as.info-2 [plain text]

This is ../.././gas/doc/as.info, produced by makeinfo version 4.7 from
../.././gas/doc/as.texinfo.

START-INFO-DIR-ENTRY
* As: (as). The GNU assembler.
* Gas: (as). The GNU assembler.
END-INFO-DIR-ENTRY

This file documents the GNU Assembler "as".

Permission is granted to copy, distribute and/or modify this document
under the terms of the GNU Free Documentation License, Version 1.1 or
any later version published by the Free Software Foundation; with no
Invariant Sections, with no Front-Cover Texts, and with no Back-Cover
Texts. A copy of the license is included in the section entitled "GNU
Free Documentation License".

File: as.info, Node: i386-Float, Next: i386-SIMD, Prev: i386-Jumps, Up: i386-Dependent

8.12.8 Floating Point
---------------------

All 80387 floating point types except packed BCD are supported. (BCD
support may be added without much difficulty). These data types are
16-, 32-, and 64- bit integers, and single (32-bit), double (64-bit),
and extended (80-bit) precision floating point. Each supported type
has an instruction mnemonic suffix and a constructor associated with
it. Instruction mnemonic suffixes specify the operand's data type.
Constructors build these data types into memory.

* Floating point constructors are `.float' or `.single', `.double',
and `.tfloat' for 32-, 64-, and 80-bit formats. These correspond
to instruction mnemonic suffixes `s', `l', and `t'. `t' stands for
80-bit (ten byte) real. The 80387 only supports this format via
the `fldt' (load 80-bit real to stack top) and `fstpt' (store
80-bit real and pop stack) instructions.

* Integer constructors are `.word', `.long' or `.int', and `.quad'
for the 16-, 32-, and 64-bit integer formats. The corresponding
instruction mnemonic suffixes are `s' (single), `l' (long), and
`q' (quad). As with the 80-bit real format, the 64-bit `q' format
is only present in the `fildq' (load quad integer to stack top)
and `fistpq' (store quad integer and pop stack) instructions.

Register to register operations should not use instruction mnemonic
suffixes. `fstl %st, %st(1)' will give a warning, and be assembled as
if you wrote `fst %st, %st(1)', since all register to register
operations use 80-bit floating point operands. (Contrast this with
`fstl %st, mem', which converts `%st' from 80-bit to 64-bit floating
point format, then stores the result in the 4 byte location `mem')

File: as.info, Node: i386-SIMD, Next: i386-16bit, Prev: i386-Float, Up: i386-Dependent

8.12.9 Intel's MMX and AMD's 3DNow! SIMD Operations
---------------------------------------------------

`as' supports Intel's MMX instruction set (SIMD instructions for
integer data), available on Intel's Pentium MMX processors and Pentium
II processors, AMD's K6 and K6-2 processors, Cyrix' M2 processor, and
probably others. It also supports AMD's 3DNow! instruction set (SIMD
instructions for 32-bit floating point data) available on AMD's K6-2
processor and possibly others in the future.

Currently, `as' does not support Intel's floating point SIMD, Katmai
(KNI).

The eight 64-bit MMX operands, also used by 3DNow!, are called
`%mm0', `%mm1', ... `%mm7'. They contain eight 8-bit integers, four
16-bit integers, two 32-bit integers, one 64-bit integer, or two 32-bit
floating point values. The MMX registers cannot be used at the same
time as the floating point stack.

See Intel and AMD documentation, keeping in mind that the operand
order in instructions is reversed from the Intel syntax.

File: as.info, Node: i386-16bit, Next: i386-Arch, Prev: i386-SIMD, Up: i386-Dependent

8.12.10 Writing 16-bit Code
---------------------------

While `as' normally writes only "pure" 32-bit i386 code or 64-bit
x86-64 code depending on the default configuration, it also supports
writing code to run in real mode or in 16-bit protected mode code
segments. To do this, put a `.code16' or `.code16gcc' directive before
the assembly language instructions to be run in 16-bit mode. You can
switch `as' back to writing normal 32-bit code with the `.code32'
directive.

`.code16gcc' provides experimental support for generating 16-bit
code from gcc, and differs from `.code16' in that `call', `ret',
`enter', `leave', `push', `pop', `pusha', `popa', `pushf', and `popf'
instructions default to 32-bit size. This is so that the stack pointer
is manipulated in the same way over function calls, allowing access to
function parameters at the same stack offsets as in 32-bit mode.
`.code16gcc' also automatically adds address size prefixes where
necessary to use the 32-bit addressing modes that gcc generates.

The code which `as' generates in 16-bit mode will not necessarily
run on a 16-bit pre-80386 processor. To write code that runs on such a
processor, you must refrain from using _any_ 32-bit constructs which
require `as' to output address or operand size prefixes.

Note that writing 16-bit code instructions by explicitly specifying a
prefix or an instruction mnemonic suffix within a 32-bit code section
generates different machine instructions than those generated for a
16-bit code segment. In a 32-bit code section, the following code
generates the machine opcode bytes `66 6a 04', which pushes the value
`4' onto the stack, decrementing `%esp' by 2.

pushw $4

The same code in a 16-bit code section would generate the machine
opcode bytes `6a 04' (ie. without the operand size prefix), which is
correct since the processor default operand size is assumed to be 16
bits in a 16-bit code section.

File: as.info, Node: i386-Bugs, Next: i386-Notes, Prev: i386-Arch, Up: i386-Dependent

8.12.11 AT&T Syntax bugs
------------------------

The UnixWare assembler, and probably other AT&T derived ix86 Unix
assemblers, generate floating point instructions with reversed source
and destination registers in certain cases. Unfortunately, gcc and
possibly many other programs use this reversed syntax, so we're stuck
with it.

For example

fsub %st,%st(3)
results in `%st(3)' being updated to `%st - %st(3)' rather than the
expected `%st(3) - %st'. This happens with all the non-commutative
arithmetic floating point operations with two register operands where
the source register is `%st' and the destination register is `%st(i)'.

File: as.info, Node: i386-Arch, Next: i386-Bugs, Prev: i386-16bit, Up: i386-Dependent

8.12.12 Specifying CPU Architecture
-----------------------------------

`as' may be told to assemble for a particular CPU (sub-)architecture
with the `.arch CPU_TYPE' directive. This directive enables a warning
when gas detects an instruction that is not supported on the CPU
specified. The choices for CPU_TYPE are:

`i8086' `i186' `i286' `i386'
`i486' `i586' `i686' `pentium'
`pentiumpro' `pentiumii' `pentiumiii' `pentium4'
`k6' `athlon'
`sledgehammer'
`.mmx' `.sse'
`.sse2'
`.3dnow'

Apart from the warning, there are only two other effects on `as'
operation; Firstly, if you specify a CPU other than `i486', then shift
by one instructions such as `sarl $1, %eax' will automatically use a
two byte opcode sequence. The larger three byte opcode sequence is
used on the 486 (and when no architecture is specified) because it
executes faster on the 486. Note that you can explicitly request the
two byte opcode by writing `sarl %eax'. Secondly, if you specify
`i8086', `i186', or `i286', _and_ `.code16' or `.code16gcc' then byte
offset conditional jumps will be promoted when necessary to a two
instruction sequence consisting of a conditional jump of the opposite
sense around an unconditional jump to the target.

Following the CPU architecture (but not a sub-architecture, which
are those starting with a dot), you may specify `jumps' or `nojumps' to
control automatic promotion of conditional jumps. `jumps' is the
default, and enables jump promotion; All external jumps will be of the
long variety, and file-local jumps will be promoted as necessary.
(*note i386-Jumps::) `nojumps' leaves external conditional jumps as
byte offset jumps, and warns about file-local conditional jumps that
`as' promotes. Unconditional jumps are treated as for `jumps'.

For example

.arch i8086,nojumps

File: as.info, Node: i386-Notes, Prev: i386-Bugs, Up: i386-Dependent

8.12.13 Notes
-------------

There is some trickery concerning the `mul' and `imul' instructions
that deserves mention. The 16-, 32-, 64- and 128-bit expanding
multiplies (base opcode `0xf6'; extension 4 for `mul' and 5 for `imul')
can be output only in the one operand form. Thus, `imul %ebx, %eax'
does _not_ select the expanding multiply; the expanding multiply would
clobber the `%edx' register, and this would confuse `gcc' output. Use
`imul %ebx' to get the 64-bit product in `%edx:%eax'.

We have added a two operand form of `imul' when the first operand is
an immediate mode expression and the second operand is a register.
This is just a shorthand, so that, multiplying `%eax' by 69, for
example, can be done with `imul $69, %eax' rather than `imul $69, %eax,
%eax'.

File: as.info, Node: i860-Dependent, Next: i960-Dependent, Prev: i386-Dependent, Up: Machine Dependencies

8.13 Intel i860 Dependent Features
==================================

* Menu:

* Notes-i860:: i860 Notes
* Options-i860:: i860 Command-line Options
* Directives-i860:: i860 Machine Directives
* Opcodes for i860:: i860 Opcodes

File: as.info, Node: Notes-i860, Next: Options-i860, Up: i860-Dependent

8.13.1 i860 Notes
-----------------

This is a fairly complete i860 assembler which is compatible with the
UNIX System V/860 Release 4 assembler. However, it does not currently
support SVR4 PIC (i.e., `@GOT, @GOTOFF, @PLT').

Like the SVR4/860 assembler, the output object format is ELF32.
Currently, this is the only supported object format. If there is
sufficient interest, other formats such as COFF may be implemented.

Both the Intel and AT&T/SVR4 syntaxes are supported, with the latter
being the default. One difference is that AT&T syntax requires the '%'
prefix on register names while Intel syntax does not. Another
difference is in the specification of relocatable expressions. The
Intel syntax is `ha%expression' whereas the SVR4 syntax is
`[expression]@ha' (and similarly for the "l" and "h" selectors).

File: as.info, Node: Options-i860, Next: Directives-i860, Prev: Notes-i860, Up: i860-Dependent

8.13.2 i860 Command-line Options
--------------------------------

8.13.2.1 SVR4 compatibility options
...................................

`-V'
Print assembler version.

`-Qy'
Ignored.

`-Qn'
Ignored.

8.13.2.2 Other options
......................

`-EL'
Select little endian output (this is the default).

`-EB'
Select big endian output. Note that the i860 always reads
instructions as little endian data, so this option only effects
data and not instructions.

`-mwarn-expand'
Emit a warning message if any pseudo-instruction expansions
occurred. For example, a `or' instruction with an immediate
larger than 16-bits will be expanded into two instructions. This
is a very undesirable feature to rely on, so this flag can help
detect any code where it happens. One use of it, for instance, has
been to find and eliminate any place where `gcc' may emit these
pseudo-instructions.

`-mxp'
Enable support for the i860XP instructions and control registers.
By default, this option is disabled so that only the base
instruction set (i.e., i860XR) is supported.

`-mintel-syntax'
The i860 assembler defaults to AT&T/SVR4 syntax. This option
enables the Intel syntax.

File: as.info, Node: Directives-i860, Next: Opcodes for i860, Prev: Options-i860, Up: i860-Dependent

8.13.3 i860 Machine Directives
------------------------------

`.dual'
Enter dual instruction mode. While this directive is supported, the
preferred way to use dual instruction mode is to explicitly code
the dual bit with the `d.' prefix.

`.enddual'
Exit dual instruction mode. While this directive is supported, the
preferred way to use dual instruction mode is to explicitly code
the dual bit with the `d.' prefix.

`.atmp'
Change the temporary register used when expanding pseudo
operations. The default register is `r31'.

The `.dual', `.enddual', and `.atmp' directives are available only
in the Intel syntax mode.

Both syntaxes allow for the standard `.align' directive. However,
the Intel syntax additionally allows keywords for the alignment
parameter: "`.align type'", where `type' is one of `.short', `.long',
`.quad', `.single', `.double' representing alignments of 2, 4, 16, 4,
and 8, respectively.

File: as.info, Node: Opcodes for i860, Prev: Directives-i860, Up: i860-Dependent

8.13.4 i860 Opcodes
-------------------

All of the Intel i860XR and i860XP machine instructions are supported.
Please see either _i860 Microprocessor Programmer's Reference Manual_
or _i860 Microprocessor Architecture_ for more information.

8.13.4.1 Other instruction support (pseudo-instructions)
........................................................

For compatibility with some other i860 assemblers, a number of
pseudo-instructions are supported. While these are supported, they are
a very undesirable feature that should be avoided - in particular, when
they result in an expansion to multiple actual i860 instructions. Below
are the pseudo-instructions that result in expansions.
* Load large immediate into general register:

The pseudo-instruction `mov imm,%rn' (where the immediate does not
fit within a signed 16-bit field) will be expanded into:
orh large_imm@h,%r0,%rn
or large_imm@l,%rn,%rn

* Load/store with relocatable address expression:

For example, the pseudo-instruction `ld.b addr_exp(%rx),%rn' will
be expanded into:
orh addr_exp@ha,%rx,%r31
ld.l addr_exp@l(%r31),%rn

The analogous expansions apply to `ld.x, st.x, fld.x, pfld.x,
fst.x', and `pst.x' as well.

* Signed large immediate with add/subtract:

If any of the arithmetic operations `adds, addu, subs, subu' are
used with an immediate larger than 16-bits (signed), then they
will be expanded. For instance, the pseudo-instruction `adds
large_imm,%rx,%rn' expands to:
orh large_imm@h,%r0,%r31
or large_imm@l,%r31,%r31
adds %r31,%rx,%rn

* Unsigned large immediate with logical operations:

Logical operations (`or, andnot, or, xor') also result in
expansions. The pseudo-instruction `or large_imm,%rx,%rn' results
in:
orh large_imm@h,%rx,%r31
or large_imm@l,%r31,%rn

Similarly for the others, except for `and' which expands to:
andnot (-1 - large_imm)@h,%rx,%r31
andnot (-1 - large_imm)@l,%r31,%rn

File: as.info, Node: i960-Dependent, Next: IA-64-Dependent, Prev: i860-Dependent, Up: Machine Dependencies

8.14 Intel 80960 Dependent Features
===================================

* Menu:

* Options-i960:: i960 Command-line Options
* Floating Point-i960:: Floating Point
* Directives-i960:: i960 Machine Directives
* Opcodes for i960:: i960 Opcodes

File: as.info, Node: Options-i960, Next: Floating Point-i960, Up: i960-Dependent

8.14.1 i960 Command-line Options
--------------------------------

`-ACA | -ACA_A | -ACB | -ACC | -AKA | -AKB | -AKC | -AMC'
Select the 80960 architecture. Instructions or features not
supported by the selected architecture cause fatal errors.

`-ACA' is equivalent to `-ACA_A'; `-AKC' is equivalent to `-AMC'.
Synonyms are provided for compatibility with other tools.

If you do not specify any of these options, `as' generates code
for any instruction or feature that is supported by _some_ version
of the 960 (even if this means mixing architectures!). In
principle, `as' attempts to deduce the minimal sufficient
processor type if none is specified; depending on the object code
format, the processor type may be recorded in the object file. If
it is critical that the `as' output match a specific architecture,
specify that architecture explicitly.

`-b'
Add code to collect information about conditional branches taken,
for later optimization using branch prediction bits. (The
conditional branch instructions have branch prediction bits in the
CA, CB, and CC architectures.) If BR represents a conditional
branch instruction, the following represents the code generated by
the assembler when `-b' is specified:

call INCREMENT ROUTINE
.word 0 # pre-counter
Label: BR
call INCREMENT ROUTINE
.word 0 # post-counter

The counter following a branch records the number of times that
branch was _not_ taken; the differenc between the two counters is
the number of times the branch _was_ taken.

A table of every such `Label' is also generated, so that the
external postprocessor `gbr960' (supplied by Intel) can locate all
the counters. This table is always labeled `__BRANCH_TABLE__';
this is a local symbol to permit collecting statistics for many
separate object files. The table is word aligned, and begins with
a two-word header. The first word, initialized to 0, is used in
maintaining linked lists of branch tables. The second word is a
count of the number of entries in the table, which follow
immediately: each is a word, pointing to one of the labels
illustrated above.

__BRANCH_TABLE__ layout

The first word of the header is used to locate multiple branch
tables, since each object file may contain one. Normally the links
are maintained with a call to an initialization routine, placed at
the beginning of each function in the file. The GNU C compiler
generates these calls automatically when you give it a `-b' option.
For further details, see the documentation of `gbr960'.

`-no-relax'
Normally, Compare-and-Branch instructions with targets that require
displacements greater than 13 bits (or that have external targets)
are replaced with the corresponding compare (or `chkbit') and
branch instructions. You can use the `-no-relax' option to
specify that `as' should generate errors instead, if the target
displacement is larger than 13 bits.

This option does not affect the Compare-and-Jump instructions; the
code emitted for them is _always_ adjusted when necessary
(depending on displacement size), regardless of whether you use
`-no-relax'.

File: as.info, Node: Floating Point-i960, Next: Directives-i960, Prev: Options-i960, Up: i960-Dependent

8.14.2 Floating Point
---------------------

`as' generates IEEE floating-point numbers for the directives `.float',
`.double', `.extended', and `.single'.

File: as.info, Node: Directives-i960, Next: Opcodes for i960, Prev: Floating Point-i960, Up: i960-Dependent

8.14.3 i960 Machine Directives
------------------------------

`.bss SYMBOL, LENGTH, ALIGN'
Reserve LENGTH bytes in the bss section for a local SYMBOL,
aligned to the power of two specified by ALIGN. LENGTH and ALIGN
must be positive absolute expressions. This directive differs
from `.lcomm' only in that it permits you to specify an alignment.
*Note `.lcomm': Lcomm.

`.extended FLONUMS'
`.extended' expects zero or more flonums, separated by commas; for
each flonum, `.extended' emits an IEEE extended-format (80-bit)
floating-point number.

`.leafproc CALL-LAB, BAL-LAB'
You can use the `.leafproc' directive in conjunction with the
optimized `callj' instruction to enable faster calls of leaf
procedures. If a procedure is known to call no other procedures,
you may define an entry point that skips procedure prolog code
(and that does not depend on system-supplied saved context), and
declare it as the BAL-LAB using `.leafproc'. If the procedure
also has an entry point that goes through the normal prolog, you
can specify that entry point as CALL-LAB.

A `.leafproc' declaration is meant for use in conjunction with the
optimized call instruction `callj'; the directive records the data
needed later to choose between converting the `callj' into a `bal'
or a `call'.

CALL-LAB is optional; if only one argument is present, or if the
two arguments are identical, the single argument is assumed to be
the `bal' entry point.

`.sysproc NAME, INDEX'
The `.sysproc' directive defines a name for a system procedure.
After you define it using `.sysproc', you can use NAME to refer to
the system procedure identified by INDEX when calling procedures
with the optimized call instruction `callj'.

Both arguments are required; INDEX must be between 0 and 31
(inclusive).

File: as.info, Node: Opcodes for i960, Prev: Directives-i960, Up: i960-Dependent

8.14.4 i960 Opcodes
-------------------

All Intel 960 machine instructions are supported; *note i960
Command-line Options: Options-i960. for a discussion of selecting the
instruction subset for a particular 960 architecture.

Some opcodes are processed beyond simply emitting a single
corresponding instruction: `callj', and Compare-and-Branch or
Compare-and-Jump instructions with target displacements larger than 13
bits.

* Menu:

* callj-i960:: `callj'
* Compare-and-branch-i960:: Compare-and-Branch

File: as.info, Node: callj-i960, Next: Compare-and-branch-i960, Up: Opcodes for i960

8.14.4.1 `callj'
................

You can write `callj' to have the assembler or the linker determine the
most appropriate form of subroutine call: `call', `bal', or `calls'.
If the assembly source contains enough information--a `.leafproc' or
`.sysproc' directive defining the operand--then `as' translates the
`callj'; if not, it simply emits the `callj', leaving it for the linker
to resolve.

File: as.info, Node: Compare-and-branch-i960, Prev: callj-i960, Up: Opcodes for i960

8.14.4.2 Compare-and-Branch
...........................

The 960 architectures provide combined Compare-and-Branch instructions
that permit you to store the branch target in the lower 13 bits of the
instruction word itself. However, if you specify a branch target far
enough away that its address won't fit in 13 bits, the assembler can
either issue an error, or convert your Compare-and-Branch instruction
into separate instructions to do the compare and the branch.

Whether `as' gives an error or expands the instruction depends on
two choices you can make: whether you use the `-no-relax' option, and
whether you use a "Compare and Branch" instruction or a "Compare and
Jump" instruction. The "Jump" instructions are _always_ expanded if
necessary; the "Branch" instructions are expanded when necessary
_unless_ you specify `-no-relax'--in which case `as' gives an error
instead.

These are the Compare-and-Branch instructions, their "Jump" variants,
and the instruction pairs they may expand into:

Compare and
Branch Jump Expanded to
------ ------ ------------
bbc chkbit; bno
bbs chkbit; bo
cmpibe cmpije cmpi; be
cmpibg cmpijg cmpi; bg
cmpibge cmpijge cmpi; bge
cmpibl cmpijl cmpi; bl
cmpible cmpijle cmpi; ble
cmpibno cmpijno cmpi; bno
cmpibne cmpijne cmpi; bne
cmpibo cmpijo cmpi; bo
cmpobe cmpoje cmpo; be
cmpobg cmpojg cmpo; bg
cmpobge cmpojge cmpo; bge
cmpobl cmpojl cmpo; bl
cmpoble cmpojle cmpo; ble
cmpobne cmpojne cmpo; bne

File: as.info, Node: IA-64-Dependent, Next: IP2K-Dependent, Prev: i960-Dependent, Up: Machine Dependencies

8.15 IA-64 Dependent Features
=============================

* Menu:

* IA-64 Options:: Options
* IA-64 Syntax:: Syntax
* IA-64 Opcodes:: Opcodes

File: as.info, Node: IA-64 Options, Next: IA-64 Syntax, Up: IA-64-Dependent

8.15.1 Options
--------------

`-mconstant-gp'
This option instructs the assembler to mark the resulting object
file as using the "constant GP" model. With this model, it is
assumed that the entire program uses a single global pointer (GP)
value. Note that this option does not in any fashion affect the
machine code emitted by the assembler. All it does is turn on the
EF_IA_64_CONS_GP flag in the ELF file header.

`-mauto-pic'
This option instructs the assembler to mark the resulting object
file as using the "constant GP without function descriptor" data
model. This model is like the "constant GP" model, except that it
additionally does away with function descriptors. What this means
is that the address of a function refers directly to the
function's code entry-point. Normally, such an address would
refer to a function descriptor, which contains both the code
entry-point and the GP-value needed by the function. Note that
this option does not in any fashion affect the machine code
emitted by the assembler. All it does is turn on the
EF_IA_64_NOFUNCDESC_CONS_GP flag in the ELF file header.

`-milp32'

`-milp64'

`-mlp64'

`-mp64'
These options select the data model. The assembler defaults to
`-mlp64' (LP64 data model).

`-mle'

`-mbe'
These options select the byte order. The `-mle' option selects
little-endian byte order (default) and `-mbe' selects big-endian
byte order. Note that IA-64 machine code always uses
little-endian byte order.

`-munwind-check=warning'

`-munwind-check=error'
These options control what the assembler will do when performing
consistency checks on unwind directives. `-munwind-check=warning'
will make the assembler issue a warning when an unwind directive
check fails. This is the default. `-munwind-check=error' will
make the assembler issue an error when an unwind directive check
fails.

`-mhint.b=ok'

`-mhint.b=warning'

`-mhint.b=error'
These options control what the assembler will do when the `hint.b'
instruction is used. `-mhint.b=ok' will make the assembler accept
`hint.b'. `-mint.b=warning' will make the assembler issue a
warning when `hint.b' is used. `-mhint.b=error' will make the
assembler treat `hint.b' as an error, which is the default.

`-x'

`-xexplicit'
These options turn on dependency violation checking.

`-xauto'
This option instructs the assembler to automatically insert stop
bits where necessary to remove dependency violations. This is the
default mode.

`-xnone'
This option turns off dependency violation checking.

`-xdebug'
This turns on debug output intended to help tracking down bugs in
the dependency violation checker.

`-xdebugn'
This is a shortcut for -xnone -xdebug.

`-xdebugx'
This is a shortcut for -xexplicit -xdebug.

File: as.info, Node: IA-64 Syntax, Next: IA-64 Opcodes, Prev: IA-64 Options, Up: IA-64-Dependent

8.15.2 Syntax
-------------

The assembler syntax closely follows the IA-64 Assembly Language
Reference Guide.

* Menu:

* IA-64-Chars:: Special Characters
* IA-64-Regs:: Register Names
* IA-64-Bits:: Bit Names

File: as.info, Node: IA-64-Chars, Next: IA-64-Regs, Up: IA-64 Syntax

8.15.2.1 Special Characters
...........................

`//' is the line comment token.

`;' can be used instead of a newline to separate statements.

File: as.info, Node: IA-64-Regs, Next: IA-64-Bits, Prev: IA-64-Chars, Up: IA-64 Syntax

8.15.2.2 Register Names
.......................

The 128 integer registers are referred to as `rN'. The 128
floating-point registers are referred to as `fN'. The 128 application
registers are referred to as `arN'. The 128 control registers are
referred to as `crN'. The 64 one-bit predicate registers are referred
to as `pN'. The 8 branch registers are referred to as `bN'. In
addition, the assembler defines a number of aliases: `gp' (`r1'), `sp'
(`r12'), `rp' (`b0'), `ret0' (`r8'), `ret1' (`r9'), `ret2' (`r10'),
`ret3' (`r9'), `fargN' (`f8+N'), and `fretN' (`f8+N').

For convenience, the assembler also defines aliases for all named
application and control registers. For example, `ar.bsp' refers to the
register backing store pointer (`ar17'). Similarly, `cr.eoi' refers to
the end-of-interrupt register (`cr67').

File: as.info, Node: IA-64-Bits, Prev: IA-64-Regs, Up: IA-64 Syntax

8.15.2.3 IA-64 Processor-Status-Register (PSR) Bit Names
........................................................

The assembler defines bit masks for each of the bits in the IA-64
processor status register. For example, `psr.ic' corresponds to a
value of 0x2000. These masks are primarily intended for use with the
`ssm'/`sum' and `rsm'/`rum' instructions, but they can be used anywhere
else where an integer constant is expected.

File: as.info, Node: IA-64 Opcodes, Prev: IA-64 Syntax, Up: IA-64-Dependent

8.15.3 Opcodes
--------------

For detailed information on the IA-64 machine instruction set, see the
IA-64 Architecture Handbook
(http://developer.intel.com/design/itanium/arch_spec.htm).

File: as.info, Node: IP2K-Dependent, Next: M32R-Dependent, Prev: IA-64-Dependent, Up: Machine Dependencies

8.16 IP2K Dependent Features
============================

* Menu:

* IP2K-Opts:: IP2K Options

File: as.info, Node: IP2K-Opts, Up: IP2K-Dependent

8.16.1 IP2K Options
-------------------

The Ubicom IP2K version of `as' has a few machine dependent options:

`-mip2022ext'
`as' can assemble the extended IP2022 instructions, but it will
only do so if this is specifically allowed via this command line
option.

`-mip2022'
This option restores the assembler's default behaviour of not
permitting the extended IP2022 instructions to be assembled.

File: as.info, Node: M32R-Dependent, Next: M68K-Dependent, Prev: IP2K-Dependent, Up: Machine Dependencies

8.17 M32R Dependent Features
============================

* Menu:

* M32R-Opts:: M32R Options
* M32R-Directives:: M32R Directives
* M32R-Warnings:: M32R Warnings

File: as.info, Node: M32R-Opts, Next: M32R-Directives, Up: M32R-Dependent

8.17.1 M32R Options
-------------------

The Renease M32R version of `as' has a few machine dependent options:

`-m32rx'
`as' can assemble code for several different members of the
Renesas M32R family. Normally the default is to assemble code for
the M32R microprocessor. This option may be used to change the
default to the M32RX microprocessor, which adds some more
instructions to the basic M32R instruction set, and some
additional parameters to some of the original instructions.

`-m32r2'
This option changes the target processor to the the M32R2
microprocessor.

`-m32r'
This option can be used to restore the assembler's default
behaviour of assembling for the M32R microprocessor. This can be
useful if the default has been changed by a previous command line
option.

`-little'
This option tells the assembler to produce little-endian code and
data. The default is dependent upon how the toolchain was
configured.

`-EL'
This is a synonum for _-little_.

`-big'
This option tells the assembler to produce big-endian code and
data.

`-EB'
This is a synonum for _-big_.

`-KPIC'
This option specifies that the output of the assembler should be
marked as position-independent code (PIC).

`-parallel'
This option tells the assembler to attempts to combine two
sequential instructions into a single, parallel instruction, where
it is legal to do so.

`-no-parallel'
This option disables a previously enabled _-parallel_ option.

`-no-bitinst'
This option disables the support for the extended bit-field
instructions provided by the M32R2. If this support needs to be
re-enabled the _-bitinst_ switch can be used to restore it.

`-O'
This option tells the assembler to attempt to optimize the
instructions that it produces. This includes filling delay slots
and converting sequential instructions into parallel ones. This
option implies _-parallel_.

`-warn-explicit-parallel-conflicts'
Instructs `as' to produce warning messages when questionable
parallel instructions are encountered. This option is enabled by
default, but `gcc' disables it when it invokes `as' directly.
Questionable instructions are those whoes behaviour would be
different if they were executed sequentially. For example the
code fragment `mv r1, r2 || mv r3, r1' produces a different result
from `mv r1, r2 \n mv r3, r1' since the former moves r1 into r3
and then r2 into r1, whereas the later moves r2 into r1 and r3.

