* doc/md.texi (msub@var{m}@var{n}4, usub@var{m}@var{n}4): Document.

[pf3gnuchains/gcc-fork.git] / gcc / doc / md.texi

@c Copyright (C) 1988, 1989, 1992, 1993, 1994, 1996, 1998, 1999, 2000, 2001,
@c 2002, 2003, 2004, 2005, 2006 Free Software Foundation, Inc.
@c This is part of the GCC manual.
@c For copying conditions, see the file gcc.texi.

@ifset INTERNALS
@node Machine Desc
@chapter Machine Descriptions
@cindex machine descriptions

A machine description has two parts: a file of instruction patterns
(@file{.md} file) and a C header file of macro definitions.

The @file{.md} file for a target machine contains a pattern for each
instruction that the target machine supports (or at least each instruction
that is worth telling the compiler about).  It may also contain comments.
A semicolon causes the rest of the line to be a comment, unless the semicolon
is inside a quoted string.

See the next chapter for information on the C header file.

@menu
* Overview::            How the machine description is used.
* Patterns::            How to write instruction patterns.
* Example::             An explained example of a @code{define_insn} pattern.
* RTL Template::        The RTL template defines what insns match a pattern.
* Output Template::     The output template says how to make assembler code
                          from such an insn.
* Output Statement::    For more generality, write C code to output
                          the assembler code.
* Predicates::          Controlling what kinds of operands can be used
                          for an insn.
* Constraints::         Fine-tuning operand selection.
* Standard Names::      Names mark patterns to use for code generation.
* Pattern Ordering::    When the order of patterns makes a difference.
* Dependent Patterns::  Having one pattern may make you need another.
* Jump Patterns::       Special considerations for patterns for jump insns.
* Looping Patterns::    How to define patterns for special looping insns.
* Insn Canonicalizations::Canonicalization of Instructions
* Expander Definitions::Generating a sequence of several RTL insns
                          for a standard operation.
* Insn Splitting::      Splitting Instructions into Multiple Instructions.
* Including Patterns::      Including Patterns in Machine Descriptions.
* Peephole Definitions::Defining machine-specific peephole optimizations.
* Insn Attributes::     Specifying the value of attributes for generated insns.
* Conditional Execution::Generating @code{define_insn} patterns for
                           predication.
* Constant Definitions::Defining symbolic constants that can be used in the
                        md file.
* Macros::              Using macros to generate patterns from a template.
@end menu

@node Overview
@section Overview of How the Machine Description is Used

There are three main conversions that happen in the compiler:

@enumerate

@item
The front end reads the source code and builds a parse tree.

@item
The parse tree is used to generate an RTL insn list based on named
instruction patterns.

@item
The insn list is matched against the RTL templates to produce assembler
code.

@end enumerate

For the generate pass, only the names of the insns matter, from either a
named @code{define_insn} or a @code{define_expand}.  The compiler will
choose the pattern with the right name and apply the operands according
to the documentation later in this chapter, without regard for the RTL
template or operand constraints.  Note that the names the compiler looks
for are hard-coded in the compiler---it will ignore unnamed patterns and
patterns with names it doesn't know about, but if you don't provide a
named pattern it needs, it will abort.

If a @code{define_insn} is used, the template given is inserted into the
insn list.  If a @code{define_expand} is used, one of three things
happens, based on the condition logic.  The condition logic may manually
create new insns for the insn list, say via @code{emit_insn()}, and
invoke @code{DONE}.  For certain named patterns, it may invoke @code{FAIL} to tell the
compiler to use an alternate way of performing that task.  If it invokes
neither @code{DONE} nor @code{FAIL}, the template given in the pattern
is inserted, as if the @code{define_expand} were a @code{define_insn}.

Once the insn list is generated, various optimization passes convert,
replace, and rearrange the insns in the insn list.  This is where the
@code{define_split} and @code{define_peephole} patterns get used, for
example.

Finally, the insn list's RTL is matched up with the RTL templates in the
@code{define_insn} patterns, and those patterns are used to emit the
final assembly code.  For this purpose, each named @code{define_insn}
acts like it's unnamed, since the names are ignored.

@node Patterns
@section Everything about Instruction Patterns
@cindex patterns
@cindex instruction patterns

@findex define_insn
Each instruction pattern contains an incomplete RTL expression, with pieces
to be filled in later, operand constraints that restrict how the pieces can
be filled in, and an output pattern or C code to generate the assembler
output, all wrapped up in a @code{define_insn} expression.

A @code{define_insn} is an RTL expression containing four or five operands:

@enumerate
@item
An optional name.  The presence of a name indicate that this instruction
pattern can perform a certain standard job for the RTL-generation
pass of the compiler.  This pass knows certain names and will use
the instruction patterns with those names, if the names are defined
in the machine description.

The absence of a name is indicated by writing an empty string
where the name should go.  Nameless instruction patterns are never
used for generating RTL code, but they may permit several simpler insns
to be combined later on.

Names that are not thus known and used in RTL-generation have no
effect; they are equivalent to no name at all.

For the purpose of debugging the compiler, you may also specify a
name beginning with the @samp{*} character.  Such a name is used only
for identifying the instruction in RTL dumps; it is entirely equivalent
to having a nameless pattern for all other purposes.

@item
The @dfn{RTL template} (@pxref{RTL Template}) is a vector of incomplete
RTL expressions which show what the instruction should look like.  It is
incomplete because it may contain @code{match_operand},
@code{match_operator}, and @code{match_dup} expressions that stand for
operands of the instruction.

If the vector has only one element, that element is the template for the
instruction pattern.  If the vector has multiple elements, then the
instruction pattern is a @code{parallel} expression containing the
elements described.

@item
@cindex pattern conditions
@cindex conditions, in patterns
A condition.  This is a string which contains a C expression that is
the final test to decide whether an insn body matches this pattern.

@cindex named patterns and conditions
For a named pattern, the condition (if present) may not depend on
the data in the insn being matched, but only the target-machine-type
flags.  The compiler needs to test these conditions during
initialization in order to learn exactly which named instructions are
available in a particular run.

@findex operands
For nameless patterns, the condition is applied only when matching an
individual insn, and only after the insn has matched the pattern's
recognition template.  The insn's operands may be found in the vector
@code{operands}.  For an insn where the condition has once matched, it
can't be used to control register allocation, for example by excluding
certain hard registers or hard register combinations.

@item
The @dfn{output template}: a string that says how to output matching
insns as assembler code.  @samp{%} in this string specifies where
to substitute the value of an operand.  @xref{Output Template}.

When simple substitution isn't general enough, you can specify a piece
of C code to compute the output.  @xref{Output Statement}.

@item
Optionally, a vector containing the values of attributes for insns matching
this pattern.  @xref{Insn Attributes}.
@end enumerate

@node Example
@section Example of @code{define_insn}
@cindex @code{define_insn} example

Here is an actual example of an instruction pattern, for the 68000/68020.

@smallexample
(define_insn "tstsi"
  [(set (cc0)
        (match_operand:SI 0 "general_operand" "rm"))]
  ""
  "*
@{
  if (TARGET_68020 || ! ADDRESS_REG_P (operands[0]))
    return \"tstl %0\";
  return \"cmpl #0,%0\";
@}")
@end smallexample

@noindent
This can also be written using braced strings:

@smallexample
(define_insn "tstsi"
  [(set (cc0)
        (match_operand:SI 0 "general_operand" "rm"))]
  ""
@{
  if (TARGET_68020 || ! ADDRESS_REG_P (operands[0]))
    return "tstl %0";
  return "cmpl #0,%0";
@})
@end smallexample

This is an instruction that sets the condition codes based on the value of
a general operand.  It has no condition, so any insn whose RTL description
has the form shown may be handled according to this pattern.  The name
@samp{tstsi} means ``test a @code{SImode} value'' and tells the RTL generation
pass that, when it is necessary to test such a value, an insn to do so
can be constructed using this pattern.

The output control string is a piece of C code which chooses which
output template to return based on the kind of operand and the specific
type of CPU for which code is being generated.

@samp{"rm"} is an operand constraint.  Its meaning is explained below.

@node RTL Template
@section RTL Template
@cindex RTL insn template
@cindex generating insns
@cindex insns, generating
@cindex recognizing insns
@cindex insns, recognizing

The RTL template is used to define which insns match the particular pattern
and how to find their operands.  For named patterns, the RTL template also
says how to construct an insn from specified operands.

Construction involves substituting specified operands into a copy of the
template.  Matching involves determining the values that serve as the
operands in the insn being matched.  Both of these activities are
controlled by special expression types that direct matching and
substitution of the operands.

@table @code
@findex match_operand
@item (match_operand:@var{m} @var{n} @var{predicate} @var{constraint})
This expression is a placeholder for operand number @var{n} of
the insn.  When constructing an insn, operand number @var{n}
will be substituted at this point.  When matching an insn, whatever
appears at this position in the insn will be taken as operand
number @var{n}; but it must satisfy @var{predicate} or this instruction
pattern will not match at all.

Operand numbers must be chosen consecutively counting from zero in
each instruction pattern.  There may be only one @code{match_operand}
expression in the pattern for each operand number.  Usually operands
are numbered in the order of appearance in @code{match_operand}
expressions.  In the case of a @code{define_expand}, any operand numbers
used only in @code{match_dup} expressions have higher values than all
other operand numbers.

@var{predicate} is a string that is the name of a function that
accepts two arguments, an expression and a machine mode.
@xref{Predicates}.  During matching, the function will be called with
the putative operand as the expression and @var{m} as the mode
argument (if @var{m} is not specified, @code{VOIDmode} will be used,
which normally causes @var{predicate} to accept any mode).  If it
returns zero, this instruction pattern fails to match.
@var{predicate} may be an empty string; then it means no test is to be
done on the operand, so anything which occurs in this position is
valid.

Most of the time, @var{predicate} will reject modes other than @var{m}---but
not always.  For example, the predicate @code{address_operand} uses
@var{m} as the mode of memory ref that the address should be valid for.
Many predicates accept @code{const_int} nodes even though their mode is
@code{VOIDmode}.

@var{constraint} controls reloading and the choice of the best register
class to use for a value, as explained later (@pxref{Constraints}).
If the constraint would be an empty string, it can be omitted.

People are often unclear on the difference between the constraint and the
predicate.  The predicate helps decide whether a given insn matches the
pattern.  The constraint plays no role in this decision; instead, it
controls various decisions in the case of an insn which does match.

@findex match_scratch
@item (match_scratch:@var{m} @var{n} @var{constraint})
This expression is also a placeholder for operand number @var{n}
and indicates that operand must be a @code{scratch} or @code{reg}
expression.

When matching patterns, this is equivalent to

@smallexample
(match_operand:@var{m} @var{n} "scratch_operand" @var{pred})
@end smallexample

but, when generating RTL, it produces a (@code{scratch}:@var{m})
expression.

If the last few expressions in a @code{parallel} are @code{clobber}
expressions whose operands are either a hard register or
@code{match_scratch}, the combiner can add or delete them when
necessary.  @xref{Side Effects}.

@findex match_dup
@item (match_dup @var{n})
This expression is also a placeholder for operand number @var{n}.
It is used when the operand needs to appear more than once in the
insn.

In construction, @code{match_dup} acts just like @code{match_operand}:
the operand is substituted into the insn being constructed.  But in
matching, @code{match_dup} behaves differently.  It assumes that operand
number @var{n} has already been determined by a @code{match_operand}
appearing earlier in the recognition template, and it matches only an
identical-looking expression.

Note that @code{match_dup} should not be used to tell the compiler that
a particular register is being used for two operands (example:
@code{add} that adds one register to another; the second register is
both an input operand and the output operand).  Use a matching
constraint (@pxref{Simple Constraints}) for those.  @code{match_dup} is for the cases where one
operand is used in two places in the template, such as an instruction
that computes both a quotient and a remainder, where the opcode takes
two input operands but the RTL template has to refer to each of those
twice; once for the quotient pattern and once for the remainder pattern.

@findex match_operator
@item (match_operator:@var{m} @var{n} @var{predicate} [@var{operands}@dots{}])
This pattern is a kind of placeholder for a variable RTL expression
code.

When constructing an insn, it stands for an RTL expression whose
expression code is taken from that of operand @var{n}, and whose
operands are constructed from the patterns @var{operands}.

When matching an expression, it matches an expression if the function
@var{predicate} returns nonzero on that expression @emph{and} the
patterns @var{operands} match the operands of the expression.

Suppose that the function @code{commutative_operator} is defined as
follows, to match any expression whose operator is one of the
commutative arithmetic operators of RTL and whose mode is @var{mode}:

@smallexample
int
commutative_integer_operator (x, mode)
     rtx x;
     enum machine_mode mode;
@{
  enum rtx_code code = GET_CODE (x);
  if (GET_MODE (x) != mode)
    return 0;
  return (GET_RTX_CLASS (code) == RTX_COMM_ARITH
          || code == EQ || code == NE);
@}
@end smallexample

Then the following pattern will match any RTL expression consisting
of a commutative operator applied to two general operands:

@smallexample
(match_operator:SI 3 "commutative_operator"
  [(match_operand:SI 1 "general_operand" "g")
   (match_operand:SI 2 "general_operand" "g")])
@end smallexample

Here the vector @code{[@var{operands}@dots{}]} contains two patterns
because the expressions to be matched all contain two operands.

When this pattern does match, the two operands of the commutative
operator are recorded as operands 1 and 2 of the insn.  (This is done
by the two instances of @code{match_operand}.)  Operand 3 of the insn
will be the entire commutative expression: use @code{GET_CODE
(operands[3])} to see which commutative operator was used.

The machine mode @var{m} of @code{match_operator} works like that of
@code{match_operand}: it is passed as the second argument to the
predicate function, and that function is solely responsible for
deciding whether the expression to be matched ``has'' that mode.

When constructing an insn, argument 3 of the gen-function will specify
the operation (i.e.@: the expression code) for the expression to be
made.  It should be an RTL expression, whose expression code is copied
into a new expression whose operands are arguments 1 and 2 of the
gen-function.  The subexpressions of argument 3 are not used;
only its expression code matters.

When @code{match_operator} is used in a pattern for matching an insn,
it usually best if the operand number of the @code{match_operator}
is higher than that of the actual operands of the insn.  This improves
register allocation because the register allocator often looks at
operands 1 and 2 of insns to see if it can do register tying.

There is no way to specify constraints in @code{match_operator}.  The
operand of the insn which corresponds to the @code{match_operator}
never has any constraints because it is never reloaded as a whole.
However, if parts of its @var{operands} are matched by
@code{match_operand} patterns, those parts may have constraints of
their own.

@findex match_op_dup
@item (match_op_dup:@var{m} @var{n}[@var{operands}@dots{}])
Like @code{match_dup}, except that it applies to operators instead of
operands.  When constructing an insn, operand number @var{n} will be
substituted at this point.  But in matching, @code{match_op_dup} behaves
differently.  It assumes that operand number @var{n} has already been
determined by a @code{match_operator} appearing earlier in the
recognition template, and it matches only an identical-looking
expression.

@findex match_parallel
@item (match_parallel @var{n} @var{predicate} [@var{subpat}@dots{}])
This pattern is a placeholder for an insn that consists of a
@code{parallel} expression with a variable number of elements.  This
expression should only appear at the top level of an insn pattern.

When constructing an insn, operand number @var{n} will be substituted at
this point.  When matching an insn, it matches if the body of the insn
is a @code{parallel} expression with at least as many elements as the
vector of @var{subpat} expressions in the @code{match_parallel}, if each
@var{subpat} matches the corresponding element of the @code{parallel},
@emph{and} the function @var{predicate} returns nonzero on the
@code{parallel} that is the body of the insn.  It is the responsibility
of the predicate to validate elements of the @code{parallel} beyond
those listed in the @code{match_parallel}.

A typical use of @code{match_parallel} is to match load and store
multiple expressions, which can contain a variable number of elements
in a @code{parallel}.  For example,

@smallexample
(define_insn ""
  [(match_parallel 0 "load_multiple_operation"
     [(set (match_operand:SI 1 "gpc_reg_operand" "=r")
           (match_operand:SI 2 "memory_operand" "m"))
      (use (reg:SI 179))
      (clobber (reg:SI 179))])]
  ""
  "loadm 0,0,%1,%2")
@end smallexample

This example comes from @file{a29k.md}.  The function
@code{load_multiple_operation} is defined in @file{a29k.c} and checks
that subsequent elements in the @code{parallel} are the same as the
@code{set} in the pattern, except that they are referencing subsequent
registers and memory locations.

An insn that matches this pattern might look like:

@smallexample
(parallel
 [(set (reg:SI 20) (mem:SI (reg:SI 100)))
  (use (reg:SI 179))
  (clobber (reg:SI 179))
  (set (reg:SI 21)
       (mem:SI (plus:SI (reg:SI 100)
                        (const_int 4))))
  (set (reg:SI 22)
       (mem:SI (plus:SI (reg:SI 100)
                        (const_int 8))))])
@end smallexample

@findex match_par_dup
@item (match_par_dup @var{n} [@var{subpat}@dots{}])
Like @code{match_op_dup}, but for @code{match_parallel} instead of
@code{match_operator}.

@end table

@node Output Template
@section Output Templates and Operand Substitution
@cindex output templates
@cindex operand substitution

@cindex @samp{%} in template
@cindex percent sign
The @dfn{output template} is a string which specifies how to output the
assembler code for an instruction pattern.  Most of the template is a
fixed string which is output literally.  The character @samp{%} is used
to specify where to substitute an operand; it can also be used to
identify places where different variants of the assembler require
different syntax.

In the simplest case, a @samp{%} followed by a digit @var{n} says to output
operand @var{n} at that point in the string.

@samp{%} followed by a letter and a digit says to output an operand in an
alternate fashion.  Four letters have standard, built-in meanings described
below.  The machine description macro @code{PRINT_OPERAND} can define
additional letters with nonstandard meanings.

@samp{%c@var{digit}} can be used to substitute an operand that is a
constant value without the syntax that normally indicates an immediate
operand.

@samp{%n@var{digit}} is like @samp{%c@var{digit}} except that the value of
the constant is negated before printing.

