiu $4,$4,%higher(sym) | |
332 | daddiu $1,$1,%lo(sym) | |
333 | dsll32 $4,$4,0 | |
334 | daddu $4,$4,$1 | |
335 | @end smallexample | |
336 | ||
337 | whereas the 32-bit expansion is simply: | |
338 | ||
339 | @smallexample | |
340 | lui $4,%hi(sym) | |
341 | daddiu $4,$4,%lo(sym) | |
342 | @end smallexample | |
343 | ||
344 | n64 code is sometimes constructed in such a way that all symbolic | |
345 | constants are known to have 32-bit values, and in such cases, it's | |
346 | preferable to use the 32-bit expansion instead of the 64-bit | |
347 | expansion. | |
348 | ||
349 | You can use the @code{.set sym32} directive to tell the assembler | |
350 | that, from this point on, all expressions of the form | |
351 | @samp{@var{symbol}} or @samp{@var{symbol} + @var{offset}} | |
352 | have 32-bit values. For example: | |
353 | ||
354 | @smallexample | |
355 | .set sym32 | |
356 | dla $4,sym | |
357 | lw $4,sym+16 | |
358 | sw $4,sym+0x8000($4) | |
359 | @end smallexample | |
360 | ||
361 | will cause the assembler to treat @samp{sym}, @code{sym+16} and | |
362 | @code{sym+0x8000} as 32-bit values. The handling of non-symbolic | |
363 | addresses is not affected. | |
364 | ||
365 | The directive @code{.set nosym32} ends a @code{.set sym32} block and | |
366 | reverts to the normal behavior. It is also possible to change the | |
367 | symbol size using the command-line options @option{-msym32} and | |
368 | @option{-mno-sym32}. | |
369 | ||
370 | These options and directives are always accepted, but at present, | |
371 | they have no effect for anything other than n64. | |
372 | ||
252b5132 RH |
373 | @node MIPS ISA |
374 | @section Directives to override the ISA level | |
375 | ||
376 | @cindex MIPS ISA override | |
377 | @kindex @code{.set mips@var{n}} | |
378 | @sc{gnu} @code{@value{AS}} supports an additional directive to change | |
379 | the @sc{mips} Instruction Set Architecture level on the fly: @code{.set | |
5f74bc13 CD |
380 | mips@var{n}}. @var{n} should be a number from 0 to 5, or 32, 32r2, 64 |
381 | or 64r2. | |
071742cf | 382 | The values other than 0 make the assembler accept instructions |
584da044 NC |
383 | for the corresponding @sc{isa} level, from that point on in the |
384 | assembly. @code{.set mips@var{n}} affects not only which instructions | |
385 | are permitted, but also how certain macros are expanded. @code{.set | |
386 | mips0} restores the @sc{isa} level to its original level: either the | |
387 | level you selected with command line options, or the default for your | |
388 | configuration. You can use this feature to permit specific @sc{r4000} | |
389 | instructions while assembling in 32 bit mode. Use this directive with | |
ec68c924 | 390 | care! |
252b5132 RH |
391 | |
392 | The directive @samp{.set mips16} puts the assembler into MIPS 16 mode, | |
393 | in which it will assemble instructions for the MIPS 16 processor. Use | |
394 | @samp{.set nomips16} to return to normal 32 bit mode. | |
395 | ||
ec68c924 | 396 | Traditional @sc{mips} assemblers do not support this directive. |
252b5132 RH |
397 | |
398 | @node MIPS autoextend | |
399 | @section Directives for extending MIPS 16 bit instructions | |
400 | ||
401 | @kindex @code{.set autoextend} | |
402 | @kindex @code{.set noautoextend} | |
403 | By default, MIPS 16 instructions are automatically extended to 32 bits | |
404 | when necessary. The directive @samp{.set noautoextend} will turn this | |
405 | off. When @samp{.set noautoextend} is in effect, any 32 bit instruction | |
406 | must be explicitly extended with the @samp{.e} modifier (e.g., | |
407 | @samp{li.e $4,1000}). The directive @samp{.set autoextend} may be used | |
408 | to once again automatically extend instructions when necessary. | |
409 | ||
410 | This directive is only meaningful when in MIPS 16 mode. Traditional | |
411 | @sc{mips} assemblers do not support this directive. | |
412 | ||
413 | @node MIPS insn | |
414 | @section Directive to mark data as an instruction | |
415 | ||
416 | @kindex @code{.insn} | |
417 | The @code{.insn} directive tells @code{@value{AS}} that the following | |
418 | data is actually instructions. This makes a difference in MIPS 16 mode: | |
419 | when loading the address of a label which precedes instructions, | |
420 | @code{@value{AS}} automatically adds 1 to the value, so that jumping to | |
421 | the loaded address will do the right thing. | |
422 | ||
423 | @node MIPS option stack | |
424 | @section Directives to save and restore options | |
425 | ||
426 | @cindex MIPS option stack | |
427 | @kindex @code{.set push} | |
428 | @kindex @code{.set pop} | |
429 | The directives @code{.set push} and @code{.set pop} may be used to save | |
430 | and restore the current settings for all the options which are | |
431 | controlled by @code{.set}. The @code{.set push} directive saves the | |
432 | current settings on a stack. The @code{.set pop} directive pops the | |
433 | stack and restores the settings. | |
434 | ||
435 | These directives can be useful inside an macro which must change an | |
436 | option such as the ISA level or instruction reordering but does not want | |
437 | to change the state of the code which invoked the macro. | |
438 | ||
439 | Traditional @sc{mips} assemblers do not support these directives. | |
1f25f5d3 CD |
440 | |
441 | @node MIPS ASE instruction generation overrides | |
442 | @section Directives to control generation of MIPS ASE instructions | |
443 | ||
444 | @cindex MIPS MIPS-3D instruction generation override | |
445 | @kindex @code{.set mips3d} | |
446 | @kindex @code{.set nomips3d} | |
447 | The directive @code{.set mips3d} makes the assembler accept instructions | |
448 | from the MIPS-3D Application Specific Extension from that point on | |
449 | in the assembly. The @code{.set nomips3d} directive prevents MIPS-3D | |
450 | instructions from being accepted. | |
451 | ||
deec1734 CD |
452 | @cindex MIPS MDMX instruction generation override |
453 | @kindex @code{.set mdmx} | |
454 | @kindex @code{.set nomdmx} | |
455 | The directive @code{.set mdmx} makes the assembler accept instructions | |
456 | from the MDMX Application Specific Extension from that point on | |
457 | in the assembly. The @code{.set nomdmx} directive prevents MDMX | |
458 | instructions from being accepted. | |
459 | ||
ef2e4d86 CF |
460 | @cindex MIPS MT instruction generation override |
461 | @kindex @code{.set mt} | |
462 | @kindex @code{.set nomt} | |
463 | The directive @code{.set mt} makes the assembler accept instructions | |
464 | from the MT Application Specific Extension from that point on | |
465 | in the assembly. The @code{.set nomt} directive prevents MT | |
466 | instructions from being accepted. | |
467 | ||
1f25f5d3 | 468 | Traditional @sc{mips} assemblers do not support these directives. |