1 @c Copyright 1996, 1997, 1998, 1999, 2000, 2001
2 @c Free Software Foundation, Inc.
3 @c This is part of the GAS manual.
4 @c For copying conditions, see the file as.texinfo.
9 @chapter ARM Dependent Features
13 @node Machine Dependencies
14 @chapter ARM Dependent Features
20 * ARM Options:: Options
22 * ARM Floating Point:: Floating Point
23 * ARM Directives:: ARM Machine Directives
24 * ARM Opcodes:: Opcodes
29 @cindex ARM options (none)
30 @cindex options for ARM (none)
34 @cindex @code{-marm} command line option, ARM
35 @item -marm@code{[2|250|3|6|60|600|610|620|7|7m|7d|7dm|7di|7dmi|70|700|700i|710|710c|7100|7500|7500fe|7tdmi|8|810|9|9tdmi|920|strongarm|strongarm110|strongarm1100]}
38 This option specifies the target processor. The assembler will issue an
39 error message if an attempt is made to assemble an instruction which
40 will not execute on the target processor.
42 The option @code{-marm9e} specifies that the target processor is the
43 Cirrus ARM processor with the Maverick DSP co-processor.
45 @cindex @code{-marmv} command line option, ARM
46 @item -marmv@code{[2|2a|3|3m|4|4t|5|5t|5te]}
47 This option specifies the target architecture. The assembler will issue
48 an error message if an attempt is made to assemble an instruction which
49 will not execute on the target architecture.
50 The option @code{-marmv5te} specifies that v5t architecture should be
51 used with the El Segundo extensions enabled.
53 @cindex @code{-mthumb} command line option, ARM
55 This option specifies that only Thumb instructions should be assembled.
57 @cindex @code{-mall} command line option, ARM
59 This option specifies that any Arm or Thumb instruction should be assembled.
61 @cindex @code{-mfpa} command line option, ARM
62 @item -mfpa @code{[10|11]}
63 This option specifies the floating point architecture in use on the
66 @cindex @code{-mfpe-old} command line option, ARM
68 Do not allow the assembly of floating point multiple instructions.
70 @cindex @code{-mno-fpu} command line option, ARM
72 Do not allow the assembly of any floating point instructions.
74 @cindex @code{-mthumb-interwork} command line option, ARM
75 @item -mthumb-interwork
76 This option specifies that the output generated by the assembler should
77 be marked as supporting interworking.
79 @cindex @code{-mapcs} command line option, ARM
80 @item -mapcs @code{[26|32]}
81 This option specifies that the output generated by the assembler should
82 be marked as supporting the indicated version of the Arm Procedure.
85 @cindex @code{-matpcs} command line option, ARM
87 This option specifies that the output generated by the assembler should
88 be marked as supporting the Arm/Thumb Procedure Calling Standard. If
89 enabled this option will cause the assembler to create an empty
90 debugging section in the object file called .arm.atpcs. Debuggers can
91 use this to determine the ABI being used by.
93 @cindex @code{-mapcs-float} command line option, ARM
95 This indicates the the floating point variant of the APCS should be
96 used. In this variant floating point arguments are passed in FP
97 registers rather than integer registers.
99 @cindex @code{-mapcs-reentrant} command line option, ARM
100 @item -mapcs-reentrant
101 This indicates that the reentrant variant of the APCS should be used.
102 This variant supports position independent code.
104 @cindex @code{-EB} command line option, ARM
106 This option specifies that the output generated by the assembler should
107 be marked as being encoded for a big-endian processor.
109 @cindex @code{-EL} command line option, ARM
111 This option specifies that the output generated by the assembler should
112 be marked as being encoded for a little-endian processor.
114 @cindex @code{-k} command line option, ARM
115 @cindex PIC code generation for ARM
117 This option specifies that the output of the assembler should be marked
118 as position-independent code (PIC).
120 @cindex @code{-moabi} command line option, ARM
122 This indicates that the code should be assembled using the old ARM ELF
123 conventions, based on a beta release release of the ARM-ELF
124 specifications, rather than the default conventions which are based on
125 the final release of the ARM-ELF specifications.
133 * ARM-Chars:: Special Characters
134 * ARM-Regs:: Register Names
138 @subsection Special Characters
140 @cindex line comment character, ARM
141 @cindex ARM line comment character
142 The presence of a @samp{@@} on a line indicates the start of a comment
143 that extends to the end of the current line. If a @samp{#} appears as
144 the first character of a line, the whole line is treated as a comment.