@samp{%a@var{digit}} can be used to substitute an operand as if it were a
memory reference, with the actual operand treated as the address.  This may
be useful when outputting a ``load address'' instruction, because often the
assembler syntax for such an instruction requires you to write the operand
as if it were a memory reference.

@samp{%l@var{digit}} is used to substitute a @code{label_ref} into a jump
instruction.

@samp{%=} outputs a number which is unique to each instruction in the
entire compilation.  This is useful for making local labels to be
referred to more than once in a single template that generates multiple
assembler instructions.

@samp{%} followed by a punctuation character specifies a substitution that
does not use an operand.  Only one case is standard: @samp{%%} outputs a
@samp{%} into the assembler code.  Other nonstandard cases can be
defined in the @code{PRINT_OPERAND} macro.  You must also define
which punctuation characters are valid with the
@code{PRINT_OPERAND_PUNCT_VALID_P} macro.

@cindex \
@cindex backslash
The template may generate multiple assembler instructions.  Write the text
for the instructions, with @samp{\;} between them.

@cindex matching operands
When the RTL contains two operands which are required by constraint to match
each other, the output template must refer only to the lower-numbered operand.
Matching operands are not always identical, and the rest of the compiler
arranges to put the proper RTL expression for printing into the lower-numbered
operand.

One use of nonstandard letters or punctuation following @samp{%} is to
distinguish between different assembler languages for the same machine; for
example, Motorola syntax versus MIT syntax for the 68000.  Motorola syntax
requires periods in most opcode names, while MIT syntax does not.  For
example, the opcode @samp{movel} in MIT syntax is @samp{move.l} in Motorola
syntax.  The same file of patterns is used for both kinds of output syntax,
but the character sequence @samp{%.} is used in each place where Motorola
syntax wants a period.  The @code{PRINT_OPERAND} macro for Motorola syntax
defines the sequence to output a period; the macro for MIT syntax defines
it to do nothing.

@cindex @code{#} in template
As a special case, a template consisting of the single character @code{#}
instructs the compiler to first split the insn, and then output the
resulting instructions separately.  This helps eliminate redundancy in the
output templates.   If you have a @code{define_insn} that needs to emit
multiple assembler instructions, and there is an matching @code{define_split}
already defined, then you can simply use @code{#} as the output template
instead of writing an output template that emits the multiple assembler
instructions.

If the macro @code{ASSEMBLER_DIALECT} is defined, you can use construct
of the form @samp{@{option0|option1|option2@}} in the templates.  These
describe multiple variants of assembler language syntax.
@xref{Instruction Output}.

@node Output Statement
@section C Statements for Assembler Output
@cindex output statements
@cindex C statements for assembler output
@cindex generating assembler output

Often a single fixed template string cannot produce correct and efficient
assembler code for all the cases that are recognized by a single
instruction pattern.  For example, the opcodes may depend on the kinds of
operands; or some unfortunate combinations of operands may require extra
machine instructions.

If the output control string starts with a @samp{@@}, then it is actually
a series of templates, each on a separate line.  (Blank lines and
leading spaces and tabs are ignored.)  The templates correspond to the
pattern's constraint alternatives (@pxref{Multi-Alternative}).  For example,
if a target machine has a two-address add instruction @samp{addr} to add
into a register and another @samp{addm} to add a register to memory, you
might write this pattern:

@smallexample
(define_insn "addsi3"
  [(set (match_operand:SI 0 "general_operand" "=r,m")
        (plus:SI (match_operand:SI 1 "general_operand" "0,0")
                 (match_operand:SI 2 "general_operand" "g,r")))]
  ""
  "@@
   addr %2,%0
   addm %2,%0")
@end smallexample

@cindex @code{*} in template
@cindex asterisk in template
If the output control string starts with a @samp{*}, then it is not an
output template but rather a piece of C program that should compute a
template.  It should execute a @code{return} statement to return the
template-string you want.  Most such templates use C string literals, which
require doublequote characters to delimit them.  To include these
doublequote characters in the string, prefix each one with @samp{\}.

If the output control string is written as a brace block instead of a
double-quoted string, it is automatically assumed to be C code.  In that
case, it is not necessary to put in a leading asterisk, or to escape the
doublequotes surrounding C string literals.

The operands may be found in the array @code{operands}, whose C data type
is @code{rtx []}.

It is very common to select different ways of generating assembler code
based on whether an immediate operand is within a certain range.  Be
careful when doing this, because the result of @code{INTVAL} is an
integer on the host machine.  If the host machine has more bits in an
@code{int} than the target machine has in the mode in which the constant
will be used, then some of the bits you get from @code{INTVAL} will be
superfluous.  For proper results, you must carefully disregard the
values of those bits.

@findex output_asm_insn
It is possible to output an assembler instruction and then go on to output
or compute more of them, using the subroutine @code{output_asm_insn}.  This
receives two arguments: a template-string and a vector of operands.  The
vector may be @code{operands}, or it may be another array of @code{rtx}
that you declare locally and initialize yourself.

@findex which_alternative
When an insn pattern has multiple alternatives in its constraints, often
the appearance of the assembler code is determined mostly by which alternative
was matched.  When this is so, the C code can test the variable
@code{which_alternative}, which is the ordinal number of the alternative
that was actually satisfied (0 for the first, 1 for the second alternative,
etc.).

For example, suppose there are two opcodes for storing zero, @samp{clrreg}
for registers and @samp{clrmem} for memory locations.  Here is how
a pattern could use @code{which_alternative} to choose between them:

@smallexample
(define_insn ""
  [(set (match_operand:SI 0 "general_operand" "=r,m")
        (const_int 0))]
  ""
  @{
  return (which_alternative == 0
          ? "clrreg %0" : "clrmem %0");
  @})
@end smallexample

The example above, where the assembler code to generate was
@emph{solely} determined by the alternative, could also have been specified
as follows, having the output control string start with a @samp{@@}:

@smallexample
@group
(define_insn ""
  [(set (match_operand:SI 0 "general_operand" "=r,m")
        (const_int 0))]
  ""
  "@@
   clrreg %0
   clrmem %0")
@end group
@end smallexample

@node Predicates
@section Predicates
@cindex predicates
@cindex operand predicates
@cindex operator predicates

A predicate determines whether a @code{match_operand} or
@code{match_operator} expression matches, and therefore whether the
surrounding instruction pattern will be used for that combination of
operands.  GCC has a number of machine-independent predicates, and you
can define machine-specific predicates as needed.  By convention,
predicates used with @code{match_operand} have names that end in
@samp{_operand}, and those used with @code{match_operator} have names
that end in @samp{_operator}.

All predicates are Boolean functions (in the mathematical sense) of
two arguments: the RTL expression that is being considered at that
position in the instruction pattern, and the machine mode that the
@code{match_operand} or @code{match_operator} specifies.  In this
section, the first argument is called @var{op} and the second argument
@var{mode}.  Predicates can be called from C as ordinary two-argument
functions; this can be useful in output templates or other
machine-specific code.

Operand predicates can allow operands that are not actually acceptable
to the hardware, as long as the constraints give reload the ability to
fix them up (@pxref{Constraints}).  However, GCC will usually generate
better code if the predicates specify the requirements of the machine
instructions as closely as possible.  Reload cannot fix up operands
that must be constants (``immediate operands''); you must use a
predicate that allows only constants, or else enforce the requirement
in the extra condition.

@cindex predicates and machine modes
@cindex normal predicates
@cindex special predicates
Most predicates handle their @var{mode} argument in a uniform manner.
If @var{mode} is @code{VOIDmode} (unspecified), then @var{op} can have
any mode.  If @var{mode} is anything else, then @var{op} must have the
same mode, unless @var{op} is a @code{CONST_INT} or integer
@code{CONST_DOUBLE}.  These RTL expressions always have
@code{VOIDmode}, so it would be counterproductive to check that their
mode matches.  Instead, predicates that accept @code{CONST_INT} and/or
integer @code{CONST_DOUBLE} check that the value stored in the
constant will fit in the requested mode.

Predicates with this behavior are called @dfn{normal}.
@command{genrecog} can optimize the instruction recognizer based on
knowledge of how normal predicates treat modes.  It can also diagnose
certain kinds of common errors in the use of normal predicates; for
instance, it is almost always an error to use a normal predicate
without specifying a mode.

Predicates that do something different with their @var{mode} argument
are called @dfn{special}.  The generic predicates
@code{address_operand} and @code{pmode_register_operand} are special
predicates.  @command{genrecog} does not do any optimizations or
diagnosis when special predicates are used.

@menu
* Machine-Independent Predicates::  Predicates available to all back ends.
* Defining Predicates::             How to write machine-specific predicate
                                    functions.
@end menu

@node Machine-Independent Predicates
@subsection Machine-Independent Predicates
@cindex machine-independent predicates
@cindex generic predicates

These are the generic predicates available to all back ends.  They are
defined in @file{recog.c}.  The first category of predicates allow
only constant, or @dfn{immediate}, operands.

@defun immediate_operand
This predicate allows any sort of constant that fits in @var{mode}.
It is an appropriate choice for instructions that take operands that
must be constant.
@end defun

@defun const_int_operand
This predicate allows any @code{CONST_INT} expression that fits in
@var{mode}.  It is an appropriate choice for an immediate operand that
does not allow a symbol or label.
@end defun

@defun const_double_operand
This predicate accepts any @code{CONST_DOUBLE} expression that has
exactly @var{mode}.  If @var{mode} is @code{VOIDmode}, it will also
accept @code{CONST_INT}.  It is intended for immediate floating point
constants.
@end defun

@noindent
The second category of predicates allow only some kind of machine
register.

@defun register_operand
This predicate allows any @code{REG} or @code{SUBREG} expression that
is valid for @var{mode}.  It is often suitable for arithmetic
instruction operands on a RISC machine.
@end defun

@defun pmode_register_operand
This is a slight variant on @code{register_operand} which works around
a limitation in the machine-description reader.

@smallexample
(match_operand @var{n} "pmode_register_operand" @var{constraint})
@end smallexample

@noindent
means exactly what

@smallexample
(match_operand:P @var{n} "register_operand" @var{constraint})
@end smallexample

@noindent
would mean, if the machine-description reader accepted @samp{:P}
mode suffixes.  Unfortunately, it cannot, because @code{Pmode} is an
alias for some other mode, and might vary with machine-specific
options.  @xref{Misc}.
@end defun

@defun scratch_operand
This predicate allows hard registers and @code{SCRATCH} expressions,
but not pseudo-registers.  It is used internally by @code{match_scratch};
it should not be used directly.
@end defun

@noindent
The third category of predicates allow only some kind of memory reference.

@defun memory_operand
This predicate allows any valid reference to a quantity of mode
@var{mode} in memory, as determined by the weak form of
@code{GO_IF_LEGITIMATE_ADDRESS} (@pxref{Addressing Modes}).
@end defun

@defun address_operand
This predicate is a little unusual; it allows any operand that is a
valid expression for the @emph{address} of a quantity of mode
@var{mode}, again determined by the weak form of
@code{GO_IF_LEGITIMATE_ADDRESS}.  To first order, if
@samp{@w{(mem:@var{mode} (@var{exp}))}} is acceptable to
@code{memory_operand}, then @var{exp} is acceptable to
@code{address_operand}.  Note that @var{exp} does not necessarily have
the mode @var{mode}.
@end defun

@defun indirect_operand
This is a stricter form of @code{memory_operand} which allows only
memory references with a @code{general_operand} as the address
expression.  New uses of this predicate are discouraged, because
@code{general_operand} is very permissive, so it's hard to tell what
an @code{indirect_operand} does or does not allow.  If a target has
different requirements for memory operands for different instructions,
it is better to define target-specific predicates which enforce the
hardware's requirements explicitly.
@end defun

@defun push_operand
This predicate allows a memory reference suitable for pushing a value
onto the stack.  This will be a @code{MEM} which refers to
@code{stack_pointer_rtx}, with a side-effect in its address expression
(@pxref{Incdec}); which one is determined by the
@code{STACK_PUSH_CODE} macro (@pxref{Frame Layout}).
@end defun

@defun pop_operand
This predicate allows a memory reference suitable for popping a value
off the stack.  Again, this will be a @code{MEM} referring to
@code{stack_pointer_rtx}, with a side-effect in its address
expression.  However, this time @code{STACK_POP_CODE} is expected.
@end defun

@noindent
The fourth category of predicates allow some combination of the above
operands.

@defun nonmemory_operand
This predicate allows any immediate or register operand valid for @var{mode}.
@end defun

@defun nonimmediate_operand
This predicate allows any register or memory operand valid for @var{mode}.
@end defun

@defun general_operand
This predicate allows any immediate, register, or memory operand
valid for @var{mode}.
@end defun

@noindent
Finally, there is one generic operator predicate.

@defun comparison_operator
This predicate matches any expression which performs an arithmetic
comparison in @var{mode}; that is, @code{COMPARISON_P} is true for the
expression code.
@end defun

@node Defining Predicates
@subsection Defining Machine-Specific Predicates
@cindex defining predicates
@findex define_predicate
@findex define_special_predicate

Many machines have requirements for their operands that cannot be
expressed precisely using the generic predicates.  You can define
additional predicates using @code{define_predicate} and
@code{define_special_predicate} expressions.  These expressions have
three operands:

@itemize @bullet
@item
The name of the predicate, as it will be referred to in
@code{match_operand} or @code{match_operator} expressions.

@item
An RTL expression which evaluates to true if the predicate allows the
operand @var{op}, false if it does not.  This expression can only use
the following RTL codes:

@table @code
@item MATCH_OPERAND
When written inside a predicate expression, a @code{MATCH_OPERAND}
expression evaluates to true if the predicate it names would allow
@var{op}.  The operand number and constraint are ignored.  Due to
limitations in @command{genrecog}, you can only refer to generic
predicates and predicates that have already been defined.

@item MATCH_CODE
This expression evaluates to true if @var{op} or a specified
subexpression of @var{op} has one of a given list of RTX codes.

The first operand of this expression is a string constant containing a
comma-separated list of RTX code names (in lower case).  These are the
codes for which the @code{MATCH_CODE} will be true.

The second operand is a string constant which indicates what
subexpression of @var{op} to examine.  If it is absent or the empty
string, @var{op} itself is examined.  Otherwise, the string constant
must be a sequence of digits and/or lowercase letters.  Each character
indicates a subexpression to extract from the current expression; for
the first character this is @var{op}, for the second and subsequent
characters it is the result of the previous character.  A digit
@var{n} extracts @samp{@w{XEXP (@var{e}, @var{n})}}; a letter @var{l}
extracts @samp{@w{XVECEXP (@var{e}, 0, @var{n})}} where @var{n} is the
alphabetic ordinal of @var{l} (0 for `a', 1 for 'b', and so on).  The
@code{MATCH_CODE} then examines the RTX code of the subexpression
extracted by the complete string.  It is not possible to extract
components of an @code{rtvec} that is not at position 0 within its RTX
object.

@item MATCH_TEST
This expression has one operand, a string constant containing a C
expression.  The predicate's arguments, @var{op} and @var{mode}, are
available with those names in the C expression.  The @code{MATCH_TEST}
evaluates to true if the C expression evaluates to a nonzero value.
@code{MATCH_TEST} expressions must not have side effects.

@item  AND
@itemx IOR
@itemx NOT
@itemx IF_THEN_ELSE
The basic @samp{MATCH_} expressions can be combined using these
logical operators, which have the semantics of the C operators
@samp{&&}, @samp{||}, @samp{!}, and @samp{@w{? :}} respectively.  As
in Common Lisp, you may give an @code{AND} or @code{IOR} expression an
arbitrary number of arguments; this has exactly the same effect as
writing a chain of two-argument @code{AND} or @code{IOR} expressions.
@end table

@item
An optional block of C code, which should execute
@samp{@w{return true}} if the predicate is found to match and
@samp{@w{return false}} if it does not.  It must not have any side
effects.  The predicate arguments, @var{op} and @var{mode}, are
available with those names.

If a code block is present in a predicate definition, then the RTL
expression must evaluate to true @emph{and} the code block must
execute @samp{@w{return true}} for the predicate to allow the operand.
The RTL expression is evaluated first; do not re-check anything in the
code block that was checked in the RTL expression.
@end itemize

The program @command{genrecog} scans @code{define_predicate} and
@code{define_special_predicate} expressions to determine which RTX
codes are possibly allowed.  You should always make this explicit in
the RTL predicate expression, using @code{MATCH_OPERAND} and
@code{MATCH_CODE}.

Here is an example of a simple predicate definition, from the IA64
machine description:

@smallexample
@group
;; @r{True if @var{op} is a @code{SYMBOL_REF} which refers to the sdata section.}
(define_predicate "small_addr_symbolic_operand"
  (and (match_code "symbol_ref")
       (match_test "SYMBOL_REF_SMALL_ADDR_P (op)")))
@end group
@end smallexample

@noindent
And here is another, showing the use of the C block.

@smallexample
@group
;; @r{True if @var{op} is a register operand that is (or could be) a GR reg.}
(define_predicate "gr_register_operand"
  (match_operand 0 "register_operand")
@{
  unsigned int regno;
  if (GET_CODE (op) == SUBREG)
    op = SUBREG_REG (op);

  regno = REGNO (op);
  return (regno >= FIRST_PSEUDO_REGISTER || GENERAL_REGNO_P (regno));
@})
@end group
@end smallexample

Predicates written with @code{define_predicate} automatically include
a test that @var{mode} is @code{VOIDmode}, or @var{op} has the same
mode as @var{mode}, or @var{op} is a @code{CONST_INT} or
@code{CONST_DOUBLE}.  They do @emph{not} check specifically for
integer @code{CONST_DOUBLE}, nor do they test that the value of either
kind of constant fits in the requested mode.  This is because
target-specific predicates that take constants usually have to do more
stringent value checks anyway.  If you need the exact same treatment
of @code{CONST_INT} or @code{CONST_DOUBLE} that the generic predicates
provide, use a @code{MATCH_OPERAND} subexpression to call
@code{const_int_operand}, @code{const_double_operand}, or
@code{immediate_operand}.

Predicates written with @code{define_special_predicate} do not get any
automatic mode checks, and are treated as having special mode handling
by @command{genrecog}.

