c-arm.texi revision 77298
1@c Copyright (C) 1996, 1998, 1999, 2000 Free Software Foundation, Inc.
2@c This is part of the GAS manual.
3@c For copying conditions, see the file as.texinfo.
4
5@ifset GENERIC
6@page
7@node ARM-Dependent
8@chapter ARM Dependent Features
9@end ifset
10
11@ifclear GENERIC
12@node Machine Dependencies
13@chapter ARM Dependent Features
14@end ifclear
15
16@cindex ARM support
17@cindex Thumb support
18@menu
19* ARM Options::              Options
20* ARM Syntax::               Syntax
21* ARM Floating Point::       Floating Point
22* ARM Directives::           ARM Machine Directives
23* ARM Opcodes::              Opcodes
24@end menu
25
26@node ARM Options
27@section Options
28@cindex ARM options (none)
29@cindex options for ARM (none)
30
31@table @code
32
33@cindex @code{-marm} command line option, ARM
34@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]}
35@itemx -mxscale
36This option specifies the target processor.  The assembler will issue an
37error message if an attempt is made to assemble an instruction which
38will not execute on the target processor.
39
40@cindex @code{-marmv} command line option, ARM
41@item -marmv@code{[2|2a|3|3m|4|4t|5|5t|5te]}
42This option specifies the target architecture.  The assembler will issue
43an error message if an attempt is made to assemble an instruction which
44will not execute on the target architecture.
45The option @code{-marmv5te} specifies that v5t architecture should be
46used with the El Segundo extensions enabled.
47
48@cindex @code{-mthumb} command line option, ARM
49@item -mthumb
50This option specifies that only Thumb instructions should be assembled.
51
52@cindex @code{-mall} command line option, ARM
53@item -mall
54This option specifies that any Arm or Thumb instruction should be assembled.
55
56@cindex @code{-mfpa} command line option, ARM
57@item -mfpa @code{[10|11]}
58This option specifies the floating point architecture in use on the
59target processor. 
60
61@cindex @code{-mfpe-old} command line option, ARM
62@item -mfpe-old
63Do not allow the assembly of floating point multiple instructions.
64
65@cindex @code{-mno-fpu} command line option, ARM
66@item -mno-fpu
67Do not allow the assembly of any floating point instructions.
68
69@cindex @code{-mthumb-interwork} command line option, ARM
70@item -mthumb-interwork
71This option specifies that the output generated by the assembler should
72be marked as supporting interworking.
73
74@cindex @code{-mapcs} command line option, ARM
75@item -mapcs @code{[26|32]}
76This option specifies that the output generated by the assembler should
77be marked as supporting the indicated version of the Arm Procedure.
78Calling Standard.
79
80@cindex @code{-matpcs} command line option, ARM
81@item -matpcs
82This option specifies that the output generated by the assembler should 
83be marked as supporting the Arm/Thumb Procedure Calling Standard.  If
84enabled this option will cause the assembler to create an empty
85debugging section in the object file called .arm.atpcs.  Debuggers can
86use this to determine the ABI being used by.
87
88@cindex @code{-mapcs-float} command line option, ARM
89@item -mapcs-float
90This indicates the the floating point variant of the APCS should be
91used.  In this variant floating point arguments are passed in FP
92registers rather than integer registers.
93
94@cindex @code{-mapcs-reentrant} command line option, ARM
95@item -mapcs-reentrant
96This indicates that the reentrant variant of the APCS should be used.
97This variant supports position independent code.
98
99@cindex @code{-EB} command line option, ARM
100@item -EB
101This option specifies that the output generated by the assembler should
102be marked as being encoded for a big-endian processor.
103
104@cindex @code{-EL} command line option, ARM
105@item -EL
106This option specifies that the output generated by the assembler should
107be marked as being encoded for a little-endian processor.
108
109@cindex @code{-k} command line option, ARM
110@cindex PIC code generation for ARM
111@item -k
112This option specifies that the output of the assembler should be marked
113as position-independent code (PIC).
114
115@cindex @code{-moabi} command line option, ARM
116@item -moabi
117This indicates that the code should be assembled using the old ARM ELF
118conventions, based on a beta release release of the ARM-ELF
119specifications, rather than the default conventions which are based on
120the final release of the ARM-ELF specifications.
121
122@end table
123
124
125@node ARM Syntax
126@section Syntax
127@menu
128* ARM-Chars::                Special Characters
129* ARM-Regs::                 Register Names
130@end menu
131
132@node ARM-Chars
133@subsection Special Characters
134
135@cindex line comment character, ARM
136@cindex ARM line comment character
137The presence of a @samp{@@} on a line indicates the start of a comment
138that extends to the end of the current line.  If a @samp{#} appears as
139the first character of a line, the whole line is treated as a comment.
140
141@cindex line separator, ARM
142@cindex statement separator, ARM
143@cindex ARM line separator
144The @samp{;} character can be used instead of a newline to separate
145statements.
146
147@cindex immediate character, ARM
148@cindex ARM immediate character
149Either @samp{#} or @samp{$} can be used to indicate immediate operands.
150
151@cindex identifiers, ARM
152@cindex ARM identifiers
153*TODO* Explain about /data modifier on symbols.