146 @cindex line separator, ARM
147 @cindex statement separator, ARM
148 @cindex ARM line separator
149 The @samp{;} character can be used instead of a newline to separate
152 @cindex immediate character, ARM
153 @cindex ARM immediate character
154 Either @samp{#} or @samp{$} can be used to indicate immediate operands.
156 @cindex identifiers, ARM
157 @cindex ARM identifiers
158 *TODO* Explain about /data modifier on symbols.
161 @subsection Register Names
163 @cindex ARM register names
164 @cindex register names, ARM
165 *TODO* Explain about ARM register naming, and the predefined names.
167 @node ARM Floating Point
168 @section Floating Point
170 @cindex floating point, ARM (@sc{ieee})
171 @cindex ARM floating point (@sc{ieee})
172 The ARM family uses @sc{ieee} floating-point numbers.
177 @section ARM Machine Directives
179 @cindex machine directives, ARM
180 @cindex ARM machine directives
183 @cindex @code{align} directive, ARM
184 @item .align @var{expression} [, @var{expression}]
185 This is the generic @var{.align} directive. For the ARM however if the
186 first argument is zero (ie no alignment is needed) the assembler will
187 behave as if the argument had been 2 (ie pad to the next four byte
188 boundary). This is for compatability with ARM's own assembler.
190 @cindex @code{req} directive, ARM
191 @item @var{name} .req @var{register name}
192 This creates an alias for @var{register name} called @var{name}. For
199 @cindex @code{code} directive, ARM
200 @item .code @code{[16|32]}
201 This directive selects the instruction set being generated. The value 16
202 selects Thumb, with the value 32 selecting ARM.
204 @cindex @code{thumb} directive, ARM
206 This performs the same action as @var{.code 16}.
208 @cindex @code{arm} directive, ARM
210 This performs the same action as @var{.code 32}.
212 @cindex @code{force_thumb} directive, ARM
214 This directive forces the selection of Thumb instructions, even if the
215 target processor does not support those instructions
217 @cindex @code{thumb_func} directive, ARM
219 This directive specifies that the following symbol is the name of a
220 Thumb encoded function. This information is necessary in order to allow
221 the assembler and linker to generate correct code for interworking
222 between Arm and Thumb instructions and should be used even if
223 interworking is not going to be performed. The presence of this
224 directive also implies @code{.thumb}
226 @cindex @code{thumb_set} directive, ARM
228 This performs the equivalent of a @code{.set} directive in that it
229 creates a symbol which is an alias for another symbol (possibly not yet
230 defined). This directive also has the added property in that it marks
231 the aliased symbol as being a thumb function entry point, in the same
232 way that the @code{.thumb_func} directive does.
234 @cindex @code{.ltorg} directive, ARM
236 This directive causes the current contents of the literal pool to be
237 dumped into the current section (which is assumed to be the .text
238 section) at the current location (aligned to a word boundary).
240 @cindex @code{.pool} directive, ARM
242 This is a synonym for .ltorg.
250 @cindex opcodes for ARM
251 @code{@value{AS}} implements all the standard ARM opcodes. It also
252 implements several pseudo opcodes, including several synthetic load
257 @cindex @code{NOP} pseudo op, ARM
263 This pseudo op will always evaluate to a legal ARM instruction that does
264 nothing. Currently it will evaluate to MOV r0, r0.
266 @cindex @code{LDR reg,=<label>} pseudo op, ARM
269 ldr <register> , = <expression>
272 If expression evaluates to a numeric constant then a MOV or MVN
273 instruction will be used in place of the LDR instruction, if the
274 constant can be generated by either of these instructions. Otherwise
275 the constant will be placed into the nearest literal pool (if it not
276 already there) and a PC relative LDR instruction will be generated.
278 @cindex @code{ADR reg,<label>} pseudo op, ARM
281 adr <register> <label>
284 This instruction will load the address of @var{label} into the indicated
285 register. The instruction will evaluate to a PC relative ADD or SUB
286 instruction depending upon where the label is located. If the label is
287 out of range, or if it is not defined in the same file (and section) as
288 the ADR instruction, then an error will be generated. This instruction
289 will not make use of the literal pool.
291 @cindex @code{ADRL reg,<label>} pseudo op, ARM
294 adrl <register> <label>
297 This instruction will load the address of @var{label} into the indicated
298 register. The instruction will evaluate to one or two PC relative ADD
299 or SUB instructions depending upon where the label is located. If a
300 second instruction is not needed a NOP instruction will be generated in
301 its place, so that this instruction is always 8 bytes long.
303 If the label is out of range, or if it is not defined in the same file
304 (and section) as the ADRL instruction, then an error will be generated.
305 This instruction will not make use of the literal pool.
309 For information on the ARM or Thumb instruction sets, see @cite{ARM
310 Software Development Toolkit Reference Manual}, Advanced RISC Machines