`-Wp'
This is a shorter synonym for the
_-warn-explicit-parallel-conflicts_ option.

`-no-warn-explicit-parallel-conflicts'
Instructs `as' not to produce warning messages when questionable
parallel instructions are encountered.

`-Wnp'
This is a shorter synonym for the
_-no-warn-explicit-parallel-conflicts_ option.

`-ignore-parallel-conflicts'
This option tells the assembler's to stop checking parallel
instructions for contraint violations. This ability is provided
for hardware vendors testing chip designs and should not be used
under normal circumstances.

`-no-ignore-parallel-conflicts'
This option restores the assembler's default behaviour of checking
parallel instructions to detect constraint violations.

`-Ip'
This is a shorter synonym for the _-ignore-parallel-conflicts_
option.

`-nIp'
This is a shorter synonym for the _-no-ignore-parallel-conflicts_
option.

`-warn-unmatched-high'
This option tells the assembler to produce a warning message if a
`.high' pseudo op is encountered without a mathcing `.low' pseudo
op. The presence of such an unmatches pseudo op usually indicates
a programming error.

`-no-warn-unmatched-high'
Disables a previously enabled _-warn-unmatched-high_ option.

`-Wuh'
This is a shorter synonym for the _-warn-unmatched-high_ option.

`-Wnuh'
This is a shorter synonym for the _-no-warn-unmatched-high_ option.

File: as.info, Node: M32R-Directives, Next: M32R-Warnings, Prev: M32R-Opts, Up: M32R-Dependent

8.17.2 M32R Directives
----------------------

The Renease M32R version of `as' has a few architecture specific
directives:

`low EXPRESSION'
The `low' directive computes the value of its expression and
places the lower 16-bits of the result into the immediate-field of
the instruction. For example:

or3 r0, r0, #low(0x12345678) ; compute r0 = r0 | 0x5678
add3, r0, r0, #low(fred) ; compute r0 = r0 + low 16-bits of address of fred

`high EXPRESSION'
The `high' directive computes the value of its expression and
places the upper 16-bits of the result into the immediate-field of
the instruction. For example:

seth r0, #high(0x12345678) ; compute r0 = 0x12340000
seth, r0, #high(fred) ; compute r0 = upper 16-bits of address of fred

`shigh EXPRESSION'
The `shigh' directive is very similar to the `high' directive. It
also computes the value of its expression and places the upper
16-bits of the result into the immediate-field of the instruction.
The difference is that `shigh' also checks to see if the lower
16-bits could be interpreted as a signed number, and if so it
assumes that a borrow will occur from the upper-16 bits. To
compensate for this the `shigh' directive pre-biases the upper 16
bit value by adding one to it. For example:

For example:

seth r0, #shigh(0x12345678) ; compute r0 = 0x12340000
seth r0, #shigh(0x00008000) ; compute r0 = 0x00010000

In the second example the lower 16-bits are 0x8000. If these are
treated as a signed value and sign extended to 32-bits then the
value becomes 0xffff8000. If this value is then added to
0x00010000 then the result is 0x00008000.

This behaviour is to allow for the different semantics of the
`or3' and `add3' instructions. The `or3' instruction treats its
16-bit immediate argument as unsigned whereas the `add3' treats
its 16-bit immediate as a signed value. So for example:

seth r0, #shigh(0x00008000)
add3 r0, r0, #low(0x00008000)

Produces the correct result in r0, whereas:

seth r0, #shigh(0x00008000)
or3 r0, r0, #low(0x00008000)

Stores 0xffff8000 into r0.

Note - the `shigh' directive does not know where in the assembly
source code the lower 16-bits of the value are going set, so it
cannot check to make sure that an `or3' instruction is being used
rather than an `add3' instruction. It is up to the programmer to
make sure that correct directives are used.

`.m32r'
The directive performs a similar thing as the _-m32r_ command line
option. It tells the assembler to only accept M32R instructions
from now on. An instructions from later M32R architectures are
refused.

`.m32rx'
The directive performs a similar thing as the _-m32rx_ command
line option. It tells the assembler to start accepting the extra
instructions in the M32RX ISA as well as the ordinary M32R ISA.

`.m32r2'
The directive performs a similar thing as the _-m32r2_ command
line option. It tells the assembler to start accepting the extra
instructions in the M32R2 ISA as well as the ordinary M32R ISA.

`.little'
The directive performs a similar thing as the _-little_ command
line option. It tells the assembler to start producing
little-endian code and data. This option should be used with care
as producing mixed-endian binary files is frought with danger.

`.big'
The directive performs a similar thing as the _-big_ command line
option. It tells the assembler to start producing big-endian code
and data. This option should be used with care as producing
mixed-endian binary files is frought with danger.

File: as.info, Node: M32R-Warnings, Prev: M32R-Directives, Up: M32R-Dependent

8.17.3 M32R Warnings
--------------------

There are several warning and error messages that can be produced by
`as' which are specific to the M32R:

`output of 1st instruction is the same as an input to 2nd instruction - is this intentional ?'
This message is only produced if warnings for explicit parallel
conflicts have been enabled. It indicates that the assembler has
encountered a parallel instruction in which the destination
register of the left hand instruction is used as an input register
in the right hand instruction. For example in this code fragment
`mv r1, r2 || neg r3, r1' register r1 is the destination of the
move instruction and the input to the neg instruction.

`output of 2nd instruction is the same as an input to 1st instruction - is this intentional ?'
This message is only produced if warnings for explicit parallel
conflicts have been enabled. It indicates that the assembler has
encountered a parallel instruction in which the destination
register of the right hand instruction is used as an input
register in the left hand instruction. For example in this code
fragment `mv r1, r2 || neg r2, r3' register r2 is the destination
of the neg instruction and the input to the move instruction.

`instruction `...' is for the M32RX only'
This message is produced when the assembler encounters an
instruction which is only supported by the M32Rx processor, and
the `-m32rx' command line flag has not been specified to allow
assembly of such instructions.

`unknown instruction `...''
This message is produced when the assembler encounters an
instruction which it does not recognise.

`only the NOP instruction can be issued in parallel on the m32r'
This message is produced when the assembler encounters a parallel
instruction which does not involve a NOP instruction and the
`-m32rx' command line flag has not been specified. Only the M32Rx
processor is able to execute two instructions in parallel.

`instruction `...' cannot be executed in parallel.'
This message is produced when the assembler encounters a parallel
instruction which is made up of one or two instructions which
cannot be executed in parallel.

`Instructions share the same execution pipeline'
This message is produced when the assembler encounters a parallel
instruction whoes components both use the same execution pipeline.

`Instructions write to the same destination register.'
This message is produced when the assembler encounters a parallel
instruction where both components attempt to modify the same
register. For example these code fragments will produce this
message: `mv r1, r2 || neg r1, r3' `jl r0 || mv r14, r1' `st r2,
@-r1 || mv r1, r3' `mv r1, r2 || ld r0, @r1+' `cmp r1, r2 || addx
r3, r4' (Both write to the condition bit)

File: as.info, Node: M68K-Dependent, Next: M68HC11-Dependent, Prev: M32R-Dependent, Up: Machine Dependencies

8.18 M680x0 Dependent Features
==============================

* Menu:

* M68K-Opts:: M680x0 Options
* M68K-Syntax:: Syntax
* M68K-Moto-Syntax:: Motorola Syntax
* M68K-Float:: Floating Point
* M68K-Directives:: 680x0 Machine Directives
* M68K-opcodes:: Opcodes

File: as.info, Node: M68K-Opts, Next: M68K-Syntax, Up: M68K-Dependent

8.18.1 M680x0 Options
---------------------

The Motorola 680x0 version of `as' has a few machine dependent options:

`-l'
You can use the `-l' option to shorten the size of references to
undefined symbols. If you do not use the `-l' option, references
to undefined symbols are wide enough for a full `long' (32 bits).
(Since `as' cannot know where these symbols end up, `as' can only
allocate space for the linker to fill in later. Since `as' does
not know how far away these symbols are, it allocates as much
space as it can.) If you use this option, the references are only
one word wide (16 bits). This may be useful if you want the
object file to be as small as possible, and you know that the
relevant symbols are always less than 17 bits away.

`--register-prefix-optional'
For some configurations, especially those where the compiler
normally does not prepend an underscore to the names of user
variables, the assembler requires a `%' before any use of a
register name. This is intended to let the assembler distinguish
between C variables and functions named `a0' through `a7', and so
on. The `%' is always accepted, but is not required for certain
configurations, notably `sun3'. The `--register-prefix-optional'
option may be used to permit omitting the `%' even for
configurations for which it is normally required. If this is
done, it will generally be impossible to refer to C variables and
functions with the same names as register names.

`--bitwise-or'
Normally the character `|' is treated as a comment character, which
means that it can not be used in expressions. The `--bitwise-or'
option turns `|' into a normal character. In this mode, you must
either use C style comments, or start comments with a `#' character
at the beginning of a line.

`--base-size-default-16 --base-size-default-32'
If you use an addressing mode with a base register without
specifying the size, `as' will normally use the full 32 bit value.
For example, the addressing mode `%a0@(%d0)' is equivalent to
`%a0@(%d0:l)'. You may use the `--base-size-default-16' option to
tell `as' to default to using the 16 bit value. In this case,
`%a0@(%d0)' is equivalent to `%a0@(%d0:w)'. You may use the
`--base-size-default-32' option to restore the default behaviour.

`--disp-size-default-16 --disp-size-default-32'
If you use an addressing mode with a displacement, and the value
of the displacement is not known, `as' will normally assume that
the value is 32 bits. For example, if the symbol `disp' has not
been defined, `as' will assemble the addressing mode
`%a0@(disp,%d0)' as though `disp' is a 32 bit value. You may use
the `--disp-size-default-16' option to tell `as' to instead assume
that the displacement is 16 bits. In this case, `as' will
assemble `%a0@(disp,%d0)' as though `disp' is a 16 bit value. You
may use the `--disp-size-default-32' option to restore the default
behaviour.

`--pcrel'
Always keep branches PC-relative. In the M680x0 architecture all
branches are defined as PC-relative. However, on some processors
they are limited to word displacements maximum. When `as' needs a
long branch that is not available, it normally emits an absolute
jump instead. This option disables this substitution. When this
option is given and no long branches are available, only word
branches will be emitted. An error message will be generated if a
word branch cannot reach its target. This option has no effect on
68020 and other processors that have long branches. *note Branch
Improvement: M68K-Branch.

`-m68000'
`as' can assemble code for several different members of the
Motorola 680x0 family. The default depends upon how `as' was
configured when it was built; normally, the default is to assemble
code for the 68020 microprocessor. The following options may be
used to change the default. These options control which
instructions and addressing modes are permitted. The members of
the 680x0 family are very similar. For detailed information about
the differences, see the Motorola manuals.

`-m68000'
`-m68ec000'
`-m68hc000'
`-m68hc001'
`-m68008'
`-m68302'
`-m68306'
`-m68307'
`-m68322'
`-m68356'
Assemble for the 68000. `-m68008', `-m68302', and so on are
synonyms for `-m68000', since the chips are the same from the
point of view of the assembler.

`-m68010'
Assemble for the 68010.

`-m68020'
`-m68ec020'
Assemble for the 68020. This is normally the default.

`-m68030'
`-m68ec030'
Assemble for the 68030.

`-m68040'
`-m68ec040'
Assemble for the 68040.

`-m68060'
`-m68ec060'
Assemble for the 68060.

`-mcpu32'
`-m68330'
`-m68331'
`-m68332'
`-m68333'
`-m68334'
`-m68336'
`-m68340'
`-m68341'
`-m68349'
`-m68360'
Assemble for the CPU32 family of chips.

`-m5200'

`-m5202'

`-m5204'

`-m5206'

`-m5206e'

`-m521x'

`-m5249'

`-m528x'

`-m5307'

`-m5407'

`-m547x'

`-m548x'

`-mcfv4'

`-mcfv4e'
Assemble for the ColdFire family of chips.

`-m68881'
`-m68882'
Assemble 68881 floating point instructions. This is the
default for the 68020, 68030, and the CPU32. The 68040 and
68060 always support floating point instructions.

`-mno-68881'
Do not assemble 68881 floating point instructions. This is
the default for 68000 and the 68010. The 68040 and 68060
always support floating point instructions, even if this
option is used.

`-m68851'
Assemble 68851 MMU instructions. This is the default for the
68020, 68030, and 68060. The 68040 accepts a somewhat
different set of MMU instructions; `-m68851' and `-m68040'
should not be used together.

`-mno-68851'
Do not assemble 68851 MMU instructions. This is the default
for the 68000, 68010, and the CPU32. The 68040 accepts a
somewhat different set of MMU instructions.

File: as.info, Node: M68K-Syntax, Next: M68K-Moto-Syntax, Prev: M68K-Opts, Up: M68K-Dependent

8.18.2 Syntax
-------------

This syntax for the Motorola 680x0 was developed at MIT.

The 680x0 version of `as' uses instructions names and syntax
compatible with the Sun assembler. Intervening periods are ignored;
for example, `movl' is equivalent to `mov.l'.

In the following table APC stands for any of the address registers
(`%a0' through `%a7'), the program counter (`%pc'), the zero-address
relative to the program counter (`%zpc'), a suppressed address register
(`%za0' through `%za7'), or it may be omitted entirely. The use of
SIZE means one of `w' or `l', and it may be omitted, along with the
leading colon, unless a scale is also specified. The use of SCALE
means one of `1', `2', `4', or `8', and it may always be omitted along
with the leading colon.

The following addressing modes are understood:
"Immediate"
`#NUMBER'

"Data Register"
`%d0' through `%d7'

"Address Register"
`%a0' through `%a7'
`%a7' is also known as `%sp', i.e. the Stack Pointer. `%a6' is
also known as `%fp', the Frame Pointer.

"Address Register Indirect"
`%a0@' through `%a7@'

"Address Register Postincrement"
`%a0@+' through `%a7@+'

"Address Register Predecrement"
`%a0@-' through `%a7@-'

"Indirect Plus Offset"
`APC@(NUMBER)'

"Index"
`APC@(NUMBER,REGISTER:SIZE:SCALE)'

The NUMBER may be omitted.

"Postindex"
`APC@(NUMBER)@(ONUMBER,REGISTER:SIZE:SCALE)'

The ONUMBER or the REGISTER, but not both, may be omitted.

"Preindex"
`APC@(NUMBER,REGISTER:SIZE:SCALE)@(ONUMBER)'

The NUMBER may be omitted. Omitting the REGISTER produces the
Postindex addressing mode.

"Absolute"
`SYMBOL', or `DIGITS', optionally followed by `:b', `:w', or `:l'.

File: as.info, Node: M68K-Moto-Syntax, Next: M68K-Float, Prev: M68K-Syntax, Up: M68K-Dependent

8.18.3 Motorola Syntax
----------------------

The standard Motorola syntax for this chip differs from the syntax
already discussed (*note Syntax: M68K-Syntax.). `as' can accept
Motorola syntax for operands, even if MIT syntax is used for other
operands in the same instruction. The two kinds of syntax are fully
compatible.

In the following table APC stands for any of the address registers
(`%a0' through `%a7'), the program counter (`%pc'), the zero-address
relative to the program counter (`%zpc'), or a suppressed address
register (`%za0' through `%za7'). The use of SIZE means one of `w' or
`l', and it may always be omitted along with the leading dot. The use
of SCALE means one of `1', `2', `4', or `8', and it may always be
omitted along with the leading asterisk.

The following additional addressing modes are understood:

"Address Register Indirect"
`(%a0)' through `(%a7)'
`%a7' is also known as `%sp', i.e. the Stack Pointer. `%a6' is
also known as `%fp', the Frame Pointer.

"Address Register Postincrement"
`(%a0)+' through `(%a7)+'

"Address Register Predecrement"
`-(%a0)' through `-(%a7)'

"Indirect Plus Offset"
`NUMBER(%A0)' through `NUMBER(%A7)', or `NUMBER(%PC)'.

The NUMBER may also appear within the parentheses, as in
`(NUMBER,%A0)'. When used with the PC, the NUMBER may be omitted
(with an address register, omitting the NUMBER produces Address
Register Indirect mode).

"Index"
`NUMBER(APC,REGISTER.SIZE*SCALE)'

The NUMBER may be omitted, or it may appear within the
parentheses. The APC may be omitted. The REGISTER and the APC
may appear in either order. If both APC and REGISTER are address
registers, and the SIZE and SCALE are omitted, then the first
register is taken as the base register, and the second as the
index register.

"Postindex"
`([NUMBER,APC],REGISTER.SIZE*SCALE,ONUMBER)'

The ONUMBER, or the REGISTER, or both, may be omitted. Either the
NUMBER or the APC may be omitted, but not both.

"Preindex"
`([NUMBER,APC,REGISTER.SIZE*SCALE],ONUMBER)'

The NUMBER, or the APC, or the REGISTER, or any two of them, may
be omitted. The ONUMBER may be omitted. The REGISTER and the APC
may appear in either order. If both APC and REGISTER are address
registers, and the SIZE and SCALE are omitted, then the first
register is taken as the base register, and the second as the
index register.

File: as.info, Node: M68K-Float, Next: M68K-Directives, Prev: M68K-Moto-Syntax, Up: M68K-Dependent

8.18.4 Floating Point
---------------------

Packed decimal (P) format floating literals are not supported. Feel
free to add the code!

The floating point formats generated by directives are these.

`.float'
`Single' precision floating point constants.

`.double'
`Double' precision floating point constants.

`.extend'
`.ldouble'
`Extended' precision (`long double') floating point constants.

File: as.info, Node: M68K-Directives, Next: M68K-opcodes, Prev: M68K-Float, Up: M68K-Dependent

8.18.5 680x0 Machine Directives
-------------------------------

In order to be compatible with the Sun assembler the 680x0 assembler
understands the following directives.

`.data1'
This directive is identical to a `.data 1' directive.

`.data2'
This directive is identical to a `.data 2' directive.

`.even'
This directive is a special case of the `.align' directive; it
aligns the output to an even byte boundary.

`.skip'
This directive is identical to a `.space' directive.

File: as.info, Node: M68K-opcodes, Prev: M68K-Directives, Up: M68K-Dependent

8.18.6 Opcodes
--------------

* Menu:

* M68K-Branch:: Branch Improvement
* M68K-Chars:: Special Characters

File: as.info, Node: M68K-Branch, Next: M68K-Chars, Up: M68K-opcodes

8.18.6.1 Branch Improvement
...........................

Certain pseudo opcodes are permitted for branch instructions. They
expand to the shortest branch instruction that reach the target.
Generally these mnemonics are made by substituting `j' for `b' at the
start of a Motorola mnemonic.

The following table summarizes the pseudo-operations. A `*' flags
cases that are more fully described after the table:

Displacement
+------------------------------------------------------------
| 68020 68000/10, not PC-relative OK
Pseudo-Op |BYTE WORD LONG ABSOLUTE LONG JUMP **
+------------------------------------------------------------
jbsr |bsrs bsrw bsrl jsr
jra |bras braw bral jmp
* jXX |bXXs bXXw bXXl bNXs;jmp
* dbXX | N/A dbXXw dbXX;bras;bral dbXX;bras;jmp
fjXX | N/A fbXXw fbXXl N/A

XX: condition
NX: negative of condition XX
`*'--see full description below
`**'--this expansion mode is disallowed by `--pcrel'

`jbsr'
`jra'
These are the simplest jump pseudo-operations; they always map to
one particular machine instruction, depending on the displacement
to the branch target. This instruction will be a byte or word
branch is that is sufficient. Otherwise, a long branch will be
emitted if available. If no long branches are available and the
`--pcrel' option is not given, an absolute long jump will be
emitted instead. If no long branches are available, the `--pcrel'
option is given, and a word branch cannot reach the target, an
error message is generated.

In addition to standard branch operands, `as' allows these
pseudo-operations to have all operands that are allowed for jsr
and jmp, substituting these instructions if the operand given is
not valid for a branch instruction.

`jXX'
Here, `jXX' stands for an entire family of pseudo-operations,
where XX is a conditional branch or condition-code test. The full
list of pseudo-ops in this family is:
jhi jls jcc jcs jne jeq jvc
jvs jpl jmi jge jlt jgt jle

Usually, each of these pseudo-operations expands to a single branch
instruction. However, if a word branch is not sufficient, no long
branches are available, and the `--pcrel' option is not given, `as'
issues a longer code fragment in terms of NX, the opposite
condition to XX. For example, under these conditions:
jXX foo
gives
bNXs oof
jmp foo
oof:

`dbXX'
The full family of pseudo-operations covered here is
dbhi dbls dbcc dbcs dbne dbeq dbvc
dbvs dbpl dbmi dbge dblt dbgt dble
dbf dbra dbt

Motorola `dbXX' instructions allow word displacements only. When
a word displacement is sufficient, each of these pseudo-operations
expands to the corresponding Motorola instruction. When a word
displacement is not sufficient and long branches are available,
when the source reads `dbXX foo', `as' emits
dbXX oo1
bras oo2
oo1:bral foo
oo2:

If, however, long branches are not available and the `--pcrel'
option is not given, `as' emits
dbXX oo1
bras oo2
oo1:jmp foo
oo2:

`fjXX'
This family includes
fjne fjeq fjge fjlt fjgt fjle fjf
fjt fjgl fjgle fjnge fjngl fjngle fjngt
fjnle fjnlt fjoge fjogl fjogt fjole fjolt
fjor fjseq fjsf fjsne fjst fjueq fjuge
fjugt fjule fjult fjun

Each of these pseudo-operations always expands to a single Motorola
coprocessor branch instruction, word or long. All Motorola
coprocessor branch instructions allow both word and long
displacements.

File: as.info, Node: M68K-Chars, Prev: M68K-Branch, Up: M68K-opcodes

8.18.6.2 Special Characters
...........................

The immediate character is `#' for Sun compatibility. The line-comment
character is `|' (unless the `--bitwise-or' option is used). If a `#'
appears at the beginning of a line, it is treated as a comment unless
it looks like `# line file', in which case it is treated normally.

File: as.info, Node: M68HC11-Dependent, Next: M88K-Dependent, Prev: M68K-Dependent, Up: Machine Dependencies

8.19 M68HC11 and M68HC12 Dependent Features
===========================================

* Menu:

* M68HC11-Opts:: M68HC11 and M68HC12 Options
* M68HC11-Syntax:: Syntax
* M68HC11-Modifiers:: Symbolic Operand Modifiers
* M68HC11-Directives:: Assembler Directives
* M68HC11-Float:: Floating Point
* M68HC11-opcodes:: Opcodes

File: as.info, Node: M68HC11-Opts, Next: M68HC11-Syntax, Up: M68HC11-Dependent

8.19.1 M68HC11 and M68HC12 Options
----------------------------------

The Motorola 68HC11 and 68HC12 version of `as' have a few machine
dependent options.

`-m68hc11'
This option switches the assembler in the M68HC11 mode. In this
mode, the assembler only accepts 68HC11 operands and mnemonics. It
produces code for the 68HC11.

`-m68hc12'
This option switches the assembler in the M68HC12 mode. In this
mode, the assembler also accepts 68HC12 operands and mnemonics. It
produces code for the 68HC12. A few 68HC11 instructions are
replaced by some 68HC12 instructions as recommended by Motorola
specifications.

`-m68hcs12'
This option switches the assembler in the M68HCS12 mode. This
mode is similar to `-m68hc12' but specifies to assemble for the
68HCS12 series. The only difference is on the assembling of the
`movb' and `movw' instruction when a PC-relative operand is used.

`-mshort'
This option controls the ABI and indicates to use a 16-bit integer
ABI. It has no effect on the assembled instructions. This is the
default.

`-mlong'
This option controls the ABI and indicates to use a 32-bit integer
ABI.

`-mshort-double'
This option controls the ABI and indicates to use a 32-bit float
ABI. This is the default.

`-mlong-double'
This option controls the ABI and indicates to use a 64-bit float
ABI.

`--strict-direct-mode'
You can use the `--strict-direct-mode' option to disable the
automatic translation of direct page mode addressing into extended
mode when the instruction does not support direct mode. For
example, the `clr' instruction does not support direct page mode
addressing. When it is used with the direct page mode, `as' will
ignore it and generate an absolute addressing. This option
prevents `as' from doing this, and the wrong usage of the direct
page mode will raise an error.

`--short-branchs'
The `--short-branchs' option turns off the translation of relative
branches into absolute branches when the branch offset is out of
range. By default `as' transforms the relative branch (`bsr',
`bgt', `bge', `beq', `bne', `ble', `blt', `bhi', `bcc', `bls',
`bcs', `bmi', `bvs', `bvs', `bra') into an absolute branch when
the offset is out of the -128 .. 127 range. In that case, the
`bsr' instruction is translated into a `jsr', the `bra'
instruction is translated into a `jmp' and the conditional branchs
instructions are inverted and followed by a `jmp'. This option
disables these translations and `as' will generate an error if a
relative branch is out of range. This option does not affect the
optimization associated to the `jbra', `jbsr' and `jbXX' pseudo
opcodes.

`--force-long-branchs'
The `--force-long-branchs' option forces the translation of
relative branches into absolute branches. This option does not
affect the optimization associated to the `jbra', `jbsr' and
`jbXX' pseudo opcodes.

`--print-insn-syntax'
You can use the `--print-insn-syntax' option to obtain the syntax
description of the instruction when an error is detected.

`--print-opcodes'
The `--print-opcodes' option prints the list of all the
instructions with their syntax. The first item of each line
represents the instruction name and the rest of the line indicates
the possible operands for that instruction. The list is printed in
alphabetical order. Once the list is printed `as' exits.

`--generate-example'
The `--generate-example' option is similar to `--print-opcodes'
but it generates an example for each instruction instead.

File: as.info, Node: M68HC11-Syntax, Next: M68HC11-Modifiers, Prev: M68HC11-Opts, Up: M68HC11-Dependent

8.19.2 Syntax
-------------

In the M68HC11 syntax, the instruction name comes first and it may be
followed by one or several operands (up to three). Operands are
separated by comma (`,'). In the normal mode, `as' will complain if too
many operands are specified for a given instruction. In the MRI mode
(turned on with `-M' option), it will treat them as comments. Example:

inx
lda #23
bset 2,x #4
brclr *bot #8 foo

The following addressing modes are understood for 68HC11 and 68HC12:
"Immediate"
`#NUMBER'

"Address Register"
`NUMBER,X', `NUMBER,Y'

The NUMBER may be omitted in which case 0 is assumed.

"Direct Addressing mode"
`*SYMBOL', or `*DIGITS'

"Absolute"
`SYMBOL', or `DIGITS'

The M68HC12 has other more complex addressing modes. All of them are
supported and they are represented below:

"Constant Offset Indexed Addressing Mode"
`NUMBER,REG'

The NUMBER may be omitted in which case 0 is assumed. The
register can be either `X', `Y', `SP' or `PC'. The assembler will
use the smaller post-byte definition according to the constant
value (5-bit constant offset, 9-bit constant offset or 16-bit
constant offset). If the constant is not known by the assembler
it will use the 16-bit constant offset post-byte and the value
will be resolved at link time.

"Offset Indexed Indirect"
`[NUMBER,REG]'

The register can be either `X', `Y', `SP' or `PC'.

"Auto Pre-Increment/Pre-Decrement/Post-Increment/Post-Decrement"
`NUMBER,-REG' `NUMBER,+REG' `NUMBER,REG-' `NUMBER,REG+'

The number must be in the range `-8'..`+8' and must not be 0. The
register can be either `X', `Y', `SP' or `PC'.

"Accumulator Offset"
`ACC,REG'

The accumulator register can be either `A', `B' or `D'. The
register can be either `X', `Y', `SP' or `PC'.

"Accumulator D offset indexed-indirect"
`[D,REG]'

The register can be either `X', `Y', `SP' or `PC'.

For example:

ldab 1024,sp
ldd [10,x]
orab 3,+x
stab -2,y-
ldx a,pc
sty [d,sp]

File: as.info, Node: M68HC11-Modifiers, Next: M68HC11-Directives, Prev: M68HC11-Syntax, Up: M68HC11-Dependent

8.19.3 Symbolic Operand Modifiers
---------------------------------

The assembler supports several modifiers when using symbol addresses in
68HC11 and 68HC12 instruction operands. The general syntax is the
following:

%modifier(symbol)

`%addr'
This modifier indicates to the assembler and linker to use the
16-bit physical address corresponding to the symbol. This is
intended to be used on memory window systems to map a symbol in
the memory bank window. If the symbol is in a memory expansion
part, the physical address corresponds to the symbol address
within the memory bank window. If the symbol is not in a memory
expansion part, this is the symbol address (using or not using the
%addr modifier has no effect in that case).

`%page'
This modifier indicates to use the memory page number corresponding
to the symbol. If the symbol is in a memory expansion part, its
page number is computed by the linker as a number used to map the
page containing the symbol in the memory bank window. If the
symbol is not in a memory expansion part, the page number is 0.

`%hi'
This modifier indicates to use the 8-bit high part of the physical
address of the symbol.

`%lo'
This modifier indicates to use the 8-bit low part of the physical
address of the symbol.