The program @command{genpreds} is responsible for generating code to
test predicates.  It also writes a header file containing function
declarations for all machine-specific predicates.  It is not necessary
to declare these predicates in @file{@var{cpu}-protos.h}.
@end ifset

@c Most of this node appears by itself (in a different place) even
@c when the INTERNALS flag is clear.  Passages that require the internals
@c manual's context are conditionalized to appear only in the internals manual.
@ifset INTERNALS
@node Constraints
@section Operand Constraints
@cindex operand constraints
@cindex constraints

Each @code{match_operand} in an instruction pattern can specify
constraints for the operands allowed.  The constraints allow you to
fine-tune matching within the set of operands allowed by the
predicate.

@end ifset
@ifclear INTERNALS
@node Constraints
@section Constraints for @code{asm} Operands
@cindex operand constraints, @code{asm}
@cindex constraints, @code{asm}
@cindex @code{asm} constraints

Here are specific details on what constraint letters you can use with
@code{asm} operands.
@end ifclear
Constraints can say whether
an operand may be in a register, and which kinds of register; whether the
operand can be a memory reference, and which kinds of address; whether the
operand may be an immediate constant, and which possible values it may
have.  Constraints can also require two operands to match.

@ifset INTERNALS
@menu
* Simple Constraints::  Basic use of constraints.
* Multi-Alternative::   When an insn has two alternative constraint-patterns.
* Class Preferences::   Constraints guide which hard register to put things in.
* Modifiers::           More precise control over effects of constraints.
* Machine Constraints:: Existing constraints for some particular machines.
* Define Constraints::  How to define machine-specific constraints.
* C Constraint Interface:: How to test constraints from C code.
@end menu
@end ifset

@ifclear INTERNALS
@menu
* Simple Constraints::  Basic use of constraints.
* Multi-Alternative::   When an insn has two alternative constraint-patterns.
* Modifiers::           More precise control over effects of constraints.
* Machine Constraints:: Special constraints for some particular machines.
@end menu
@end ifclear

@node Simple Constraints
@subsection Simple Constraints
@cindex simple constraints

The simplest kind of constraint is a string full of letters, each of
which describes one kind of operand that is permitted.  Here are
the letters that are allowed:

@table @asis
@item whitespace
Whitespace characters are ignored and can be inserted at any position
except the first.  This enables each alternative for different operands to
be visually aligned in the machine description even if they have different
number of constraints and modifiers.

@cindex @samp{m} in constraint
@cindex memory references in constraints
@item @samp{m}
A memory operand is allowed, with any kind of address that the machine
supports in general.

@cindex offsettable address
@cindex @samp{o} in constraint
@item @samp{o}
A memory operand is allowed, but only if the address is
@dfn{offsettable}.  This means that adding a small integer (actually,
the width in bytes of the operand, as determined by its machine mode)
may be added to the address and the result is also a valid memory
address.

@cindex autoincrement/decrement addressing
For example, an address which is constant is offsettable; so is an
address that is the sum of a register and a constant (as long as a
slightly larger constant is also within the range of address-offsets
supported by the machine); but an autoincrement or autodecrement
address is not offsettable.  More complicated indirect/indexed
addresses may or may not be offsettable depending on the other
addressing modes that the machine supports.

Note that in an output operand which can be matched by another
operand, the constraint letter @samp{o} is valid only when accompanied
by both @samp{<} (if the target machine has predecrement addressing)
and @samp{>} (if the target machine has preincrement addressing).

@cindex @samp{V} in constraint
@item @samp{V}
A memory operand that is not offsettable.  In other words, anything that
would fit the @samp{m} constraint but not the @samp{o} constraint.

@cindex @samp{<} in constraint
@item @samp{<}
A memory operand with autodecrement addressing (either predecrement or
postdecrement) is allowed.

@cindex @samp{>} in constraint
@item @samp{>}
A memory operand with autoincrement addressing (either preincrement or
postincrement) is allowed.

@cindex @samp{r} in constraint
@cindex registers in constraints
@item @samp{r}
A register operand is allowed provided that it is in a general
register.

@cindex constants in constraints
@cindex @samp{i} in constraint
@item @samp{i}
An immediate integer operand (one with constant value) is allowed.
This includes symbolic constants whose values will be known only at
assembly time or later.

@cindex @samp{n} in constraint
@item @samp{n}
An immediate integer operand with a known numeric value is allowed.
Many systems cannot support assembly-time constants for operands less
than a word wide.  Constraints for these operands should use @samp{n}
rather than @samp{i}.

@cindex @samp{I} in constraint
@item @samp{I}, @samp{J}, @samp{K}, @dots{} @samp{P}
Other letters in the range @samp{I} through @samp{P} may be defined in
a machine-dependent fashion to permit immediate integer operands with
explicit integer values in specified ranges.  For example, on the
68000, @samp{I} is defined to stand for the range of values 1 to 8.
This is the range permitted as a shift count in the shift
instructions.

@cindex @samp{E} in constraint
@item @samp{E}
An immediate floating operand (expression code @code{const_double}) is
allowed, but only if the target floating point format is the same as
that of the host machine (on which the compiler is running).

@cindex @samp{F} in constraint
@item @samp{F}
An immediate floating operand (expression code @code{const_double} or
@code{const_vector}) is allowed.

@cindex @samp{G} in constraint
@cindex @samp{H} in constraint
@item @samp{G}, @samp{H}
@samp{G} and @samp{H} may be defined in a machine-dependent fashion to
permit immediate floating operands in particular ranges of values.

@cindex @samp{s} in constraint
@item @samp{s}
An immediate integer operand whose value is not an explicit integer is
allowed.

This might appear strange; if an insn allows a constant operand with a
value not known at compile time, it certainly must allow any known
value.  So why use @samp{s} instead of @samp{i}?  Sometimes it allows
better code to be generated.

For example, on the 68000 in a fullword instruction it is possible to
use an immediate operand; but if the immediate value is between @minus{}128
and 127, better code results from loading the value into a register and
using the register.  This is because the load into the register can be
done with a @samp{moveq} instruction.  We arrange for this to happen
by defining the letter @samp{K} to mean ``any integer outside the
range @minus{}128 to 127'', and then specifying @samp{Ks} in the operand
constraints.

@cindex @samp{g} in constraint
@item @samp{g}
Any register, memory or immediate integer operand is allowed, except for
registers that are not general registers.

@cindex @samp{X} in constraint
@item @samp{X}
@ifset INTERNALS
Any operand whatsoever is allowed, even if it does not satisfy
@code{general_operand}.  This is normally used in the constraint of
a @code{match_scratch} when certain alternatives will not actually
require a scratch register.
@end ifset
@ifclear INTERNALS
Any operand whatsoever is allowed.
@end ifclear

@cindex @samp{0} in constraint
@cindex digits in constraint
@item @samp{0}, @samp{1}, @samp{2}, @dots{} @samp{9}
An operand that matches the specified operand number is allowed.  If a
digit is used together with letters within the same alternative, the
digit should come last.

This number is allowed to be more than a single digit.  If multiple
digits are encountered consecutively, they are interpreted as a single
decimal integer.  There is scant chance for ambiguity, since to-date
it has never been desirable that @samp{10} be interpreted as matching
either operand 1 @emph{or} operand 0.  Should this be desired, one
can use multiple alternatives instead.

@cindex matching constraint
@cindex constraint, matching
This is called a @dfn{matching constraint} and what it really means is
that the assembler has only a single operand that fills two roles
@ifset INTERNALS
considered separate in the RTL insn.  For example, an add insn has two
input operands and one output operand in the RTL, but on most CISC
@end ifset
@ifclear INTERNALS
which @code{asm} distinguishes.  For example, an add instruction uses
two input operands and an output operand, but on most CISC
@end ifclear
machines an add instruction really has only two operands, one of them an
input-output operand:

@smallexample
addl #35,r12
@end smallexample

Matching constraints are used in these circumstances.
More precisely, the two operands that match must include one input-only
operand and one output-only operand.  Moreover, the digit must be a
smaller number than the number of the operand that uses it in the
constraint.

@ifset INTERNALS
For operands to match in a particular case usually means that they
are identical-looking RTL expressions.  But in a few special cases
specific kinds of dissimilarity are allowed.  For example, @code{*x}
as an input operand will match @code{*x++} as an output operand.
For proper results in such cases, the output template should always
use the output-operand's number when printing the operand.
@end ifset

@cindex load address instruction
@cindex push address instruction
@cindex address constraints
@cindex @samp{p} in constraint
@item @samp{p}
An operand that is a valid memory address is allowed.  This is
for ``load address'' and ``push address'' instructions.

@findex address_operand
@samp{p} in the constraint must be accompanied by @code{address_operand}
as the predicate in the @code{match_operand}.  This predicate interprets
the mode specified in the @code{match_operand} as the mode of the memory
reference for which the address would be valid.

@cindex other register constraints
@cindex extensible constraints
@item @var{other-letters}
Other letters can be defined in machine-dependent fashion to stand for
particular classes of registers or other arbitrary operand types.
@samp{d}, @samp{a} and @samp{f} are defined on the 68000/68020 to stand
for data, address and floating point registers.
@end table

@ifset INTERNALS
In order to have valid assembler code, each operand must satisfy
its constraint.  But a failure to do so does not prevent the pattern
from applying to an insn.  Instead, it directs the compiler to modify
the code so that the constraint will be satisfied.  Usually this is
done by copying an operand into a register.

Contrast, therefore, the two instruction patterns that follow:

@smallexample
(define_insn ""
  [(set (match_operand:SI 0 "general_operand" "=r")
        (plus:SI (match_dup 0)
                 (match_operand:SI 1 "general_operand" "r")))]
  ""
  "@dots{}")
@end smallexample

@noindent
which has two operands, one of which must appear in two places, and

@smallexample
(define_insn ""
  [(set (match_operand:SI 0 "general_operand" "=r")
        (plus:SI (match_operand:SI 1 "general_operand" "0")
                 (match_operand:SI 2 "general_operand" "r")))]
  ""
  "@dots{}")
@end smallexample

@noindent
which has three operands, two of which are required by a constraint to be
identical.  If we are considering an insn of the form

@smallexample
(insn @var{n} @var{prev} @var{next}
  (set (reg:SI 3)
       (plus:SI (reg:SI 6) (reg:SI 109)))
  @dots{})
@end smallexample

@noindent
the first pattern would not apply at all, because this insn does not
contain two identical subexpressions in the right place.  The pattern would
say, ``That does not look like an add instruction; try other patterns''.
The second pattern would say, ``Yes, that's an add instruction, but there
is something wrong with it''.  It would direct the reload pass of the
compiler to generate additional insns to make the constraint true.  The
results might look like this:

@smallexample
(insn @var{n2} @var{prev} @var{n}
  (set (reg:SI 3) (reg:SI 6))
  @dots{})

(insn @var{n} @var{n2} @var{next}
  (set (reg:SI 3)
       (plus:SI (reg:SI 3) (reg:SI 109)))
  @dots{})
@end smallexample

It is up to you to make sure that each operand, in each pattern, has
constraints that can handle any RTL expression that could be present for
that operand.  (When multiple alternatives are in use, each pattern must,
for each possible combination of operand expressions, have at least one
alternative which can handle that combination of operands.)  The
constraints don't need to @emph{allow} any possible operand---when this is
the case, they do not constrain---but they must at least point the way to
reloading any possible operand so that it will fit.

@itemize @bullet
@item
If the constraint accepts whatever operands the predicate permits,
there is no problem: reloading is never necessary for this operand.

For example, an operand whose constraints permit everything except
registers is safe provided its predicate rejects registers.

An operand whose predicate accepts only constant values is safe
provided its constraints include the letter @samp{i}.  If any possible
constant value is accepted, then nothing less than @samp{i} will do;
if the predicate is more selective, then the constraints may also be
more selective.

@item
Any operand expression can be reloaded by copying it into a register.
So if an operand's constraints allow some kind of register, it is
certain to be safe.  It need not permit all classes of registers; the
compiler knows how to copy a register into another register of the
proper class in order to make an instruction valid.

@cindex nonoffsettable memory reference
@cindex memory reference, nonoffsettable
@item
A nonoffsettable memory reference can be reloaded by copying the
address into a register.  So if the constraint uses the letter
@samp{o}, all memory references are taken care of.

@item
A constant operand can be reloaded by allocating space in memory to
hold it as preinitialized data.  Then the memory reference can be used
in place of the constant.  So if the constraint uses the letters
@samp{o} or @samp{m}, constant operands are not a problem.

@item
If the constraint permits a constant and a pseudo register used in an insn
was not allocated to a hard register and is equivalent to a constant,
the register will be replaced with the constant.  If the predicate does
not permit a constant and the insn is re-recognized for some reason, the
compiler will crash.  Thus the predicate must always recognize any
objects allowed by the constraint.
@end itemize

If the operand's predicate can recognize registers, but the constraint does
not permit them, it can make the compiler crash.  When this operand happens
to be a register, the reload pass will be stymied, because it does not know
how to copy a register temporarily into memory.

If the predicate accepts a unary operator, the constraint applies to the
operand.  For example, the MIPS processor at ISA level 3 supports an
instruction which adds two registers in @code{SImode} to produce a
@code{DImode} result, but only if the registers are correctly sign
extended.  This predicate for the input operands accepts a
@code{sign_extend} of an @code{SImode} register.  Write the constraint
to indicate the type of register that is required for the operand of the
@code{sign_extend}.
@end ifset

@node Multi-Alternative
@subsection Multiple Alternative Constraints
@cindex multiple alternative constraints

Sometimes a single instruction has multiple alternative sets of possible
operands.  For example, on the 68000, a logical-or instruction can combine
register or an immediate value into memory, or it can combine any kind of
operand into a register; but it cannot combine one memory location into
another.

These constraints are represented as multiple alternatives.  An alternative
can be described by a series of letters for each operand.  The overall
constraint for an operand is made from the letters for this operand
from the first alternative, a comma, the letters for this operand from
the second alternative, a comma, and so on until the last alternative.
@ifset INTERNALS
Here is how it is done for fullword logical-or on the 68000:

@smallexample
(define_insn "iorsi3"
  [(set (match_operand:SI 0 "general_operand" "=m,d")
        (ior:SI (match_operand:SI 1 "general_operand" "%0,0")
                (match_operand:SI 2 "general_operand" "dKs,dmKs")))]
  @dots{})
@end smallexample

The first alternative has @samp{m} (memory) for operand 0, @samp{0} for
operand 1 (meaning it must match operand 0), and @samp{dKs} for operand
2.  The second alternative has @samp{d} (data register) for operand 0,
@samp{0} for operand 1, and @samp{dmKs} for operand 2.  The @samp{=} and
@samp{%} in the constraints apply to all the alternatives; their
meaning is explained in the next section (@pxref{Class Preferences}).
@end ifset

@c FIXME Is this ? and ! stuff of use in asm()?  If not, hide unless INTERNAL
If all the operands fit any one alternative, the instruction is valid.
Otherwise, for each alternative, the compiler counts how many instructions
must be added to copy the operands so that that alternative applies.
The alternative requiring the least copying is chosen.  If two alternatives
need the same amount of copying, the one that comes first is chosen.
These choices can be altered with the @samp{?} and @samp{!} characters:

@table @code
@cindex @samp{?} in constraint
@cindex question mark
@item ?
Disparage slightly the alternative that the @samp{?} appears in,
as a choice when no alternative applies exactly.  The compiler regards
this alternative as one unit more costly for each @samp{?} that appears
in it.

@cindex @samp{!} in constraint
@cindex exclamation point
@item !
Disparage severely the alternative that the @samp{!} appears in.
This alternative can still be used if it fits without reloading,
but if reloading is needed, some other alternative will be used.
@end table

@ifset INTERNALS
When an insn pattern has multiple alternatives in its constraints, often
the appearance of the assembler code is determined mostly by which
alternative was matched.  When this is so, the C code for writing the
assembler code can use the variable @code{which_alternative}, which is
the ordinal number of the alternative that was actually satisfied (0 for
the first, 1 for the second alternative, etc.).  @xref{Output Statement}.
@end ifset

@ifset INTERNALS
@node Class Preferences
@subsection Register Class Preferences
@cindex class preference constraints
@cindex register class preference constraints

@cindex voting between constraint alternatives
The operand constraints have another function: they enable the compiler
to decide which kind of hardware register a pseudo register is best
allocated to.  The compiler examines the constraints that apply to the
insns that use the pseudo register, looking for the machine-dependent
letters such as @samp{d} and @samp{a} that specify classes of registers.
The pseudo register is put in whichever class gets the most ``votes''.
The constraint letters @samp{g} and @samp{r} also vote: they vote in
favor of a general register.  The machine description says which registers
are considered general.

Of course, on some machines all registers are equivalent, and no register
classes are defined.  Then none of this complexity is relevant.
@end ifset

@node Modifiers
@subsection Constraint Modifier Characters
@cindex modifiers in constraints
@cindex constraint modifier characters

@c prevent bad page break with this line
Here are constraint modifier characters.

@table @samp
@cindex @samp{=} in constraint
@item =
Means that this operand is write-only for this instruction: the previous
value is discarded and replaced by output data.

@cindex @samp{+} in constraint
@item +
Means that this operand is both read and written by the instruction.

When the compiler fixes up the operands to satisfy the constraints,
it needs to know which operands are inputs to the instruction and
which are outputs from it.  @samp{=} identifies an output; @samp{+}
identifies an operand that is both input and output; all other operands
are assumed to be input only.

If you specify @samp{=} or @samp{+} in a constraint, you put it in the
first character of the constraint string.

@cindex @samp{&} in constraint
@cindex earlyclobber operand
@item &
Means (in a particular alternative) that this operand is an
@dfn{earlyclobber} operand, which is modified before the instruction is
finished using the input operands.  Therefore, this operand may not lie
in a register that is used as an input operand or as part of any memory
address.

@samp{&} applies only to the alternative in which it is written.  In
constraints with multiple alternatives, sometimes one alternative
requires @samp{&} while others do not.  See, for example, the
@samp{movdf} insn of the 68000.

An input operand can be tied to an earlyclobber operand if its only
use as an input occurs before the early result is written.  Adding
alternatives of this form often allows GCC to produce better code
when only some of the inputs can be affected by the earlyclobber.
See, for example, the @samp{mulsi3} insn of the ARM@.