154
155@node ARM-Regs
156@subsection Register Names
157
158@cindex ARM register names
159@cindex register names, ARM
160*TODO* Explain about ARM register naming, and the predefined names.
161
162@node ARM Floating Point
163@section Floating Point
164
165@cindex floating point, ARM (@sc{ieee})
166@cindex ARM floating point (@sc{ieee})
167The ARM family uses @sc{ieee} floating-point numbers.
168
169
170
171@node ARM Directives
172@section ARM Machine Directives
173
174@cindex machine directives, ARM
175@cindex ARM machine directives
176@table @code
177
178@cindex @code{align} directive, ARM
179@item .align @var{expression} [, @var{expression}]
180This is the generic @var{.align} directive.  For the ARM however if the
181first argument is zero (ie no alignment is needed) the assembler will
182behave as if the argument had been 2 (ie pad to the next four byte
183boundary).  This is for compatability with ARM's own assembler.
184
185@cindex @code{req} directive, ARM
186@item @var{name} .req @var{register name}
187This creates an alias for @var{register name} called @var{name}.  For
188example:
189
190@smallexample
191        foo .req r0
192@end smallexample
193
194@cindex @code{code} directive, ARM
195@item .code @code{[16|32]}
196This directive selects the instruction set being generated. The value 16
197selects Thumb, with the value 32 selecting ARM.
198
199@cindex @code{thumb} directive, ARM
200@item .thumb
201This performs the same action as @var{.code 16}.
202
203@cindex @code{arm} directive, ARM
204@item .arm
205This performs the same action as @var{.code 32}.
206
207@cindex @code{force_thumb} directive, ARM
208@item .force_thumb
209This directive forces the selection of Thumb instructions, even if the
210target processor does not support those instructions
211
212@cindex @code{thumb_func} directive, ARM
213@item .thumb_func
214This directive specifies that the following symbol is the name of a
215Thumb encoded function.  This information is necessary in order to allow
216the assembler and linker to generate correct code for interworking
217between Arm and Thumb instructions and should be used even if
218interworking is not going to be performed.  The presence of this
219directive also implies @code{.thumb}
220
221@cindex @code{thumb_set} directive, ARM
222@item .thumb_set
223This performs the equivalent of a @code{.set} directive in that it
224creates a symbol which is an alias for another symbol (possibly not yet
225defined).  This directive also has the added property in that it marks
226the aliased symbol as being a thumb function entry point, in the same
227way that the @code{.thumb_func} directive does.
228
229@cindex @code{.ltorg} directive, ARM
230@item .ltorg
231This directive causes the current contents of the literal pool to be
232dumped into the current section (which is assumed to be the .text
233section) at the current location (aligned to a word boundary).
234
235@cindex @code{.pool} directive, ARM
236@item .pool
237This is a synonym for .ltorg.
238
239@end table
240
241@node ARM Opcodes
242@section Opcodes
243
244@cindex ARM opcodes
245@cindex opcodes for ARM
246@code{@value{AS}} implements all the standard ARM opcodes.  It also
247implements several pseudo opcodes, including several synthetic load
248instructions. 
249
250@table @code
251
252@cindex @code{NOP} pseudo op, ARM
253@item NOP
254@smallexample
255  nop
256@end smallexample
257
258This pseudo op will always evaluate to a legal ARM instruction that does
259nothing.  Currently it will evaluate to MOV r0, r0.
260
261@cindex @code{LDR reg,=<label>} pseudo op, ARM
262@item LDR 
263@smallexample
264  ldr <register> , = <expression>
265@end smallexample
266
267If expression evaluates to a numeric constant then a MOV or MVN
268instruction will be used in place of the LDR instruction, if the
269constant can be generated by either of these instructions.  Otherwise
270the constant will be placed into the nearest literal pool (if it not
271already there) and a PC relative LDR instruction will be generated.
272
273@cindex @code{ADR reg,<label>} pseudo op, ARM
274@item ADR
275@smallexample
276  adr <register> <label>
277@end smallexample
278
279This instruction will load the address of @var{label} into the indicated
280register.  The instruction will evaluate to a PC relative ADD or SUB
281instruction depending upon where the label is located.  If the label is
282out of range, or if it is not defined in the same file (and section) as
283the ADR instruction, then an error will be generated.  This instruction
284will not make use of the literal pool.
285
286@cindex @code{ADRL reg,<label>} pseudo op, ARM
287@item ADRL 
288@smallexample
289  adrl <register> <label>
290@end smallexample
291
292This instruction will load the address of @var{label} into the indicated
293register.  The instruction will evaluate to one or two PC relative ADD
294or SUB instructions depending upon where the label is located.  If a
295second instruction is not needed a NOP instruction will be generated in
296its place, so that this instruction is always 8 bytes long.
297
298If the label is out of range, or if it is not defined in the same file
299(and section) as the ADRL instruction, then an error will be generated.
300This instruction will not make use of the literal pool.
301
302@end table
303
304For information on the ARM or Thumb instruction sets, see @cite{ARM
305Software Development Toolkit Reference Manual}, Advanced RISC Machines
306Ltd.
307
308