For example a 68HC12 call to a function `foo_example' stored in
memory expansion part could be written as follows:

call %addr(foo_example),%page(foo_example)

and this is equivalent to

call foo_example

And for 68HC11 it could be written as follows:

ldab #%page(foo_example)
stab _page_switch
jsr %addr(foo_example)

File: as.info, Node: M68HC11-Directives, Next: M68HC11-Float, Prev: M68HC11-Modifiers, Up: M68HC11-Dependent

8.19.4 Assembler Directives
---------------------------

The 68HC11 and 68HC12 version of `as' have the following specific
assembler directives:

`.relax'
The relax directive is used by the `GNU Compiler' to emit a
specific relocation to mark a group of instructions for linker
relaxation. The sequence of instructions within the group must be
known to the linker so that relaxation can be performed.

`.mode [mshort|mlong|mshort-double|mlong-double]'
This directive specifies the ABI. It overrides the `-mshort',
`-mlong', `-mshort-double' and `-mlong-double' options.

`.far SYMBOL'
This directive marks the symbol as a `far' symbol meaning that it
uses a `call/rtc' calling convention as opposed to `jsr/rts'.
During a final link, the linker will identify references to the
`far' symbol and will verify the proper calling convention.

`.interrupt SYMBOL'
This directive marks the symbol as an interrupt entry point. This
information is then used by the debugger to correctly unwind the
frame across interrupts.

`.xrefb SYMBOL'
This directive is defined for compatibility with the
`Specification for Motorola 8 and 16-Bit Assembly Language Input
Standard' and is ignored.

File: as.info, Node: M68HC11-Float, Next: M68HC11-opcodes, Prev: M68HC11-Directives, Up: M68HC11-Dependent

8.19.5 Floating Point
---------------------

Packed decimal (P) format floating literals are not supported. Feel
free to add the code!

The floating point formats generated by directives are these.

`.float'
`Single' precision floating point constants.

`.double'
`Double' precision floating point constants.

`.extend'
`.ldouble'
`Extended' precision (`long double') floating point constants.

File: as.info, Node: M68HC11-opcodes, Prev: M68HC11-Float, Up: M68HC11-Dependent

8.19.6 Opcodes
--------------

* Menu:

* M68HC11-Branch:: Branch Improvement

File: as.info, Node: M68HC11-Branch, Up: M68HC11-opcodes

8.19.6.1 Branch Improvement
...........................

Certain pseudo opcodes are permitted for branch instructions. They
expand to the shortest branch instruction that reach the target.
Generally these mnemonics are made by prepending `j' to the start of
Motorola mnemonic. These pseudo opcodes are not affected by the
`--short-branchs' or `--force-long-branchs' options.

The following table summarizes the pseudo-operations.

`jbsr'
`jbra'
These are the simplest jump pseudo-operations; they always map to
one particular machine instruction, depending on the displacement
to the branch target.

`jbXX'
Here, `jbXX' stands for an entire family of pseudo-operations,
where XX is a conditional branch or condition-code test. The full
list of pseudo-ops in this family is:
jbcc jbeq jbge jbgt jbhi jbvs jbpl jblo
jbcs jbne jblt jble jbls jbvc jbmi

For the cases of non-PC relative displacements and long
displacements, `as' issues a longer code fragment in terms of NX,
the opposite condition to XX. For example, for the non-PC
relative case:
jbXX foo
gives
bNXs oof
jmp foo
oof:

File: as.info, Node: M88K-Dependent, Next: MIPS-Dependent, Prev: M68HC11-Dependent, Up: Machine Dependencies

8.20 Motorola M88K Dependent Features
=====================================

* Menu:

* M88K Directives:: M88K Machine Directives

File: as.info, Node: M88K Directives, Up: M88K-Dependent

8.20.1 M88K Machine Directives
------------------------------

The M88K version of the assembler supports the following machine
directives:

`.align'
This directive aligns the section program counter on the next
4-byte boundary.

`.dfloat EXPR'
This assembles a double precision (64-bit) floating point constant.

`.ffloat EXPR'
This assembles a single precision (32-bit) floating point constant.

`.half EXPR'
This directive assembles a half-word (16-bit) constant.

`.word EXPR'
This assembles a word (32-bit) constant.

`.string "STR"'
This directive behaves like the standard `.ascii' directive for
copying STR into the object file. The string is not terminated
with a null byte.

`.set SYMBOL, VALUE'
This directive creates a symbol named SYMBOL which is an alias for
another symbol (possibly not yet defined). This should not be
confused with the mnemonic `set', which is a legitimate M88K
instruction.

`.def SYMBOL, VALUE'
This directive is synonymous with `.set' and is presumably provided
for compatibility with other M88K assemblers.

File: as.info, Node: MIPS-Dependent, Next: MMIX-Dependent, Prev: M88K-Dependent, Up: Machine Dependencies

8.21 MIPS Dependent Features
============================

GNU `as' for MIPS architectures supports several different MIPS
processors, and MIPS ISA levels I through V, MIPS32, and MIPS64. For
information about the MIPS instruction set, see `MIPS RISC
Architecture', by Kane and Heindrich (Prentice-Hall). For an overview
of MIPS assembly conventions, see "Appendix D: Assembly Language
Programming" in the same work.

* Menu:

* MIPS Opts:: Assembler options
* MIPS Object:: ECOFF object code
* MIPS Stabs:: Directives for debugging information
* MIPS ISA:: Directives to override the ISA level
* MIPS symbol sizes:: Directives to override the size of symbols
* MIPS autoextend:: Directives for extending MIPS 16 bit instructions
* MIPS insn:: Directive to mark data as an instruction
* MIPS option stack:: Directives to save and restore options
* MIPS ASE instruction generation overrides:: Directives to control
generation of MIPS ASE instructions

File: as.info, Node: MIPS Opts, Next: MIPS Object, Up: MIPS-Dependent

8.21.1 Assembler options
------------------------

The MIPS configurations of GNU `as' support these special options:

`-G NUM'
This option sets the largest size of an object that can be
referenced implicitly with the `gp' register. It is only accepted
for targets that use ECOFF format. The default value is 8.

`-EB'
`-EL'
Any MIPS configuration of `as' can select big-endian or
little-endian output at run time (unlike the other GNU development
tools, which must be configured for one or the other). Use `-EB'
to select big-endian output, and `-EL' for little-endian.

`-mips1'
`-mips2'
`-mips3'
`-mips4'
`-mips5'
`-mips32'
`-mips32r2'
`-mips64'
`-mips64r2'
Generate code for a particular MIPS Instruction Set Architecture
level. `-mips1' corresponds to the R2000 and R3000 processors,
`-mips2' to the R6000 processor, `-mips3' to the R4000 processor,
and `-mips4' to the R8000 and R10000 processors. `-mips5',
`-mips32', `-mips32r2', `-mips64', and `-mips64r2' correspond to
generic MIPS V, MIPS32, MIPS32 RELEASE 2, MIPS64, and MIPS64
RELEASE 2 ISA processors, respectively. You can also switch
instruction sets during the assembly; see *Note Directives to
override the ISA level: MIPS ISA.

`-mgp32'
`-mfp32'
Some macros have different expansions for 32-bit and 64-bit
registers. The register sizes are normally inferred from the ISA
and ABI, but these flags force a certain group of registers to be
treated as 32 bits wide at all times. `-mgp32' controls the size
of general-purpose registers and `-mfp32' controls the size of
floating-point registers.

On some MIPS variants there is a 32-bit mode flag; when this flag
is set, 64-bit instructions generate a trap. Also, some 32-bit
OSes only save the 32-bit registers on a context switch, so it is
essential never to use the 64-bit registers.

`-mgp64'
Assume that 64-bit general purpose registers are available. This
is provided in the interests of symmetry with -gp32.

`-mips16'
`-no-mips16'
Generate code for the MIPS 16 processor. This is equivalent to
putting `.set mips16' at the start of the assembly file.
`-no-mips16' turns off this option.

`-mips3d'
`-no-mips3d'
Generate code for the MIPS-3D Application Specific Extension.
This tells the assembler to accept MIPS-3D instructions.
`-no-mips3d' turns off this option.

`-mdmx'
`-no-mdmx'
Generate code for the MDMX Application Specific Extension. This
tells the assembler to accept MDMX instructions. `-no-mdmx' turns
off this option.

`-mfix7000'
`-mno-fix7000'
Cause nops to be inserted if the read of the destination register
of an mfhi or mflo instruction occurs in the following two
instructions.

`-mfix-vr4120'
`-no-mfix-vr4120'
Insert nops to work around certain VR4120 errata. This option is
intended to be used on GCC-generated code: it is not designed to
catch all problems in hand-written assembler code.

`-mfix-vr4130'
`-no-mfix-vr4130'
Insert nops to work around the VR4130 `mflo'/`mfhi' errata.

`-m4010'
`-no-m4010'
Generate code for the LSI R4010 chip. This tells the assembler to
accept the R4010 specific instructions (`addciu', `ffc', etc.),
and to not schedule `nop' instructions around accesses to the `HI'
and `LO' registers. `-no-m4010' turns off this option.

`-m4650'
`-no-m4650'
Generate code for the MIPS R4650 chip. This tells the assembler
to accept the `mad' and `madu' instruction, and to not schedule
`nop' instructions around accesses to the `HI' and `LO' registers.
`-no-m4650' turns off this option.

`-m3900'
`-no-m3900'
`-m4100'
`-no-m4100'
For each option `-mNNNN', generate code for the MIPS RNNNN chip.
This tells the assembler to accept instructions specific to that
chip, and to schedule for that chip's hazards.

`-march=CPU'
Generate code for a particular MIPS cpu. It is exactly equivalent
to `-mCPU', except that there are more value of CPU understood.
Valid CPU value are:

2000, 3000, 3900, 4000, 4010, 4100, 4111, vr4120, vr4130,
vr4181, 4300, 4400, 4600, 4650, 5000, rm5200, rm5230, rm5231,
rm5261, rm5721, vr5400, vr5500, 6000, rm7000, 8000, rm9000,
10000, 12000, mips32-4k, sb1

`-mtune=CPU'
Schedule and tune for a particular MIPS cpu. Valid CPU values are
identical to `-march=CPU'.

`-mabi=ABI'
Record which ABI the source code uses. The recognized arguments
are: `32', `n32', `o64', `64' and `eabi'.

`-msym32'
`-mno-sym32'
Equivalent to adding `.set sym32' or `.set nosym32' to the
beginning of the assembler input. *Note MIPS symbol sizes::.

`-nocpp'
This option is ignored. It is accepted for command-line
compatibility with other assemblers, which use it to turn off C
style preprocessing. With GNU `as', there is no need for
`-nocpp', because the GNU assembler itself never runs the C
preprocessor.

`--construct-floats'
`--no-construct-floats'
The `--no-construct-floats' option disables the construction of
double width floating point constants by loading the two halves of
the value into the two single width floating point registers that
make up the double width register. This feature is useful if the
processor support the FR bit in its status register, and this bit
is known (by the programmer) to be set. This bit prevents the
aliasing of the double width register by the single width
registers.

By default `--construct-floats' is selected, allowing construction
of these floating point constants.

`--trap'
`--no-break'
`as' automatically macro expands certain division and
multiplication instructions to check for overflow and division by
zero. This option causes `as' to generate code to take a trap
exception rather than a break exception when an error is detected.
The trap instructions are only supported at Instruction Set
Architecture level 2 and higher.

`--break'
`--no-trap'
Generate code to take a break exception rather than a trap
exception when an error is detected. This is the default.

`-mpdr'
`-mno-pdr'
Control generation of `.pdr' sections. Off by default on IRIX, on
elsewhere.

`-mshared'
`-mno-shared'
When generating code using the Unix calling conventions (selected
by `-KPIC' or `-mcall_shared'), gas will normally generate code
which can go into a shared library. The `-mno-shared' option
tells gas to generate code which uses the calling convention, but
can not go into a shared library. The resulting code is slightly
more efficient. This option only affects the handling of the
`.cpload' and `.cpsetup' pseudo-ops.

File: as.info, Node: MIPS Object, Next: MIPS Stabs, Prev: MIPS Opts, Up: MIPS-Dependent

8.21.2 MIPS ECOFF object code
-----------------------------

Assembling for a MIPS ECOFF target supports some additional sections
besides the usual `.text', `.data' and `.bss'. The additional sections
are `.rdata', used for read-only data, `.sdata', used for small data,
and `.sbss', used for small common objects.

When assembling for ECOFF, the assembler uses the `$gp' (`$28')
register to form the address of a "small object". Any object in the
`.sdata' or `.sbss' sections is considered "small" in this sense. For
external objects, or for objects in the `.bss' section, you can use the
`gcc' `-G' option to control the size of objects addressed via `$gp';
the default value is 8, meaning that a reference to any object eight
bytes or smaller uses `$gp'. Passing `-G 0' to `as' prevents it from
using the `$gp' register on the basis of object size (but the assembler
uses `$gp' for objects in `.sdata' or `sbss' in any case). The size of
an object in the `.bss' section is set by the `.comm' or `.lcomm'
directive that defines it. The size of an external object may be set
with the `.extern' directive. For example, `.extern sym,4' declares
that the object at `sym' is 4 bytes in length, whie leaving `sym'
otherwise undefined.

Using small ECOFF objects requires linker support, and assumes that
the `$gp' register is correctly initialized (normally done
automatically by the startup code). MIPS ECOFF assembly code must not
modify the `$gp' register.

File: as.info, Node: MIPS Stabs, Next: MIPS ISA, Prev: MIPS Object, Up: MIPS-Dependent

8.21.3 Directives for debugging information
-------------------------------------------

MIPS ECOFF `as' supports several directives used for generating
debugging information which are not support by traditional MIPS
assemblers. These are `.def', `.endef', `.dim', `.file', `.scl',
`.size', `.tag', `.type', `.val', `.stabd', `.stabn', and `.stabs'.
The debugging information generated by the three `.stab' directives can
only be read by GDB, not by traditional MIPS debuggers (this
enhancement is required to fully support C++ debugging). These
directives are primarily used by compilers, not assembly language
programmers!

File: as.info, Node: MIPS symbol sizes, Next: MIPS autoextend, Prev: MIPS ISA, Up: MIPS-Dependent

8.21.4 Directives to override the size of symbols
-------------------------------------------------

The n64 ABI allows symbols to have any 64-bit value. Although this
provides a great deal of flexibility, it means that some macros have
much longer expansions than their 32-bit counterparts. For example,
the non-PIC expansion of `dla $4,sym' is usually:

lui $4,%highest(sym)
lui $1,%hi(sym)
daddiu $4,$4,%higher(sym)
daddiu $1,$1,%lo(sym)
dsll32 $4,$4,0
daddu $4,$4,$1

whereas the 32-bit expansion is simply:

lui $4,%hi(sym)
daddiu $4,$4,%lo(sym)

n64 code is sometimes constructed in such a way that all symbolic
constants are known to have 32-bit values, and in such cases, it's
preferable to use the 32-bit expansion instead of the 64-bit expansion.

You can use the `.set sym32' directive to tell the assembler that,
from this point on, all expressions of the form `SYMBOL' or `SYMBOL +
OFFSET' have 32-bit values. For example:

.set sym32
dla $4,sym
lw $4,sym+16
sw $4,sym+0x8000($4)

will cause the assembler to treat `sym', `sym+16' and `sym+0x8000'
as 32-bit values. The handling of non-symbolic addresses is not
affected.

The directive `.set nosym32' ends a `.set sym32' block and reverts
to the normal behavior. It is also possible to change the symbol size
using the command-line options `-msym32' and `-mno-sym32'.

These options and directives are always accepted, but at present,
they have no effect for anything other than n64.

File: as.info, Node: MIPS ISA, Next: MIPS symbol sizes, Prev: MIPS Stabs, Up: MIPS-Dependent

8.21.5 Directives to override the ISA level
-------------------------------------------

GNU `as' supports an additional directive to change the MIPS
Instruction Set Architecture level on the fly: `.set mipsN'. N should
be a number from 0 to 5, or 32, 32r2, 64 or 64r2. The values other
than 0 make the assembler accept instructions for the corresponding ISA
level, from that point on in the assembly. `.set mipsN' affects not
only which instructions are permitted, but also how certain macros are
expanded. `.set mips0' restores the ISA level to its original level:
either the level you selected with command line options, or the default
for your configuration. You can use this feature to permit specific
R4000 instructions while assembling in 32 bit mode. Use this directive
with care!

The directive `.set mips16' puts the assembler into MIPS 16 mode, in
which it will assemble instructions for the MIPS 16 processor. Use
`.set nomips16' to return to normal 32 bit mode.

Traditional MIPS assemblers do not support this directive.

File: as.info, Node: MIPS autoextend, Next: MIPS insn, Prev: MIPS symbol sizes, Up: MIPS-Dependent

8.21.6 Directives for extending MIPS 16 bit instructions
--------------------------------------------------------

By default, MIPS 16 instructions are automatically extended to 32 bits
when necessary. The directive `.set noautoextend' will turn this off.
When `.set noautoextend' is in effect, any 32 bit instruction must be
explicitly extended with the `.e' modifier (e.g., `li.e $4,1000'). The
directive `.set autoextend' may be used to once again automatically
extend instructions when necessary.

This directive is only meaningful when in MIPS 16 mode. Traditional
MIPS assemblers do not support this directive.

File: as.info, Node: MIPS insn, Next: MIPS option stack, Prev: MIPS autoextend, Up: MIPS-Dependent

8.21.7 Directive to mark data as an instruction
-----------------------------------------------

The `.insn' directive tells `as' that the following data is actually
instructions. This makes a difference in MIPS 16 mode: when loading
the address of a label which precedes instructions, `as' automatically
adds 1 to the value, so that jumping to the loaded address will do the
right thing.

File: as.info, Node: MIPS option stack, Next: MIPS ASE instruction generation overrides, Prev: MIPS insn, Up: MIPS-Dependent

8.21.8 Directives to save and restore options
---------------------------------------------

The directives `.set push' and `.set pop' may be used to save and
restore the current settings for all the options which are controlled
by `.set'. The `.set push' directive saves the current settings on a
stack. The `.set pop' directive pops the stack and restores the
settings.

These directives can be useful inside an macro which must change an
option such as the ISA level or instruction reordering but does not want
to change the state of the code which invoked the macro.

Traditional MIPS assemblers do not support these directives.

File: as.info, Node: MIPS ASE instruction generation overrides, Prev: MIPS option stack, Up: MIPS-Dependent

8.21.9 Directives to control generation of MIPS ASE instructions
----------------------------------------------------------------

The directive `.set mips3d' makes the assembler accept instructions
from the MIPS-3D Application Specific Extension from that point on in
the assembly. The `.set nomips3d' directive prevents MIPS-3D
instructions from being accepted.

The directive `.set mdmx' makes the assembler accept instructions
from the MDMX Application Specific Extension from that point on in the
assembly. The `.set nomdmx' directive prevents MDMX instructions from
being accepted.

Traditional MIPS assemblers do not support these directives.

File: as.info, Node: MMIX-Dependent, Next: MSP430-Dependent, Prev: MIPS-Dependent, Up: Machine Dependencies

8.22 MMIX Dependent Features
============================

* Menu:

* MMIX-Opts:: Command-line Options
* MMIX-Expand:: Instruction expansion
* MMIX-Syntax:: Syntax
* MMIX-mmixal:: Differences to `mmixal' syntax and semantics

File: as.info, Node: MMIX-Opts, Next: MMIX-Expand, Up: MMIX-Dependent

8.22.1 Command-line Options
---------------------------

The MMIX version of `as' has some machine-dependent options.

When `--fixed-special-register-names' is specified, only the register
names specified in *Note MMIX-Regs:: are recognized in the instructions
`PUT' and `GET'.

You can use the `--globalize-symbols' to make all symbols global.
This option is useful when splitting up a `mmixal' program into several
files.

The `--gnu-syntax' turns off most syntax compatibility with
`mmixal'. Its usability is currently doubtful.

The `--relax' option is not fully supported, but will eventually make
the object file prepared for linker relaxation.

If you want to avoid inadvertently calling a predefined symbol and
would rather get an error, for example when using `as' with a compiler
or other machine-generated code, specify `--no-predefined-syms'. This
turns off built-in predefined definitions of all such symbols,
including rounding-mode symbols, segment symbols, `BIT' symbols, and
`TRAP' symbols used in `mmix' "system calls". It also turns off
predefined special-register names, except when used in `PUT' and `GET'
instructions.

By default, some instructions are expanded to fit the size of the
operand or an external symbol (*note MMIX-Expand::). By passing
`--no-expand', no such expansion will be done, instead causing errors
at link time if the operand does not fit.

The `mmixal' documentation (*note mmixsite::) specifies that global
registers allocated with the `GREG' directive (*note MMIX-greg::) and
initialized to the same non-zero value, will refer to the same global
register. This isn't strictly enforceable in `as' since the final
addresses aren't known until link-time, but it will do an effort unless
the `--no-merge-gregs' option is specified. (Register merging isn't
yet implemented in `ld'.)

`as' will warn every time it expands an instruction to fit an
operand unless the option `-x' is specified. It is believed that this
behaviour is more useful than just mimicking `mmixal''s behaviour, in
which instructions are only expanded if the `-x' option is specified,
and assembly fails otherwise, when an instruction needs to be expanded.
It needs to be kept in mind that `mmixal' is both an assembler and
linker, while `as' will expand instructions that at link stage can be
contracted. (Though linker relaxation isn't yet implemented in `ld'.)
The option `-x' also imples `--linker-allocated-gregs'.

If instruction expansion is enabled, `as' can expand a `PUSHJ'
instruction into a series of instructions. The shortest expansion is
to not expand it, but just mark the call as redirectable to a stub,
which `ld' creates at link-time, but only if the original `PUSHJ'
instruction is found not to reach the target. The stub consists of the
necessary instructions to form a jump to the target. This happens if
`as' can assert that the `PUSHJ' instruction can reach such a stub.
The option `--no-pushj-stubs' disables this shorter expansion, and the
longer series of instructions is then created at assembly-time. The
option `--no-stubs' is a synonym, intended for compatibility with
future releases, where generation of stubs for other instructions may
be implemented.

Usually a two-operand-expression (*note GREG-base::) without a
matching `GREG' directive is treated as an error by `as'. When the
option `--linker-allocated-gregs' is in effect, they are instead passed
through to the linker, which will allocate as many global registers as
is needed.

File: as.info, Node: MMIX-Expand, Next: MMIX-Syntax, Prev: MMIX-Opts, Up: MMIX-Dependent

8.22.2 Instruction expansion
----------------------------

When `as' encounters an instruction with an operand that is either not
known or does not fit the operand size of the instruction, `as' (and
`ld') will expand the instruction into a sequence of instructions
semantically equivalent to the operand fitting the instruction.
Expansion will take place for the following instructions:

`GETA'
Expands to a sequence of four instructions: `SETL', `INCML',
`INCMH' and `INCH'. The operand must be a multiple of four.

Conditional branches
A branch instruction is turned into a branch with the complemented
condition and prediction bit over five instructions; four
instructions setting `$255' to the operand value, which like with
`GETA' must be a multiple of four, and a final `GO $255,$255,0'.

`PUSHJ'
Similar to expansion for conditional branches; four instructions
set `$255' to the operand value, followed by a `PUSHGO
$255,$255,0'.

`JMP'
Similar to conditional branches and `PUSHJ'. The final instruction
is `GO $255,$255,0'.

The linker `ld' is expected to shrink these expansions for code
assembled with `--relax' (though not currently implemented).

File: as.info, Node: MMIX-Syntax, Next: MMIX-mmixal, Prev: MMIX-Expand, Up: MMIX-Dependent

8.22.3 Syntax
-------------

The assembly syntax is supposed to be upward compatible with that
described in Sections 1.3 and 1.4 of `The Art of Computer Programming,
Volume 1'. Draft versions of those chapters as well as other MMIX
information is located at
`http://www-cs-faculty.stanford.edu/~knuth/mmix-news.html'. Most code
examples from the mmixal package located there should work unmodified
when assembled and linked as single files, with a few noteworthy
exceptions (*note MMIX-mmixal::).

Before an instruction is emitted, the current location is aligned to
the next four-byte boundary. If a label is defined at the beginning of
the line, its value will be the aligned value.

In addition to the traditional hex-prefix `0x', a hexadecimal number
can also be specified by the prefix character `#'.

After all operands to an MMIX instruction or directive have been
specified, the rest of the line is ignored, treated as a comment.

* Menu:

* MMIX-Chars:: Special Characters
* MMIX-Symbols:: Symbols
* MMIX-Regs:: Register Names
* MMIX-Pseudos:: Assembler Directives

File: as.info, Node: MMIX-Chars, Next: MMIX-Symbols, Up: MMIX-Syntax

8.22.3.1 Special Characters
...........................

The characters `*' and `#' are line comment characters; each start a
comment at the beginning of a line, but only at the beginning of a
line. A `#' prefixes a hexadecimal number if found elsewhere on a line.

Two other characters, `%' and `!', each start a comment anywhere on
the line. Thus you can't use the `modulus' and `not' operators in
expressions normally associated with these two characters.

A `;' is a line separator, treated as a new-line, so separate
instructions can be specified on a single line.

File: as.info, Node: MMIX-Symbols, Next: MMIX-Regs, Prev: MMIX-Chars, Up: MMIX-Syntax

8.22.3.2 Symbols
................

The character `:' is permitted in identifiers. There are two
exceptions to it being treated as any other symbol character: if a
symbol begins with `:', it means that the symbol is in the global
namespace and that the current prefix should not be prepended to that
symbol (*note MMIX-prefix::). The `:' is then not considered part of
the symbol. For a symbol in the label position (first on a line), a `:'
at the end of a symbol is silently stripped off. A label is permitted,
but not required, to be followed by a `:', as with many other assembly
formats.

The character `@' in an expression, is a synonym for `.', the
current location.

In addition to the common forward and backward local symbol formats
(*note Symbol Names::), they can be specified with upper-case `B' and
`F', as in `8B' and `9F'. A local label defined for the current
position is written with a `H' appended to the number:
3H LDB $0,$1,2
This and traditional local-label formats cannot be mixed: a label
must be defined and referred to using the same format.

There's a minor caveat: just as for the ordinary local symbols, the
local symbols are translated into ordinary symbols using control
characters are to hide the ordinal number of the symbol.
Unfortunately, these symbols are not translated back in error messages.
Thus you may see confusing error messages when local symbols are used.
Control characters `\003' (control-C) and `\004' (control-D) are used
for the MMIX-specific local-symbol syntax.

The symbol `Main' is handled specially; it is always global.

By defining the symbols `__.MMIX.start..text' and
`__.MMIX.start..data', the address of respectively the `.text' and
`.data' segments of the final program can be defined, though when
linking more than one object file, the code or data in the object file
containing the symbol is not guaranteed to be start at that position;
just the final executable. *Note MMIX-loc::.

File: as.info, Node: MMIX-Regs, Next: MMIX-Pseudos, Prev: MMIX-Symbols, Up: MMIX-Syntax

8.22.3.3 Register names
.......................

Local and global registers are specified as `$0' to `$255'. The
recognized special register names are `rJ', `rA', `rB', `rC', `rD',
`rE', `rF', `rG', `rH', `rI', `rK', `rL', `rM', `rN', `rO', `rP', `rQ',
`rR', `rS', `rT', `rU', `rV', `rW', `rX', `rY', `rZ', `rBB', `rTT',
`rWW', `rXX', `rYY' and `rZZ'. A leading `:' is optional for special
register names.

Local and global symbols can be equated to register names and used in
place of ordinary registers.

Similarly for special registers, local and global symbols can be
used. Also, symbols equated from numbers and constant expressions are
allowed in place of a special register, except when either of the
options `--no-predefined-syms' and `--fixed-special-register-names' are
specified. Then only the special register names above are allowed for
the instructions having a special register operand; `GET' and `PUT'.

File: as.info, Node: MMIX-Pseudos, Prev: MMIX-Regs, Up: MMIX-Syntax

8.22.3.4 Assembler Directives
.............................

`LOC'
The `LOC' directive sets the current location to the value of the
operand field, which may include changing sections. If the
operand is a constant, the section is set to either `.data' if the
value is `0x2000000000000000' or larger, else it is set to `.text'.
Within a section, the current location may only be changed to
monotonically higher addresses. A LOC expression must be a
previously defined symbol or a "pure" constant.

An example, which sets the label PREV to the current location, and
updates the current location to eight bytes forward:
prev LOC @+8

When a LOC has a constant as its operand, a symbol
`__.MMIX.start..text' or `__.MMIX.start..data' is defined
depending on the address as mentioned above. Each such symbol is
interpreted as special by the linker, locating the section at that
address. Note that if multiple files are linked, the first object
file with that section will be mapped to that address (not
necessarily the file with the LOC definition).

`LOCAL'
Example:
LOCAL external_symbol
LOCAL 42
.local asymbol

This directive-operation generates a link-time assertion that the
operand does not correspond to a global register. The operand is
an expression that at link-time resolves to a register symbol or a
number. A number is treated as the register having that number.
There is one restriction on the use of this directive: the
pseudo-directive must be placed in a section with contents, code
or data.