@samp{&} does not obviate the need to write @samp{=}.

@cindex @samp{%} in constraint
@item %
Declares the instruction to be commutative for this operand and the
following operand.  This means that the compiler may interchange the
two operands if that is the cheapest way to make all operands fit the
constraints.
@ifset INTERNALS
This is often used in patterns for addition instructions
that really have only two operands: the result must go in one of the
arguments.  Here for example, is how the 68000 halfword-add
instruction is defined:

@smallexample
(define_insn "addhi3"
  [(set (match_operand:HI 0 "general_operand" "=m,r")
     (plus:HI (match_operand:HI 1 "general_operand" "%0,0")
              (match_operand:HI 2 "general_operand" "di,g")))]
  @dots{})
@end smallexample
@end ifset
GCC can only handle one commutative pair in an asm; if you use more,
the compiler may fail.  Note that you need not use the modifier if
the two alternatives are strictly identical; this would only waste
time in the reload pass.  The modifier is not operational after
register allocation, so the result of @code{define_peephole2}
and @code{define_split}s performed after reload cannot rely on
@samp{%} to make the intended insn match.

@cindex @samp{#} in constraint
@item #
Says that all following characters, up to the next comma, are to be
ignored as a constraint.  They are significant only for choosing
register preferences.

@cindex @samp{*} in constraint
@item *
Says that the following character should be ignored when choosing
register preferences.  @samp{*} has no effect on the meaning of the
constraint as a constraint, and no effect on reloading.

@ifset INTERNALS
Here is an example: the 68000 has an instruction to sign-extend a
halfword in a data register, and can also sign-extend a value by
copying it into an address register.  While either kind of register is
acceptable, the constraints on an address-register destination are
less strict, so it is best if register allocation makes an address
register its goal.  Therefore, @samp{*} is used so that the @samp{d}
constraint letter (for data register) is ignored when computing
register preferences.

@smallexample
(define_insn "extendhisi2"
  [(set (match_operand:SI 0 "general_operand" "=*d,a")
        (sign_extend:SI
         (match_operand:HI 1 "general_operand" "0,g")))]
  @dots{})
@end smallexample
@end ifset
@end table

@node Machine Constraints
@subsection Constraints for Particular Machines
@cindex machine specific constraints
@cindex constraints, machine specific

Whenever possible, you should use the general-purpose constraint letters
in @code{asm} arguments, since they will convey meaning more readily to
people reading your code.  Failing that, use the constraint letters
that usually have very similar meanings across architectures.  The most
commonly used constraints are @samp{m} and @samp{r} (for memory and
general-purpose registers respectively; @pxref{Simple Constraints}), and
@samp{I}, usually the letter indicating the most common
immediate-constant format.

Each architecture defines additional constraints.  These constraints
are used by the compiler itself for instruction generation, as well as
for @code{asm} statements; therefore, some of the constraints are not
particularly useful for @code{asm}.  Here is a summary of some of the
machine-dependent constraints available on some particular machines;
it includes both constraints that are useful for @code{asm} and
constraints that aren't.  The compiler source file mentioned in the
table heading for each architecture is the definitive reference for
the meanings of that architecture's constraints.
 
@table @emph
@item ARM family---@file{config/arm/arm.h}
@table @code
@item f
Floating-point register

@item w
VFP floating-point register

@item F
One of the floating-point constants 0.0, 0.5, 1.0, 2.0, 3.0, 4.0, 5.0
or 10.0

@item G
Floating-point constant that would satisfy the constraint @samp{F} if it
were negated

@item I
Integer that is valid as an immediate operand in a data processing
instruction.  That is, an integer in the range 0 to 255 rotated by a
multiple of 2

@item J
Integer in the range @minus{}4095 to 4095

@item K
Integer that satisfies constraint @samp{I} when inverted (ones complement)

@item L
Integer that satisfies constraint @samp{I} when negated (twos complement)

@item M
Integer in the range 0 to 32

@item Q
A memory reference where the exact address is in a single register
(`@samp{m}' is preferable for @code{asm} statements)

@item R
An item in the constant pool

@item S
A symbol in the text segment of the current file

@item Uv
A memory reference suitable for VFP load/store insns (reg+constant offset)

@item Uy
A memory reference suitable for iWMMXt load/store instructions.

@item Uq
A memory reference suitable for the ARMv4 ldrsb instruction.
@end table

@item AVR family---@file{config/avr/constraints.md}
@table @code
@item l
Registers from r0 to r15

@item a
Registers from r16 to r23

@item d
Registers from r16 to r31

@item w
Registers from r24 to r31.  These registers can be used in @samp{adiw} command

@item e
Pointer register (r26--r31)

@item b
Base pointer register (r28--r31)

@item q
Stack pointer register (SPH:SPL)

@item t
Temporary register r0

@item x
Register pair X (r27:r26)

@item y
Register pair Y (r29:r28)

@item z
Register pair Z (r31:r30)

@item I
Constant greater than @minus{}1, less than 64

@item J
Constant greater than @minus{}64, less than 1

@item K
Constant integer 2

@item L
Constant integer 0

@item M
Constant that fits in 8 bits

@item N
Constant integer @minus{}1

@item O
Constant integer 8, 16, or 24

@item P
Constant integer 1

@item G
A floating point constant 0.0
@end table

@item CRX Architecture---@file{config/crx/crx.h}
@table @code

@item b
Registers from r0 to r14 (registers without stack pointer)

@item l
Register r16 (64-bit accumulator lo register)

@item h
Register r17 (64-bit accumulator hi register)

@item k
Register pair r16-r17. (64-bit accumulator lo-hi pair)

@item I
Constant that fits in 3 bits

@item J
Constant that fits in 4 bits

@item K
Constant that fits in 5 bits

@item L
Constant that is one of -1, 4, -4, 7, 8, 12, 16, 20, 32, 48

@item G
Floating point constant that is legal for store immediate
@end table

@item PowerPC and IBM RS6000---@file{config/rs6000/rs6000.h}
@table @code
@item b
Address base register

@item f
Floating point register

@item v
Vector register

@item h
@samp{MQ}, @samp{CTR}, or @samp{LINK} register

@item q
@samp{MQ} register

@item c
@samp{CTR} register

@item l
@samp{LINK} register

@item x
@samp{CR} register (condition register) number 0

@item y
@samp{CR} register (condition register)

@item z
@samp{FPMEM} stack memory for FPR-GPR transfers

@item I
Signed 16-bit constant

@item J
Unsigned 16-bit constant shifted left 16 bits (use @samp{L} instead for
@code{SImode} constants)

@item K
Unsigned 16-bit constant

@item L
Signed 16-bit constant shifted left 16 bits

@item M
Constant larger than 31

@item N
Exact power of 2

@item O
Zero

@item P
Constant whose negation is a signed 16-bit constant

@item G
Floating point constant that can be loaded into a register with one
instruction per word

@item H
Integer/Floating point constant that can be loaded into a register using
three instructions

@item Q
Memory operand that is an offset from a register (@samp{m} is preferable
for @code{asm} statements)

@item Z
Memory operand that is an indexed or indirect from a register (@samp{m} is
preferable for @code{asm} statements)

@item R
AIX TOC entry

@item a
Address operand that is an indexed or indirect from a register (@samp{p} is
preferable for @code{asm} statements)

@item S
Constant suitable as a 64-bit mask operand

@item T
Constant suitable as a 32-bit mask operand

@item U
System V Release 4 small data area reference

@item t
AND masks that can be performed by two rldic@{l, r@} instructions

@item W
Vector constant that does not require memory

@end table

@item MorphoTech family---@file{config/mt/mt.h}
@table @code
@item I
Constant for an arithmetic insn (16-bit signed integer).

@item J
The constant 0.

@item K
Constant for a logical insn (16-bit zero-extended integer).

@item L
A constant that can be loaded with @code{lui} (i.e.@: the bottom 16
bits are zero).

@item M
A constant that takes two words to load (i.e.@: not matched by
@code{I}, @code{K}, or @code{L}).

@item N
Negative 16-bit constants other than -65536.

@item O
A 15-bit signed integer constant.

@item P
A positive 16-bit constant.
@end table

@item Intel 386---@file{config/i386/constraints.md}
@table @code
@item R
Legacy register---the eight integer registers available on all
i386 processors (@code{a}, @code{b}, @code{c}, @code{d},
@code{si}, @code{di}, @code{bp}, @code{sp}).

@item q
Any register accessible as @code{@var{r}l}.  In 32-bit mode, @code{a},
@code{b}, @code{c}, and @code{d}; in 64-bit mode, any integer register.

@item Q
Any register accessible as @code{@var{r}h}: @code{a}, @code{b},
@code{c}, and @code{d}.

@ifset INTERNALS
@item l
Any register that can be used as the index in a base+index memory
access: that is, any general register except the stack pointer.
@end ifset

@item a
The @code{a} register.

@item b
The @code{b} register.

@item c
The @code{c} register.

@item d
The @code{d} register.

@item S
The @code{si} register.

@item D
The @code{di} register.

@item A
The @code{a} and @code{d} registers, as a pair (for instructions that
return half the result in one and half in the other).

@item f
Any 80387 floating-point (stack) register.

@item t
Top of 80387 floating-point stack (@code{%st(0)}).

@item u
Second from top of 80387 floating-point stack (@code{%st(1)}).

@item y
Any MMX register.

@item x
Any SSE register.

@ifset INTERNALS
@item Y
Any SSE2 register.
@end ifset

@item I
Integer constant in the range 0 @dots{} 31, for 32-bit shifts.

@item J
Integer constant in the range 0 @dots{} 63, for 64-bit shifts.

@item K
Signed 8-bit integer constant.

@item L
@code{0xFF} or @code{0xFFFF}, for andsi as a zero-extending move.

@item M
0, 1, 2, or 3 (shifts for the @code{lea} instruction).

@item N
Unsigned 8-bit integer constant (for @code{in} and @code{out} 
instructions).

@ifset INTERNALS
@item O
Integer constant in the range 0 @dots{} 127, for 128-bit shifts.
@end ifset

@item G
Standard 80387 floating point constant.

@item C
Standard SSE floating point constant.

@item e
32-bit signed integer constant, or a symbolic reference known
to fit that range (for immediate operands in sign-extending x86-64
instructions).

@item Z
32-bit unsigned integer constant, or a symbolic reference known
to fit that range (for immediate operands in zero-extending x86-64
instructions).

@end table

@item Intel IA-64---@file{config/ia64/ia64.h}
@table @code
@item a
General register @code{r0} to @code{r3} for @code{addl} instruction

@item b
Branch register

@item c
Predicate register (@samp{c} as in ``conditional'')

@item d
Application register residing in M-unit

@item e
Application register residing in I-unit

@item f
Floating-point register

@item m
Memory operand.
Remember that @samp{m} allows postincrement and postdecrement which
require printing with @samp{%Pn} on IA-64.
Use @samp{S} to disallow postincrement and postdecrement.

@item G
Floating-point constant 0.0 or 1.0

@item I
14-bit signed integer constant

@item J
22-bit signed integer constant

@item K
8-bit signed integer constant for logical instructions

@item L
8-bit adjusted signed integer constant for compare pseudo-ops

@item M
6-bit unsigned integer constant for shift counts

@item N
9-bit signed integer constant for load and store postincrements

@item O
The constant zero

@item P
0 or @minus{}1 for @code{dep} instruction

@item Q
Non-volatile memory for floating-point loads and stores

@item R
Integer constant in the range 1 to 4 for @code{shladd} instruction

@item S
Memory operand except postincrement and postdecrement
@end table

@item FRV---@file{config/frv/frv.h}
@table @code
@item a
Register in the class @code{ACC_REGS} (@code{acc0} to @code{acc7}).

@item b
Register in the class @code{EVEN_ACC_REGS} (@code{acc0} to @code{acc7}).

@item c
Register in the class @code{CC_REGS} (@code{fcc0} to @code{fcc3} and
@code{icc0} to @code{icc3}).

@item d
Register in the class @code{GPR_REGS} (@code{gr0} to @code{gr63}).

@item e
Register in the class @code{EVEN_REGS} (@code{gr0} to @code{gr63}).
Odd registers are excluded not in the class but through the use of a machine
mode larger than 4 bytes.

@item f
Register in the class @code{FPR_REGS} (@code{fr0} to @code{fr63}).

@item h
Register in the class @code{FEVEN_REGS} (@code{fr0} to @code{fr63}).
Odd registers are excluded not in the class but through the use of a machine
mode larger than 4 bytes.

@item l
Register in the class @code{LR_REG} (the @code{lr} register).

@item q
Register in the class @code{QUAD_REGS} (@code{gr2} to @code{gr63}).
Register numbers not divisible by 4 are excluded not in the class but through
the use of a machine mode larger than 8 bytes.

@item t
Register in the class @code{ICC_REGS} (@code{icc0} to @code{icc3}).

@item u
Register in the class @code{FCC_REGS} (@code{fcc0} to @code{fcc3}).

@item v
Register in the class @code{ICR_REGS} (@code{cc4} to @code{cc7}).

@item w
Register in the class @code{FCR_REGS} (@code{cc0} to @code{cc3}).

@item x
Register in the class @code{QUAD_FPR_REGS} (@code{fr0} to @code{fr63}).
Register numbers not divisible by 4 are excluded not in the class but through
the use of a machine mode larger than 8 bytes.

@item z
Register in the class @code{SPR_REGS} (@code{lcr} and @code{lr}).

@item A
Register in the class @code{QUAD_ACC_REGS} (@code{acc0} to @code{acc7}).

@item B
Register in the class @code{ACCG_REGS} (@code{accg0} to @code{accg7}).

@item C
Register in the class @code{CR_REGS} (@code{cc0} to @code{cc7}).

@item G
Floating point constant zero

@item I
6-bit signed integer constant

@item J
10-bit signed integer constant

@item L
16-bit signed integer constant

@item M
16-bit unsigned integer constant

@item N
12-bit signed integer constant that is negative---i.e.@: in the
range of @minus{}2048 to @minus{}1

@item O
Constant zero

@item P
12-bit signed integer constant that is greater than zero---i.e.@: in the
range of 1 to 2047.

@end table

@item Blackfin family---@file{config/bfin/bfin.h}
@table @code
@item a
P register

@item d
D register

@item z
A call clobbered P register.

@item D
Even-numbered D register

@item W
Odd-numbered D register

@item e
Accumulator register.

@item A
Even-numbered accumulator register.

@item B
Odd-numbered accumulator register.

@item b
I register

@item v
B register

@item f
M register

@item c
Registers used for circular buffering, i.e. I, B, or L registers.

@item C
The CC register.

@item t
LT0 or LT1.

@item k
LC0 or LC1.

@item u
LB0 or LB1.

@item x
Any D, P, B, M, I or L register.

@item y
Additional registers typically used only in prologues and epilogues: RETS,
RETN, RETI, RETX, RETE, ASTAT, SEQSTAT and USP.

@item w
Any register except accumulators or CC.

@item Ksh
Signed 16 bit integer (in the range -32768 to 32767)

@item Kuh
Unsigned 16 bit integer (in the range 0 to 65535)

@item Ks7
Signed 7 bit integer (in the range -64 to 63)

@item Ku7
Unsigned 7 bit integer (in the range 0 to 127)

@item Ku5
Unsigned 5 bit integer (in the range 0 to 31)

@item Ks4
Signed 4 bit integer (in the range -8 to 7)

@item Ks3
Signed 3 bit integer (in the range -3 to 4)

@item Ku3
Unsigned 3 bit integer (in the range 0 to 7)

@item P@var{n}
Constant @var{n}, where @var{n} is a single-digit constant in the range 0 to 4.

@item PA
An integer equal to one of the MACFLAG_XXX constants that is suitable for
use with either accumulator.

@item PB
An integer equal to one of the MACFLAG_XXX constants that is suitable for
use only with accumulator A1.

@item M1
Constant 255.

@item M2
Constant 65535.

@item J
An integer constant with exactly a single bit set.

@item L
An integer constant with all bits set except exactly one.

@item H

@item Q
Any SYMBOL_REF.
@end table

@item M32C---@file{config/m32c/m32c.c}
@table @code
@item Rsp
@itemx Rfb
@itemx Rsb
@samp{$sp}, @samp{$fb}, @samp{$sb}.

@item Rcr
Any control register, when they're 16 bits wide (nothing if control
registers are 24 bits wide)

@item Rcl
Any control register, when they're 24 bits wide.

@item R0w
@itemx R1w
@itemx R2w
@itemx R3w
$r0, $r1, $r2, $r3.

@item R02
$r0 or $r2, or $r2r0 for 32 bit values.

@item R13
$r1 or $r3, or $r3r1 for 32 bit values.

@item Rdi
A register that can hold a 64 bit value.

@item Rhl
$r0 or $r1 (registers with addressable high/low bytes)

@item R23
$r2 or $r3

@item Raa
Address registers

@item Raw
Address registers when they're 16 bits wide.

@item Ral
Address registers when they're 24 bits wide.

@item Rqi
Registers that can hold QI values.

@item Rad
Registers that can be used with displacements ($a0, $a1, $sb).

@item Rsi
Registers that can hold 32 bit values.

@item Rhi
Registers that can hold 16 bit values.

@item Rhc
Registers chat can hold 16 bit values, including all control
registers.

@item Rra
$r0 through R1, plus $a0 and $a1.

@item Rfl
The flags register.

@item Rmm
The memory-based pseudo-registers $mem0 through $mem15.

@item Rpi
Registers that can hold pointers (16 bit registers for r8c, m16c; 24
bit registers for m32cm, m32c).

@item Rpa
Matches multiple registers in a PARALLEL to form a larger register.
Used to match function return values.

@item Is3
-8 @dots{} 7

@item IS1
-128 @dots{} 127

@item IS2
-32768 @dots{} 32767

@item IU2
0 @dots{} 65535

@item In4
-8 @dots{} -1 or 1 @dots{} 8

@item In5
-16 @dots{} -1 or 1 @dots{} 16

@item In6
-32 @dots{} -1 or 1 @dots{} 32

@item IM2
-65536 @dots{} -1

@item Ilb
An 8 bit value with exactly one bit set.

@item Ilw
A 16 bit value with exactly one bit set.