`IS'
The `IS' directive:
asymbol IS an_expression
sets the symbol `asymbol' to `an_expression'. A symbol may not be
set more than once using this directive. Local labels may be set
using this directive, for example:
5H IS @+4

`GREG'
This directive reserves a global register, gives it an initial
value and optionally gives it a symbolic name. Some examples:

areg GREG
breg GREG data_value
GREG data_buffer
.greg creg, another_data_value

The symbolic register name can be used in place of a (non-special)
register. If a value isn't provided, it defaults to zero. Unless
the option `--no-merge-gregs' is specified, non-zero registers
allocated with this directive may be eliminated by `as'; another
register with the same value used in its place. Any of the
instructions `CSWAP', `GO', `LDA', `LDBU', `LDB', `LDHT', `LDOU',
`LDO', `LDSF', `LDTU', `LDT', `LDUNC', `LDVTS', `LDWU', `LDW',
`PREGO', `PRELD', `PREST', `PUSHGO', `STBU', `STB', `STCO', `STHT',
`STOU', `STSF', `STTU', `STT', `STUNC', `SYNCD', `SYNCID', can
have a value nearby an initial value in place of its second and
third operands. Here, "nearby" is defined as within the range
0...255 from the initial value of such an allocated register.

buffer1 BYTE 0,0,0,0,0
buffer2 BYTE 0,0,0,0,0
...
GREG buffer1
LDOU $42,buffer2
In the example above, the `Y' field of the `LDOUI' instruction
(LDOU with a constant Z) will be replaced with the global register
allocated for `buffer1', and the `Z' field will have the value 5,
the offset from `buffer1' to `buffer2'. The result is equivalent
to this code:
buffer1 BYTE 0,0,0,0,0
buffer2 BYTE 0,0,0,0,0
...
tmpreg GREG buffer1
LDOU $42,tmpreg,(buffer2-buffer1)

Global registers allocated with this directive are allocated in
order higher-to-lower within a file. Other than that, the exact
order of register allocation and elimination is undefined. For
example, the order is undefined when more than one file with such
directives are linked together. With the options `-x' and
`--linker-allocated-gregs', `GREG' directives for two-operand
cases like the one mentioned above can be omitted. Sufficient
global registers will then be allocated by the linker.

`BYTE'
The `BYTE' directive takes a series of operands separated by a
comma. If an operand is a string (*note Strings::), each
character of that string is emitted as a byte. Other operands
must be constant expressions without forward references, in the
range 0...255. If you need operands having expressions with
forward references, use `.byte' (*note Byte::). An operand can be
omitted, defaulting to a zero value.

`WYDE'
`TETRA'
`OCTA'
The directives `WYDE', `TETRA' and `OCTA' emit constants of two,
four and eight bytes size respectively. Before anything else
happens for the directive, the current location is aligned to the
respective constant-size boundary. If a label is defined at the
beginning of the line, its value will be that after the alignment.
A single operand can be omitted, defaulting to a zero value
emitted for the directive. Operands can be expressed as strings
(*note Strings::), in which case each character in the string is
emitted as a separate constant of the size indicated by the
directive.

`PREFIX'
The `PREFIX' directive sets a symbol name prefix to be prepended to
all symbols (except local symbols, *note MMIX-Symbols::), that are
not prefixed with `:', until the next `PREFIX' directive. Such
prefixes accumulate. For example,
PREFIX a
PREFIX b
c IS 0
defines a symbol `abc' with the value 0.

`BSPEC'
`ESPEC'
A pair of `BSPEC' and `ESPEC' directives delimit a section of
special contents (without specified semantics). Example:
BSPEC 42
TETRA 1,2,3
ESPEC
The single operand to `BSPEC' must be number in the range 0...255.
The `BSPEC' number 80 is used by the GNU binutils implementation.

File: as.info, Node: MMIX-mmixal, Prev: MMIX-Syntax, Up: MMIX-Dependent

8.22.4 Differences to `mmixal'
------------------------------

The binutils `as' and `ld' combination has a few differences in
function compared to `mmixal' (*note mmixsite::).

The replacement of a symbol with a GREG-allocated register (*note
GREG-base::) is not handled the exactly same way in `as' as in
`mmixal'. This is apparent in the `mmixal' example file `inout.mms',
where different registers with different offsets, eventually yielding
the same address, are used in the first instruction. This type of
difference should however not affect the function of any program unless
it has specific assumptions about the allocated register number.

Line numbers (in the `mmo' object format) are currently not
supported.

Expression operator precedence is not that of mmixal: operator
precedence is that of the C programming language. It's recommended to
use parentheses to explicitly specify wanted operator precedence
whenever more than one type of operators are used.

The serialize unary operator `&', the fractional division operator
`//', the logical not operator `!' and the modulus operator `%' are not
available.

Symbols are not global by default, unless the option
`--globalize-symbols' is passed. Use the `.global' directive to
globalize symbols (*note Global::).

Operand syntax is a bit stricter with `as' than `mmixal'. For
example, you can't say `addu 1,2,3', instead you must write `addu
$1,$2,3'.

You can't LOC to a lower address than those already visited (i.e.
"backwards").

A LOC directive must come before any emitted code.

Predefined symbols are visible as file-local symbols after use. (In
the ELF file, that is--the linked mmo file has no notion of a file-local
symbol.)

Some mapping of constant expressions to sections in LOC expressions
is attempted, but that functionality is easily confused and should be
avoided unless compatibility with `mmixal' is required. A LOC
expression to `0x2000000000000000' or higher, maps to the `.data'
section and lower addresses map to the `.text' section (*note
MMIX-loc::).

The code and data areas are each contiguous. Sparse programs with
far-away LOC directives will take up the same amount of space as a
contiguous program with zeros filled in the gaps between the LOC
directives. If you need sparse programs, you might try and get the
wanted effect with a linker script and splitting up the code parts into
sections (*note Section::). Assembly code for this, to be compatible
with `mmixal', would look something like:
.if 0
LOC away_expression
.else
.section away,"ax"
.fi
`as' will not execute the LOC directive and `mmixal' ignores the
lines with `.'. This construct can be used generally to help
compatibility.

Symbols can't be defined twice-not even to the same value.

Instruction mnemonics are recognized case-insensitive, though the
`IS' and `GREG' pseudo-operations must be specified in upper-case
characters.

There's no unicode support.

The following is a list of programs in `mmix.tar.gz', available at
`http://www-cs-faculty.stanford.edu/~knuth/mmix-news.html', last
checked with the version dated 2001-08-25 (md5sum
c393470cfc86fac040487d22d2bf0172) that assemble with `mmixal' but do
not assemble with `as':

`silly.mms'
LOC to a previous address.

`sim.mms'
Redefines symbol `Done'.

`test.mms'
Uses the serial operator `&'.

File: as.info, Node: MSP430-Dependent, Next: SH-Dependent, Prev: MMIX-Dependent, Up: Machine Dependencies

8.23 MSP 430 Dependent Features
===============================

* Menu:

* MSP430 Options:: Options
* MSP430 Syntax:: Syntax
* MSP430 Floating Point:: Floating Point
* MSP430 Directives:: MSP 430 Machine Directives
* MSP430 Opcodes:: Opcodes
* MSP430 Profiling Capability:: Profiling Capability

File: as.info, Node: MSP430 Options, Next: MSP430 Syntax, Up: MSP430-Dependent

8.23.1 Options
--------------

`as' has only -m flag which selects the mpu arch. Currently has no
effect.

File: as.info, Node: MSP430 Syntax, Next: MSP430 Floating Point, Prev: MSP430 Options, Up: MSP430-Dependent

8.23.2 Syntax
-------------

* Menu:

* MSP430-Macros:: Macros
* MSP430-Chars:: Special Characters
* MSP430-Regs:: Register Names
* MSP430-Ext:: Assembler Extensions

File: as.info, Node: MSP430-Macros, Next: MSP430-Chars, Up: MSP430 Syntax

8.23.2.1 Macros
...............

The macro syntax used on the MSP 430 is like that described in the MSP
430 Family Assembler Specification. Normal `as' macros should still
work.

Additional built-in macros are:

`llo(exp)'
Extracts least significant word from 32-bit expression 'exp'.

`lhi(exp)'
Extracts most significant word from 32-bit expression 'exp'.

`hlo(exp)'
Extracts 3rd word from 64-bit expression 'exp'.

`hhi(exp)'
Extracts 4rd word from 64-bit expression 'exp'.

They normally being used as an immediate source operand.
mov #llo(1), r10 ; == mov #1, r10
mov #lhi(1), r10 ; == mov #0, r10

File: as.info, Node: MSP430-Chars, Next: MSP430-Regs, Prev: MSP430-Macros, Up: MSP430 Syntax

8.23.2.2 Special Characters
...........................

`;' is the line comment character.

The character `$' in jump instructions indicates current location and
implemented only for TI syntax compatibility.

File: as.info, Node: MSP430-Regs, Next: MSP430-Ext, Prev: MSP430-Chars, Up: MSP430 Syntax

8.23.2.3 Register Names
.......................

General-purpose registers are represented by predefined symbols of the
form `rN' (for global registers), where N represents a number between
`0' and `15'. The leading letters may be in either upper or lower
case; for example, `r13' and `R7' are both valid register names.

Register names `PC', `SP' and `SR' cannot be used as register names
and will be treated as variables. Use `r0', `r1', and `r2' instead.

File: as.info, Node: MSP430-Ext, Prev: MSP430-Regs, Up: MSP430 Syntax

8.23.2.4 Assembler Extensions
.............................

`@rN'
As destination operand being treated as `0(rn)'

`0(rN)'
As source operand being treated as `@rn'

`jCOND +N'
Skips next N bytes followed by jump instruction and equivalent to
`jCOND $+N+2'

Also, there are some instructions, which cannot be found in other
assemblers. These are branch instructions, which has different opcodes
upon jump distance. They all got PC relative addressing mode.

`beq label'
A polymorph instruction which is `jeq label' in case if jump
distance within allowed range for cpu's jump instruction. If not,
this unrolls into a sequence of
jne $+6
br label

`bne label'
A polymorph instruction which is `jne label' or `jeq +4; br label'

`blt label'
A polymorph instruction which is `jl label' or `jge +4; br label'

`bltn label'
A polymorph instruction which is `jn label' or `jn +2; jmp +4; br
label'

`bltu label'
A polymorph instruction which is `jlo label' or `jhs +2; br label'

`bge label'
A polymorph instruction which is `jge label' or `jl +4; br label'

`bgeu label'
A polymorph instruction which is `jhs label' or `jlo +4; br label'

`bgt label'
A polymorph instruction which is `jeq +2; jge label' or `jeq +6;
jl +4; br label'

`bgtu label'
A polymorph instruction which is `jeq +2; jhs label' or `jeq +6;
jlo +4; br label'

`bleu label'
A polymorph instruction which is `jeq label; jlo label' or `jeq
+2; jhs +4; br label'

`ble label'
A polymorph instruction which is `jeq label; jl label' or `jeq
+2; jge +4; br label'

`jump label'
A polymorph instruction which is `jmp label' or `br label'

File: as.info, Node: MSP430 Floating Point, Next: MSP430 Directives, Prev: MSP430 Syntax, Up: MSP430-Dependent

8.23.3 Floating Point
---------------------

The MSP 430 family uses IEEE 32-bit floating-point numbers.

File: as.info, Node: MSP430 Directives, Next: MSP430 Opcodes, Prev: MSP430 Floating Point, Up: MSP430-Dependent

8.23.4 MSP 430 Machine Directives
---------------------------------

`.file'
This directive is ignored; it is accepted for compatibility with
other MSP 430 assemblers.

_Warning:_ in other versions of the GNU assembler, `.file' is
used for the directive called `.app-file' in the MSP 430
support.

`.line'
This directive is ignored; it is accepted for compatibility with
other MSP 430 assemblers.

`.arch'
Currently this directive is ignored; it is accepted for
compatibility with other MSP 430 assemblers.

`.profiler'
This directive instructs assembler to add new profile entry to the
object file.

File: as.info, Node: MSP430 Opcodes, Next: MSP430 Profiling Capability, Prev: MSP430 Directives, Up: MSP430-Dependent

8.23.5 Opcodes
--------------

`as' implements all the standard MSP 430 opcodes. No additional
pseudo-instructions are needed on this family.

For information on the 430 machine instruction set, see `MSP430
User's Manual, document slau049b', Texas Instrument, Inc.

File: as.info, Node: MSP430 Profiling Capability, Prev: MSP430 Opcodes, Up: MSP430-Dependent

8.23.6 Profiling Capability
---------------------------

It is a performance hit to use gcc's profiling approach for this tiny
target. Even more - jtag hardware facility does not perform any
profiling functions. However we've got gdb's built-in simulator where
we can do anything.

We define new section `.profiler' which holds all profiling
information. We define new pseudo operation `.profiler' which will
instruct assembler to add new profile entry to the object file. Profile
should take place at the present address.

Pseudo operation format:

`.profiler flags,function_to_profile [, cycle_corrector, extra]'

where:

`flags' is a combination of the following characters:

`s'
function entry

`x'
function exit

`i'
function is in init section

`f'
function is in fini section

`l'
library call

`c'
libc standard call

`d'
stack value demand

`I'
interrupt service routine

`P'
prologue start

`p'
prologue end

`E'
epilogue start

`e'
epilogue end

`j'
long jump / sjlj unwind

`a'
an arbitrary code fragment

`t'
extra parameter saved (a constant value like frame size)

`function_to_profile'
a function address

`cycle_corrector'
a value which should be added to the cycle counter, zero if
omitted.

`extra'
any extra parameter, zero if omitted.

For example:
.global fxx
.type fxx,@function
fxx:
.LFrameOffset_fxx=0x08
.profiler "scdP", fxx ; function entry.
; we also demand stack value to be saved
push r11
push r10
push r9
push r8
.profiler "cdpt",fxx,0, .LFrameOffset_fxx ; check stack value at this point
; (this is a prologue end)
; note, that spare var filled with
; the farme size
mov r15,r8
...
.profiler cdE,fxx ; check stack
pop r8
pop r9
pop r10
pop r11
.profiler xcde,fxx,3 ; exit adds 3 to the cycle counter
ret ; cause 'ret' insn takes 3 cycles

File: as.info, Node: PDP-11-Dependent, Next: PJ-Dependent, Prev: SH64-Dependent, Up: Machine Dependencies

8.24 PDP-11 Dependent Features
==============================

* Menu:

* PDP-11-Options:: Options
* PDP-11-Pseudos:: Assembler Directives
* PDP-11-Syntax:: DEC Syntax versus BSD Syntax
* PDP-11-Mnemonics:: Instruction Naming
* PDP-11-Synthetic:: Synthetic Instructions

File: as.info, Node: PDP-11-Options, Next: PDP-11-Pseudos, Up: PDP-11-Dependent

8.24.1 Options
--------------

The PDP-11 version of `as' has a rich set of machine dependent options.

8.24.1.1 Code Generation Options
................................

`-mpic | -mno-pic'
Generate position-independent (or position-dependent) code.

The default is to generate position-independent code.

8.24.1.2 Instruction Set Extension Options
..........................................

These options enables or disables the use of extensions over the base
line instruction set as introduced by the first PDP-11 CPU: the KA11.
Most options come in two variants: a `-m'EXTENSION that enables
EXTENSION, and a `-mno-'EXTENSION that disables EXTENSION.

The default is to enable all extensions.

`-mall | -mall-extensions'
Enable all instruction set extensions.

`-mno-extensions'
Disable all instruction set extensions.

`-mcis | -mno-cis'
Enable (or disable) the use of the commercial instruction set,
which consists of these instructions: `ADDNI', `ADDN', `ADDPI',
`ADDP', `ASHNI', `ASHN', `ASHPI', `ASHP', `CMPCI', `CMPC',
`CMPNI', `CMPN', `CMPPI', `CMPP', `CVTLNI', `CVTLN', `CVTLPI',
`CVTLP', `CVTNLI', `CVTNL', `CVTNPI', `CVTNP', `CVTPLI', `CVTPL',
`CVTPNI', `CVTPN', `DIVPI', `DIVP', `L2DR', `L3DR', `LOCCI',
`LOCC', `MATCI', `MATC', `MOVCI', `MOVC', `MOVRCI', `MOVRC',
`MOVTCI', `MOVTC', `MULPI', `MULP', `SCANCI', `SCANC', `SKPCI',
`SKPC', `SPANCI', `SPANC', `SUBNI', `SUBN', `SUBPI', and `SUBP'.

`-mcsm | -mno-csm'
Enable (or disable) the use of the `CSM' instruction.

`-meis | -mno-eis'
Enable (or disable) the use of the extended instruction set, which
consists of these instructions: `ASHC', `ASH', `DIV', `MARK',
`MUL', `RTT', `SOB' `SXT', and `XOR'.

`-mfis | -mkev11'
`-mno-fis | -mno-kev11'
Enable (or disable) the use of the KEV11 floating-point
instructions: `FADD', `FDIV', `FMUL', and `FSUB'.

`-mfpp | -mfpu | -mfp-11'
`-mno-fpp | -mno-fpu | -mno-fp-11'
Enable (or disable) the use of FP-11 floating-point instructions:
`ABSF', `ADDF', `CFCC', `CLRF', `CMPF', `DIVF', `LDCFF', `LDCIF',
`LDEXP', `LDF', `LDFPS', `MODF', `MULF', `NEGF', `SETD', `SETF',
`SETI', `SETL', `STCFF', `STCFI', `STEXP', `STF', `STFPS', `STST',
`SUBF', and `TSTF'.

`-mlimited-eis | -mno-limited-eis'
Enable (or disable) the use of the limited extended instruction
set: `MARK', `RTT', `SOB', `SXT', and `XOR'.

The -mno-limited-eis options also implies -mno-eis.

`-mmfpt | -mno-mfpt'
Enable (or disable) the use of the `MFPT' instruction.

`-mmultiproc | -mno-multiproc'
Enable (or disable) the use of multiprocessor instructions:
`TSTSET' and `WRTLCK'.

`-mmxps | -mno-mxps'
Enable (or disable) the use of the `MFPS' and `MTPS' instructions.

`-mspl | -mno-spl'
Enable (or disable) the use of the `SPL' instruction.

Enable (or disable) the use of the microcode instructions: `LDUB',
`MED', and `XFC'.

8.24.1.3 CPU Model Options
..........................

These options enable the instruction set extensions supported by a
particular CPU, and disables all other extensions.

`-mka11'
KA11 CPU. Base line instruction set only.

`-mkb11'
KB11 CPU. Enable extended instruction set and `SPL'.

`-mkd11a'
KD11-A CPU. Enable limited extended instruction set.

`-mkd11b'
KD11-B CPU. Base line instruction set only.

`-mkd11d'
KD11-D CPU. Base line instruction set only.

`-mkd11e'
KD11-E CPU. Enable extended instruction set, `MFPS', and `MTPS'.

`-mkd11f | -mkd11h | -mkd11q'
KD11-F, KD11-H, or KD11-Q CPU. Enable limited extended
instruction set, `MFPS', and `MTPS'.

`-mkd11k'
KD11-K CPU. Enable extended instruction set, `LDUB', `MED',
`MFPS', `MFPT', `MTPS', and `XFC'.

`-mkd11z'
KD11-Z CPU. Enable extended instruction set, `CSM', `MFPS',
`MFPT', `MTPS', and `SPL'.

`-mf11'
F11 CPU. Enable extended instruction set, `MFPS', `MFPT', and
`MTPS'.

`-mj11'
J11 CPU. Enable extended instruction set, `CSM', `MFPS', `MFPT',
`MTPS', `SPL', `TSTSET', and `WRTLCK'.

`-mt11'
T11 CPU. Enable limited extended instruction set, `MFPS', and
`MTPS'.

8.24.1.4 Machine Model Options
..............................

These options enable the instruction set extensions supported by a
particular machine model, and disables all other extensions.

`-m11/03'
Same as `-mkd11f'.

`-m11/04'
Same as `-mkd11d'.

`-m11/05 | -m11/10'
Same as `-mkd11b'.

`-m11/15 | -m11/20'
Same as `-mka11'.

`-m11/21'
Same as `-mt11'.

`-m11/23 | -m11/24'
Same as `-mf11'.

`-m11/34'
Same as `-mkd11e'.

`-m11/34a'
Ame as `-mkd11e' `-mfpp'.

`-m11/35 | -m11/40'
Same as `-mkd11a'.

`-m11/44'
Same as `-mkd11z'.

`-m11/45 | -m11/50 | -m11/55 | -m11/70'
Same as `-mkb11'.

`-m11/53 | -m11/73 | -m11/83 | -m11/84 | -m11/93 | -m11/94'
Same as `-mj11'.

`-m11/60'
Same as `-mkd11k'.

File: as.info, Node: PDP-11-Pseudos, Next: PDP-11-Syntax, Prev: PDP-11-Options, Up: PDP-11-Dependent

8.24.2 Assembler Directives
---------------------------

The PDP-11 version of `as' has a few machine dependent assembler
directives.

`.bss'
Switch to the `bss' section.

`.even'
Align the location counter to an even number.

File: as.info, Node: PDP-11-Syntax, Next: PDP-11-Mnemonics, Prev: PDP-11-Pseudos, Up: PDP-11-Dependent

8.24.3 PDP-11 Assembly Language Syntax
--------------------------------------

`as' supports both DEC syntax and BSD syntax. The only difference is
that in DEC syntax, a `#' character is used to denote an immediate
constants, while in BSD syntax the character for this purpose is `$'.

eneral-purpose registers are named `r0' through `r7'. Mnemonic
alternatives for `r6' and `r7' are `sp' and `pc', respectively.

Floating-point registers are named `ac0' through `ac3', or
alternatively `fr0' through `fr3'.

Comments are started with a `#' or a `/' character, and extend to
the end of the line. (FIXME: clash with immediates?)

File: as.info, Node: PDP-11-Mnemonics, Next: PDP-11-Synthetic, Prev: PDP-11-Syntax, Up: PDP-11-Dependent

8.24.4 Instruction Naming
-------------------------

Some instructions have alternative names.

`BCC'
`BHIS'

`BCS'
`BLO'

`L2DR'
`L2D'

`L3DR'
`L3D'

`SYS'
`TRAP'

File: as.info, Node: PDP-11-Synthetic, Prev: PDP-11-Mnemonics, Up: PDP-11-Dependent

8.24.5 Synthetic Instructions
-----------------------------

The `JBR' and `J'CC synthetic instructions are not supported yet.

File: as.info, Node: PJ-Dependent, Next: PPC-Dependent, Prev: PDP-11-Dependent, Up: Machine Dependencies

8.25 picoJava Dependent Features
================================

* Menu:

* PJ Options:: Options

File: as.info, Node: PJ Options, Up: PJ-Dependent

8.25.1 Options
--------------

`as' has two additional command-line options for the picoJava
architecture.
`-ml'
This option selects little endian data output.

`-mb'
This option selects big endian data output.

File: as.info, Node: PPC-Dependent, Next: Sparc-Dependent, Prev: PJ-Dependent, Up: Machine Dependencies

8.26 PowerPC Dependent Features
===============================

* Menu:

* PowerPC-Opts:: Options
* PowerPC-Pseudo:: PowerPC Assembler Directives

File: as.info, Node: PowerPC-Opts, Next: PowerPC-Pseudo, Up: PPC-Dependent

8.26.1 Options
--------------

The PowerPC chip family includes several successive levels, using the
same core instruction set, but including a few additional instructions
at each level. There are exceptions to this however. For details on
what instructions each variant supports, please see the chip's
architecture reference manual.

The following table lists all available PowerPC options.

`-mpwrx | -mpwr2'
Generate code for POWER/2 (RIOS2).

`-mpwr'
Generate code for POWER (RIOS1)

`-m601'
Generate code for PowerPC 601.

`-mppc, -mppc32, -m603, -m604'
Generate code for PowerPC 603/604.

`-m403, -m405'
Generate code for PowerPC 403/405.

`-m440'
Generate code for PowerPC 440. BookE and some 405 instructions.

`-m7400, -m7410, -m7450, -m7455'
Generate code for PowerPC 7400/7410/7450/7455.

`-mppc64, -m620'
Generate code for PowerPC 620/625/630.

`-mppc64bridge'
Generate code for PowerPC 64, including bridge insns.

`-mbooke64'
Generate code for 64-bit BookE.

`-mbooke, mbooke32'
Generate code for 32-bit BookE.

`-maltivec'
Generate code for processors with AltiVec instructions.

`-mpower4'
Generate code for Power4 architecture.

`-mcom'
Generate code Power/PowerPC common instructions.

`-many'
Generate code for any architecture (PWR/PWRX/PPC).

`-mregnames'
Allow symbolic names for registers.

`-mno-regnames'
Do not allow symbolic names for registers.

`-mrelocatable'
Support for GCC's -mrelocatble option.

`-mrelocatable-lib'
Support for GCC's -mrelocatble-lib option.

`-memb'
Set PPC_EMB bit in ELF flags.

`-mlittle, -mlittle-endian'
Generate code for a little endian machine.

`-mbig, -mbig-endian'
Generate code for a big endian machine.

`-msolaris'
Generate code for Solaris.

`-mno-solaris'
Do not generate code for Solaris.

File: as.info, Node: PowerPC-Pseudo, Prev: PowerPC-Opts, Up: PPC-Dependent

8.26.2 PowerPC Assembler Directives
-----------------------------------

A number of assembler directives are available for PowerPC. The
following table is far from complete.

`.machine "string"'
This directive allows you to change the machine for which code is
generated. `"string"' may be any of the -m cpu selection options
(without the -m) enclosed in double quotes, `"push"', or `"pop"'.
`.machine "push"' saves the currently selected cpu, which may be
restored with `.machine "pop"'.

File: as.info, Node: SH-Dependent, Next: SH64-Dependent, Prev: MSP430-Dependent, Up: Machine Dependencies

8.27 Renesas / SuperH SH Dependent Features
===========================================

* Menu:

* SH Options:: Options
* SH Syntax:: Syntax
* SH Floating Point:: Floating Point
* SH Directives:: SH Machine Directives
* SH Opcodes:: Opcodes

File: as.info, Node: SH Options, Next: SH Syntax, Up: SH-Dependent

8.27.1 Options
--------------

`as' has following command-line options for the Renesas (formerly
Hitachi) / SuperH SH family.

`-little'
Generate little endian code.

`-big'
Generate big endian code.

`-relax'
Alter jump instructions for long displacements.

`-small'
Align sections to 4 byte boundaries, not 16.

`-dsp'
Enable sh-dsp insns, and disable sh3e / sh4 insns.

`-renesas'
Disable optimization with section symbol for compatibility with
Renesas assembler.

`-isa=sh4 | sh4a'
Specify the sh4 or sh4a instruction set.

`-isa=dsp'
Enable sh-dsp insns, and disable sh3e / sh4 insns.

`-isa=fp'
Enable sh2e, sh3e, sh4, and sh4a insn sets.

`-isa=all'
Enable sh1, sh2, sh2e, sh3, sh3e, sh4, sh4a, and sh-dsp insn sets.

File: as.info, Node: SH Syntax, Next: SH Floating Point, Prev: SH Options, Up: SH-Dependent

8.27.2 Syntax
-------------

* Menu:

* SH-Chars:: Special Characters
* SH-Regs:: Register Names
* SH-Addressing:: Addressing Modes

File: as.info, Node: SH-Chars, Next: SH-Regs, Up: SH Syntax

8.27.2.1 Special Characters
...........................

`!' is the line comment character.

You can use `;' instead of a newline to separate statements.

Since `$' has no special meaning, you may use it in symbol names.

File: as.info, Node: SH-Regs, Next: SH-Addressing, Prev: SH-Chars, Up: SH Syntax

8.27.2.2 Register Names
.......................

You can use the predefined symbols `r0', `r1', `r2', `r3', `r4', `r5',
`r6', `r7', `r8', `r9', `r10', `r11', `r12', `r13', `r14', and `r15' to
refer to the SH registers.

The SH also has these control registers:

`pr'
procedure register (holds return address)

`pc'
program counter

`mach'
`macl'
high and low multiply accumulator registers

`sr'
status register

`gbr'
global base register

`vbr'
vector base register (for interrupt vectors)

File: as.info, Node: SH-Addressing, Prev: SH-Regs, Up: SH Syntax

8.27.2.3 Addressing Modes
.........................

`as' understands the following addressing modes for the SH. `RN' in
the following refers to any of the numbered registers, but _not_ the
control registers.

`RN'
Register direct

`@RN'
Register indirect

`@-RN'
Register indirect with pre-decrement

`@RN+'
Register indirect with post-increment

`@(DISP, RN)'
Register indirect with displacement

`@(R0, RN)'
Register indexed

`@(DISP, GBR)'
`GBR' offset

`@(R0, GBR)'
GBR indexed

`ADDR'
`@(DISP, PC)'
PC relative address (for branch or for addressing memory). The
`as' implementation allows you to use the simpler form ADDR
anywhere a PC relative address is called for; the alternate form
is supported for compatibility with other assemblers.

`#IMM'
Immediate data

File: as.info, Node: SH Floating Point, Next: SH Directives, Prev: SH Syntax, Up: SH-Dependent

8.27.3 Floating Point
---------------------

SH2E, SH3E and SH4 groups have on-chip floating-point unit (FPU). Other
SH groups can use `.float' directive to generate IEEE floating-point
numbers.

SH2E and SH3E support single-precision floating point calculations as
well as entirely PCAPI compatible emulation of double-precision
floating point calculations. SH2E and SH3E instructions are a subset of
the floating point calculations conforming to the IEEE754 standard.

In addition to single-precision and double-precision floating-point
operation capability, the on-chip FPU of SH4 has a 128-bit graphic
engine that enables 32-bit floating-point data to be processed 128 bits
at a time. It also supports 4 * 4 array operations and inner product
operations. Also, a superscalar architecture is employed that enables
simultaneous execution of two instructions (including FPU
instructions), providing performance of up to twice that of
conventional architectures at the same frequency.

File: as.info, Node: SH Directives, Next: SH Opcodes, Prev: SH Floating Point, Up: SH-Dependent

8.27.4 SH Machine Directives
----------------------------

`uaword'
`ualong'
`as' will issue a warning when a misaligned `.word' or `.long'
directive is used. You may use `.uaword' or `.ualong' to indicate
that the value is intentionally misaligned.

File: as.info, Node: SH Opcodes, Prev: SH Directives, Up: SH-Dependent

8.27.5 Opcodes
--------------

For detailed information on the SH machine instruction set, see
`SH-Microcomputer User's Manual' (Renesas) or `SH-4 32-bit CPU Core
Architecture' (SuperH) and `SuperH (SH) 64-Bit RISC Series' (SuperH).

`as' implements all the standard SH opcodes. No additional
pseudo-instructions are needed on this family. Note, however, that
because `as' supports a simpler form of PC-relative addressing, you may
simply write (for example)

mov.l bar,r0

where other assemblers might require an explicit displacement to `bar'
from the program counter:

mov.l @(DISP, PC)

Here is a summary of SH opcodes:

Legend:
Rn a numbered register
Rm another numbered register
#imm immediate data
disp displacement
disp8 8-bit displacement
disp12 12-bit displacement

add #imm,Rn lds.l @Rn+,PR
add Rm,Rn mac.w @Rm+,@Rn+
addc Rm,Rn mov #imm,Rn
addv Rm,Rn mov Rm,Rn
and #imm,R0 mov.b Rm,@(R0,Rn)
and Rm,Rn mov.b Rm,@-Rn
and.b #imm,@(R0,GBR) mov.b Rm,@Rn
bf disp8 mov.b @(disp,Rm),R0
bra disp12 mov.b @(disp,GBR),R0
bsr disp12 mov.b @(R0,Rm),Rn
bt disp8 mov.b @Rm+,Rn
clrmac mov.b @Rm,Rn
clrt mov.b R0,@(disp,Rm)
cmp/eq #imm,R0 mov.b R0,@(disp,GBR)
cmp/eq Rm,Rn mov.l Rm,@(disp,Rn)
cmp/ge Rm,Rn mov.l Rm,@(R0,Rn)
cmp/gt Rm,Rn mov.l Rm,@-Rn
cmp/hi Rm,Rn mov.l Rm,@Rn
cmp/hs Rm,Rn mov.l @(disp,Rn),Rm
cmp/pl Rn mov.l @(disp,GBR),R0
cmp/pz Rn mov.l @(disp,PC),Rn
cmp/str Rm,Rn mov.l @(R0,Rm),Rn
div0s Rm,Rn mov.l @Rm+,Rn
div0u mov.l @Rm,Rn
div1 Rm,Rn mov.l R0,@(disp,GBR)
exts.b Rm,Rn mov.w Rm,@(R0,Rn)
exts.w Rm,Rn mov.w Rm,@-Rn
extu.b Rm,Rn mov.w Rm,@Rn
extu.w Rm,Rn mov.w @(disp,Rm),R0
jmp @Rn mov.w @(disp,GBR),R0
jsr @Rn mov.w @(disp,PC),Rn
ldc Rn,GBR mov.w @(R0,Rm),Rn
ldc Rn,SR mov.w @Rm+,Rn
ldc Rn,VBR mov.w @Rm,Rn
ldc.l @Rn+,GBR mov.w R0,@(disp,Rm)
ldc.l @Rn+,SR mov.w R0,@(disp,GBR)
ldc.l @Rn+,VBR mova @(disp,PC),R0
lds Rn,MACH movt Rn
lds Rn,MACL muls Rm,Rn
lds Rn,PR mulu Rm,Rn
lds.l @Rn+,MACH neg Rm,Rn
lds.l @Rn+,MACL negc Rm,Rn

nop stc VBR,Rn
not Rm,Rn stc.l GBR,@-Rn
or #imm,R0 stc.l SR,@-Rn
or Rm,Rn stc.l VBR,@-Rn
or.b #imm,@(R0,GBR) sts MACH,Rn
rotcl Rn sts MACL,Rn
rotcr Rn sts PR,Rn
rotl Rn sts.l MACH,@-Rn
rotr Rn sts.l MACL,@-Rn
rte sts.l PR,@-Rn
rts sub Rm,Rn
sett subc Rm,Rn
shal Rn subv Rm,Rn
shar Rn swap.b Rm,Rn
shll Rn swap.w Rm,Rn
shll16 Rn tas.b @Rn
shll2 Rn trapa #imm
shll8 Rn tst #imm,R0
shlr Rn tst Rm,Rn
shlr16 Rn tst.b #imm,@(R0,GBR)
shlr2 Rn xor #imm,R0
shlr8 Rn xor Rm,Rn
sleep xor.b #imm,@(R0,GBR)
stc GBR,Rn xtrct Rm,Rn
stc SR,Rn

File: as.info, Node: SH64-Dependent, Next: PDP-11-Dependent, Prev: SH-Dependent, Up: Machine Dependencies

8.28 SuperH SH64 Dependent Features
===================================

* Menu:

* SH64 Options:: Options
* SH64 Syntax:: Syntax
* SH64 Directives:: SH64 Machine Directives
* SH64 Opcodes:: Opcodes

File: as.info, Node: SH64 Options, Next: SH64 Syntax, Up: SH64-Dependent

8.28.1 Options
--------------

`-isa=sh4 | sh4a'
Specify the sh4 or sh4a instruction set.

`-isa=dsp'
Enable sh-dsp insns, and disable sh3e / sh4 insns.

`-isa=fp'
Enable sh2e, sh3e, sh4, and sh4a insn sets.

`-isa=all'
Enable sh1, sh2, sh2e, sh3, sh3e, sh4, sh4a, and sh-dsp insn sets.

`-isa=shmedia | -isa=shcompact'
Specify the default instruction set. `SHmedia' specifies the
32-bit opcodes, and `SHcompact' specifies the 16-bit opcodes
compatible with previous SH families. The default depends on the
ABI selected; the default for the 64-bit ABI is SHmedia, and the
default for the 32-bit ABI is SHcompact. If neither the ABI nor
the ISA is specified, the default is 32-bit SHcompact.

Note that the `.mode' pseudo-op is not permitted if the ISA is not
specified on the command line.

`-abi=32 | -abi=64'
Specify the default ABI. If the ISA is specified and the ABI is
not, the default ABI depends on the ISA, with SHmedia defaulting
to 64-bit and SHcompact defaulting to 32-bit.

Note that the `.abi' pseudo-op is not permitted if the ABI is not
specified on the command line. When the ABI is specified on the
command line, any `.abi' pseudo-ops in the source must match it.

`-shcompact-const-crange'
Emit code-range descriptors for constants in SHcompact code
sections.

`-no-mix'
Disallow SHmedia code in the same section as constants and
SHcompact code.

`-no-expand'
Do not expand MOVI, PT, PTA or PTB instructions.

`-expand-pt32'
With -abi=64, expand PT, PTA and PTB instructions to 32 bits only.

File: as.info, Node: SH64 Syntax, Next: SH64 Directives, Prev: SH64 Options, Up: SH64-Dependent

8.28.2 Syntax
-------------

* Menu:

* SH64-Chars:: Special Characters
* SH64-Regs:: Register Names
* SH64-Addressing:: Addressing Modes

File: as.info, Node: SH64-Chars, Next: SH64-Regs, Up: SH64 Syntax

8.28.2.1 Special Characters
...........................

`!' is the line comment character.

You can use `;' instead of a newline to separate statements.

Since `$' has no special meaning, you may use it in symbol names.

File: as.info, Node: SH64-Regs, Next: SH64-Addressing, Prev: SH64-Chars, Up: SH64 Syntax

8.28.2.2 Register Names
.......................

You can use the predefined symbols `r0' through `r63' to refer to the
SH64 general registers, `cr0' through `cr63' for control registers,
`tr0' through `tr7' for target address registers, `fr0' through `fr63'
for single-precision floating point registers, `dr0' through `dr62'
(even numbered registers only) for double-precision floating point
registers, `fv0' through `fv60' (multiples of four only) for
single-precision floating point vectors, `fp0' through `fp62' (even
numbered registers only) for single-precision floating point pairs,
`mtrx0' through `mtrx48' (multiples of 16 only) for 4x4 matrices of
single-precision floating point registers, `pc' for the program
counter, and `fpscr' for the floating point status and control register.

You can also refer to the control registers by the mnemonics `sr',
`ssr', `pssr', `intevt', `expevt', `pexpevt', `tra', `spc', `pspc',
`resvec', `vbr', `tea', `dcr', `kcr0', `kcr1', `ctc', and `usr'.

File: as.info, Node: SH64-Addressing, Prev: SH64-Regs, Up: SH64 Syntax

8.28.2.3 Addressing Modes
.........................

SH64 operands consist of either a register or immediate value. The
immediate value can be a constant or label reference (or portion of a
label reference), as in this example:

movi 4,r2
pt function, tr4
movi (function >> 16) & 65535,r0
shori function & 65535, r0
ld.l r0,4,r0

Instruction label references can reference labels in either SHmedia
or SHcompact. To differentiate between the two, labels in SHmedia
sections will always have the least significant bit set (i.e. they will
be odd), which SHcompact labels will have the least significant bit
reset (i.e. they will be even). If you need to reference the actual
address of a label, you can use the `datalabel' modifier, as in this
example:

.long function
.long datalabel function

In that example, the first longword may or may not have the least
significant bit set depending on whether the label is an SHmedia label
or an SHcompact label. The second longword will be the actual address
of the label, regardless of what type of label it is.

File: as.info, Node: SH64 Directives, Next: SH64 Opcodes, Prev: SH64 Syntax, Up: SH64-Dependent

8.28.3 SH64 Machine Directives
------------------------------

In addition to the SH directives, the SH64 provides the following
directives:

`.mode [shmedia|shcompact]'
`.isa [shmedia|shcompact]'
Specify the ISA for the following instructions (the two directives
are equivalent). Note that programs such as `objdump' rely on
symbolic labels to determine when such mode switches occur (by
checking the least significant bit of the label's address), so
such mode/isa changes should always be followed by a label (in
practice, this is true anyway). Note that you cannot use these
directives if you didn't specify an ISA on the command line.

`.abi [32|64]'
Specify the ABI for the following instructions. Note that you
cannot use this directive unless you specified an ABI on the
command line, and the ABIs specified must match.

`.uaquad'
Like .uaword and .ualong, this allows you to specify an
intentionally unaligned quadword (64 bit word).

File: as.info, Node: SH64 Opcodes, Prev: SH64 Directives, Up: SH64-Dependent

8.28.4 Opcodes
--------------

For detailed information on the SH64 machine instruction set, see
`SuperH 64 bit RISC Series Architecture Manual' (SuperH, Inc.).

`as' implements all the standard SH64 opcodes. In addition, the
following pseudo-opcodes may be expanded into one or more alternate
opcodes:

`movi'
If the value doesn't fit into a standard `movi' opcode, `as' will
replace the `movi' with a sequence of `movi' and `shori' opcodes.

`pt'
This expands to a sequence of `movi' and `shori' opcode, followed
by a `ptrel' opcode, or to a `pta' or `ptb' opcode, depending on
the label referenced.

File: as.info, Node: Sparc-Dependent, Next: TIC54X-Dependent, Prev: PPC-Dependent, Up: Machine Dependencies

8.29 SPARC Dependent Features
=============================

* Menu:

* Sparc-Opts:: Options
* Sparc-Aligned-Data:: Option to enforce aligned data
* Sparc-Float:: Floating Point
* Sparc-Directives:: Sparc Machine Directives

File: as.info, Node: Sparc-Opts, Next: Sparc-Aligned-Data, Up: Sparc-Dependent

8.29.1 Options
--------------

The SPARC chip family includes several successive levels, using the same
core instruction set, but including a few additional instructions at
each level. There are exceptions to this however. For details on what
instructions each variant supports, please see the chip's architecture
reference manual.

By default, `as' assumes the core instruction set (SPARC v6), but
"bumps" the architecture level as needed: it switches to successively
higher architectures as it encounters instructions that only exist in
the higher levels.

If not configured for SPARC v9 (`sparc64-*-*') GAS will not bump
passed sparclite by default, an option must be passed to enable the v9
instructions.

GAS treats sparclite as being compatible with v8, unless an
architecture is explicitly requested. SPARC v9 is always incompatible
with sparclite.

`-Av6 | -Av7 | -Av8 | -Asparclet | -Asparclite'
`-Av8plus | -Av8plusa | -Av9 | -Av9a'
Use one of the `-A' options to select one of the SPARC
architectures explicitly. If you select an architecture
explicitly, `as' reports a fatal error if it encounters an
instruction or feature requiring an incompatible or higher level.

`-Av8plus' and `-Av8plusa' select a 32 bit environment.

`-Av9' and `-Av9a' select a 64 bit environment and are not
available unless GAS is explicitly configured with 64 bit
environment support.

`-Av8plusa' and `-Av9a' enable the SPARC V9 instruction set with
UltraSPARC extensions.

`-xarch=v8plus | -xarch=v8plusa'
For compatibility with the Solaris v9 assembler. These options are
equivalent to -Av8plus and -Av8plusa, respectively.

`-bump'
Warn whenever it is necessary to switch to another level. If an
architecture level is explicitly requested, GAS will not issue
warnings until that level is reached, and will then bump the level
as required (except between incompatible levels).

`-32 | -64'
Select the word size, either 32 bits or 64 bits. These options
are only available with the ELF object file format, and require
that the necessary BFD support has been included.

File: as.info, Node: Sparc-Aligned-Data, Next: Sparc-Float, Prev: Sparc-Opts, Up: Sparc-Dependent

8.29.2 Enforcing aligned data
-----------------------------

SPARC GAS normally permits data to be misaligned. For example, it
permits the `.long' pseudo-op to be used on a byte boundary. However,
the native SunOS and Solaris assemblers issue an error when they see
misaligned data.

You can use the `--enforce-aligned-data' option to make SPARC GAS
also issue an error about misaligned data, just as the SunOS and Solaris
assemblers do.

The `--enforce-aligned-data' option is not the default because gcc
issues misaligned data pseudo-ops when it initializes certain packed
data structures (structures defined using the `packed' attribute). You
may have to assemble with GAS in order to initialize packed data
structures in your own code.

File: as.info, Node: Sparc-Float, Next: Sparc-Directives, Prev: Sparc-Aligned-Data, Up: Sparc-Dependent

8.29.3 Floating Point
---------------------

The Sparc uses IEEE floating-point numbers.

File: as.info, Node: Sparc-Directives, Prev: Sparc-Float, Up: Sparc-Dependent

8.29.4 Sparc Machine Directives
-------------------------------

The Sparc version of `as' supports the following additional machine
directives:

`.align'
This must be followed by the desired alignment in bytes.

`.common'
This must be followed by a symbol name, a positive number, and
`"bss"'. This behaves somewhat like `.comm', but the syntax is
different.

`.half'
This is functionally identical to `.short'.

`.nword'
On the Sparc, the `.nword' directive produces native word sized
value, ie. if assembling with -32 it is equivalent to `.word', if
assembling with -64 it is equivalent to `.xword'.

`.proc'
This directive is ignored. Any text following it on the same line
is also ignored.

`.register'
This directive declares use of a global application or system
register. It must be followed by a register name %g2, %g3, %g6 or
%g7, comma and the symbol name for that register. If symbol name
is `#scratch', it is a scratch register, if it is `#ignore', it
just suppresses any errors about using undeclared global register,
but does not emit any information about it into the object file.
This can be useful e.g. if you save the register before use and
restore it after.

`.reserve'
This must be followed by a symbol name, a positive number, and
`"bss"'. This behaves somewhat like `.lcomm', but the syntax is
different.

`.seg'
This must be followed by `"text"', `"data"', or `"data1"'. It
behaves like `.text', `.data', or `.data 1'.

`.skip'
This is functionally identical to the `.space' directive.

`.word'
On the Sparc, the `.word' directive produces 32 bit values,
instead of the 16 bit values it produces on many other machines.

`.xword'
On the Sparc V9 processor, the `.xword' directive produces 64 bit
values.

File: as.info, Node: TIC54X-Dependent, Next: V850-Dependent, Prev: Sparc-Dependent, Up: Machine Dependencies

8.30 TIC54X Dependent Features
==============================

* Menu:

* TIC54X-Opts:: Command-line Options
* TIC54X-Block:: Blocking
* TIC54X-Env:: Environment Settings
* TIC54X-Constants:: Constants Syntax
* TIC54X-Subsyms:: String Substitution
* TIC54X-Locals:: Local Label Syntax
* TIC54X-Builtins:: Builtin Assembler Math Functions
* TIC54X-Ext:: Extended Addressing Support
* TIC54X-Directives:: Directives
* TIC54X-Macros:: Macro Features
* TIC54X-MMRegs:: Memory-mapped Registers

File: as.info, Node: TIC54X-Opts, Next: TIC54X-Block, Up: TIC54X-Dependent

8.30.1 Options
--------------

The TMS320C54x version of `as' has a few machine-dependent options.

You can use the `-mfar-mode' option to enable extended addressing
mode. All addresses will be assumed to be > 16 bits, and the
appropriate relocation types will be used. This option is equivalent
to using the `.far_mode' directive in the assembly code. If you do not
use the `-mfar-mode' option, all references will be assumed to be 16
bits. This option may be abbreviated to `-mf'.

You can use the `-mcpu' option to specify a particular CPU. This
option is equivalent to using the `.version' directive in the assembly
code. For recognized CPU codes, see *Note `.version':
TIC54X-Directives. The default CPU version is `542'.

You can use the `-merrors-to-file' option to redirect error output
to a file (this provided for those deficient environments which don't
provide adequate output redirection). This option may be abbreviated to
`-me'.

File: as.info, Node: TIC54X-Block, Next: TIC54X-Env, Prev: TIC54X-Opts, Up: TIC54X-Dependent

8.30.2 Blocking
---------------

A blocked section or memory block is guaranteed not to cross the
blocking boundary (usually a page, or 128 words) if it is smaller than
the blocking size, or to start on a page boundary if it is larger than
the blocking size.

File: as.info, Node: TIC54X-Env, Next: TIC54X-Constants, Prev: TIC54X-Block, Up: TIC54X-Dependent

8.30.3 Environment Settings
---------------------------

`C54XDSP_DIR' and `A_DIR' are semicolon-separated paths which are added
to the list of directories normally searched for source and include
files. `C54XDSP_DIR' will override `A_DIR'.

File: as.info, Node: TIC54X-Constants, Next: TIC54X-Subsyms, Prev: TIC54X-Env, Up: TIC54X-Dependent

8.30.4 Constants Syntax
-----------------------

The TIC54X version of `as' allows the following additional constant
formats, using a suffix to indicate the radix:

Binary `000000B, 011000b'
Octal `10Q, 224q'
Hexadecimal `45h, 0FH'

File: as.info, Node: TIC54X-Subsyms, Next: TIC54X-Locals, Prev: TIC54X-Constants, Up: TIC54X-Dependent

8.30.5 String Substitution
--------------------------

A subset of allowable symbols (which we'll call subsyms) may be assigned
arbitrary string values. This is roughly equivalent to C preprocessor
#define macros. When `as' encounters one of these symbols, the symbol
is replaced in the input stream by its string value. Subsym names
*must* begin with a letter.

Subsyms may be defined using the `.asg' and `.eval' directives
(*Note `.asg': TIC54X-Directives, *Note `.eval': TIC54X-Directives.

Expansion is recursive until a previously encountered symbol is
seen, at which point substitution stops.

In this example, x is replaced with SYM2; SYM2 is replaced with
SYM1, and SYM1 is replaced with x. At this point, x has already been
encountered and the substitution stops.

.asg "x",SYM1
.asg "SYM1",SYM2
.asg "SYM2",x
add x,a ; final code assembled is "add x, a"

Macro parameters are converted to subsyms; a side effect of this is
the normal `as' '\ARG' dereferencing syntax is unnecessary. Subsyms
defined within a macro will have global scope, unless the `.var'
directive is used to identify the subsym as a local macro variable
*note `.var': TIC54X-Directives.

Substitution may be forced in situations where replacement might be
ambiguous by placing colons on either side of the subsym. The following
code:

.eval "10",x
LAB:X: add #x, a

When assembled becomes:

LAB10 add #10, a

Smaller parts of the string assigned to a subsym may be accessed with
the following syntax:

``:SYMBOL(CHAR_INDEX):''
Evaluates to a single-character string, the character at
CHAR_INDEX.

``:SYMBOL(START,LENGTH):''
Evaluates to a substring of SYMBOL beginning at START with length
LENGTH.

File: as.info, Node: TIC54X-Locals, Next: TIC54X-Builtins, Prev: TIC54X-Subsyms, Up: TIC54X-Dependent

8.30.6 Local Labels
-------------------

Local labels may be defined in two ways:

* $N, where N is a decimal number between 0 and 9

* LABEL?, where LABEL is any legal symbol name.

Local labels thus defined may be redefined or automatically
generated. The scope of a local label is based on when it may be
undefined or reset. This happens when one of the following situations
is encountered:

* .newblock directive *note `.newblock': TIC54X-Directives.

* The current section is changed (.sect, .text, or .data)

* Entering or leaving an included file

* The macro scope where the label was defined is exited

File: as.info, Node: TIC54X-Builtins, Next: TIC54X-Ext, Prev: TIC54X-Locals, Up: TIC54X-Dependent

8.30.7 Math Builtins
--------------------

The following built-in functions may be used to generate a
floating-point value. All return a floating-point value except `$cvi',
`$int', and `$sgn', which return an integer value.

``$acos(EXPR)''
Returns the floating point arccosine of EXPR.

``$asin(EXPR)''
Returns the floating point arcsine of EXPR.

``$atan(EXPR)''
Returns the floating point arctangent of EXPR.

``$atan2(EXPR1,EXPR2)''
Returns the floating point arctangent of EXPR1 / EXPR2.

``$ceil(EXPR)''
Returns the smallest integer not less than EXPR as floating point.

``$cosh(EXPR)''
Returns the floating point hyperbolic cosine of EXPR.

``$cos(EXPR)''
Returns the floating point cosine of EXPR.

``$cvf(EXPR)''
Returns the integer value EXPR converted to floating-point.

``$cvi(EXPR)''
Returns the floating point value EXPR converted to integer.

``$exp(EXPR)''
Returns the floating point value e ^ EXPR.

``$fabs(EXPR)''
Returns the floating point absolute value of EXPR.

``$floor(EXPR)''
Returns the largest integer that is not greater than EXPR as
floating point.

``$fmod(EXPR1,EXPR2)''
Returns the floating point remainder of EXPR1 / EXPR2.

``$int(EXPR)''
Returns 1 if EXPR evaluates to an integer, zero otherwise.

``$ldexp(EXPR1,EXPR2)''
Returns the floating point value EXPR1 * 2 ^ EXPR2.

``$log10(EXPR)''
Returns the base 10 logarithm of EXPR.

``$log(EXPR)''
Returns the natural logarithm of EXPR.

``$max(EXPR1,EXPR2)''
Returns the floating point maximum of EXPR1 and EXPR2.

``$min(EXPR1,EXPR2)''
Returns the floating point minimum of EXPR1 and EXPR2.

``$pow(EXPR1,EXPR2)''
Returns the floating point value EXPR1 ^ EXPR2.

``$round(EXPR)''
Returns the nearest integer to EXPR as a floating point number.

``$sgn(EXPR)''
Returns -1, 0, or 1 based on the sign of EXPR.

``$sin(EXPR)''
Returns the floating point sine of EXPR.

``$sinh(EXPR)''
Returns the floating point hyperbolic sine of EXPR.

``$sqrt(EXPR)''
Returns the floating point square root of EXPR.

``$tan(EXPR)''
Returns the floating point tangent of EXPR.

``$tanh(EXPR)''
Returns the floating point hyperbolic tangent of EXPR.

``$trunc(EXPR)''
Returns the integer value of EXPR truncated towards zero as
floating point.

File: as.info, Node: TIC54X-Ext, Next: TIC54X-Directives, Prev: TIC54X-Builtins, Up: TIC54X-Dependent

8.30.8 Extended Addressing
--------------------------

The `LDX' pseudo-op is provided for loading the extended addressing bits
of a label or address. For example, if an address `_label' resides in
extended program memory, the value of `_label' may be loaded as follows:
ldx #_label,16,a ; loads extended bits of _label
or #_label,a ; loads lower 16 bits of _label
bacc a ; full address is in accumulator A

File: as.info, Node: TIC54X-Directives, Next: TIC54X-Macros, Prev: TIC54X-Ext, Up: TIC54X-Dependent

8.30.9 Directives
-----------------

`.align [SIZE]'
`.even'
Align the section program counter on the next boundary, based on
SIZE. SIZE may be any power of 2. `.even' is equivalent to
`.align' with a SIZE of 2.
`1'
Align SPC to word boundary

`2'
Align SPC to longword boundary (same as .even)

`128'
Align SPC to page boundary

`.asg STRING, NAME'
Assign NAME the string STRING. String replacement is performed on
STRING before assignment.

`.eval STRING, NAME'
Evaluate the contents of string STRING and assign the result as a
string to the subsym NAME. String replacement is performed on
STRING before assignment.

`.bss SYMBOL, SIZE [, [BLOCKING_FLAG] [,ALIGNMENT_FLAG]]'
Reserve space for SYMBOL in the .bss section. SIZE is in words.
If present, BLOCKING_FLAG indicates the allocated space should be
aligned on a page boundary if it would otherwise cross a page
boundary. If present, ALIGNMENT_FLAG causes the assembler to
allocate SIZE on a long word boundary.

`.byte VALUE [,...,VALUE_N]'
`.ubyte VALUE [,...,VALUE_N]'
`.char VALUE [,...,VALUE_N]'
`.uchar VALUE [,...,VALUE_N]'
Place one or more bytes into consecutive words of the current
section. The upper 8 bits of each word is zero-filled. If a
label is used, it points to the word allocated for the first byte
encountered.

`.clink ["SECTION_NAME"]'
Set STYP_CLINK flag for this section, which indicates to the
linker that if no symbols from this section are referenced, the
section should not be included in the link. If SECTION_NAME is
omitted, the current section is used.

`.c_mode'
TBD.

`.copy "FILENAME" | FILENAME'
`.include "FILENAME" | FILENAME'
Read source statements from FILENAME. The normal include search
path is used. Normally .copy will cause statements from the
included file to be printed in the assembly listing and .include
will not, but this distinction is not currently implemented.

`.data'
Begin assembling code into the .data section.

`.double VALUE [,...,VALUE_N]'
`.ldouble VALUE [,...,VALUE_N]'
`.float VALUE [,...,VALUE_N]'
`.xfloat VALUE [,...,VALUE_N]'
Place an IEEE single-precision floating-point representation of
one or more floating-point values into the current section. All
but `.xfloat' align the result on a longword boundary. Values are
stored most-significant word first.

`.drlist'
`.drnolist'
Control printing of directives to the listing file. Ignored.

`.emsg STRING'
`.mmsg STRING'
`.wmsg STRING'
Emit a user-defined error, message, or warning, respectively.

`.far_mode'
Use extended addressing when assembling statements. This should
appear only once per file, and is equivalent to the -mfar-mode
option *note `-mfar-mode': TIC54X-Opts.

`.fclist'
`.fcnolist'
Control printing of false conditional blocks to the listing file.

`.field VALUE [,SIZE]'
Initialize a bitfield of SIZE bits in the current section. If
VALUE is relocatable, then SIZE must be 16. SIZE defaults to 16
bits. If VALUE does not fit into SIZE bits, the value will be
truncated. Successive `.field' directives will pack starting at
the current word, filling the most significant bits first, and
aligning to the start of the next word if the field size does not
fit into the space remaining in the current word. A `.align'
directive with an operand of 1 will force the next `.field'
directive to begin packing into a new word. If a label is used, it
points to the word that contains the specified field.

`.global SYMBOL [,...,SYMBOL_N]'
`.def SYMBOL [,...,SYMBOL_N]'
`.ref SYMBOL [,...,SYMBOL_N]'
`.def' nominally identifies a symbol defined in the current file
and availalbe to other files. `.ref' identifies a symbol used in
the current file but defined elsewhere. Both map to the standard
`.global' directive.

`.half VALUE [,...,VALUE_N]'
`.uhalf VALUE [,...,VALUE_N]'
`.short VALUE [,...,VALUE_N]'
`.ushort VALUE [,...,VALUE_N]'
`.int VALUE [,...,VALUE_N]'
`.uint VALUE [,...,VALUE_N]'
`.word VALUE [,...,VALUE_N]'
`.uword VALUE [,...,VALUE_N]'
Place one or more values into consecutive words of the current
section. If a label is used, it points to the word allocated for
the first value encountered.

`.label SYMBOL'
Define a special SYMBOL to refer to the load time address of the
current section program counter.

`.length'
`.width'
Set the page length and width of the output listing file. Ignored.

`.list'
`.nolist'
Control whether the source listing is printed. Ignored.

`.long VALUE [,...,VALUE_N]'
`.ulong VALUE [,...,VALUE_N]'
`.xlong VALUE [,...,VALUE_N]'
Place one or more 32-bit values into consecutive words in the
current section. The most significant word is stored first.
`.long' and `.ulong' align the result on a longword boundary;
`xlong' does not.