@item Sd
The common src/dest memory addressing modes.

@item Sa
Memory addressed using $a0 or $a1.

@item Si
Memory addressed with immediate addresses.

@item Ss
Memory addressed using the stack pointer ($sp).

@item Sf
Memory addressed using the frame base register ($fb).

@item Ss
Memory addressed using the small base register ($sb).

@item S1
$r1h
@end table

@item MIPS---@file{config/mips/constraints.md}
@table @code
@item d
An address register.  This is equivalent to @code{r} unless
generating MIPS16 code.

@item f
A floating-point register (if available).

@item h
The @code{hi} register.

@item l
The @code{lo} register.

@item x
The @code{hi} and @code{lo} registers.

@item c
A register suitable for use in an indirect jump.  This will always be
@code{$25} for @option{-mabicalls}.

@item y
Equivalent to @code{r}; retained for backwards compatibility.

@item z
A floating-point condition code register.

@item I
A signed 16-bit constant (for arithmetic instructions).

@item J
Integer zero.

@item K
An unsigned 16-bit constant (for logic instructions).

@item L
A signed 32-bit constant in which the lower 16 bits are zero.
Such constants can be loaded using @code{lui}.

@item M
A constant that cannot be loaded using @code{lui}, @code{addiu}
or @code{ori}.

@item N
A constant in the range -65535 to -1 (inclusive).

@item O
A signed 15-bit constant.

@item P
A constant in the range 1 to 65535 (inclusive).

@item G
Floating-point zero.

@item R
An address that can be used in a non-macro load or store.
@end table

@item Motorola 680x0---@file{config/m68k/m68k.h}
@table @code
@item a
Address register

@item d
Data register

@item f
68881 floating-point register, if available

@item I
Integer in the range 1 to 8

@item J
16-bit signed number

@item K
Signed number whose magnitude is greater than 0x80

@item L
Integer in the range @minus{}8 to @minus{}1

@item M
Signed number whose magnitude is greater than 0x100

@item G
Floating point constant that is not a 68881 constant
@end table

@item Motorola 68HC11 & 68HC12 families---@file{config/m68hc11/m68hc11.h}
@table @code
@item a
Register `a'

@item b
Register `b'

@item d
Register `d'

@item q
An 8-bit register

@item t
Temporary soft register _.tmp

@item u
A soft register _.d1 to _.d31

@item w
Stack pointer register

@item x
Register `x'

@item y
Register `y'

@item z
Pseudo register `z' (replaced by `x' or `y' at the end)

@item A
An address register: x, y or z

@item B
An address register: x or y

@item D
Register pair (x:d) to form a 32-bit value

@item L
Constants in the range @minus{}65536 to 65535

@item M
Constants whose 16-bit low part is zero

@item N
Constant integer 1 or @minus{}1

@item O
Constant integer 16

@item P
Constants in the range @minus{}8 to 2

@end table

@need 1000
@item SPARC---@file{config/sparc/sparc.h}
@table @code
@item f
Floating-point register on the SPARC-V8 architecture and
lower floating-point register on the SPARC-V9 architecture.

@item e
Floating-point register.  It is equivalent to @samp{f} on the
SPARC-V8 architecture and contains both lower and upper
floating-point registers on the SPARC-V9 architecture.

@item c
Floating-point condition code register.

@item d
Lower floating-point register.  It is only valid on the SPARC-V9
architecture when the Visual Instruction Set is available.

@item b
Floating-point register.  It is only valid on the SPARC-V9 architecture
when the Visual Instruction Set is available.

@item h
64-bit global or out register for the SPARC-V8+ architecture.

@item I
Signed 13-bit constant

@item J
Zero

@item K
32-bit constant with the low 12 bits clear (a constant that can be
loaded with the @code{sethi} instruction)

@item L
A constant in the range supported by @code{movcc} instructions

@item M
A constant in the range supported by @code{movrcc} instructions

@item N
Same as @samp{K}, except that it verifies that bits that are not in the
lower 32-bit range are all zero.  Must be used instead of @samp{K} for
modes wider than @code{SImode}

@item O
The constant 4096

@item G
Floating-point zero

@item H
Signed 13-bit constant, sign-extended to 32 or 64 bits

@item Q
Floating-point constant whose integral representation can
be moved into an integer register using a single sethi
instruction

@item R
Floating-point constant whose integral representation can
be moved into an integer register using a single mov
instruction

@item S
Floating-point constant whose integral representation can
be moved into an integer register using a high/lo_sum
instruction sequence

@item T
Memory address aligned to an 8-byte boundary

@item U
Even register

@item W
Memory address for @samp{e} constraint registers

@item Y
Vector zero

@end table

@item SPU---@file{config/spu/spu.h}
@table @code
@item a
An immediate which can be loaded with the il/ila/ilh/ilhu instructions.  const_int is treated as a 64 bit value.  

@item c
An immediate for and/xor/or instructions.  const_int is treated as a 64 bit value.  

@item d
An immediate for the @code{iohl} instruction.  const_int is treated as a 64 bit value.  

@item f
An immediate which can be loaded with @code{fsmbi}.  

@item A
An immediate which can be loaded with the il/ila/ilh/ilhu instructions.  const_int is treated as a 32 bit value.  

@item B
An immediate for most arithmetic instructions.  const_int is treated as a 32 bit value.  

@item C
An immediate for and/xor/or instructions.  const_int is treated as a 32 bit value.  

@item D
An immediate for the @code{iohl} instruction.  const_int is treated as a 32 bit value.  

@item I
A constant in the range [-64, 63] for shift/rotate instructions.  

@item J
An unsigned 7-bit constant for conversion/nop/channel instructions.  

@item K
A signed 10-bit constant for most arithmetic instructions.  

@item M
A signed 16 bit immediate for @code{stop}.  

@item N
An unsigned 16-bit constant for @code{iohl} and @code{fsmbi}.  

@item O
An unsigned 7-bit constant whose 3 least significant bits are 0.  

@item P
An unsigned 3-bit constant for 16-byte rotates and shifts 

@item R
Call operand, reg, for indirect calls 

@item S
Call operand, symbol, for relative calls.  

@item T
Call operand, const_int, for absolute calls.  

@item U
An immediate which can be loaded with the il/ila/ilh/ilhu instructions.  const_int is sign extended to 128 bit.  

@item W
An immediate for shift and rotate instructions.  const_int is treated as a 32 bit value.  

@item Y
An immediate for and/xor/or instructions.  const_int is sign extended as a 128 bit.  

@item Z
An immediate for the @code{iohl} instruction.  const_int is sign extended to 128 bit.  

@end table

@item TMS320C3x/C4x---@file{config/c4x/c4x.h}
@table @code
@item a
Auxiliary (address) register (ar0-ar7)

@item b
Stack pointer register (sp)

@item c
Standard (32-bit) precision integer register

@item f
Extended (40-bit) precision register (r0-r11)

@item k
Block count register (bk)

@item q
Extended (40-bit) precision low register (r0-r7)

@item t
Extended (40-bit) precision register (r0-r1)

@item u
Extended (40-bit) precision register (r2-r3)

@item v
Repeat count register (rc)

@item x
Index register (ir0-ir1)

@item y
Status (condition code) register (st)

@item z
Data page register (dp)

@item G
Floating-point zero

@item H
Immediate 16-bit floating-point constant

@item I
Signed 16-bit constant

@item J
Signed 8-bit constant

@item K
Signed 5-bit constant

@item L
Unsigned 16-bit constant

@item M
Unsigned 8-bit constant

@item N
Ones complement of unsigned 16-bit constant

@item O
High 16-bit constant (32-bit constant with 16 LSBs zero)

@item Q
Indirect memory reference with signed 8-bit or index register displacement

@item R
Indirect memory reference with unsigned 5-bit displacement

@item S
Indirect memory reference with 1 bit or index register displacement

@item T
Direct memory reference

@item U
Symbolic address

@end table

@item S/390 and zSeries---@file{config/s390/s390.h}
@table @code
@item a
Address register (general purpose register except r0)

@item c
Condition code register

@item d
Data register (arbitrary general purpose register)

@item f
Floating-point register

@item I
Unsigned 8-bit constant (0--255)

@item J
Unsigned 12-bit constant (0--4095)

@item K
Signed 16-bit constant (@minus{}32768--32767)

@item L
Value appropriate as displacement.
@table @code
       @item (0..4095)
       for short displacement
       @item (-524288..524287)
       for long displacement
@end table

@item M
Constant integer with a value of 0x7fffffff.

@item N
Multiple letter constraint followed by 4 parameter letters.
@table @code
         @item 0..9:
         number of the part counting from most to least significant
         @item H,Q:
         mode of the part
         @item D,S,H:
         mode of the containing operand
         @item 0,F:
         value of the other parts (F---all bits set)
@end table
The constraint matches if the specified part of a constant
has a value different from it's other parts.

@item Q
Memory reference without index register and with short displacement.

@item R
Memory reference with index register and short displacement.

@item S
Memory reference without index register but with long displacement.

@item T
Memory reference with index register and long displacement.

@item U
Pointer with short displacement.

@item W
Pointer with long displacement.

@item Y
Shift count operand.

@end table

@item Score family---@file{config/score/score.h}
@table @code
@item d
Registers from r0 to r32.

@item e
Registers from r0 to r16.

@item t
r8---r11 or r22---r27 registers.

@item h
hi register.

@item l
lo register.

@item x
hi + lo register.

@item q
cnt register.

@item y
lcb register.

@item z
scb register.

@item a
cnt + lcb + scb register.

@item c
cr0---cr15 register.

@item b
cp1 registers.

@item f
cp2 registers.

@item i
cp3 registers.

@item j
cp1 + cp2 + cp3 registers.

@item I
High 16-bit constant (32-bit constant with 16 LSBs zero).

@item J
Unsigned 5 bit integer (in the range 0 to 31).

@item K
Unsigned 16 bit integer (in the range 0 to 65535).

@item L
Signed 16 bit integer (in the range @minus{}32768 to 32767).

@item M
Unsigned 14 bit integer (in the range 0 to 16383).

@item N
Signed 14 bit integer (in the range @minus{}8192 to 8191).

@item Z
Any SYMBOL_REF.
@end table

@item Xstormy16---@file{config/stormy16/stormy16.h}
@table @code
@item a
Register r0.

@item b
Register r1.

@item c
Register r2.

@item d
Register r8.

@item e
Registers r0 through r7.

@item t
Registers r0 and r1.

@item y
The carry register.

@item z
Registers r8 and r9.

@item I
A constant between 0 and 3 inclusive.

@item J
A constant that has exactly one bit set.

@item K
A constant that has exactly one bit clear.

@item L
A constant between 0 and 255 inclusive.

@item M
A constant between @minus{}255 and 0 inclusive.

@item N
A constant between @minus{}3 and 0 inclusive.

@item O
A constant between 1 and 4 inclusive.

@item P
A constant between @minus{}4 and @minus{}1 inclusive.

@item Q
A memory reference that is a stack push.

@item R
A memory reference that is a stack pop.

@item S
A memory reference that refers to a constant address of known value.

@item T
The register indicated by Rx (not implemented yet).

@item U
A constant that is not between 2 and 15 inclusive.

@item Z
The constant 0.

@end table

@item Xtensa---@file{config/xtensa/constraints.md}
@table @code
@item a
General-purpose 32-bit register

@item b
One-bit boolean register

@item A
MAC16 40-bit accumulator register

@item I
Signed 12-bit integer constant, for use in MOVI instructions

@item J
Signed 8-bit integer constant, for use in ADDI instructions

@item K
Integer constant valid for BccI instructions

@item L
Unsigned constant valid for BccUI instructions

@end table

@end table

@ifset INTERNALS
@node Define Constraints
@subsection Defining Machine-Specific Constraints
@cindex defining constraints
@cindex constraints, defining

Machine-specific constraints fall into two categories: register and
non-register constraints.  Within the latter category, constraints
which allow subsets of all possible memory or address operands should
be specially marked, to give @code{reload} more information.

Machine-specific constraints can be given names of arbitrary length,
but they must be entirely composed of letters, digits, underscores
(@samp{_}), and angle brackets (@samp{< >}).  Like C identifiers, they
must begin with a letter or underscore. 

In order to avoid ambiguity in operand constraint strings, no
constraint can have a name that begins with any other constraint's
name.  For example, if @code{x} is defined as a constraint name,
@code{xy} may not be, and vice versa.  As a consequence of this rule,
no constraint may begin with one of the generic constraint letters:
@samp{E F V X g i m n o p r s}.

Register constraints correspond directly to register classes.
@xref{Register Classes}.  There is thus not much flexibility in their
definitions.

@deffn {MD Expression} define_register_constraint name regclass docstring
All three arguments are string constants.
@var{name} is the name of the constraint, as it will appear in
@code{match_operand} expressions.  If @var{name} is a multi-letter
constraint its length shall be the same for all constraints starting
with the same letter.  @var{regclass} can be either the
name of the corresponding register class (@pxref{Register Classes}),
or a C expression which evaluates to the appropriate register class.
If it is an expression, it must have no side effects, and it cannot
look at the operand.  The usual use of expressions is to map some
register constraints to @code{NO_REGS} when the register class
is not available on a given subarchitecture.

@var{docstring} is a sentence documenting the meaning of the
constraint.  Docstrings are explained further below.
@end deffn

Non-register constraints are more like predicates: the constraint
definition gives a Boolean expression which indicates whether the
constraint matches.

@deffn {MD Expression} define_constraint name docstring exp
The @var{name} and @var{docstring} arguments are the same as for
@code{define_register_constraint}, but note that the docstring comes
immediately after the name for these expressions.  @var{exp} is an RTL
expression, obeying the same rules as the RTL expressions in predicate
definitions.  @xref{Defining Predicates}, for details.  If it
evaluates true, the constraint matches; if it evaluates false, it
doesn't. Constraint expressions should indicate which RTL codes they
might match, just like predicate expressions.

@code{match_test} C expressions have access to the
following variables:

@table @var
@item op
The RTL object defining the operand.
@item mode
The machine mode of @var{op}.
@item ival
@samp{INTVAL (@var{op})}, if @var{op} is a @code{const_int}.
@item hval
@samp{CONST_DOUBLE_HIGH (@var{op})}, if @var{op} is an integer
@code{const_double}.
@item lval
@samp{CONST_DOUBLE_LOW (@var{op})}, if @var{op} is an integer
@code{const_double}.
@item rval
@samp{CONST_DOUBLE_REAL_VALUE (@var{op})}, if @var{op} is a floating-point
@code{const_double}.
@end table

The @var{*val} variables should only be used once another piece of the
expression has verified that @var{op} is the appropriate kind of RTL
object.
@end deffn

Most non-register constraints should be defined with
@code{define_constraint}.  The remaining two definition expressions
are only appropriate for constraints that should be handled specially
by @code{reload} if they fail to match.

@deffn {MD Expression} define_memory_constraint name docstring exp
Use this expression for constraints that match a subset of all memory
operands: that is, @code{reload} can make them match by converting the
operand to the form @samp{@w{(mem (reg @var{X}))}}, where @var{X} is a
base register (from the register class specified by
@code{BASE_REG_CLASS}, @pxref{Register Classes}).

For example, on the S/390, some instructions do not accept arbitrary
memory references, but only those that do not make use of an index
register.  The constraint letter @samp{Q} is defined to represent a
memory address of this type.  If @samp{Q} is defined with
@code{define_memory_constraint}, a @samp{Q} constraint can handle any
memory operand, because @code{reload} knows it can simply copy the
memory address into a base register if required.  This is analogous to
the way a @samp{o} constraint can handle any memory operand.

The syntax and semantics are otherwise identical to
@code{define_constraint}.
@end deffn

@deffn {MD Expression} define_address_constraint name docstring exp
Use this expression for constraints that match a subset of all address
operands: that is, @code{reload} can make the constraint match by
converting the operand to the form @samp{@w{(reg @var{X})}}, again
with @var{X} a base register.

Constraints defined with @code{define_address_constraint} can only be
used with the @code{address_operand} predicate, or machine-specific
predicates that work the same way.  They are treated analogously to
the generic @samp{p} constraint.

The syntax and semantics are otherwise identical to
@code{define_constraint}.
@end deffn

For historical reasons, names beginning with the letters @samp{G H}
are reserved for constraints that match only @code{const_double}s, and
names beginning with the letters @samp{I J K L M N O P} are reserved
for constraints that match only @code{const_int}s.  This may change in
the future.  For the time being, constraints with these names must be
written in a stylized form, so that @code{genpreds} can tell you did
it correctly:

@smallexample
@group
(define_constraint "[@var{GHIJKLMNOP}]@dots{}"
  "@var{doc}@dots{}"
  (and (match_code "const_int")  ; @r{@code{const_double} for G/H}
       @var{condition}@dots{}))            ; @r{usually a @code{match_test}}
@end group
@end smallexample
@c the semicolons line up in the formatted manual

It is fine to use names beginning with other letters for constraints
that match @code{const_double}s or @code{const_int}s.

Each docstring in a constraint definition should be one or more complete
sentences, marked up in Texinfo format.  @emph{They are currently unused.}
In the future they will be copied into the GCC manual, in @ref{Machine
Constraints}, replacing the hand-maintained tables currently found in
that section.  Also, in the future the compiler may use this to give
more helpful diagnostics when poor choice of @code{asm} constraints
causes a reload failure.

If you put the pseudo-Texinfo directive @samp{@@internal} at the
beginning of a docstring, then (in the future) it will appear only in
the internals manual's version of the machine-specific constraint tables.
Use this for constraints that should not appear in @code{asm} statements.

@node C Constraint Interface
@subsection Testing constraints from C
@cindex testing constraints
@cindex constraints, testing

It is occasionally useful to test a constraint from C code rather than
implicitly via the constraint string in a @code{match_operand}.  The
generated file @file{tm_p.h} declares a few interfaces for working
with machine-specific constraints.  None of these interfaces work with
the generic constraints described in @ref{Simple Constraints}.  This
may change in the future.

@strong{Warning:} @file{tm_p.h} may declare other functions that
operate on constraints, besides the ones documented here.  Do not use
those functions from machine-dependent code.  They exist to implement
the old constraint interface that machine-independent components of
the compiler still expect.  They will change or disappear in the
future.