`.loop [COUNT]'
`.break [CONDITION]'
`.endloop'
Repeatedly assemble a block of code. `.loop' begins the block, and
`.endloop' marks its termination. COUNT defaults to 1024, and
indicates the number of times the block should be repeated.
`.break' terminates the loop so that assembly begins after the
`.endloop' directive. The optional CONDITION will cause the loop
to terminate only if it evaluates to zero.

`MACRO_NAME .macro [PARAM1][,...PARAM_N]'
`[.mexit]'
`.endm'
See the section on macros for more explanation (*Note
TIC54X-Macros::.

`.mlib "FILENAME" | FILENAME'
Load the macro library FILENAME. FILENAME must be an archived
library (BFD ar-compatible) of text files, expected to contain
only macro definitions. The standard include search path is used.

`.mlist'

`.mnolist'
Control whether to include macro and loop block expansions in the
listing output. Ignored.

`.mmregs'
Define global symbolic names for the 'c54x registers. Supposedly
equivalent to executing `.set' directives for each register with
its memory-mapped value, but in reality is provided only for
compatibility and does nothing.

`.newblock'
This directive resets any TIC54X local labels currently defined.
Normal `as' local labels are unaffected.

`.option OPTION_LIST'
Set listing options. Ignored.

`.sblock "SECTION_NAME" | SECTION_NAME [,"NAME_N" | NAME_N]'
Designate SECTION_NAME for blocking. Blocking guarantees that a
section will start on a page boundary (128 words) if it would
otherwise cross a page boundary. Only initialized sections may be
designated with this directive. See also *Note TIC54X-Block::.

`.sect "SECTION_NAME"'
Define a named initialized section and make it the current section.

`SYMBOL .set "VALUE"'
`SYMBOL .equ "VALUE"'
Equate a constant VALUE to a SYMBOL, which is placed in the symbol
table. SYMBOL may not be previously defined.

`.space SIZE_IN_BITS'
`.bes SIZE_IN_BITS'
Reserve the given number of bits in the current section and
zero-fill them. If a label is used with `.space', it points to the
*first* word reserved. With `.bes', the label points to the
*last* word reserved.

`.sslist'
`.ssnolist'
Controls the inclusion of subsym replacement in the listing
output. Ignored.

`.string "STRING" [,...,"STRING_N"]'
`.pstring "STRING" [,...,"STRING_N"]'
Place 8-bit characters from STRING into the current section.
`.string' zero-fills the upper 8 bits of each word, while
`.pstring' puts two characters into each word, filling the
most-significant bits first. Unused space is zero-filled. If a
label is used, it points to the first word initialized.

`[STAG] .struct [OFFSET]'
`[NAME_1] element [COUNT_1]'
`[NAME_2] element [COUNT_2]'
`[TNAME] .tag STAGX [TCOUNT]'
`...'
`[NAME_N] element [COUNT_N]'
`[SSIZE] .endstruct'
`LABEL .tag [STAG]'
Assign symbolic offsets to the elements of a structure. STAG
defines a symbol to use to reference the structure. OFFSET
indicates a starting value to use for the first element
encountered; otherwise it defaults to zero. Each element can have
a named offset, NAME, which is a symbol assigned the value of the
element's offset into the structure. If STAG is missing, these
become global symbols. COUNT adjusts the offset that many times,
as if `element' were an array. `element' may be one of `.byte',
`.word', `.long', `.float', or any equivalent of those, and the
structure offset is adjusted accordingly. `.field' and `.string'
are also allowed; the size of `.field' is one bit, and `.string'
is considered to be one word in size. Only element descriptors,
structure/union tags, `.align' and conditional assembly directives
are allowed within `.struct'/`.endstruct'. `.align' aligns member
offsets to word boundaries only. SSIZE, if provided, will always
be assigned the size of the structure.

The `.tag' directive, in addition to being used to define a
structure/union element within a structure, may be used to apply a
structure to a symbol. Once applied to LABEL, the individual
structure elements may be applied to LABEL to produce the desired
offsets using LABEL as the structure base.

`.tab'
Set the tab size in the output listing. Ignored.

`[UTAG] .union'
`[NAME_1] element [COUNT_1]'
`[NAME_2] element [COUNT_2]'
`[TNAME] .tag UTAGX[,TCOUNT]'
`...'
`[NAME_N] element [COUNT_N]'
`[USIZE] .endstruct'
`LABEL .tag [UTAG]'
Similar to `.struct', but the offset after each element is reset to
zero, and the USIZE is set to the maximum of all defined elements.
Starting offset for the union is always zero.

`[SYMBOL] .usect "SECTION_NAME", SIZE, [,[BLOCKING_FLAG] [,ALIGNMENT_FLAG]]'
Reserve space for variables in a named, uninitialized section
(similar to .bss). `.usect' allows definitions sections
independent of .bss. SYMBOL points to the first location reserved
by this allocation. The symbol may be used as a variable name.
SIZE is the allocated size in words. BLOCKING_FLAG indicates
whether to block this section on a page boundary (128 words)
(*note TIC54X-Block::). ALIGNMENT FLAG indicates whether the
section should be longword-aligned.

`.var SYM[,..., SYM_N]'
Define a subsym to be a local variable within a macro. See *Note
TIC54X-Macros::.

`.version VERSION'
Set which processor to build instructions for. Though the
following values are accepted, the op is ignored.
`541'
`542'
`543'
`545'
`545LP'
`546LP'
`548'
`549'

File: as.info, Node: TIC54X-Macros, Next: TIC54X-MMRegs, Prev: TIC54X-Directives, Up: TIC54X-Dependent

8.30.10 Macros
--------------

Macros do not require explicit dereferencing of arguments (i.e. \ARG).

During macro expansion, the macro parameters are converted to
subsyms. If the number of arguments passed the macro invocation
exceeds the number of parameters defined, the last parameter is
assigned the string equivalent of all remaining arguments. If fewer
arguments are given than parameters, the missing parameters are
assigned empty strings. To include a comma in an argument, you must
enclose the argument in quotes.

The following built-in subsym functions allow examination of the
string value of subsyms (or ordinary strings). The arguments are
strings unless otherwise indicated (subsyms passed as args will be
replaced by the strings they represent).
``$symlen(STR)''
Returns the length of STR.

``$symcmp(STR1,STR2)''
Returns 0 if STR1 == STR2, non-zero otherwise.

``$firstch(STR,CH)''
Returns index of the first occurrence of character constant CH in
STR.

``$lastch(STR,CH)''
Returns index of the last occurrence of character constant CH in
STR.

``$isdefed(SYMBOL)''
Returns zero if the symbol SYMBOL is not in the symbol table,
non-zero otherwise.

``$ismember(SYMBOL,LIST)''
Assign the first member of comma-separated string LIST to SYMBOL;
LIST is reassigned the remainder of the list. Returns zero if
LIST is a null string. Both arguments must be subsyms.

``$iscons(EXPR)''
Returns 1 if string EXPR is binary, 2 if octal, 3 if hexadecimal,
4 if a character, 5 if decimal, and zero if not an integer.

``$isname(NAME)''
Returns 1 if NAME is a valid symbol name, zero otherwise.

``$isreg(REG)''
Returns 1 if REG is a valid predefined register name (AR0-AR7
only).

``$structsz(STAG)''
Returns the size of the structure or union represented by STAG.

``$structacc(STAG)''
Returns the reference point of the structure or union represented
by STAG. Always returns zero.

File: as.info, Node: TIC54X-MMRegs, Prev: TIC54X-Macros, Up: TIC54X-Dependent

8.30.11 Memory-mapped Registers
-------------------------------

The following symbols are recognized as memory-mapped registers:

File: as.info, Node: Z8000-Dependent, Next: Vax-Dependent, Prev: Xtensa-Dependent, Up: Machine Dependencies

8.31 Z8000 Dependent Features
=============================

The Z8000 as supports both members of the Z8000 family: the
unsegmented Z8002, with 16 bit addresses, and the segmented Z8001 with
24 bit addresses.

When the assembler is in unsegmented mode (specified with the
`unsegm' directive), an address takes up one word (16 bit) sized
register. When the assembler is in segmented mode (specified with the
`segm' directive), a 24-bit address takes up a long (32 bit) register.
*Note Assembler Directives for the Z8000: Z8000 Directives, for a list
of other Z8000 specific assembler directives.

* Menu:

* Z8000 Options:: Command-line options for the Z8000
* Z8000 Syntax:: Assembler syntax for the Z8000
* Z8000 Directives:: Special directives for the Z8000
* Z8000 Opcodes:: Opcodes

File: as.info, Node: Z8000 Options, Next: Z8000 Syntax, Up: Z8000-Dependent

8.31.1 Options
--------------

`-z8001'
Generate segmented code by default.

`-z8002'
Generate unsegmented code by default.

File: as.info, Node: Z8000 Syntax, Next: Z8000 Directives, Prev: Z8000 Options, Up: Z8000-Dependent

8.31.2 Syntax
-------------

* Menu:

* Z8000-Chars:: Special Characters
* Z8000-Regs:: Register Names
* Z8000-Addressing:: Addressing Modes

File: as.info, Node: Z8000-Chars, Next: Z8000-Regs, Up: Z8000 Syntax

8.31.2.1 Special Characters
...........................

`!' is the line comment character.

You can use `;' instead of a newline to separate statements.

File: as.info, Node: Z8000-Regs, Next: Z8000-Addressing, Prev: Z8000-Chars, Up: Z8000 Syntax

8.31.2.2 Register Names
.......................

The Z8000 has sixteen 16 bit registers, numbered 0 to 15. You can refer
to different sized groups of registers by register number, with the
prefix `r' for 16 bit registers, `rr' for 32 bit registers and `rq' for
64 bit registers. You can also refer to the contents of the first
eight (of the sixteen 16 bit registers) by bytes. They are named `rlN'
and `rhN'.

_byte registers_
rl0 rh0 rl1 rh1 rl2 rh2 rl3 rh3
rl4 rh4 rl5 rh5 rl6 rh6 rl7 rh7

_word registers_
r0 r1 r2 r3 r4 r5 r6 r7 r8 r9 r10 r11 r12 r13 r14 r15

_long word registers_
rr0 rr2 rr4 rr6 rr8 rr10 rr12 rr14

_quad word registers_
rq0 rq4 rq8 rq12

File: as.info, Node: Z8000-Addressing, Prev: Z8000-Regs, Up: Z8000 Syntax

8.31.2.3 Addressing Modes
.........................

as understands the following addressing modes for the Z8000:

`rlN'
`rhN'
`rN'
`rrN'
`rqN'
Register direct: 8bit, 16bit, 32bit, and 64bit registers.

`@rN'
`@rrN'
Indirect register: @rrN in segmented mode, @rN in unsegmented
mode.

`ADDR'
Direct: the 16 bit or 24 bit address (depending on whether the
assembler is in segmented or unsegmented mode) of the operand is
in the instruction.

`address(rN)'
Indexed: the 16 or 24 bit address is added to the 16 bit register
to produce the final address in memory of the operand.

`rN(#IMM)'
`rrN(#IMM)'
Base Address: the 16 or 24 bit register is added to the 16 bit sign
extended immediate displacement to produce the final address in
memory of the operand.

`rN(rM)'
`rrN(rM)'
Base Index: the 16 or 24 bit register rN or rrN is added to the
sign extended 16 bit index register rM to produce the final
address in memory of the operand.

`#XX'
Immediate data XX.

File: as.info, Node: Z8000 Directives, Next: Z8000 Opcodes, Prev: Z8000 Syntax, Up: Z8000-Dependent

8.31.3 Assembler Directives for the Z8000
-----------------------------------------

The Z8000 port of as includes additional assembler directives, for
compatibility with other Z8000 assemblers. These do not begin with `.'
(unlike the ordinary as directives).

`segm'
`.z8001'
Generate code for the segmented Z8001.

`unsegm'
`.z8002'
Generate code for the unsegmented Z8002.

`name'
Synonym for `.file'

`global'
Synonym for `.global'

`wval'
Synonym for `.word'

`lval'
Synonym for `.long'

`bval'
Synonym for `.byte'

`sval'
Assemble a string. `sval' expects one string literal, delimited by
single quotes. It assembles each byte of the string into
consecutive addresses. You can use the escape sequence `%XX'
(where XX represents a two-digit hexadecimal number) to represent
the character whose ASCII value is XX. Use this feature to
describe single quote and other characters that may not appear in
string literals as themselves. For example, the C statement
`char *a = "he said \"it's 50% off\"";' is represented in Z8000
assembly language (shown with the assembler output in hex at the
left) as

68652073 sval 'he said %22it%27s 50%25 off%22%00'
61696420
22697427
73203530
25206F66
662200

`rsect'
synonym for `.section'

`block'
synonym for `.space'

`even'
special case of `.align'; aligns output to even byte boundary.

File: as.info, Node: Z8000 Opcodes, Prev: Z8000 Directives, Up: Z8000-Dependent

8.31.4 Opcodes
--------------

For detailed information on the Z8000 machine instruction set, see
`Z8000 Technical Manual'.

The following table summarizes the opcodes and their arguments:

rs 16 bit source register
rd 16 bit destination register
rbs 8 bit source register
rbd 8 bit destination register
rrs 32 bit source register
rrd 32 bit destination register
rqs 64 bit source register
rqd 64 bit destination register
addr 16/24 bit address
imm immediate data

adc rd,rs clrb addr cpsir @rd,@rs,rr,cc
adcb rbd,rbs clrb addr(rd) cpsirb @rd,@rs,rr,cc
add rd,@rs clrb rbd dab rbd
add rd,addr com @rd dbjnz rbd,disp7
add rd,addr(rs) com addr dec @rd,imm4m1
add rd,imm16 com addr(rd) dec addr(rd),imm4m1
add rd,rs com rd dec addr,imm4m1
addb rbd,@rs comb @rd dec rd,imm4m1
addb rbd,addr comb addr decb @rd,imm4m1
addb rbd,addr(rs) comb addr(rd) decb addr(rd),imm4m1
addb rbd,imm8 comb rbd decb addr,imm4m1
addb rbd,rbs comflg flags decb rbd,imm4m1
addl rrd,@rs cp @rd,imm16 di i2
addl rrd,addr cp addr(rd),imm16 div rrd,@rs
addl rrd,addr(rs) cp addr,imm16 div rrd,addr
addl rrd,imm32 cp rd,@rs div rrd,addr(rs)
addl rrd,rrs cp rd,addr div rrd,imm16
and rd,@rs cp rd,addr(rs) div rrd,rs
and rd,addr cp rd,imm16 divl rqd,@rs
and rd,addr(rs) cp rd,rs divl rqd,addr
and rd,imm16 cpb @rd,imm8 divl rqd,addr(rs)
and rd,rs cpb addr(rd),imm8 divl rqd,imm32
andb rbd,@rs cpb addr,imm8 divl rqd,rrs
andb rbd,addr cpb rbd,@rs djnz rd,disp7
andb rbd,addr(rs) cpb rbd,addr ei i2
andb rbd,imm8 cpb rbd,addr(rs) ex rd,@rs
andb rbd,rbs cpb rbd,imm8 ex rd,addr
bit @rd,imm4 cpb rbd,rbs ex rd,addr(rs)
bit addr(rd),imm4 cpd rd,@rs,rr,cc ex rd,rs
bit addr,imm4 cpdb rbd,@rs,rr,cc exb rbd,@rs
bit rd,imm4 cpdr rd,@rs,rr,cc exb rbd,addr
bit rd,rs cpdrb rbd,@rs,rr,cc exb rbd,addr(rs)
bitb @rd,imm4 cpi rd,@rs,rr,cc exb rbd,rbs
bitb addr(rd),imm4 cpib rbd,@rs,rr,cc ext0e imm8
bitb addr,imm4 cpir rd,@rs,rr,cc ext0f imm8
bitb rbd,imm4 cpirb rbd,@rs,rr,cc ext8e imm8
bitb rbd,rs cpl rrd,@rs ext8f imm8
bpt cpl rrd,addr exts rrd
call @rd cpl rrd,addr(rs) extsb rd
call addr cpl rrd,imm32 extsl rqd
call addr(rd) cpl rrd,rrs halt
calr disp12 cpsd @rd,@rs,rr,cc in rd,@rs
clr @rd cpsdb @rd,@rs,rr,cc in rd,imm16
clr addr cpsdr @rd,@rs,rr,cc inb rbd,@rs
clr addr(rd) cpsdrb @rd,@rs,rr,cc inb rbd,imm16
clr rd cpsi @rd,@rs,rr,cc inc @rd,imm4m1
clrb @rd cpsib @rd,@rs,rr,cc inc addr(rd),imm4m1
inc addr,imm4m1 ldb rbd,rs(rx) mult rrd,addr(rs)
inc rd,imm4m1 ldb rd(imm16),rbs mult rrd,imm16
incb @rd,imm4m1 ldb rd(rx),rbs mult rrd,rs
incb addr(rd),imm4m1 ldctl ctrl,rs multl rqd,@rs
incb addr,imm4m1 ldctl rd,ctrl multl rqd,addr
incb rbd,imm4m1 ldd @rs,@rd,rr multl rqd,addr(rs)
ind @rd,@rs,ra lddb @rs,@rd,rr multl rqd,imm32
indb @rd,@rs,rba lddr @rs,@rd,rr multl rqd,rrs
inib @rd,@rs,ra lddrb @rs,@rd,rr neg @rd
inibr @rd,@rs,ra ldi @rd,@rs,rr neg addr
iret ldib @rd,@rs,rr neg addr(rd)
jp cc,@rd ldir @rd,@rs,rr neg rd
jp cc,addr ldirb @rd,@rs,rr negb @rd
jp cc,addr(rd) ldk rd,imm4 negb addr
jr cc,disp8 ldl @rd,rrs negb addr(rd)
ld @rd,imm16 ldl addr(rd),rrs negb rbd
ld @rd,rs ldl addr,rrs nop
ld addr(rd),imm16 ldl rd(imm16),rrs or rd,@rs
ld addr(rd),rs ldl rd(rx),rrs or rd,addr
ld addr,imm16 ldl rrd,@rs or rd,addr(rs)
ld addr,rs ldl rrd,addr or rd,imm16
ld rd(imm16),rs ldl rrd,addr(rs) or rd,rs
ld rd(rx),rs ldl rrd,imm32 orb rbd,@rs
ld rd,@rs ldl rrd,rrs orb rbd,addr
ld rd,addr ldl rrd,rs(imm16) orb rbd,addr(rs)
ld rd,addr(rs) ldl rrd,rs(rx) orb rbd,imm8
ld rd,imm16 ldm @rd,rs,n orb rbd,rbs
ld rd,rs ldm addr(rd),rs,n out @rd,rs
ld rd,rs(imm16) ldm addr,rs,n out imm16,rs
ld rd,rs(rx) ldm rd,@rs,n outb @rd,rbs
lda rd,addr ldm rd,addr(rs),n outb imm16,rbs
lda rd,addr(rs) ldm rd,addr,n outd @rd,@rs,ra
lda rd,rs(imm16) ldps @rs outdb @rd,@rs,rba
lda rd,rs(rx) ldps addr outib @rd,@rs,ra
ldar rd,disp16 ldps addr(rs) outibr @rd,@rs,ra
ldb @rd,imm8 ldr disp16,rs pop @rd,@rs
ldb @rd,rbs ldr rd,disp16 pop addr(rd),@rs
ldb addr(rd),imm8 ldrb disp16,rbs pop addr,@rs
ldb addr(rd),rbs ldrb rbd,disp16 pop rd,@rs
ldb addr,imm8 ldrl disp16,rrs popl @rd,@rs
ldb addr,rbs ldrl rrd,disp16 popl addr(rd),@rs
ldb rbd,@rs mbit popl addr,@rs
ldb rbd,addr mreq rd popl rrd,@rs
ldb rbd,addr(rs) mres push @rd,@rs
ldb rbd,imm8 mset push @rd,addr
ldb rbd,rbs mult rrd,@rs push @rd,addr(rs)
ldb rbd,rs(imm16) mult rrd,addr push @rd,imm16
push @rd,rs set addr,imm4 subl rrd,imm32
pushl @rd,@rs set rd,imm4 subl rrd,rrs
pushl @rd,addr set rd,rs tcc cc,rd
pushl @rd,addr(rs) setb @rd,imm4 tccb cc,rbd
pushl @rd,rrs setb addr(rd),imm4 test @rd
res @rd,imm4 setb addr,imm4 test addr
res addr(rd),imm4 setb rbd,imm4 test addr(rd)
res addr,imm4 setb rbd,rs test rd
res rd,imm4 setflg imm4 testb @rd
res rd,rs sinb rbd,imm16 testb addr
resb @rd,imm4 sinb rd,imm16 testb addr(rd)
resb addr(rd),imm4 sind @rd,@rs,ra testb rbd
resb addr,imm4 sindb @rd,@rs,rba testl @rd
resb rbd,imm4 sinib @rd,@rs,ra testl addr
resb rbd,rs sinibr @rd,@rs,ra testl addr(rd)
resflg imm4 sla rd,imm8 testl rrd
ret cc slab rbd,imm8 trdb @rd,@rs,rba
rl rd,imm1or2 slal rrd,imm8 trdrb @rd,@rs,rba
rlb rbd,imm1or2 sll rd,imm8 trib @rd,@rs,rbr
rlc rd,imm1or2 sllb rbd,imm8 trirb @rd,@rs,rbr
rlcb rbd,imm1or2 slll rrd,imm8 trtdrb @ra,@rb,rbr
rldb rbb,rba sout imm16,rs trtib @ra,@rb,rr
rr rd,imm1or2 soutb imm16,rbs trtirb @ra,@rb,rbr
rrb rbd,imm1or2 soutd @rd,@rs,ra trtrb @ra,@rb,rbr
rrc rd,imm1or2 soutdb @rd,@rs,rba tset @rd
rrcb rbd,imm1or2 soutib @rd,@rs,ra tset addr
rrdb rbb,rba soutibr @rd,@rs,ra tset addr(rd)
rsvd36 sra rd,imm8 tset rd
rsvd38 srab rbd,imm8 tsetb @rd
rsvd78 sral rrd,imm8 tsetb addr
rsvd7e srl rd,imm8 tsetb addr(rd)
rsvd9d srlb rbd,imm8 tsetb rbd
rsvd9f srll rrd,imm8 xor rd,@rs
rsvdb9 sub rd,@rs xor rd,addr
rsvdbf sub rd,addr xor rd,addr(rs)
sbc rd,rs sub rd,addr(rs) xor rd,imm16
sbcb rbd,rbs sub rd,imm16 xor rd,rs
sc imm8 sub rd,rs xorb rbd,@rs
sda rd,rs subb rbd,@rs xorb rbd,addr
sdab rbd,rs subb rbd,addr xorb rbd,addr(rs)
sdal rrd,rs subb rbd,addr(rs) xorb rbd,imm8
sdl rd,rs subb rbd,imm8 xorb rbd,rbs
sdlb rbd,rs subb rbd,rbs xorb rbd,rbs
sdll rrd,rs subl rrd,@rs
set @rd,imm4 subl rrd,addr
set addr(rd),imm4 subl rrd,addr(rs)

File: as.info, Node: Vax-Dependent, Prev: Z8000-Dependent, Up: Machine Dependencies

8.32 VAX Dependent Features
===========================

* Menu:

* VAX-Opts:: VAX Command-Line Options
* VAX-float:: VAX Floating Point
* VAX-directives:: Vax Machine Directives
* VAX-opcodes:: VAX Opcodes
* VAX-branch:: VAX Branch Improvement
* VAX-operands:: VAX Operands
* VAX-no:: Not Supported on VAX

File: as.info, Node: VAX-Opts, Next: VAX-float, Up: Vax-Dependent

8.32.1 VAX Command-Line Options
-------------------------------

The Vax version of `as' accepts any of the following options, gives a
warning message that the option was ignored and proceeds. These
options are for compatibility with scripts designed for other people's
assemblers.

``-D' (Debug)'
``-S' (Symbol Table)'
``-T' (Token Trace)'
These are obsolete options used to debug old assemblers.

``-d' (Displacement size for JUMPs)'
This option expects a number following the `-d'. Like options
that expect filenames, the number may immediately follow the `-d'
(old standard) or constitute the whole of the command line
argument that follows `-d' (GNU standard).

``-V' (Virtualize Interpass Temporary File)'
Some other assemblers use a temporary file. This option commanded
them to keep the information in active memory rather than in a
disk file. `as' always does this, so this option is redundant.

``-J' (JUMPify Longer Branches)'
Many 32-bit computers permit a variety of branch instructions to
do the same job. Some of these instructions are short (and fast)
but have a limited range; others are long (and slow) but can
branch anywhere in virtual memory. Often there are 3 flavors of
branch: short, medium and long. Some other assemblers would emit
short and medium branches, unless told by this option to emit
short and long branches.

``-t' (Temporary File Directory)'
Some other assemblers may use a temporary file, and this option
takes a filename being the directory to site the temporary file.
Since `as' does not use a temporary disk file, this option makes
no difference. `-t' needs exactly one filename.

The Vax version of the assembler accepts additional options when
compiled for VMS:

`-h N'
External symbol or section (used for global variables) names are
not case sensitive on VAX/VMS and always mapped to upper case.
This is contrary to the C language definition which explicitly
distinguishes upper and lower case. To implement a standard
conforming C compiler, names must be changed (mapped) to preserve
the case information. The default mapping is to convert all lower
case characters to uppercase and adding an underscore followed by
a 6 digit hex value, representing a 24 digit binary value. The
one digits in the binary value represent which characters are
uppercase in the original symbol name.

The `-h N' option determines how we map names. This takes several
values. No `-h' switch at all allows case hacking as described
above. A value of zero (`-h0') implies names should be upper
case, and inhibits the case hack. A value of 2 (`-h2') implies
names should be all lower case, with no case hack. A value of 3
(`-h3') implies that case should be preserved. The value 1 is
unused. The `-H' option directs `as' to display every mapped
symbol during assembly.

Symbols whose names include a dollar sign `$' are exceptions to the
general name mapping. These symbols are normally only used to
reference VMS library names. Such symbols are always mapped to
upper case.

`-+'
The `-+' option causes `as' to truncate any symbol name larger
than 31 characters. The `-+' option also prevents some code
following the `_main' symbol normally added to make the object
file compatible with Vax-11 "C".

`-1'
This option is ignored for backward compatibility with `as'
version 1.x.

`-H'
The `-H' option causes `as' to print every symbol which was
changed by case mapping.

File: as.info, Node: VAX-float, Next: VAX-directives, Prev: VAX-Opts, Up: Vax-Dependent

8.32.2 VAX Floating Point
-------------------------

Conversion of flonums to floating point is correct, and compatible with
previous assemblers. Rounding is towards zero if the remainder is
exactly half the least significant bit.

`D', `F', `G' and `H' floating point formats are understood.

Immediate floating literals (_e.g._ `S`$6.9') are rendered
correctly. Again, rounding is towards zero in the boundary case.

The `.float' directive produces `f' format numbers. The `.double'
directive produces `d' format numbers.

File: as.info, Node: VAX-directives, Next: VAX-opcodes, Prev: VAX-float, Up: Vax-Dependent

8.32.3 Vax Machine Directives
-----------------------------

The Vax version of the assembler supports four directives for
generating Vax floating point constants. They are described in the
table below.

`.dfloat'
This expects zero or more flonums, separated by commas, and
assembles Vax `d' format 64-bit floating point constants.

`.ffloat'
This expects zero or more flonums, separated by commas, and
assembles Vax `f' format 32-bit floating point constants.

`.gfloat'
This expects zero or more flonums, separated by commas, and
assembles Vax `g' format 64-bit floating point constants.

`.hfloat'
This expects zero or more flonums, separated by commas, and
assembles Vax `h' format 128-bit floating point constants.

File: as.info, Node: VAX-opcodes, Next: VAX-branch, Prev: VAX-directives, Up: Vax-Dependent

8.32.4 VAX Opcodes
------------------

All DEC mnemonics are supported. Beware that `case...' instructions
have exactly 3 operands. The dispatch table that follows the `case...'
instruction should be made with `.word' statements. This is compatible
with all unix assemblers we know of.

File: as.info, Node: VAX-branch, Next: VAX-operands, Prev: VAX-opcodes, Up: Vax-Dependent

8.32.5 VAX Branch Improvement
-----------------------------

Certain pseudo opcodes are permitted. They are for branch
instructions. They expand to the shortest branch instruction that
reaches the target. Generally these mnemonics are made by substituting
`j' for `b' at the start of a DEC mnemonic. This feature is included
both for compatibility and to help compilers. If you do not need this
feature, avoid these opcodes. Here are the mnemonics, and the code
they can expand into.

`jbsb'
`Jsb' is already an instruction mnemonic, so we chose `jbsb'.
(byte displacement)
`bsbb ...'

(word displacement)
`bsbw ...'

(long displacement)
`jsb ...'

`jbr'
`jr'
Unconditional branch.
(byte displacement)
`brb ...'

(word displacement)
`brw ...'

(long displacement)
`jmp ...'

`jCOND'
COND may be any one of the conditional branches `neq', `nequ',
`eql', `eqlu', `gtr', `geq', `lss', `gtru', `lequ', `vc', `vs',
`gequ', `cc', `lssu', `cs'. COND may also be one of the bit tests
`bs', `bc', `bss', `bcs', `bsc', `bcc', `bssi', `bcci', `lbs',
`lbc'. NOTCOND is the opposite condition to COND.
(byte displacement)
`bCOND ...'

(word displacement)
`bNOTCOND foo ; brw ... ; foo:'

(long displacement)
`bNOTCOND foo ; jmp ... ; foo:'

`jacbX'
X may be one of `b d f g h l w'.
(word displacement)
`OPCODE ...'

(long displacement)
OPCODE ..., foo ;
brb bar ;
foo: jmp ... ;
bar:

`jaobYYY'
YYY may be one of `lss leq'.

`jsobZZZ'
ZZZ may be one of `geq gtr'.
(byte displacement)
`OPCODE ...'

(word displacement)
OPCODE ..., foo ;
brb bar ;
foo: brw DESTINATION ;
bar:

(long displacement)
OPCODE ..., foo ;
brb bar ;
foo: jmp DESTINATION ;
bar:

`aobleq'
`aoblss'
`sobgeq'
`sobgtr'

(byte displacement)
`OPCODE ...'

(word displacement)
OPCODE ..., foo ;
brb bar ;
foo: brw DESTINATION ;
bar:

(long displacement)
OPCODE ..., foo ;
brb bar ;
foo: jmp DESTINATION ;
bar:

File: as.info, Node: VAX-operands, Next: VAX-no, Prev: VAX-branch, Up: Vax-Dependent

8.32.6 VAX Operands
-------------------

The immediate character is `$' for Unix compatibility, not `#' as DEC
writes it.

The indirect character is `*' for Unix compatibility, not `@' as DEC
writes it.

The displacement sizing character is ``' (an accent grave) for Unix
compatibility, not `^' as DEC writes it. The letter preceding ``' may
have either case. `G' is not understood, but all other letters (`b i l
s w') are understood.

For instance
tstb *w`$4(r5)

Any expression is permitted in an operand. Operands are comma
separated.

File: as.info, Node: VAX-no, Prev: VAX-operands, Up: Vax-Dependent

8.32.7 Not Supported on VAX
---------------------------

Vax bit fields can not be assembled with `as'. Someone can add the
required code if they really need it.

File: as.info, Node: V850-Dependent, Next: Xtensa-Dependent, Prev: TIC54X-Dependent, Up: Machine Dependencies

8.33 v850 Dependent Features
============================

* Menu:

* V850 Options:: Options
* V850 Syntax:: Syntax
* V850 Floating Point:: Floating Point
* V850 Directives:: V850 Machine Directives
* V850 Opcodes:: Opcodes

File: as.info, Node: V850 Options, Next: V850 Syntax, Up: V850-Dependent

8.33.1 Options
--------------

`as' supports the following additional command-line options for the
V850 processor family:

`-wsigned_overflow'
Causes warnings to be produced when signed immediate values
overflow the space available for then within their opcodes. By
default this option is disabled as it is possible to receive
spurious warnings due to using exact bit patterns as immediate
constants.

`-wunsigned_overflow'
Causes warnings to be produced when unsigned immediate values
overflow the space available for then within their opcodes. By
default this option is disabled as it is possible to receive
spurious warnings due to using exact bit patterns as immediate
constants.

`-mv850'
Specifies that the assembled code should be marked as being
targeted at the V850 processor. This allows the linker to detect
attempts to link such code with code assembled for other
processors.

`-mv850e'
Specifies that the assembled code should be marked as being
targeted at the V850E processor. This allows the linker to detect
attempts to link such code with code assembled for other
processors.

`-mv850e1'
Specifies that the assembled code should be marked as being
targeted at the V850E1 processor. This allows the linker to
detect attempts to link such code with code assembled for other
processors.

`-mv850any'
Specifies that the assembled code should be marked as being
targeted at the V850 processor but support instructions that are
specific to the extended variants of the process. This allows the
production of binaries that contain target specific code, but
which are also intended to be used in a generic fashion. For
example libgcc.a contains generic routines used by the code
produced by GCC for all versions of the v850 architecture,
together with support routines only used by the V850E architecture.

`-mrelax'
Enables relaxation. This allows the .longcall and .longjump pseudo
ops to be used in the assembler source code. These ops label
sections of code which are either a long function call or a long
branch. The assembler will then flag these sections of code and
the linker will attempt to relax them.

File: as.info, Node: V850 Syntax, Next: V850 Floating Point, Prev: V850 Options, Up: V850-Dependent

8.33.2 Syntax
-------------

* Menu:

* V850-Chars:: Special Characters
* V850-Regs:: Register Names

File: as.info, Node: V850-Chars, Next: V850-Regs, Up: V850 Syntax

8.33.2.1 Special Characters
...........................

`#' is the line comment character.

File: as.info, Node: V850-Regs, Prev: V850-Chars, Up: V850 Syntax

8.33.2.2 Register Names
.......................

`as' supports the following names for registers:
`general register 0'
r0, zero

`general register 1'
r1

`general register 2'
r2, hp

`general register 3'
r3, sp

`general register 4'
r4, gp

`general register 5'
r5, tp

`general register 6'
r6

`general register 7'
r7

`general register 8'
r8

`general register 9'
r9

`general register 10'
r10

`general register 11'
r11

`general register 12'
r12

`general register 13'
r13

`general register 14'
r14

`general register 15'
r15

`general register 16'
r16

`general register 17'
r17

`general register 18'
r18

`general register 19'
r19

`general register 20'
r20

`general register 21'
r21

`general register 22'
r22

`general register 23'
r23

`general register 24'
r24

`general register 25'
r25

`general register 26'
r26

`general register 27'
r27

`general register 28'
r28

`general register 29'
r29

`general register 30'
r30, ep

`general register 31'
r31, lp

`system register 0'
eipc

`system register 1'
eipsw

`system register 2'
fepc

`system register 3'
fepsw

`system register 4'
ecr

`system register 5'
psw

`system register 16'
ctpc

`system register 17'
ctpsw

`system register 18'
dbpc

`system register 19'
dbpsw

`system register 20'
ctbp

File: as.info, Node: V850 Floating Point, Next: V850 Directives, Prev: V850 Syntax, Up: V850-Dependent

8.33.3 Floating Point
---------------------

The V850 family uses IEEE floating-point numbers.

File: as.info, Node: V850 Directives, Next: V850 Opcodes, Prev: V850 Floating Point, Up: V850-Dependent

8.33.4 V850 Machine Directives
------------------------------

`.offset <EXPRESSION>'
Moves the offset into the current section to the specified amount.

`.section "name", <type>'
This is an extension to the standard .section directive. It sets
the current section to be <type> and creates an alias for this
section called "name".

`.v850'
Specifies that the assembled code should be marked as being
targeted at the V850 processor. This allows the linker to detect
attempts to link such code with code assembled for other
processors.

`.v850e'
Specifies that the assembled code should be marked as being
targeted at the V850E processor. This allows the linker to detect
attempts to link such code with code assembled for other
processors.

`.v850e1'
Specifies that the assembled code should be marked as being
targeted at the V850E1 processor. This allows the linker to
detect attempts to link such code with code assembled for other
processors.

File: as.info, Node: V850 Opcodes, Prev: V850 Directives, Up: V850-Dependent

8.33.5 Opcodes
--------------

`as' implements all the standard V850 opcodes.

`as' also implements the following pseudo ops:

`hi0()'
Computes the higher 16 bits of the given expression and stores it
into the immediate operand field of the given instruction. For
example:

`mulhi hi0(here - there), r5, r6'

computes the difference between the address of labels 'here' and
'there', takes the upper 16 bits of this difference, shifts it
down 16 bits and then mutliplies it by the lower 16 bits in
register 5, putting the result into register 6.

`lo()'
Computes the lower 16 bits of the given expression and stores it
into the immediate operand field of the given instruction. For
example:

`addi lo(here - there), r5, r6'

computes the difference between the address of labels 'here' and
'there', takes the lower 16 bits of this difference and adds it to
register 5, putting the result into register 6.

`hi()'
Computes the higher 16 bits of the given expression and then adds
the value of the most significant bit of the lower 16 bits of the
expression and stores the result into the immediate operand field
of the given instruction. For example the following code can be
used to compute the address of the label 'here' and store it into
register 6:

`movhi hi(here), r0, r6' `movea lo(here), r6, r6'

The reason for this special behaviour is that movea performs a sign
extension on its immediate operand. So for example if the address
of 'here' was 0xFFFFFFFF then without the special behaviour of the
hi() pseudo-op the movhi instruction would put 0xFFFF0000 into r6,
then the movea instruction would takes its immediate operand,
0xFFFF, sign extend it to 32 bits, 0xFFFFFFFF, and then add it
into r6 giving 0xFFFEFFFF which is wrong (the fifth nibble is E).
With the hi() pseudo op adding in the top bit of the lo() pseudo
op, the movhi instruction actually stores 0 into r6 (0xFFFF + 1 =
0x0000), so that the movea instruction stores 0xFFFFFFFF into r6 -
the right value.

`hilo()'
Computes the 32 bit value of the given expression and stores it
into the immediate operand field of the given instruction (which
must be a mov instruction). For example:

`mov hilo(here), r6'

computes the absolute address of label 'here' and puts the result
into register 6.

`sdaoff()'
Computes the offset of the named variable from the start of the
Small Data Area (whoes address is held in register 4, the GP
register) and stores the result as a 16 bit signed value in the
immediate operand field of the given instruction. For example:

`ld.w sdaoff(_a_variable)[gp],r6'

loads the contents of the location pointed to by the label
'_a_variable' into register 6, provided that the label is located
somewhere within +/- 32K of the address held in the GP register.
[Note the linker assumes that the GP register contains a fixed
address set to the address of the label called '__gp'. This can
either be set up automatically by the linker, or specifically set
by using the `--defsym __gp=<value>' command line option].

`tdaoff()'
Computes the offset of the named variable from the start of the
Tiny Data Area (whoes address is held in register 30, the EP
register) and stores the result as a 4,5, 7 or 8 bit unsigned
value in the immediate operand field of the given instruction.
For example:

`sld.w tdaoff(_a_variable)[ep],r6'

loads the contents of the location pointed to by the label
'_a_variable' into register 6, provided that the label is located
somewhere within +256 bytes of the address held in the EP
register. [Note the linker assumes that the EP register contains
a fixed address set to the address of the label called '__ep'.
This can either be set up automatically by the linker, or
specifically set by using the `--defsym __ep=<value>' command line
option].

`zdaoff()'
Computes the offset of the named variable from address 0 and
stores the result as a 16 bit signed value in the immediate
operand field of the given instruction. For example:

`movea zdaoff(_a_variable),zero,r6'

puts the address of the label '_a_variable' into register 6,
assuming that the label is somewhere within the first 32K of
memory. (Strictly speaking it also possible to access the last
32K of memory as well, as the offsets are signed).

`ctoff()'
Computes the offset of the named variable from the start of the
Call Table Area (whoes address is helg in system register 20, the
CTBP register) and stores the result a 6 or 16 bit unsigned value
in the immediate field of then given instruction or piece of data.
For example:

`callt ctoff(table_func1)'

will put the call the function whoes address is held in the call
table at the location labeled 'table_func1'.

`.longcall `name''
Indicates that the following sequence of instructions is a long
call to function `name'. The linker will attempt to shorten this
call sequence if `name' is within a 22bit offset of the call. Only
valid if the `-mrelax' command line switch has been enabled.

`.longjump `name''
Indicates that the following sequence of instructions is a long
jump to label `name'. The linker will attempt to shorten this code
sequence if `name' is within a 22bit offset of the jump. Only
valid if the `-mrelax' command line switch has been enabled.

For information on the V850 instruction set, see `V850 Family
32-/16-Bit single-Chip Microcontroller Architecture Manual' from NEC.
Ltd.

File: as.info, Node: Xtensa-Dependent, Next: Z8000-Dependent, Prev: V850-Dependent, Up: Machine Dependencies

8.34 Xtensa Dependent Features
==============================

This chapter covers features of the GNU assembler that are specific
to the Xtensa architecture. For details about the Xtensa instruction
set, please consult the `Xtensa Instruction Set Architecture (ISA)
Reference Manual'.

* Menu:

* Xtensa Options:: Command-line Options.
* Xtensa Syntax:: Assembler Syntax for Xtensa Processors.
* Xtensa Optimizations:: Assembler Optimizations.
* Xtensa Relaxation:: Other Automatic Transformations.
* Xtensa Directives:: Directives for Xtensa Processors.

File: as.info, Node: Xtensa Options, Next: Xtensa Syntax, Up: Xtensa-Dependent

8.34.1 Command Line Options
---------------------------

The Xtensa version of the GNU assembler supports these special options:

`--text-section-literals | --no-text-section-literals'
Control the treatment of literal pools. The default is
`--no-text-section-literals', which places literals in a separate
section in the output file. This allows the literal pool to be
placed in a data RAM/ROM. With `--text-section-literals', the
literals are interspersed in the text section in order to keep
them as close as possible to their references. This may be
necessary for large assembly files, where the literals would
otherwise be out of range of the `L32R' instructions in the text
section. These options only affect literals referenced via
PC-relative `L32R' instructions; literals for absolute mode `L32R'
instructions are handled separately.

`--absolute-literals | --no-absolute-literals'
Indicate to the assembler whether `L32R' instructions use absolute
or PC-relative addressing. If the processor includes the absolute
addressing option, the default is to use absolute `L32R'
relocations. Otherwise, only the PC-relative `L32R' relocations
can be used.

`--target-align | --no-target-align'
Enable or disable automatic alignment to reduce branch penalties
at some expense in code size. *Note Automatic Instruction
Alignment: Xtensa Automatic Alignment. This optimization is
enabled by default. Note that the assembler will always align
instructions like `LOOP' that have fixed alignment requirements.

`--longcalls | --no-longcalls'
Enable or disable transformation of call instructions to allow
calls across a greater range of addresses. *Note Function Call
Relaxation: Xtensa Call Relaxation. This option should be used
when call targets can potentially be out of range. It may degrade
both code size and performance, but the linker can generally
optimize away the unnecessary overhead when a call ends up within
range. The default is `--no-longcalls'.

`--transform | --no-transform'
Enable or disable all assembler transformations of Xtensa
instructions, including both relaxation and optimization. The
default is `--transform'; `--no-transform' should only be used in
the rare cases when the instructions must be exactly as specified
in the assembly source. Using `--no-transform' causes out of range
instruction operands to be errors.

`--rename-section OLDNAME=NEWNAME'
Rename the OLDNAME section to NEWNAME. This option can be used
multiple times to rename multiple sections.

File: as.info, Node: Xtensa Syntax, Next: Xtensa Optimizations, Prev: Xtensa Options, Up: Xtensa-Dependent

8.34.2 Assembler Syntax
-----------------------

Block comments are delimited by `/*' and `*/'. End of line comments
may be introduced with either `#' or `//'.

Instructions consist of a leading opcode or macro name followed by
whitespace and an optional comma-separated list of operands:

OPCODE [OPERAND, ...]

Instructions must be separated by a newline or semicolon.

FLIX instructions, which bundle multiple opcodes together in a single
instruction, are specified by enclosing the bundled opcodes inside
braces:

{
[FORMAT]
OPCODE0 [OPERANDS]
OPCODE1 [OPERANDS]
OPCODE2 [OPERANDS]
...
}

The opcodes in a FLIX instruction are listed in the same order as the
corresponding instruction slots in the TIE format declaration.
Directives and labels are not allowed inside the braces of a FLIX
instruction. A particular TIE format name can optionally be specified
immediately after the opening brace, but this is usually unnecessary.
The assembler will automatically search for a format that can encode the
specified opcodes, so the format name need only be specified in rare
cases where there is more than one applicable format and where it
matters which of those formats is used. A FLIX instruction can also be
specified on a single line by separating the opcodes with semicolons:

{ [FORMAT;] OPCODE0 [OPERANDS]; OPCODE1 [OPERANDS]; OPCODE2 [OPERANDS]; ... }

The assembler can automatically bundle opcodes into FLIX
instructions. It encodes the opcodes in order, one at a time, choosing
the smallest format where each opcode can be encoded and filling unused
instruction slots with no-ops.

* Menu:

* Xtensa Opcodes:: Opcode Naming Conventions.
* Xtensa Registers:: Register Naming.

File: as.info, Node: Xtensa Opcodes, Next: Xtensa Registers, Up: Xtensa Syntax

8.34.2.1 Opcode Names
.....................

See the `Xtensa Instruction Set Architecture (ISA) Reference Manual'
for a complete list of opcodes and descriptions of their semantics.

If an opcode name is prefixed with an underscore character (`_'),
`as' will not transform that instruction in any way. The underscore
prefix disables both optimization (*note Xtensa Optimizations: Xtensa
Optimizations.) and relaxation (*note Xtensa Relaxation: Xtensa
Relaxation.) for that particular instruction. Only use the underscore
prefix when it is essential to select the exact opcode produced by the
assembler. Using this feature unnecessarily makes the code less
efficient by disabling assembler optimization and less flexible by
disabling relaxation.

Note that this special handling of underscore prefixes only applies
to Xtensa opcodes, not to either built-in macros or user-defined macros.
When an underscore prefix is used with a macro (e.g., `_MOV'), it
refers to a different macro. The assembler generally provides built-in
macros both with and without the underscore prefix, where the underscore
versions behave as if the underscore carries through to the instructions
in the macros. For example, `_MOV' may expand to `_MOV.N'.

The underscore prefix only applies to individual instructions, not to
series of instructions. For example, if a series of instructions have
underscore prefixes, the assembler will not transform the individual
instructions, but it may insert other instructions between them (e.g.,
to align a `LOOP' instruction). To prevent the assembler from
modifying a series of instructions as a whole, use the `no-transform'
directive. *Note transform: Transform Directive.

File: as.info, Node: Xtensa Registers, Prev: Xtensa Opcodes, Up: Xtensa Syntax

8.34.2.2 Register Names
.......................

The assembly syntax for a register file entry is the "short" name for a
TIE register file followed by the index into that register file. For
example, the general-purpose `AR' register file has a short name of
`a', so these registers are named `a0'...`a15'. As a special feature,
`sp' is also supported as a synonym for `a1'. Additional registers may
be added by processor configuration options and by designer-defined TIE
extensions. An initial `$' character is optional in all register names.

File: as.info, Node: Xtensa Optimizations, Next: Xtensa Relaxation, Prev: Xtensa Syntax, Up: Xtensa-Dependent

8.34.3 Xtensa Optimizations
---------------------------

The optimizations currently supported by `as' are generation of density
instructions where appropriate and automatic branch target alignment.

* Menu:

* Density Instructions:: Using Density Instructions.
* Xtensa Automatic Alignment:: Automatic Instruction Alignment.

File: as.info, Node: Density Instructions, Next: Xtensa Automatic Alignment, Up: Xtensa Optimizations

8.34.3.1 Using Density Instructions
...................................

The Xtensa instruction set has a code density option that provides
16-bit versions of some of the most commonly used opcodes. Use of these
opcodes can significantly reduce code size. When possible, the
assembler automatically translates instructions from the core Xtensa
instruction set into equivalent instructions from the Xtensa code
density option. This translation can be disabled by using underscore
prefixes (*note Opcode Names: Xtensa Opcodes.), by using the
`--no-transform' command-line option (*note Command Line Options:
Xtensa Options.), or by using the `no-transform' directive (*note
transform: Transform Directive.).

It is a good idea _not_ to use the density instructions directly.
The assembler will automatically select dense instructions where
possible. If you later need to use an Xtensa processor without the code
density option, the same assembly code will then work without
modification.

File: as.info, Node: Xtensa Automatic Alignment, Prev: Density Instructions, Up: Xtensa Optimizations

8.34.3.2 Automatic Instruction Alignment
........................................

The Xtensa assembler will automatically align certain instructions, both
to optimize performance and to satisfy architectural requirements.

As an optimization to improve performance, the assembler attempts to
align branch targets so they do not cross instruction fetch boundaries.
(Xtensa processors can be configured with either 32-bit or 64-bit
instruction fetch widths.) An instruction immediately following a call
is treated as a branch target in this context, because it will be the
target of a return from the call. This alignment has the potential to
reduce branch penalties at some expense in code size. The assembler
will not attempt to align labels with the prefixes `.Ln' and `.LM',
since these labels are used for debugging information and are not
typically branch targets. This optimization is enabled by default.
You can disable it with the `--no-target-align' command-line option
(*note Command Line Options: Xtensa Options.).

The target alignment optimization is done without adding instructions
that could increase the execution time of the program. If there are
density instructions in the code preceding a target, the assembler can
change the target alignment by widening some of those instructions to
the equivalent 24-bit instructions. Extra bytes of padding can be
inserted immediately following unconditional jump and return
instructions. This approach is usually successful in aligning many,
but not all, branch targets.

The `LOOP' family of instructions must be aligned such that the
first instruction in the loop body does not cross an instruction fetch
boundary (e.g., with a 32-bit fetch width, a `LOOP' instruction must be
on either a 1 or 2 mod 4 byte boundary). The assembler knows about
this restriction and inserts the minimal number of 2 or 3 byte no-op
instructions to satisfy it. When no-op instructions are added, any
label immediately preceding the original loop will be moved in order to
refer to the loop instruction, not the newly generated no-op
instruction. To preserve binary compatibility across processors with
different fetch widths, the assembler conservatively assumes a 32-bit
fetch width when aligning `LOOP' instructions (except if the first
instruction in the loop is a 64-bit instruction).

Similarly, the `ENTRY' instruction must be aligned on a 0 mod 4 byte
boundary. The assembler satisfies this requirement by inserting zero
bytes when required. In addition, labels immediately preceding the
`ENTRY' instruction will be moved to the newly aligned instruction
location.

File: as.info, Node: Xtensa Relaxation, Next: Xtensa Directives, Prev: Xtensa Optimizations, Up: Xtensa-Dependent

8.34.4 Xtensa Relaxation
------------------------

When an instruction operand is outside the range allowed for that
particular instruction field, `as' can transform the code to use a
functionally-equivalent instruction or sequence of instructions. This
process is known as "relaxation". This is typically done for branch
instructions because the distance of the branch targets is not known
until assembly-time. The Xtensa assembler offers branch relaxation and
also extends this concept to function calls, `MOVI' instructions and
other instructions with immediate fields.

* Menu:

* Xtensa Branch Relaxation:: Relaxation of Branches.
* Xtensa Call Relaxation:: Relaxation of Function Calls.
* Xtensa Immediate Relaxation:: Relaxation of other Immediate Fields.

File: as.info, Node: Xtensa Branch Relaxation, Next: Xtensa Call Relaxation, Up: Xtensa Relaxation

8.34.4.1 Conditional Branch Relaxation
......................................

When the target of a branch is too far away from the branch itself,
i.e., when the offset from the branch to the target is too large to fit
in the immediate field of the branch instruction, it may be necessary to
replace the branch with a branch around a jump. For example,

beqz a2, L

may result in:

bnez.n a2, M
j L
M:

(The `BNEZ.N' instruction would be used in this example only if the
density option is available. Otherwise, `BNEZ' would be used.)

This relaxation works well because the unconditional jump instruction
has a much larger offset range than the various conditional branches.
However, an error will occur if a branch target is beyond the range of a
jump instruction. `as' cannot relax unconditional jumps. Similarly,
an error will occur if the original input contains an unconditional
jump to a target that is out of range.

Branch relaxation is enabled by default. It can be disabled by using
underscore prefixes (*note Opcode Names: Xtensa Opcodes.), the
`--no-transform' command-line option (*note Command Line Options:
Xtensa Options.), or the `no-transform' directive (*note transform:
Transform Directive.).

File: as.info, Node: Xtensa Call Relaxation, Next: Xtensa Immediate Relaxation, Prev: Xtensa Branch Relaxation, Up: Xtensa Relaxation

8.34.4.2 Function Call Relaxation
.................................

Function calls may require relaxation because the Xtensa immediate call
instructions (`CALL0', `CALL4', `CALL8' and `CALL12') provide a
PC-relative offset of only 512 Kbytes in either direction. For larger
programs, it may be necessary to use indirect calls (`CALLX0',
`CALLX4', `CALLX8' and `CALLX12') where the target address is specified
in a register. The Xtensa assembler can automatically relax immediate
call instructions into indirect call instructions. This relaxation is
done by loading the address of the called function into the callee's
return address register and then using a `CALLX' instruction. So, for
example:

call8 func

might be relaxed to:

.literal .L1, func
l32r a8, .L1
callx8 a8

Because the addresses of targets of function calls are not generally
known until link-time, the assembler must assume the worst and relax all
the calls to functions in other source files, not just those that really
will be out of range. The linker can recognize calls that were
unnecessarily relaxed, and it will remove the overhead introduced by the
assembler for those cases where direct calls are sufficient.

Call relaxation is disabled by default because it can have a negative
effect on both code size and performance, although the linker can
usually eliminate the unnecessary overhead. If a program is too large
and some of the calls are out of range, function call relaxation can be
enabled using the `--longcalls' command-line option or the `longcalls'
directive (*note longcalls: Longcalls Directive.).

File: as.info, Node: Xtensa Immediate Relaxation, Prev: Xtensa Call Relaxation, Up: Xtensa Relaxation

8.34.4.3 Other Immediate Field Relaxation
.........................................

The assembler normally performs the following other relaxations. They
can be disabled by using underscore prefixes (*note Opcode Names:
Xtensa Opcodes.), the `--no-transform' command-line option (*note
Command Line Options: Xtensa Options.), or the `no-transform' directive
(*note transform: Transform Directive.).

The `MOVI' machine instruction can only materialize values in the
range from -2048 to 2047. Values outside this range are best
materialized with `L32R' instructions. Thus:

movi a0, 100000

is assembled into the following machine code:

.literal .L1, 100000
l32r a0, .L1

The `L8UI' machine instruction can only be used with immediate
offsets in the range from 0 to 255. The `L16SI' and `L16UI' machine
instructions can only be used with offsets from 0 to 510. The `L32I'
machine instruction can only be used with offsets from 0 to 1020. A
load offset outside these ranges can be materalized with an `L32R'
instruction if the destination register of the load is different than
the source address register. For example:

l32i a1, a0, 2040

is translated to:

.literal .L1, 2040
l32r a1, .L1
addi a1, a0, a1
l32i a1, a1, 0

If the load destination and source address register are the same, an
out-of-range offset causes an error.

The Xtensa `ADDI' instruction only allows immediate operands in the
range from -128 to 127. There are a number of alternate instruction
sequences for the `ADDI' operation. First, if the immediate is 0, the
`ADDI' will be turned into a `MOV.N' instruction (or the equivalent
`OR' instruction if the code density option is not available). If the
`ADDI' immediate is outside of the range -128 to 127, but inside the
range -32896 to 32639, an `ADDMI' instruction or `ADDMI'/`ADDI'
sequence will be used. Finally, if the immediate is outside of this
range and a free register is available, an `L32R'/`ADD' sequence will
be used with a literal allocated from the literal pool.

For example:

addi a5, a6, 0
addi a5, a6, 512
addi a5, a6, 513
addi a5, a6, 50000

is assembled into the following:

.literal .L1, 50000
mov.n a5, a6
addmi a5, a6, 0x200
addmi a5, a6, 0x200
addi a5, a5, 1
l32r a5, .L1
add a5, a6, a5

File: as.info, Node: Xtensa Directives, Prev: Xtensa Relaxation, Up: Xtensa-Dependent

8.34.5 Directives
-----------------

The Xtensa assember supports a region-based directive syntax:

.begin DIRECTIVE [OPTIONS]
...
.end DIRECTIVE

All the Xtensa-specific directives that apply to a region of code use
this syntax.

The directive applies to code between the `.begin' and the `.end'.
The state of the option after the `.end' reverts to what it was before
the `.begin'. A nested `.begin'/`.end' region can further change the
state of the directive without having to be aware of its outer state.
For example, consider:

.begin no-transform
L: add a0, a1, a2
.begin transform
M: add a0, a1, a2
.end transform
N: add a0, a1, a2
.end no-transform

The `ADD' opcodes at `L' and `N' in the outer `no-transform' region
both result in `ADD' machine instructions, but the assembler selects an
`ADD.N' instruction for the `ADD' at `M' in the inner `transform'
region.

The advantage of this style is that it works well inside macros
which can preserve the context of their callers.

The following directives are available:

* Menu:

* Schedule Directive:: Enable instruction scheduling.
* Longcalls Directive:: Use Indirect Calls for Greater Range.
* Transform Directive:: Disable All Assembler Transformations.
* Literal Directive:: Intermix Literals with Instructions.
* Literal Position Directive:: Specify Inline Literal Pool Locations.
* Literal Prefix Directive:: Specify Literal Section Name Prefix.
* Absolute Literals Directive:: Control PC-Relative vs. Absolute Literals.

File: as.info, Node: Schedule Directive, Next: Longcalls Directive, Up: Xtensa Directives

8.34.5.1 schedule
.................

The `schedule' directive is recognized only for compatibility with
Tensilica's assembler.

.begin [no-]schedule
.end [no-]schedule

This directive is ignored and has no effect on `as'.

File: as.info, Node: Longcalls Directive, Next: Transform Directive, Prev: Schedule Directive, Up: Xtensa Directives

8.34.5.2 longcalls
..................

The `longcalls' directive enables or disables function call relaxation.
*Note Function Call Relaxation: Xtensa Call Relaxation.

.begin [no-]longcalls
.end [no-]longcalls

Call relaxation is disabled by default unless the `--longcalls'
command-line option is specified. The `longcalls' directive overrides
the default determined by the command-line options.

File: as.info, Node: Transform Directive, Next: Literal Directive, Prev: Longcalls Directive, Up: Xtensa Directives

8.34.5.3 transform
..................

This directive enables or disables all assembler transformation,
including relaxation (*note Xtensa Relaxation: Xtensa Relaxation.) and
optimization (*note Xtensa Optimizations: Xtensa Optimizations.).