Some valid constraint names are not valid C identifiers, so there is a
mangling scheme for referring to them from C@.  Constraint names that
do not contain angle brackets or underscores are left unchanged.
Underscores are doubled, each @samp{<} is replaced with @samp{_l}, and
each @samp{>} with @samp{_g}.  Here are some examples:

@c the @c's prevent double blank lines in the printed manual.
@example
@multitable {Original} {Mangled}
@item @strong{Original} @tab @strong{Mangled}  @c
@item @code{x}     @tab @code{x}       @c
@item @code{P42x}  @tab @code{P42x}    @c
@item @code{P4_x}  @tab @code{P4__x}   @c
@item @code{P4>x}  @tab @code{P4_gx}   @c
@item @code{P4>>}  @tab @code{P4_g_g}  @c
@item @code{P4_g>} @tab @code{P4__g_g} @c
@end multitable
@end example

Throughout this section, the variable @var{c} is either a constraint
in the abstract sense, or a constant from @code{enum constraint_num};
the variable @var{m} is a mangled constraint name (usually as part of
a larger identifier).

@deftp Enum constraint_num
For each machine-specific constraint, there is a corresponding
enumeration constant: @samp{CONSTRAINT_} plus the mangled name of the
constraint.  Functions that take an @code{enum constraint_num} as an
argument expect one of these constants.

Machine-independent constraints do not have associated constants.
This may change in the future.
@end deftp

@deftypefun {inline bool} satisfies_constraint_@var{m} (rtx @var{exp})
For each machine-specific, non-register constraint @var{m}, there is
one of these functions; it returns @code{true} if @var{exp} satisfies the
constraint.  These functions are only visible if @file{rtl.h} was included
before @file{tm_p.h}.
@end deftypefun

@deftypefun bool constraint_satisfied_p (rtx @var{exp}, enum constraint_num @var{c})
Like the @code{satisfies_constraint_@var{m}} functions, but the
constraint to test is given as an argument, @var{c}.  If @var{c}
specifies a register constraint, this function will always return
@code{false}.
@end deftypefun

@deftypefun {enum reg_class} regclass_for_constraint (enum constraint_num @var{c})
Returns the register class associated with @var{c}.  If @var{c} is not
a register constraint, or those registers are not available for the
currently selected subtarget, returns @code{NO_REGS}.
@end deftypefun

Here is an example use of @code{satisfies_constraint_@var{m}}.  In
peephole optimizations (@pxref{Peephole Definitions}), operand
constraint strings are ignored, so if there are relevant constraints,
they must be tested in the C condition.  In the example, the
optimization is applied if operand 2 does @emph{not} satisfy the
@samp{K} constraint.  (This is a simplified version of a peephole
definition from the i386 machine description.)

@smallexample
(define_peephole2
  [(match_scratch:SI 3 "r")
   (set (match_operand:SI 0 "register_operand" "")
        (mult:SI (match_operand:SI 1 "memory_operand" "")
                 (match_operand:SI 2 "immediate_operand" "")))]

  "!satisfies_constraint_K (operands[2])"

  [(set (match_dup 3) (match_dup 1))
   (set (match_dup 0) (mult:SI (match_dup 3) (match_dup 2)))]

  "")
@end smallexample

@node Standard Names
@section Standard Pattern Names For Generation
@cindex standard pattern names
@cindex pattern names
@cindex names, pattern

Here is a table of the instruction names that are meaningful in the RTL
generation pass of the compiler.  Giving one of these names to an
instruction pattern tells the RTL generation pass that it can use the
pattern to accomplish a certain task.

@table @asis
@cindex @code{mov@var{m}} instruction pattern
@item @samp{mov@var{m}}
Here @var{m} stands for a two-letter machine mode name, in lowercase.
This instruction pattern moves data with that machine mode from operand
1 to operand 0.  For example, @samp{movsi} moves full-word data.

If operand 0 is a @code{subreg} with mode @var{m} of a register whose
own mode is wider than @var{m}, the effect of this instruction is
to store the specified value in the part of the register that corresponds
to mode @var{m}.  Bits outside of @var{m}, but which are within the
same target word as the @code{subreg} are undefined.  Bits which are
outside the target word are left unchanged.

This class of patterns is special in several ways.  First of all, each
of these names up to and including full word size @emph{must} be defined,
because there is no other way to copy a datum from one place to another.
If there are patterns accepting operands in larger modes,
@samp{mov@var{m}} must be defined for integer modes of those sizes.

Second, these patterns are not used solely in the RTL generation pass.
Even the reload pass can generate move insns to copy values from stack
slots into temporary registers.  When it does so, one of the operands is
a hard register and the other is an operand that can need to be reloaded
into a register.

@findex force_reg
Therefore, when given such a pair of operands, the pattern must generate
RTL which needs no reloading and needs no temporary registers---no
registers other than the operands.  For example, if you support the
pattern with a @code{define_expand}, then in such a case the
@code{define_expand} mustn't call @code{force_reg} or any other such
function which might generate new pseudo registers.

This requirement exists even for subword modes on a RISC machine where
fetching those modes from memory normally requires several insns and
some temporary registers.

@findex change_address
During reload a memory reference with an invalid address may be passed
as an operand.  Such an address will be replaced with a valid address
later in the reload pass.  In this case, nothing may be done with the
address except to use it as it stands.  If it is copied, it will not be
replaced with a valid address.  No attempt should be made to make such
an address into a valid address and no routine (such as
@code{change_address}) that will do so may be called.  Note that
@code{general_operand} will fail when applied to such an address.

@findex reload_in_progress
The global variable @code{reload_in_progress} (which must be explicitly
declared if required) can be used to determine whether such special
handling is required.

The variety of operands that have reloads depends on the rest of the
machine description, but typically on a RISC machine these can only be
pseudo registers that did not get hard registers, while on other
machines explicit memory references will get optional reloads.

If a scratch register is required to move an object to or from memory,
it can be allocated using @code{gen_reg_rtx} prior to life analysis.

If there are cases which need scratch registers during or after reload,
you must provide an appropriate secondary_reload target hook.

@findex no_new_pseudos
The global variable @code{no_new_pseudos} can be used to determine if it
is unsafe to create new pseudo registers.  If this variable is nonzero, then
it is unsafe to call @code{gen_reg_rtx} to allocate a new pseudo.

The constraints on a @samp{mov@var{m}} must permit moving any hard
register to any other hard register provided that
@code{HARD_REGNO_MODE_OK} permits mode @var{m} in both registers and
@code{REGISTER_MOVE_COST} applied to their classes returns a value of 2.

It is obligatory to support floating point @samp{mov@var{m}}
instructions into and out of any registers that can hold fixed point
values, because unions and structures (which have modes @code{SImode} or
@code{DImode}) can be in those registers and they may have floating
point members.

There may also be a need to support fixed point @samp{mov@var{m}}
instructions in and out of floating point registers.  Unfortunately, I
have forgotten why this was so, and I don't know whether it is still
true.  If @code{HARD_REGNO_MODE_OK} rejects fixed point values in
floating point registers, then the constraints of the fixed point
@samp{mov@var{m}} instructions must be designed to avoid ever trying to
reload into a floating point register.

@cindex @code{reload_in} instruction pattern
@cindex @code{reload_out} instruction pattern
@item @samp{reload_in@var{m}}
@itemx @samp{reload_out@var{m}}
These named patterns have been obsoleted by the target hook
@code{secondary_reload}.

Like @samp{mov@var{m}}, but used when a scratch register is required to
move between operand 0 and operand 1.  Operand 2 describes the scratch
register.  See the discussion of the @code{SECONDARY_RELOAD_CLASS}
macro in @pxref{Register Classes}.

There are special restrictions on the form of the @code{match_operand}s
used in these patterns.  First, only the predicate for the reload
operand is examined, i.e., @code{reload_in} examines operand 1, but not
the predicates for operand 0 or 2.  Second, there may be only one
alternative in the constraints.  Third, only a single register class
letter may be used for the constraint; subsequent constraint letters
are ignored.  As a special exception, an empty constraint string
matches the @code{ALL_REGS} register class.  This may relieve ports
of the burden of defining an @code{ALL_REGS} constraint letter just
for these patterns.

@cindex @code{movstrict@var{m}} instruction pattern
@item @samp{movstrict@var{m}}
Like @samp{mov@var{m}} except that if operand 0 is a @code{subreg}
with mode @var{m} of a register whose natural mode is wider,
the @samp{movstrict@var{m}} instruction is guaranteed not to alter
any of the register except the part which belongs to mode @var{m}.

@cindex @code{movmisalign@var{m}} instruction pattern
@item @samp{movmisalign@var{m}}
This variant of a move pattern is designed to load or store a value
from a memory address that is not naturally aligned for its mode.
For a store, the memory will be in operand 0; for a load, the memory
will be in operand 1.  The other operand is guaranteed not to be a
memory, so that it's easy to tell whether this is a load or store.

This pattern is used by the autovectorizer, and when expanding a
@code{MISALIGNED_INDIRECT_REF} expression.

@cindex @code{load_multiple} instruction pattern
@item @samp{load_multiple}
Load several consecutive memory locations into consecutive registers.
Operand 0 is the first of the consecutive registers, operand 1
is the first memory location, and operand 2 is a constant: the
number of consecutive registers.

Define this only if the target machine really has such an instruction;
do not define this if the most efficient way of loading consecutive
registers from memory is to do them one at a time.

On some machines, there are restrictions as to which consecutive
registers can be stored into memory, such as particular starting or
ending register numbers or only a range of valid counts.  For those
machines, use a @code{define_expand} (@pxref{Expander Definitions})
and make the pattern fail if the restrictions are not met.

Write the generated insn as a @code{parallel} with elements being a
@code{set} of one register from the appropriate memory location (you may
also need @code{use} or @code{clobber} elements).  Use a
@code{match_parallel} (@pxref{RTL Template}) to recognize the insn.  See
@file{rs6000.md} for examples of the use of this insn pattern.

@cindex @samp{store_multiple} instruction pattern
@item @samp{store_multiple}
Similar to @samp{load_multiple}, but store several consecutive registers
into consecutive memory locations.  Operand 0 is the first of the
consecutive memory locations, operand 1 is the first register, and
operand 2 is a constant: the number of consecutive registers.

@cindex @code{vec_set@var{m}} instruction pattern
@item @samp{vec_set@var{m}}
Set given field in the vector value.  Operand 0 is the vector to modify,
operand 1 is new value of field and operand 2 specify the field index.

@cindex @code{vec_extract@var{m}} instruction pattern
@item @samp{vec_extract@var{m}}
Extract given field from the vector value.  Operand 1 is the vector, operand 2
specify field index and operand 0 place to store value into.

@cindex @code{vec_extract_even@var{m}} instruction pattern
@item @samp{vec_extract_even@var{m}}
Extract even elements from the input vectors (operand 1 and operand 2). 
The even elements of operand 2 are concatenated to the even elements of operand
1 in their original order. The result is stored in operand 0. 
The output and input vectors should have the same modes. 

@cindex @code{vec_extract_odd@var{m}} instruction pattern
@item @samp{vec_extract_odd@var{m}}
Extract odd elements from the input vectors (operand 1 and operand 2). 
The odd elements of operand 2 are concatenated to the odd elements of operand 
1 in their original order. The result is stored in operand 0.
The output and input vectors should have the same modes.

@cindex @code{vec_interleave_high@var{m}} instruction pattern
@item @samp{vec_interleave_high@var{m}}
Merge high elements of the two input vectors into the output vector. The output
and input vectors should have the same modes (@code{N} elements). The high
@code{N/2} elements of the first input vector are interleaved with the high
@code{N/2} elements of the second input vector.

@cindex @code{vec_interleave_low@var{m}} instruction pattern
@item @samp{vec_interleave_low@var{m}}
Merge low elements of the two input vectors into the output vector. The output
and input vectors should have the same modes (@code{N} elements). The low
@code{N/2} elements of the first input vector are interleaved with the low 
@code{N/2} elements of the second input vector.

@cindex @code{vec_init@var{m}} instruction pattern
@item @samp{vec_init@var{m}}
Initialize the vector to given values.  Operand 0 is the vector to initialize
and operand 1 is parallel containing values for individual fields.

@cindex @code{push@var{m}1} instruction pattern
@item @samp{push@var{m}1}
Output a push instruction.  Operand 0 is value to push.  Used only when
@code{PUSH_ROUNDING} is defined.  For historical reason, this pattern may be
missing and in such case an @code{mov} expander is used instead, with a
@code{MEM} expression forming the push operation.  The @code{mov} expander
method is deprecated.

@cindex @code{add@var{m}3} instruction pattern
@item @samp{add@var{m}3}
Add operand 2 and operand 1, storing the result in operand 0.  All operands
must have mode @var{m}.  This can be used even on two-address machines, by
means of constraints requiring operands 1 and 0 to be the same location.

@cindex @code{sub@var{m}3} instruction pattern
@cindex @code{mul@var{m}3} instruction pattern
@cindex @code{div@var{m}3} instruction pattern
@cindex @code{udiv@var{m}3} instruction pattern
@cindex @code{mod@var{m}3} instruction pattern
@cindex @code{umod@var{m}3} instruction pattern
@cindex @code{umin@var{m}3} instruction pattern
@cindex @code{umax@var{m}3} instruction pattern
@cindex @code{and@var{m}3} instruction pattern
@cindex @code{ior@var{m}3} instruction pattern
@cindex @code{xor@var{m}3} instruction pattern
@item @samp{sub@var{m}3}, @samp{mul@var{m}3}
@itemx @samp{div@var{m}3}, @samp{udiv@var{m}3}
@itemx @samp{mod@var{m}3}, @samp{umod@var{m}3}
@itemx @samp{umin@var{m}3}, @samp{umax@var{m}3}
@itemx @samp{and@var{m}3}, @samp{ior@var{m}3}, @samp{xor@var{m}3}
Similar, for other arithmetic operations.

@cindex @code{min@var{m}3} instruction pattern
@cindex @code{max@var{m}3} instruction pattern
@item @samp{smin@var{m}3}, @samp{smax@var{m}3}
Signed minimum and maximum operations.  When used with floating point,
if both operands are zeros, or if either operand is @code{NaN}, then
it is unspecified which of the two operands is returned as the result.

@cindex @code{reduc_smin_@var{m}} instruction pattern
@cindex @code{reduc_smax_@var{m}} instruction pattern
@item @samp{reduc_smin_@var{m}}, @samp{reduc_smax_@var{m}}
Find the signed minimum/maximum of the elements of a vector. The vector is
operand 1, and the scalar result is stored in the least significant bits of
operand 0 (also a vector). The output and input vector should have the same
modes.

@cindex @code{reduc_umin_@var{m}} instruction pattern
@cindex @code{reduc_umax_@var{m}} instruction pattern
@item @samp{reduc_umin_@var{m}}, @samp{reduc_umax_@var{m}}
Find the unsigned minimum/maximum of the elements of a vector. The vector is
operand 1, and the scalar result is stored in the least significant bits of
operand 0 (also a vector). The output and input vector should have the same
modes.

@cindex @code{reduc_splus_@var{m}} instruction pattern
@item @samp{reduc_splus_@var{m}}
Compute the sum of the signed elements of a vector. The vector is operand 1,
and the scalar result is stored in the least significant bits of operand 0
(also a vector). The output and input vector should have the same modes.

@cindex @code{reduc_uplus_@var{m}} instruction pattern
@item @samp{reduc_uplus_@var{m}}
Compute the sum of the unsigned elements of a vector. The vector is operand 1,
and the scalar result is stored in the least significant bits of operand 0
(also a vector). The output and input vector should have the same modes.

@cindex @code{sdot_prod@var{m}} instruction pattern
@item @samp{sdot_prod@var{m}}
@cindex @code{udot_prod@var{m}} instruction pattern
@item @samp{udot_prod@var{m}}
Compute the sum of the products of two signed/unsigned elements. 
Operand 1 and operand 2 are of the same mode. Their product, which is of a 
wider mode, is computed and added to operand 3. Operand 3 is of a mode equal or 
wider than the mode of the product. The result is placed in operand 0, which
is of the same mode as operand 3. 

@cindex @code{ssum_widen@var{m3}} instruction pattern
@item @samp{ssum_widen@var{m3}}
@cindex @code{usum_widen@var{m3}} instruction pattern
@item @samp{usum_widen@var{m3}}
Operands 0 and 2 are of the same mode, which is wider than the mode of 
operand 1. Add operand 1 to operand 2 and place the widened result in
operand 0. (This is used express accumulation of elements into an accumulator
of a wider mode.)

@cindex @code{vec_shl_@var{m}} instruction pattern
@cindex @code{vec_shr_@var{m}} instruction pattern
@item @samp{vec_shl_@var{m}}, @samp{vec_shr_@var{m}}
Whole vector left/right shift in bits.
Operand 1 is a vector to be shifted.
Operand 2 is an integer shift amount in bits.
Operand 0 is where the resulting shifted vector is stored.
The output and input vectors should have the same modes.

@cindex @code{vec_pack_trunc_@var{m}} instruction pattern
@item @samp{vec_pack_trunc_@var{m}}
Narrow (demote) and merge the elements of two vectors. Operands 1 and 2
are vectors of the same mode having N integral or floating point elements
of size S.  Operand 0 is the resulting vector in which 2*N elements of
size N/2 are concatenated after narrowing them down using truncation.

@cindex @code{vec_pack_ssat_@var{m}} instruction pattern
@cindex @code{vec_pack_usat_@var{m}} instruction pattern
@item @samp{vec_pack_ssat_@var{m}}, @samp{vec_pack_usat_@var{m}}
Narrow (demote) and merge the elements of two vectors.  Operands 1 and 2
are vectors of the same mode having N integral elements of size S.
Operand 0 is the resulting vector in which the elements of the two input
vectors are concatenated after narrowing them down using signed/unsigned
saturating arithmetic.

@cindex @code{vec_unpacks_hi_@var{m}} instruction pattern
@cindex @code{vec_unpacks_lo_@var{m}} instruction pattern
@item @samp{vec_unpacks_hi_@var{m}}, @samp{vec_unpacks_lo_@var{m}}
Extract and widen (promote) the high/low part of a vector of signed
integral or floating point elements.  The input vector (operand 1) has N
elements of size S.  Widen (promote) the high/low elements of the vector
using signed or floating point extension and place the resulting N/2
values of size 2*S in the output vector (operand 0).