.begin [no-]transform
.end [no-]transform

Transformations are enabled by default unless the `--no-transform'
option is used. The `transform' directive overrides the default
determined by the command-line options. An underscore opcode prefix,
disabling transformation of that opcode, always takes precedence over
both directives and command-line flags.

File: as.info, Node: Literal Directive, Next: Literal Position Directive, Prev: Transform Directive, Up: Xtensa Directives

8.34.5.4 literal
................

The `.literal' directive is used to define literal pool data, i.e.,
read-only 32-bit data accessed via `L32R' instructions.

.literal LABEL, VALUE[, VALUE...]

This directive is similar to the standard `.word' directive, except
that the actual location of the literal data is determined by the
assembler and linker, not by the position of the `.literal' directive.
Using this directive gives the assembler freedom to locate the literal
data in the most appropriate place and possibly to combine identical
literals. For example, the code:

entry sp, 40
.literal .L1, sym
l32r a4, .L1

can be used to load a pointer to the symbol `sym' into register
`a4'. The value of `sym' will not be placed between the `ENTRY' and
`L32R' instructions; instead, the assembler puts the data in a literal
pool.

Literal pools for absolute mode `L32R' instructions (*note Absolute
Literals Directive::) are placed in a seperate `.lit4' section. By
default literal pools for PC-relative mode `L32R' instructions are
placed in a separate `.literal' section; however, when using the
`--text-section-literals' option (*note Command Line Options: Xtensa
Options.), the literal pools are placed in the current section. These
text section literal pools are created automatically before `ENTRY'
instructions and manually after `.literal_position' directives (*note
literal_position: Literal Position Directive.). If there are no
preceding `ENTRY' instructions, explicit `.literal_position' directives
must be used to place the text section literal pools; otherwise, `as'
will report an error.

File: as.info, Node: Literal Position Directive, Next: Literal Prefix Directive, Prev: Literal Directive, Up: Xtensa Directives

8.34.5.5 literal_position
.........................

When using `--text-section-literals' to place literals inline in the
section being assembled, the `.literal_position' directive can be used
to mark a potential location for a literal pool.

.literal_position

The `.literal_position' directive is ignored when the
`--text-section-literals' option is not used or when `L32R'
instructions use the absolute addressing mode.

The assembler will automatically place text section literal pools
before `ENTRY' instructions, so the `.literal_position' directive is
only needed to specify some other location for a literal pool. You may
need to add an explicit jump instruction to skip over an inline literal
pool.

For example, an interrupt vector does not begin with an `ENTRY'
instruction so the assembler will be unable to automatically find a good
place to put a literal pool. Moreover, the code for the interrupt
vector must be at a specific starting address, so the literal pool
cannot come before the start of the code. The literal pool for the
vector must be explicitly positioned in the middle of the vector (before
any uses of the literals, due to the negative offsets used by
PC-relative `L32R' instructions). The `.literal_position' directive
can be used to do this. In the following code, the literal for `M'
will automatically be aligned correctly and is placed after the
unconditional jump.

.global M
code_start:
j continue
.literal_position
.align 4
continue:
movi a4, M

File: as.info, Node: Literal Prefix Directive, Next: Absolute Literals Directive, Prev: Literal Position Directive, Up: Xtensa Directives

8.34.5.6 literal_prefix
.......................

The `literal_prefix' directive allows you to specify different sections
to hold literals from different portions of an assembly file. With
this directive, a single assembly file can be used to generate code
into multiple sections, including literals generated by the assembler.

.begin literal_prefix [NAME]
.end literal_prefix

By default the assembler places literal pools in sections separate
from the instructions, using the default literal section names of
`.literal' for PC-relative mode `L32R' instructions and `.lit4' for
absolute mode `L32R' instructions (*note Absolute Literals
Directive::). The `literal_prefix' directive causes different literal
sections to be used for the code inside the delimited region. The new
literal sections are determined by including NAME as a prefix to the
default literal section names. If the NAME argument is omitted, the
literal sections revert to the defaults. This directive has no effect
when using the `--text-section-literals' option (*note Command Line
Options: Xtensa Options.).

Except for two special cases, the assembler determines the new
literal sections by simply prepending NAME to the default section names,
resulting in `NAME.literal' and `NAME.lit4' sections. The
`literal_prefix' directive is often used with the name of the current
text section as the prefix argument. To facilitate this usage, the
assembler uses special case rules when it recognizes NAME as a text
section name. First, if NAME ends with `.text', that suffix is not
included in the literal section name. For example, if NAME is
`.iram0.text', then the literal sections will be `.iram0.literal' and
`.iram0.lit4'. Second, if NAME begins with `.gnu.linkonce.t.', then
the literal section names are formed by replacing the `.t' substring
with `.literal' and `.lit4'. For example, if NAME is
`.gnu.linkonce.t.func', the literal sections will be
`.gnu.linkonce.literal.func' and `.gnu.linkonce.lit4.func'.

File: as.info, Node: Absolute Literals Directive, Prev: Literal Prefix Directive, Up: Xtensa Directives

8.34.5.7 absolute-literals
..........................

The `absolute-literals' and `no-absolute-literals' directives control
the absolute vs. PC-relative mode for `L32R' instructions. These are
relevant only for Xtensa configurations that include the absolute
addressing option for `L32R' instructions.

.begin [no-]absolute-literals
.end [no-]absolute-literals

These directives do not change the `L32R' mode--they only cause the
assembler to emit the appropriate kind of relocation for `L32R'
instructions and to place the literal values in the appropriate section.
To change the `L32R' mode, the program must write the `LITBASE' special
register. It is the programmer's responsibility to keep track of the
mode and indicate to the assembler which mode is used in each region of
code.

If the Xtensa configuration includes the absolute `L32R' addressing
option, the default is to assume absolute `L32R' addressing unless the
`--no-absolute-literals' command-line option is specified. Otherwise,
the default is to assume PC-relative `L32R' addressing. The
`absolute-literals' directive can then be used to override the default
determined by the command-line options.

File: as.info, Node: Reporting Bugs, Next: Acknowledgements, Prev: Machine Dependencies, Up: Top

9 Reporting Bugs
****************

Your bug reports play an essential role in making `as' reliable.

Reporting a bug may help you by bringing a solution to your problem,
or it may not. But in any case the principal function of a bug report
is to help the entire community by making the next version of `as' work
better. Bug reports are your contribution to the maintenance of `as'.

In order for a bug report to serve its purpose, you must include the
information that enables us to fix the bug.

* Menu:

* Bug Criteria:: Have you found a bug?
* Bug Reporting:: How to report bugs

File: as.info, Node: Bug Criteria, Next: Bug Reporting, Up: Reporting Bugs

9.1 Have You Found a Bug?
=========================

If you are not sure whether you have found a bug, here are some
guidelines:

* If the assembler gets a fatal signal, for any input whatever, that
is a `as' bug. Reliable assemblers never crash.

* If `as' produces an error message for valid input, that is a bug.

* If `as' does not produce an error message for invalid input, that
is a bug. However, you should note that your idea of "invalid
input" might be our idea of "an extension" or "support for
traditional practice".

* If you are an experienced user of assemblers, your suggestions for
improvement of `as' are welcome in any case.

File: as.info, Node: Bug Reporting, Prev: Bug Criteria, Up: Reporting Bugs

9.2 How to Report Bugs
======================

A number of companies and individuals offer support for GNU products.
If you obtained `as' from a support organization, we recommend you
contact that organization first.

You can find contact information for many support companies and
individuals in the file `etc/SERVICE' in the GNU Emacs distribution.

In any event, we also recommend that you send bug reports for `as'
to `bug-binutils@gnu.org'.

The fundamental principle of reporting bugs usefully is this:
*report all the facts*. If you are not sure whether to state a fact or
leave it out, state it!

Often people omit facts because they think they know what causes the
problem and assume that some details do not matter. Thus, you might
assume that the name of a symbol you use in an example does not matter.
Well, probably it does not, but one cannot be sure. Perhaps the bug
is a stray memory reference which happens to fetch from the location
where that name is stored in memory; perhaps, if the name were
different, the contents of that location would fool the assembler into
doing the right thing despite the bug. Play it safe and give a
specific, complete example. That is the easiest thing for you to do,
and the most helpful.

Keep in mind that the purpose of a bug report is to enable us to fix
the bug if it is new to us. Therefore, always write your bug reports
on the assumption that the bug has not been reported previously.

Sometimes people give a few sketchy facts and ask, "Does this ring a
bell?" This cannot help us fix a bug, so it is basically useless. We
respond by asking for enough details to enable us to investigate. You
might as well expedite matters by sending them to begin with.

To enable us to fix the bug, you should include all these things:

* The version of `as'. `as' announces it if you start it with the
`--version' argument.

Without this, we will not know whether there is any point in
looking for the bug in the current version of `as'.

* Any patches you may have applied to the `as' source.

* The type of machine you are using, and the operating system name
and version number.

* What compiler (and its version) was used to compile `as'--e.g.
"`gcc-2.7'".

* The command arguments you gave the assembler to assemble your
example and observe the bug. To guarantee you will not omit
something important, list them all. A copy of the Makefile (or
the output from make) is sufficient.

If we were to try to guess the arguments, we would probably guess
wrong and then we might not encounter the bug.

* A complete input file that will reproduce the bug. If the bug is
observed when the assembler is invoked via a compiler, send the
assembler source, not the high level language source. Most
compilers will produce the assembler source when run with the `-S'
option. If you are using `gcc', use the options `-v
--save-temps'; this will save the assembler source in a file with
an extension of `.s', and also show you exactly how `as' is being
run.

* A description of what behavior you observe that you believe is
incorrect. For example, "It gets a fatal signal."

Of course, if the bug is that `as' gets a fatal signal, then we
will certainly notice it. But if the bug is incorrect output, we
might not notice unless it is glaringly wrong. You might as well
not give us a chance to make a mistake.

Even if the problem you experience is a fatal signal, you should
still say so explicitly. Suppose something strange is going on,
such as, your copy of `as' is out of synch, or you have
encountered a bug in the C library on your system. (This has
happened!) Your copy might crash and ours would not. If you told
us to expect a crash, then when ours fails to crash, we would know
that the bug was not happening for us. If you had not told us to
expect a crash, then we would not be able to draw any conclusion
from our observations.

* If you wish to suggest changes to the `as' source, send us context
diffs, as generated by `diff' with the `-u', `-c', or `-p' option.
Always send diffs from the old file to the new file. If you even
discuss something in the `as' source, refer to it by context, not
by line number.

The line numbers in our development sources will not match those
in your sources. Your line numbers would convey no useful
information to us.

Here are some things that are not necessary:

* A description of the envelope of the bug.

Often people who encounter a bug spend a lot of time investigating
which changes to the input file will make the bug go away and which
changes will not affect it.

This is often time consuming and not very useful, because the way
we will find the bug is by running a single example under the
debugger with breakpoints, not by pure deduction from a series of
examples. We recommend that you save your time for something else.

Of course, if you can find a simpler example to report _instead_
of the original one, that is a convenience for us. Errors in the
output will be easier to spot, running under the debugger will take
less time, and so on.

However, simplification is not vital; if you do not want to do
this, report the bug anyway and send us the entire test case you
used.

* A patch for the bug.

A patch for the bug does help us if it is a good one. But do not
omit the necessary information, such as the test case, on the
assumption that a patch is all we need. We might see problems
with your patch and decide to fix the problem another way, or we
might not understand it at all.

Sometimes with a program as complicated as `as' it is very hard to
construct an example that will make the program follow a certain
path through the code. If you do not send us the example, we will
not be able to construct one, so we will not be able to verify
that the bug is fixed.

And if we cannot understand what bug you are trying to fix, or why
your patch should be an improvement, we will not install it. A
test case will help us to understand.

* A guess about what the bug is or what it depends on.

Such guesses are usually wrong. Even we cannot guess right about
such things without first using the debugger to find the facts.

File: as.info, Node: Acknowledgements, Next: GNU Free Documentation License, Prev: Reporting Bugs, Up: Top

10 Acknowledgements
*******************

If you have contributed to GAS and your name isn't listed here, it is
not meant as a slight. We just don't know about it. Send mail to the
maintainer, and we'll correct the situation. Currently the maintainer
is Ken Raeburn (email address `raeburn@cygnus.com').

Dean Elsner wrote the original GNU assembler for the VAX.(1)

Jay Fenlason maintained GAS for a while, adding support for
GDB-specific debug information and the 68k series machines, most of the
preprocessing pass, and extensive changes in `messages.c',
`input-file.c', `write.c'.

K. Richard Pixley maintained GAS for a while, adding various
enhancements and many bug fixes, including merging support for several
processors, breaking GAS up to handle multiple object file format back
ends (including heavy rewrite, testing, an integration of the coff and
b.out back ends), adding configuration including heavy testing and
verification of cross assemblers and file splits and renaming,
converted GAS to strictly ANSI C including full prototypes, added
support for m680[34]0 and cpu32, did considerable work on i960
including a COFF port (including considerable amounts of reverse
engineering), a SPARC opcode file rewrite, DECstation, rs6000, and
hp300hpux host ports, updated "know" assertions and made them work,
much other reorganization, cleanup, and lint.

Ken Raeburn wrote the high-level BFD interface code to replace most
of the code in format-specific I/O modules.

The original VMS support was contributed by David L. Kashtan. Eric
Youngdale has done much work with it since.

The Intel 80386 machine description was written by Eliot Dresselhaus.

Minh Tran-Le at IntelliCorp contributed some AIX 386 support.

The Motorola 88k machine description was contributed by Devon Bowen
of Buffalo University and Torbjorn Granlund of the Swedish Institute of
Computer Science.

Keith Knowles at the Open Software Foundation wrote the original
MIPS back end (`tc-mips.c', `tc-mips.h'), and contributed Rose format
support (which hasn't been merged in yet). Ralph Campbell worked with
the MIPS code to support a.out format.

Support for the Zilog Z8k and Renesas H8/300 and H8/500 processors
(tc-z8k, tc-h8300, tc-h8500), and IEEE 695 object file format
(obj-ieee), was written by Steve Chamberlain of Cygnus Support. Steve
also modified the COFF back end to use BFD for some low-level
operations, for use with the H8/300 and AMD 29k targets.

John Gilmore built the AMD 29000 support, added `.include' support,
and simplified the configuration of which versions accept which
directives. He updated the 68k machine description so that Motorola's
opcodes always produced fixed-size instructions (e.g., `jsr'), while
synthetic instructions remained shrinkable (`jbsr'). John fixed many
bugs, including true tested cross-compilation support, and one bug in
relaxation that took a week and required the proverbial one-bit fix.

Ian Lance Taylor of Cygnus Support merged the Motorola and MIT
syntax for the 68k, completed support for some COFF targets (68k, i386
SVR3, and SCO Unix), added support for MIPS ECOFF and ELF targets,
wrote the initial RS/6000 and PowerPC assembler, and made a few other
minor patches.

Steve Chamberlain made GAS able to generate listings.

Hewlett-Packard contributed support for the HP9000/300.

Jeff Law wrote GAS and BFD support for the native HPPA object format
(SOM) along with a fairly extensive HPPA testsuite (for both SOM and
ELF object formats). This work was supported by both the Center for
Software Science at the University of Utah and Cygnus Support.

Support for ELF format files has been worked on by Mark Eichin of
Cygnus Support (original, incomplete implementation for SPARC), Pete
Hoogenboom and Jeff Law at the University of Utah (HPPA mainly),
Michael Meissner of the Open Software Foundation (i386 mainly), and Ken
Raeburn of Cygnus Support (sparc, and some initial 64-bit support).

Linas Vepstas added GAS support for the ESA/390 "IBM 370"
architecture.

Richard Henderson rewrote the Alpha assembler. Klaus Kaempf wrote
GAS and BFD support for openVMS/Alpha.

Timothy Wall, Michael Hayes, and Greg Smart contributed to the
various tic* flavors.

David Heine, Sterling Augustine, Bob Wilson and John Ruttenberg from
Tensilica, Inc. added support for Xtensa processors.

Several engineers at Cygnus Support have also provided many small
bug fixes and configuration enhancements.

Many others have contributed large or small bugfixes and
enhancements. If you have contributed significant work and are not
mentioned on this list, and want to be, let us know. Some of the
history has been lost; we are not intentionally leaving anyone out.

---------- Footnotes ----------

(1) Any more details?

File: as.info, Node: GNU Free Documentation License, Next: Index, Prev: Acknowledgements, Up: Top

Appendix A GNU Free Documentation License
*****************************************

Version 1.1, March 2000

Everyone is permitted to copy and distribute verbatim copies
of this license document, but changing it is not allowed.

0. PREAMBLE

The purpose of this License is to make a manual, textbook, or other
written document "free" in the sense of freedom: to assure everyone
the effective freedom to copy and redistribute it, with or without
modifying it, either commercially or noncommercially. Secondarily,
this License preserves for the author and publisher a way to get
credit for their work, while not being considered responsible for
modifications made by others.

This License is a kind of "copyleft", which means that derivative
works of the document must themselves be free in the same sense.
It complements the GNU General Public License, which is a copyleft
license designed for free software.

We have designed this License in order to use it for manuals for
free software, because free software needs free documentation: a
free program should come with manuals providing the same freedoms
that the software does. But this License is not limited to
software manuals; it can be used for any textual work, regardless
of subject matter or whether it is published as a printed book.
We recommend this License principally for works whose purpose is
instruction or reference.

1. APPLICABILITY AND DEFINITIONS

This License applies to any manual or other work that contains a
notice placed by the copyright holder saying it can be distributed
under the terms of this License. The "Document", below, refers to
any such manual or work. Any member of the public is a licensee,
and is addressed as "you."

A "Modified Version" of the Document means any work containing the
Document or a portion of it, either copied verbatim, or with
modifications and/or translated into another language.

A "Secondary Section" is a named appendix or a front-matter
section of the Document that deals exclusively with the
relationship of the publishers or authors of the Document to the
Document's overall subject (or to related matters) and contains
nothing that could fall directly within that overall subject.
(For example, if the Document is in part a textbook of
mathematics, a Secondary Section may not explain any mathematics.)
The relationship could be a matter of historical connection with
the subject or with related matters, or of legal, commercial,
philosophical, ethical or political position regarding them.

The "Invariant Sections" are certain Secondary Sections whose
titles are designated, as being those of Invariant Sections, in
the notice that says that the Document is released under this
License.

The "Cover Texts" are certain short passages of text that are
listed, as Front-Cover Texts or Back-Cover Texts, in the notice
that says that the Document is released under this License.

A "Transparent" copy of the Document means a machine-readable copy,
represented in a format whose specification is available to the
general public, whose contents can be viewed and edited directly
and straightforwardly with generic text editors or (for images
composed of pixels) generic paint programs or (for drawings) some
widely available drawing editor, and that is suitable for input to
text formatters or for automatic translation to a variety of
formats suitable for input to text formatters. A copy made in an
otherwise Transparent file format whose markup has been designed
to thwart or discourage subsequent modification by readers is not
Transparent. A copy that is not "Transparent" is called "Opaque."

Examples of suitable formats for Transparent copies include plain
ASCII without markup, Texinfo input format, LaTeX input format,
SGML or XML using a publicly available DTD, and
standard-conforming simple HTML designed for human modification.
Opaque formats include PostScript, PDF, proprietary formats that
can be read and edited only by proprietary word processors, SGML
or XML for which the DTD and/or processing tools are not generally
available, and the machine-generated HTML produced by some word
processors for output purposes only.

The "Title Page" means, for a printed book, the title page itself,
plus such following pages as are needed to hold, legibly, the
material this License requires to appear in the title page. For
works in formats which do not have any title page as such, "Title
Page" means the text near the most prominent appearance of the
work's title, preceding the beginning of the body of the text.

2. VERBATIM COPYING

You may copy and distribute the Document in any medium, either
commercially or noncommercially, provided that this License, the
copyright notices, and the license notice saying this License
applies to the Document are reproduced in all copies, and that you
add no other conditions whatsoever to those of this License. You
may not use technical measures to obstruct or control the reading
or further copying of the copies you make or distribute. However,
you may accept compensation in exchange for copies. If you
distribute a large enough number of copies you must also follow
the conditions in section 3.

You may also lend copies, under the same conditions stated above,
and you may publicly display copies.

3. COPYING IN QUANTITY

If you publish printed copies of the Document numbering more than
100, and the Document's license notice requires Cover Texts, you
must enclose the copies in covers that carry, clearly and legibly,
all these Cover Texts: Front-Cover Texts on the front cover, and
Back-Cover Texts on the back cover. Both covers must also clearly
and legibly identify you as the publisher of these copies. The
front cover must present the full title with all words of the
title equally prominent and visible. You may add other material
on the covers in addition. Copying with changes limited to the
covers, as long as they preserve the title of the Document and
satisfy these conditions, can be treated as verbatim copying in
other respects.

If the required texts for either cover are too voluminous to fit
legibly, you should put the first ones listed (as many as fit
reasonably) on the actual cover, and continue the rest onto
adjacent pages.

If you publish or distribute Opaque copies of the Document
numbering more than 100, you must either include a
machine-readable Transparent copy along with each Opaque copy, or
state in or with each Opaque copy a publicly-accessible
computer-network location containing a complete Transparent copy
of the Document, free of added material, which the general
network-using public has access to download anonymously at no
charge using public-standard network protocols. If you use the
latter option, you must take reasonably prudent steps, when you
begin distribution of Opaque copies in quantity, to ensure that
this Transparent copy will remain thus accessible at the stated
location until at least one year after the last time you
distribute an Opaque copy (directly or through your agents or
retailers) of that edition to the public.

It is requested, but not required, that you contact the authors of
the Document well before redistributing any large number of
copies, to give them a chance to provide you with an updated
version of the Document.

4. MODIFICATIONS

You may copy and distribute a Modified Version of the Document
under the conditions of sections 2 and 3 above, provided that you
release the Modified Version under precisely this License, with
the Modified Version filling the role of the Document, thus
licensing distribution and modification of the Modified Version to
whoever possesses a copy of it. In addition, you must do these
things in the Modified Version:

A. Use in the Title Page (and on the covers, if any) a title
distinct from that of the Document, and from those of previous
versions (which should, if there were any, be listed in the
History section of the Document). You may use the same title
as a previous version if the original publisher of that version
gives permission.
B. List on the Title Page, as authors, one or more persons or
entities responsible for authorship of the modifications in the
Modified Version, together with at least five of the principal
authors of the Document (all of its principal authors, if it
has less than five).
C. State on the Title page the name of the publisher of the
Modified Version, as the publisher.
D. Preserve all the copyright notices of the Document.
E. Add an appropriate copyright notice for your modifications
adjacent to the other copyright notices.
F. Include, immediately after the copyright notices, a license
notice giving the public permission to use the Modified Version
under the terms of this License, in the form shown in the
Addendum below.
G. Preserve in that license notice the full lists of Invariant
Sections and required Cover Texts given in the Document's
license notice.
H. Include an unaltered copy of this License.
I. Preserve the section entitled "History", and its title, and add
to it an item stating at least the title, year, new authors, and
publisher of the Modified Version as given on the Title Page.
If there is no section entitled "History" in the Document,
create one stating the title, year, authors, and publisher of
the Document as given on its Title Page, then add an item
describing the Modified Version as stated in the previous
sentence.
J. Preserve the network location, if any, given in the Document for
public access to a Transparent copy of the Document, and
likewise the network locations given in the Document for
previous versions it was based on. These may be placed in the
"History" section. You may omit a network location for a work
that was published at least four years before the Document
itself, or if the original publisher of the version it refers
to gives permission.
K. In any section entitled "Acknowledgements" or "Dedications",
preserve the section's title, and preserve in the section all the
substance and tone of each of the contributor acknowledgements
and/or dedications given therein.
L. Preserve all the Invariant Sections of the Document,
unaltered in their text and in their titles. Section numbers
or the equivalent are not considered part of the section titles.
M. Delete any section entitled "Endorsements." Such a section
may not be included in the Modified Version.
N. Do not retitle any existing section as "Endorsements" or to
conflict in title with any Invariant Section.

If the Modified Version includes new front-matter sections or
appendices that qualify as Secondary Sections and contain no
material copied from the Document, you may at your option
designate some or all of these sections as invariant. To do this,
add their titles to the list of Invariant Sections in the Modified
Version's license notice. These titles must be distinct from any
other section titles.

You may add a section entitled "Endorsements", provided it contains
nothing but endorsements of your Modified Version by various
parties-for example, statements of peer review or that the text has
been approved by an organization as the authoritative definition
of a standard.

You may add a passage of up to five words as a Front-Cover Text,
and a passage of up to 25 words as a Back-Cover Text, to the end
of the list of Cover Texts in the Modified Version. Only one
passage of Front-Cover Text and one of Back-Cover Text may be
added by (or through arrangements made by) any one entity. If the
Document already includes a cover text for the same cover,
previously added by you or by arrangement made by the same entity
you are acting on behalf of, you may not add another; but you may
replace the old one, on explicit permission from the previous
publisher that added the old one.

The author(s) and publisher(s) of the Document do not by this
License give permission to use their names for publicity for or to
assert or imply endorsement of any Modified Version.

5. COMBINING DOCUMENTS

You may combine the Document with other documents released under
this License, under the terms defined in section 4 above for
modified versions, provided that you include in the combination
all of the Invariant Sections of all of the original documents,
unmodified, and list them all as Invariant Sections of your
combined work in its license notice.

The combined work need only contain one copy of this License, and
multiple identical Invariant Sections may be replaced with a single
copy. If there are multiple Invariant Sections with the same name
but different contents, make the title of each such section unique
by adding at the end of it, in parentheses, the name of the
original author or publisher of that section if known, or else a
unique number. Make the same adjustment to the section titles in
the list of Invariant Sections in the license notice of the
combined work.

In the combination, you must combine any sections entitled
"History" in the various original documents, forming one section
entitled "History"; likewise combine any sections entitled
"Acknowledgements", and any sections entitled "Dedications." You
must delete all sections entitled "Endorsements."

6. COLLECTIONS OF DOCUMENTS

You may make a collection consisting of the Document and other
documents released under this License, and replace the individual
copies of this License in the various documents with a single copy
that is included in the collection, provided that you follow the
rules of this License for verbatim copying of each of the
documents in all other respects.

You may extract a single document from such a collection, and
distribute it individually under this License, provided you insert
a copy of this License into the extracted document, and follow
this License in all other respects regarding verbatim copying of
that document.

7. AGGREGATION WITH INDEPENDENT WORKS

A compilation of the Document or its derivatives with other
separate and independent documents or works, in or on a volume of
a storage or distribution medium, does not as a whole count as a
Modified Version of the Document, provided no compilation
copyright is claimed for the compilation. Such a compilation is
called an "aggregate", and this License does not apply to the
other self-contained works thus compiled with the Document, on
account of their being thus compiled, if they are not themselves
derivative works of the Document.

If the Cover Text requirement of section 3 is applicable to these
copies of the Document, then if the Document is less than one
quarter of the entire aggregate, the Document's Cover Texts may be
placed on covers that surround only the Document within the
aggregate. Otherwise they must appear on covers around the whole
aggregate.

8. TRANSLATION

Translation is considered a kind of modification, so you may
distribute translations of the Document under the terms of section
4. Replacing Invariant Sections with translations requires special
permission from their copyright holders, but you may include
translations of some or all Invariant Sections in addition to the
original versions of these Invariant Sections. You may include a
translation of this License provided that you also include the
original English version of this License. In case of a
disagreement between the translation and the original English
version of this License, the original English version will prevail.

9. TERMINATION

You may not copy, modify, sublicense, or distribute the Document
except as expressly provided for under this License. Any other
attempt to copy, modify, sublicense or distribute the Document is
void, and will automatically terminate your rights under this
License. However, parties who have received copies, or rights,
from you under this License will not have their licenses
terminated so long as such parties remain in full compliance.

10. FUTURE REVISIONS OF THIS LICENSE

The Free Software Foundation may publish new, revised versions of
the GNU Free Documentation License from time to time. Such new
versions will be similar in spirit to the present version, but may
differ in detail to address new problems or concerns. See
http://www.gnu.org/copyleft/.

Each version of the License is given a distinguishing version
number. If the Document specifies that a particular numbered
version of this License "or any later version" applies to it, you
have the option of following the terms and conditions either of
that specified version or of any later version that has been
published (not as a draft) by the Free Software Foundation. If
the Document does not specify a version number of this License,
you may choose any version ever published (not as a draft) by the
Free Software Foundation.

ADDENDUM: How to use this License for your documents
====================================================

To use this License in a document you have written, include a copy of
the License in the document and put the following copyright and license
notices just after the title page:

Copyright (C) YEAR YOUR NAME.
Permission is granted to copy, distribute and/or modify this document
under the terms of the GNU Free Documentation License, Version 1.1
or any later version published by the Free Software Foundation;
with the Invariant Sections being LIST THEIR TITLES, with the
Front-Cover Texts being LIST, and with the Back-Cover Texts being LIST.
A copy of the license is included in the section entitled "GNU
Free Documentation License."

If you have no Invariant Sections, write "with no Invariant Sections"
instead of saying which ones are invariant. If you have no Front-Cover
Texts, write "no Front-Cover Texts" instead of "Front-Cover Texts being
LIST"; likewise for Back-Cover Texts.

If your document contains nontrivial examples of program code, we
recommend releasing these examples in parallel under your choice of
free software license, such as the GNU General Public License, to
permit their use in free software.