@cindex @code{vec_unpacku_hi_@var{m}} instruction pattern
@cindex @code{vec_unpacku_lo_@var{m}} instruction pattern
@item @samp{vec_unpacku_hi_@var{m}}, @samp{vec_unpacku_lo_@var{m}}
Extract and widen (promote) the high/low part of a vector of unsigned
integral elements.  The input vector (operand 1) has N elements of size S.
Widen (promote) the high/low elements of the vector using zero extension and
place the resulting N/2 values of size 2*S in the output vector (operand 0).

@cindex @code{vec_widen_umult_hi_@var{m}} instruction pattern
@cindex @code{vec_widen_umult_lo__@var{m}} instruction pattern
@cindex @code{vec_widen_smult_hi_@var{m}} instruction pattern
@cindex @code{vec_widen_smult_lo_@var{m}} instruction pattern
@item @samp{vec_widen_umult_hi_@var{m}}, @samp{vec_widen_umult_lo_@var{m}}, @samp{vec_widen_smult_hi_@var{m}}, @samp{vec_widen_smult_lo_@var{m}}
Signed/Unsigned widening multiplication.  The two inputs (operands 1 and 2)
are vectors with N signed/unsigned elements of size S.  Multiply the high/low
elements of the two vectors, and put the N/2 products of size 2*S in the
output vector (operand 0).

@cindex @code{mulhisi3} instruction pattern
@item @samp{mulhisi3}
Multiply operands 1 and 2, which have mode @code{HImode}, and store
a @code{SImode} product in operand 0.

@cindex @code{mulqihi3} instruction pattern
@cindex @code{mulsidi3} instruction pattern
@item @samp{mulqihi3}, @samp{mulsidi3}
Similar widening-multiplication instructions of other widths.

@cindex @code{umulqihi3} instruction pattern
@cindex @code{umulhisi3} instruction pattern
@cindex @code{umulsidi3} instruction pattern
@item @samp{umulqihi3}, @samp{umulhisi3}, @samp{umulsidi3}
Similar widening-multiplication instructions that do unsigned
multiplication.

@cindex @code{usmulqihi3} instruction pattern
@cindex @code{usmulhisi3} instruction pattern
@cindex @code{usmulsidi3} instruction pattern
@item @samp{usmulqihi3}, @samp{usmulhisi3}, @samp{usmulsidi3}
Similar widening-multiplication instructions that interpret the first
operand as unsigned and the second operand as signed, then do a signed
multiplication.

@cindex @code{smul@var{m}3_highpart} instruction pattern
@item @samp{smul@var{m}3_highpart}
Perform a signed multiplication of operands 1 and 2, which have mode
@var{m}, and store the most significant half of the product in operand 0.
The least significant half of the product is discarded.

@cindex @code{umul@var{m}3_highpart} instruction pattern
@item @samp{umul@var{m}3_highpart}
Similar, but the multiplication is unsigned.

@cindex @code{madd@var{m}@var{n}4} instruction pattern
@item @samp{madd@var{m}@var{n}4}
Multiply operands 1 and 2, sign-extend them to mode @var{n}, add
operand 3, and store the result in operand 0.  Operands 1 and 2
have mode @var{m} and operands 0 and 3 have mode @var{n}.
Both modes must be integer modes and @var{n} must be twice
the size of @var{m}.

In other words, @code{madd@var{m}@var{n}4} is like
@code{mul@var{m}@var{n}3} except that it also adds operand 3.

These instructions are not allowed to @code{FAIL}.

@cindex @code{umadd@var{m}@var{n}4} instruction pattern
@item @samp{umadd@var{m}@var{n}4}
Like @code{madd@var{m}@var{n}4}, but zero-extend the multiplication
operands instead of sign-extending them.

@cindex @code{msub@var{m}@var{n}4} instruction pattern
@item @samp{msub@var{m}@var{n}4}
Multiply operands 1 and 2, sign-extend them to mode @var{n}, subtract the
result from operand 3, and store the result in operand 0.  Operands 1 and 2
have mode @var{m} and operands 0 and 3 have mode @var{n}.
Both modes must be integer modes and @var{n} must be twice
the size of @var{m}.

In other words, @code{msub@var{m}@var{n}4} is like
@code{mul@var{m}@var{n}3} except that it also subtracts the result
from operand 3.

These instructions are not allowed to @code{FAIL}.

@cindex @code{umsub@var{m}@var{n}4} instruction pattern
@item @samp{umsub@var{m}@var{n}4}
Like @code{msub@var{m}@var{n}4}, but zero-extend the multiplication
operands instead of sign-extending them.

@cindex @code{divmod@var{m}4} instruction pattern
@item @samp{divmod@var{m}4}
Signed division that produces both a quotient and a remainder.
Operand 1 is divided by operand 2 to produce a quotient stored
in operand 0 and a remainder stored in operand 3.

For machines with an instruction that produces both a quotient and a
remainder, provide a pattern for @samp{divmod@var{m}4} but do not
provide patterns for @samp{div@var{m}3} and @samp{mod@var{m}3}.  This
allows optimization in the relatively common case when both the quotient
and remainder are computed.

If an instruction that just produces a quotient or just a remainder
exists and is more efficient than the instruction that produces both,
write the output routine of @samp{divmod@var{m}4} to call
@code{find_reg_note} and look for a @code{REG_UNUSED} note on the
quotient or remainder and generate the appropriate instruction.

@cindex @code{udivmod@var{m}4} instruction pattern
@item @samp{udivmod@var{m}4}
Similar, but does unsigned division.

@anchor{shift patterns}
@cindex @code{ashl@var{m}3} instruction pattern
@item @samp{ashl@var{m}3}
Arithmetic-shift operand 1 left by a number of bits specified by operand
2, and store the result in operand 0.  Here @var{m} is the mode of
operand 0 and operand 1; operand 2's mode is specified by the
instruction pattern, and the compiler will convert the operand to that
mode before generating the instruction.  The meaning of out-of-range shift
counts can optionally be specified by @code{TARGET_SHIFT_TRUNCATION_MASK}.
@xref{TARGET_SHIFT_TRUNCATION_MASK}.

@cindex @code{ashr@var{m}3} instruction pattern
@cindex @code{lshr@var{m}3} instruction pattern
@cindex @code{rotl@var{m}3} instruction pattern
@cindex @code{rotr@var{m}3} instruction pattern
@item @samp{ashr@var{m}3}, @samp{lshr@var{m}3}, @samp{rotl@var{m}3}, @samp{rotr@var{m}3}
Other shift and rotate instructions, analogous to the
@code{ashl@var{m}3} instructions.

@cindex @code{neg@var{m}2} instruction pattern
@item @samp{neg@var{m}2}
Negate operand 1 and store the result in operand 0.

@cindex @code{abs@var{m}2} instruction pattern
@item @samp{abs@var{m}2}
Store the absolute value of operand 1 into operand 0.

@cindex @code{sqrt@var{m}2} instruction pattern
@item @samp{sqrt@var{m}2}
Store the square root of operand 1 into operand 0.

The @code{sqrt} built-in function of C always uses the mode which
corresponds to the C data type @code{double} and the @code{sqrtf}
built-in function uses the mode which corresponds to the C data
type @code{float}.

@cindex @code{fmod@var{m}3} instruction pattern
@item @samp{fmod@var{m}3}
Store the remainder of dividing operand 1 by operand 2 into
operand 0, rounded towards zero to an integer.

The @code{fmod} built-in function of C always uses the mode which
corresponds to the C data type @code{double} and the @code{fmodf}
built-in function uses the mode which corresponds to the C data
type @code{float}.

@cindex @code{remainder@var{m}3} instruction pattern
@item @samp{remainder@var{m}3}
Store the remainder of dividing operand 1 by operand 2 into
operand 0, rounded to the nearest integer.

The @code{remainder} built-in function of C always uses the mode
which corresponds to the C data type @code{double} and the
@code{remainderf} built-in function uses the mode which corresponds
to the C data type @code{float}.

@cindex @code{cos@var{m}2} instruction pattern
@item @samp{cos@var{m}2}
Store the cosine of operand 1 into operand 0.

The @code{cos} built-in function of C always uses the mode which
corresponds to the C data type @code{double} and the @code{cosf}
built-in function uses the mode which corresponds to the C data
type @code{float}.

@cindex @code{sin@var{m}2} instruction pattern
@item @samp{sin@var{m}2}
Store the sine of operand 1 into operand 0.

The @code{sin} built-in function of C always uses the mode which
corresponds to the C data type @code{double} and the @code{sinf}
built-in function uses the mode which corresponds to the C data
type @code{float}.

@cindex @code{exp@var{m}2} instruction pattern
@item @samp{exp@var{m}2}
Store the exponential of operand 1 into operand 0.

The @code{exp} built-in function of C always uses the mode which
corresponds to the C data type @code{double} and the @code{expf}
built-in function uses the mode which corresponds to the C data
type @code{float}.

@cindex @code{log@var{m}2} instruction pattern
@item @samp{log@var{m}2}
Store the natural logarithm of operand 1 into operand 0.

The @code{log} built-in function of C always uses the mode which
corresponds to the C data type @code{double} and the @code{logf}
built-in function uses the mode which corresponds to the C data
type @code{float}.

@cindex @code{pow@var{m}3} instruction pattern
@item @samp{pow@var{m}3}
Store the value of operand 1 raised to the exponent operand 2
into operand 0.

The @code{pow} built-in function of C always uses the mode which
corresponds to the C data type @code{double} and the @code{powf}
built-in function uses the mode which corresponds to the C data
type @code{float}.

@cindex @code{atan2@var{m}3} instruction pattern
@item @samp{atan2@var{m}3}
Store the arc tangent (inverse tangent) of operand 1 divided by
operand 2 into operand 0, using the signs of both arguments to
determine the quadrant of the result.

The @code{atan2} built-in function of C always uses the mode which
corresponds to the C data type @code{double} and the @code{atan2f}
built-in function uses the mode which corresponds to the C data
type @code{float}.

@cindex @code{floor@var{m}2} instruction pattern
@item @samp{floor@var{m}2}
Store the largest integral value not greater than argument.

The @code{floor} built-in function of C always uses the mode which
corresponds to the C data type @code{double} and the @code{floorf}
built-in function uses the mode which corresponds to the C data
type @code{float}.

@cindex @code{btrunc@var{m}2} instruction pattern
@item @samp{btrunc@var{m}2}
Store the argument rounded to integer towards zero.

The @code{trunc} built-in function of C always uses the mode which
corresponds to the C data type @code{double} and the @code{truncf}
built-in function uses the mode which corresponds to the C data
type @code{float}.

@cindex @code{round@var{m}2} instruction pattern
@item @samp{round@var{m}2}
Store the argument rounded to integer away from zero.

The @code{round} built-in function of C always uses the mode which
corresponds to the C data type @code{double} and the @code{roundf}
built-in function uses the mode which corresponds to the C data
type @code{float}.

@cindex @code{ceil@var{m}2} instruction pattern
@item @samp{ceil@var{m}2}
Store the argument rounded to integer away from zero.

The @code{ceil} built-in function of C always uses the mode which
corresponds to the C data type @code{double} and the @code{ceilf}
built-in function uses the mode which corresponds to the C data
type @code{float}.

@cindex @code{nearbyint@var{m}2} instruction pattern
@item @samp{nearbyint@var{m}2}
Store the argument rounded according to the default rounding mode

The @code{nearbyint} built-in function of C always uses the mode which
corresponds to the C data type @code{double} and the @code{nearbyintf}
built-in function uses the mode which corresponds to the C data
type @code{float}.

@cindex @code{rint@var{m}2} instruction pattern
@item @samp{rint@var{m}2}
Store the argument rounded according to the default rounding mode and
raise the inexact exception when the result differs in value from
the argument

The @code{rint} built-in function of C always uses the mode which
corresponds to the C data type @code{double} and the @code{rintf}
built-in function uses the mode which corresponds to the C data
type @code{float}.

@cindex @code{lrint@var{m}@var{n}2}
@item @samp{lrint@var{m}@var{n}2}
Convert operand 1 (valid for floating point mode @var{m}) to fixed
point mode @var{n} as a signed number according to the current
rounding mode and store in operand 0 (which has mode @var{n}).

@cindex @code{lround@var{m}@var{n}2}
@item @samp{lround@var{m}2}
Convert operand 1 (valid for floating point mode @var{m}) to fixed
point mode @var{n} as a signed number rounding to nearest and away
from zero and store in operand 0 (which has mode @var{n}).

@cindex @code{lfloor@var{m}@var{n}2}
@item @samp{lfloor@var{m}2}
Convert operand 1 (valid for floating point mode @var{m}) to fixed
point mode @var{n} as a signed number rounding down and store in
operand 0 (which has mode @var{n}).

@cindex @code{lceil@var{m}@var{n}2}
@item @samp{lceil@var{m}2}
Convert operand 1 (valid for floating point mode @var{m}) to fixed
point mode @var{n} as a signed number rounding up and store in
operand 0 (which has mode @var{n}).

@cindex @code{copysign@var{m}3} instruction pattern
@item @samp{copysign@var{m}3}
Store a value with the magnitude of operand 1 and the sign of operand
2 into operand 0.

The @code{copysign} built-in function of C always uses the mode which
corresponds to the C data type @code{double} and the @code{copysignf}
built-in function uses the mode which corresponds to the C data
type @code{float}.

@cindex @code{ffs@var{m}2} instruction pattern
@item @samp{ffs@var{m}2}
Store into operand 0 one plus the index of the least significant 1-bit
of operand 1.  If operand 1 is zero, store zero.  @var{m} is the mode
of operand 0; operand 1's mode is specified by the instruction
pattern, and the compiler will convert the operand to that mode before
generating the instruction.

The @code{ffs} built-in function of C always uses the mode which
corresponds to the C data type @code{int}.

@cindex @code{clz@var{m}2} instruction pattern
@item @samp{clz@var{m}2}
Store into operand 0 the number of leading 0-bits in @var{x}, starting
at the most significant bit position.  If @var{x} is 0, the result is
undefined.  @var{m} is the mode of operand 0; operand 1's mode is
specified by the instruction pattern, and the compiler will convert the
operand to that mode before generating the instruction.

@cindex @code{ctz@var{m}2} instruction pattern
@item @samp{ctz@var{m}2}
Store into operand 0 the number of trailing 0-bits in @var{x}, starting
at the least significant bit position.  If @var{x} is 0, the result is
undefined.  @var{m} is the mode of operand 0; operand 1's mode is
specified by the instruction pattern, and the compiler will convert the
operand to that mode before generating the instruction.

@cindex @code{popcount@var{m}2} instruction pattern
@item @samp{popcount@var{m}2}
Store into operand 0 the number of 1-bits in @var{x}.  @var{m} is the
mode of operand 0; operand 1's mode is specified by the instruction
pattern, and the compiler will convert the operand to that mode before
generating the instruction.

@cindex @code{parity@var{m}2} instruction pattern
@item @samp{parity@var{m}2}
Store into operand 0 the parity of @var{x}, i.e.@: the number of 1-bits
in @var{x} modulo 2.  @var{m} is the mode of operand 0; operand 1's mode
is specified by the instruction pattern, and the compiler will convert
the operand to that mode before generating the instruction.

@cindex @code{one_cmpl@var{m}2} instruction pattern
@item @samp{one_cmpl@var{m}2}
Store the bitwise-complement of operand 1 into operand 0.

@cindex @code{cmp@var{m}} instruction pattern
@item @samp{cmp@var{m}}
Compare operand 0 and operand 1, and set the condition codes.
The RTL pattern should look like this:

@smallexample
(set (cc0) (compare (match_operand:@var{m} 0 @dots{})
                    (match_operand:@var{m} 1 @dots{})))
@end smallexample

@cindex @code{tst@var{m}} instruction pattern
@item @samp{tst@var{m}}
Compare operand 0 against zero, and set the condition codes.
The RTL pattern should look like this:

@smallexample
(set (cc0) (match_operand:@var{m} 0 @dots{}))
@end smallexample

@samp{tst@var{m}} patterns should not be defined for machines that do
not use @code{(cc0)}.  Doing so would confuse the optimizer since it
would no longer be clear which @code{set} operations were comparisons.
The @samp{cmp@var{m}} patterns should be used instead.

@cindex @code{movmem@var{m}} instruction pattern
@item @samp{movmem@var{m}}
Block move instruction.  The destination and source blocks of memory
are the first two operands, and both are @code{mem:BLK}s with an
address in mode @code{Pmode}.

The number of bytes to move is the third operand, in mode @var{m}.
Usually, you specify @code{word_mode} for @var{m}.  However, if you can
generate better code knowing the range of valid lengths is smaller than
those representable in a full word, you should provide a pattern with a
mode corresponding to the range of values you can handle efficiently
(e.g., @code{QImode} for values in the range 0--127; note we avoid numbers
that appear negative) and also a pattern with @code{word_mode}.

The fourth operand is the known shared alignment of the source and
destination, in the form of a @code{const_int} rtx.  Thus, if the
compiler knows that both source and destination are word-aligned,
it may provide the value 4 for this operand.

Optional operands 5 and 6 specify expected alignment and size of block
respectively.  The expected alignment differs from alignment in operand 4
in a way that the blocks are not required to be aligned according to it in
all cases. Expected size, when unknown, is set to @code{(const_int -1)}.

Descriptions of multiple @code{movmem@var{m}} patterns can only be
beneficial if the patterns for smaller modes have fewer restrictions
on their first, second and fourth operands.  Note that the mode @var{m}
in @code{movmem@var{m}} does not impose any restriction on the mode of
individually moved data units in the block.

These patterns need not give special consideration to the possibility
that the source and destination strings might overlap.

@cindex @code{movstr} instruction pattern
@item @samp{movstr}
String copy instruction, with @code{stpcpy} semantics.  Operand 0 is
an output operand in mode @code{Pmode}.  The addresses of the
destination and source strings are operands 1 and 2, and both are
@code{mem:BLK}s with addresses in mode @code{Pmode}.  The execution of
the expansion of this pattern should store in operand 0 the address in
which the @code{NUL} terminator was stored in the destination string.

@cindex @code{setmem@var{m}} instruction pattern
@item @samp{setmem@var{m}}
Block set instruction.  The destination string is the first operand,
given as a @code{mem:BLK} whose address is in mode @code{Pmode}.  The
number of bytes to set is the second operand, in mode @var{m}.  The value to
initialize the memory with is the third operand. Targets that only support the
clearing of memory should reject any value that is not the constant 0.  See
@samp{movmem@var{m}} for a discussion of the choice of mode.

The fourth operand is the known alignment of the destination, in the form
of a @code{const_int} rtx.  Thus, if the compiler knows that the
destination is word-aligned, it may provide the value 4 for this
operand.

Optional operands 5 and 6 specify expected alignment and size of block
respectively.  The expected alignment differs from alignment in operand 4
in a way that the blocks are not required to be aligned according to it in
all cases. Expected size, when unknown, is set to @code{(const_int -1)}.

The use for multiple @code{setmem@var{m}} is as for @code{movmem@var{m}}.

@cindex @code{cmpstrn@var{m}} instruction pattern
@item @samp{cmpstrn@var{m}}
String compare instruction, with five operands.  Operand 0 is the output;
it has mode @var{m}.  The remaining four operands are like the operands
of @samp{movmem@var{m}}.  The two memory blocks specified are compared
byte by byte in lexicographic order starting at the beginning of each
string.  The instruction is not allowed to prefetch more than one byte
at a time since either string may end in the first byte and reading past
that may access an invalid page or segment and cause a fault.  The
effect of the instruction is to store a value in operand 0 whose sign
indicates the result of the comparison.

@cindex @code{cmpstr@var{m}} instruction pattern
@item @samp{cmpstr@var{m}}
String compare instruction, without known maximum length.  Operand 0 is the
output; it has mode @var{m}.  The second and third operand are the blocks of
memory to be compared; both are @code{mem:BLK} with an address in mode
@code{Pmode}.

The fourth operand is the known shared alignment of the source and
destination, in the form of a @code{const_int} rtx.  Thus, if the
compiler knows that both source and destination are word-aligned,
it may provide the value 4 for this operand.

The two memory blocks specified are compared byte by byte in lexicographic
order starting at the beginning of each string.  The instruction is not allowed
to prefetch more than one byte at a time since either string may end in the
first byte and reading past that may access an invalid page or segment and
cause a fault.  The effect of the instruction is to store a value in operand 0
whose sign indicates the result of the comparison.

@cindex @code{cmpmem@var{m}} instruction pattern
@item @samp{cmpmem@var{m}}
Block compare instruction, with five operands like the operands
of @samp{cmpstr@var{m}}.  The two memory blocks specified are compared
byte by byte in lexicographic order starting at the beginning of each
block.  Unlike @samp{cmpstr@var{m}} the instruction can prefetch
any bytes in the two memory blocks.  The effect of the instruction is
to store a value in operand 0 whose sign indicates the result of the
comparison.

@cindex @code{strlen@var{m}} instruction pattern
@item @samp{strlen@var{m}}
Compute the length of a string, with three operands.
Operand 0 is the result (of mode @var{m}), operand 1 is
a @code{mem} referring to the first character of the string,
operand 2 is the character to search for (normally zero),
and operand 3 is a constant describing the known alignment
of the beginning of the string.

@cindex @code{float@var{mn}2} instruction pattern
@item @samp{float@var{m}@var{n}2}
Convert signed integer operand 1 (valid for fixed point mode @var{m}) to
floating point mode @var{n} and store in operand 0 (which has mode
@var{n}).

@cindex @code{floatuns@var{mn}2} instruction pattern
@item @samp{floatuns@var{m}@var{n}2}
Convert unsigned integer operand 1 (valid for fixed point mode @var{m})
to floating point mode @var{n} and store in operand 0 (which has mode
@var{n}).

@cindex @code{fix@var{mn}2} instruction pattern
@item @samp{fix@var{m}@var{n}2}
Convert operand 1 (valid for floating point mode @var{m}) to fixed
point mode @var{n} as a signed number and store in operand 0 (which
has mode @var{n}).  This instruction's result is defined only when
the value of operand 1 is an integer.

If the machine description defines this pattern, it also needs to
define the @code{ftrunc} pattern.

@cindex @code{fixuns@var{mn}2} instruction pattern
@item @samp{fixuns@var{m}@var{n}2}
Convert operand 1 (valid for floating point mode @var{m}) to fixed
point mode @var{n} as an unsigned number and store in operand 0 (which
has mode @var{n}).  This instruction's result is defined only when the
value of operand 1 is an integer.

@cindex @code{ftrunc@var{m}2} instruction pattern
@item @samp{ftrunc@var{m}2}
Convert operand 1 (valid for floating point mode @var{m}) to an
integer value, still represented in floating point mode @var{m}, and
store it in operand 0 (valid for floating point mode @var{m}).

@cindex @code{fix_trunc@var{mn}2} instruction pattern
@item @samp{fix_trunc@var{m}@var{n}2}
Like @samp{fix@var{m}@var{n}2} but works for any floating point value
of mode @var{m} by converting the value to an integer.

@cindex @code{fixuns_trunc@var{mn}2} instruction pattern
@item @samp{fixuns_trunc@var{m}@var{n}2}
Like @samp{fixuns@var{m}@var{n}2} but works for any floating point
value of mode @var{m} by converting the value to an integer.

@cindex @code{trunc@var{mn}2} instruction pattern
@item @samp{trunc@var{m}@var{n}2}
Truncate operand 1 (valid for mode @var{m}) to mode @var{n} and
store in operand 0 (which has mode @var{n}).  Both modes must be fixed
point or both floating point.

@cindex @code{extend@var{mn}2} instruction pattern
@item @samp{extend@var{m}@var{n}2}
Sign-extend operand 1 (valid for mode @var{m}) to mode @var{n} and
store in operand 0 (which has mode @var{n}).  Both modes must be fixed
point or both floating point.

@cindex @code{zero_extend@var{mn}2} instruction pattern
@item @samp{zero_extend@var{m}@var{n}2}
Zero-extend operand 1 (valid for mode @var{m}) to mode @var{n} and
store in operand 0 (which has mode @var{n}).  Both modes must be fixed
point.

@cindex @code{extv} instruction pattern
@item @samp{extv}
Extract a bit-field from operand 1 (a register or memory operand), where
operand 2 specifies the width in bits and operand 3 the starting bit,
and store it in operand 0.  Operand 0 must have mode @code{word_mode}.
Operand 1 may have mode @code{byte_mode} or @code{word_mode}; often
@code{word_mode} is allowed only for registers.  Operands 2 and 3 must
be valid for @code{word_mode}.

The RTL generation pass generates this instruction only with constants
for operands 2 and 3 and the constant is never zero for operand 2.

The bit-field value is sign-extended to a full word integer
before it is stored in operand 0.

@cindex @code{extzv} instruction pattern
@item @samp{extzv}
Like @samp{extv} except that the bit-field value is zero-extended.

@cindex @code{insv} instruction pattern
@item @samp{insv}
Store operand 3 (which must be valid for @code{word_mode}) into a
bit-field in operand 0, where operand 1 specifies the width in bits and
operand 2 the starting bit.  Operand 0 may have mode @code{byte_mode} or
@code{word_mode}; often @code{word_mode} is allowed only for registers.
Operands 1 and 2 must be valid for @code{word_mode}.

The RTL generation pass generates this instruction only with constants
for operands 1 and 2 and the constant is never zero for operand 1.

@cindex @code{mov@var{mode}cc} instruction pattern
@item @samp{mov@var{mode}cc}
Conditionally move operand 2 or operand 3 into operand 0 according to the
comparison in operand 1.  If the comparison is true, operand 2 is moved
into operand 0, otherwise operand 3 is moved.

The mode of the operands being compared need not be the same as the operands
being moved.  Some machines, sparc64 for example, have instructions that
conditionally move an integer value based on the floating point condition
codes and vice versa.

If the machine does not have conditional move instructions, do not
define these patterns.

@cindex @code{add@var{mode}cc} instruction pattern
@item @samp{add@var{mode}cc}
Similar to @samp{mov@var{mode}cc} but for conditional addition.  Conditionally
move operand 2 or (operands 2 + operand 3) into operand 0 according to the
comparison in operand 1.  If the comparison is true, operand 2 is moved into
operand 0, otherwise (operand 2 + operand 3) is moved.

@cindex @code{s@var{cond}} instruction pattern
@item @samp{s@var{cond}}
Store zero or nonzero in the operand according to the condition codes.
Value stored is nonzero iff the condition @var{cond} is true.
@var{cond} is the name of a comparison operation expression code, such
as @code{eq}, @code{lt} or @code{leu}.

You specify the mode that the operand must have when you write the
@code{match_operand} expression.  The compiler automatically sees
which mode you have used and supplies an operand of that mode.

The value stored for a true condition must have 1 as its low bit, or
else must be negative.  Otherwise the instruction is not suitable and
you should omit it from the machine description.  You describe to the
compiler exactly which value is stored by defining the macro
@code{STORE_FLAG_VALUE} (@pxref{Misc}).  If a description cannot be
found that can be used for all the @samp{s@var{cond}} patterns, you
should omit those operations from the machine description.

These operations may fail, but should do so only in relatively
uncommon cases; if they would fail for common cases involving
integer comparisons, it is best to omit these patterns.

If these operations are omitted, the compiler will usually generate code
that copies the constant one to the target and branches around an
assignment of zero to the target.  If this code is more efficient than
the potential instructions used for the @samp{s@var{cond}} pattern
followed by those required to convert the result into a 1 or a zero in
@code{SImode}, you should omit the @samp{s@var{cond}} operations from
the machine description.

@cindex @code{b@var{cond}} instruction pattern
@item @samp{b@var{cond}}
Conditional branch instruction.  Operand 0 is a @code{label_ref} that
refers to the label to jump to.  Jump if the condition codes meet
condition @var{cond}.

Some machines do not follow the model assumed here where a comparison
instruction is followed by a conditional branch instruction.  In that
case, the @samp{cmp@var{m}} (and @samp{tst@var{m}}) patterns should
simply store the operands away and generate all the required insns in a
@code{define_expand} (@pxref{Expander Definitions}) for the conditional
branch operations.  All calls to expand @samp{b@var{cond}} patterns are
immediately preceded by calls to expand either a @samp{cmp@var{m}}
pattern or a @samp{tst@var{m}} pattern.

Machines that use a pseudo register for the condition code value, or
where the mode used for the comparison depends on the condition being
tested, should also use the above mechanism.  @xref{Jump Patterns}.

The above discussion also applies to the @samp{mov@var{mode}cc} and
@samp{s@var{cond}} patterns.

@cindex @code{cbranch@var{mode}4} instruction pattern
@item @samp{cbranch@var{mode}4}
Conditional branch instruction combined with a compare instruction.
Operand 0 is a comparison operator.  Operand 1 and operand 2 are the
first and second operands of the comparison, respectively.  Operand 3
is a @code{label_ref} that refers to the label to jump to.

@cindex @code{jump} instruction pattern
@item @samp{jump}
A jump inside a function; an unconditional branch.  Operand 0 is the
@code{label_ref} of the label to jump to.  This pattern name is mandatory
on all machines.

@cindex @code{call} instruction pattern
@item @samp{call}
Subroutine call instruction returning no value.  Operand 0 is the
function to call; operand 1 is the number of bytes of arguments pushed
as a @code{const_int}; operand 2 is the number of registers used as
operands.

On most machines, operand 2 is not actually stored into the RTL
pattern.  It is supplied for the sake of some RISC machines which need
to put this information into the assembler code; they can put it in
the RTL instead of operand 1.

Operand 0 should be a @code{mem} RTX whose address is the address of the
function.  Note, however, that this address can be a @code{symbol_ref}
expression even if it would not be a legitimate memory address on the
target machine.  If it is also not a valid argument for a call
instruction, the pattern for this operation should be a
@code{define_expand} (@pxref{Expander Definitions}) that places the
address into a register and uses that register in the call instruction.

@cindex @code{call_value} instruction pattern
@item @samp{call_value}
Subroutine call instruction returning a value.  Operand 0 is the hard
register in which the value is returned.  There are three more
operands, the same as the three operands of the @samp{call}
instruction (but with numbers increased by one).

Subroutines that return @code{BLKmode} objects use the @samp{call}
insn.

@cindex @code{call_pop} instruction pattern
@cindex @code{call_value_pop} instruction pattern
@item @samp{call_pop}, @samp{call_value_pop}
Similar to @samp{call} and @samp{call_value}, except used if defined and
if @code{RETURN_POPS_ARGS} is nonzero.  They should emit a @code{parallel}
that contains both the function call and a @code{set} to indicate the
adjustment made to the frame pointer.

For machines where @code{RETURN_POPS_ARGS} can be nonzero, the use of these
patterns increases the number of functions for which the frame pointer
can be eliminated, if desired.

@cindex @code{untyped_call} instruction pattern
@item @samp{untyped_call}
Subroutine call instruction returning a value of any type.  Operand 0 is
the function to call; operand 1 is a memory location where the result of
calling the function is to be stored; operand 2 is a @code{parallel}
expression where each element is a @code{set} expression that indicates
the saving of a function return value into the result block.

This instruction pattern should be defined to support
@code{__builtin_apply} on machines where special instructions are needed
to call a subroutine with arbitrary arguments or to save the value
returned.  This instruction pattern is required on machines that have
multiple registers that can hold a return value
(i.e.@: @code{FUNCTION_VALUE_REGNO_P} is true for more than one register).

@cindex @code{return} instruction pattern
@item @samp{return}
Subroutine return instruction.  This instruction pattern name should be
defined only if a single instruction can do all the work of returning
from a function.

Like the @samp{mov@var{m}} patterns, this pattern is also used after the
RTL generation phase.  In this case it is to support machines where
multiple instructions are usually needed to return from a function, but
some class of functions only requires one instruction to implement a
return.  Normally, the applicable functions are those which do not need
to save any registers or allocate stack space.

@findex reload_completed
@findex leaf_function_p
For such machines, the condition specified in this pattern should only
be true when @code{reload_completed} is nonzero and the function's
epilogue would only be a single instruction.  For machines with register
windows, the routine @code{leaf_function_p} may be used to determine if
a register window push is required.

Machines that have conditional return instructions should define patterns
such as

@smallexample
(define_insn ""
  [(set (pc)
        (if_then_else (match_operator
                         0 "comparison_operator"
                         [(cc0) (const_int 0)])
                      (return)
                      (pc)))]
  "@var{condition}"
  "@dots{}")
@end smallexample

where @var{condition} would normally be the same condition specified on the
named @samp{return} pattern.

@cindex @code{untyped_return} instruction pattern
@item @samp{untyped_return}
Untyped subroutine return instruction.  This instruction pattern should
be defined to support @code{__builtin_return} on machines where special
instructions are needed to return a value of any type.

Operand 0 is a memory location where the result of calling a function
with @code{__builtin_apply} is stored; operand 1 is a @code{parallel}
expression where each element is a @code{set} expression that indicates
the restoring of a function return value from the result block.

@cindex @code{nop} instruction pattern
@item @samp{nop}
No-op instruction.  This instruction pattern name should always be defined
to output a no-op in assembler code.  @code{(const_int 0)} will do as an
RTL pattern.

@cindex @code{indirect_jump} instruction pattern
@item @samp{indirect_jump}
An instruction to jump to an address which is operand zero.
This pattern name is mandatory on all machines.

@cindex @code{casesi} instruction pattern
@item @samp{casesi}
Instruction to jump through a dispatch table, including bounds checking.
This instruction takes five operands:

@enumerate
@item
The index to dispatch on, which has mode @code{SImode}.

@item
The lower bound for indices in the table, an integer constant.

@item
The total range of indices in the table---the largest index
minus the smallest one (both inclusive).

@item
A label that precedes the table itself.

@item
A label to jump to if the index has a value outside the bounds.
@end enumerate

The table is a @code{addr_vec} or @code{addr_diff_vec} inside of a
@code{jump_insn}.  The number of elements in the table is one plus the
difference between the upper bound and the lower bound.

@cindex @code{tablejump} instruction pattern
@item @samp{tablejump}
Instruction to jump to a variable address.  This is a low-level
capability which can be used to implement a dispatch table when there
is no @samp{casesi} pattern.

This pattern requires two operands: the address or offset, and a label
which should immediately precede the jump table.  If the macro
@code{CASE_VECTOR_PC_RELATIVE} evaluates to a nonzero value then the first
operand is an offset which counts from the address of the table; otherwise,
it is an absolute address to jump to.  In either case, the first operand has
mode @code{Pmode}.

The @samp{tablejump} insn is always the last insn before the jump
table it uses.  Its assembler code normally has no need to use the
second operand, but you should incorporate it in the RTL pattern so
that the jump optimizer will not delete the table as unreachable code.


@cindex @code{decrement_and_branch_until_zero} instruction pattern
@item @samp{decrement_and_branch_until_zero}
Conditional branch instruction that decrements a register and
jumps if the register is nonzero.  Operand 0 is the register to
decrement and test; operand 1 is the label to jump to if the
register is nonzero.  @xref{Looping Patterns}.

This optional instruction pattern is only used by the combiner,
typically for loops reversed by the loop optimizer when strength
reduction is enabled.

@cindex @code{doloop_end} instruction pattern
@item @samp{doloop_end}
Conditional branch instruction that decrements a register and jumps if
the register is nonzero.  This instruction takes five operands: Operand
0 is the register to decrement and test; operand 1 is the number of loop
iterations as a @code{const_int} or @code{const0_rtx} if this cannot be
determined until run-time; operand 2 is the actual or estimated maximum
number of iterations as a @code{const_int}; operand 3 is the number of
enclosed loops as a @code{const_int} (an innermost loop has a value of
1); operand 4 is the label to jump to if the register is nonzero.
@xref{Looping Patterns}.

This optional instruction pattern should be defined for machines with
low-overhead looping instructions as the loop optimizer will try to
modify suitable loops to utilize it.  If nested low-overhead looping is
not supported, use a @code{define_expand} (@pxref{Expander Definitions})
and make the pattern fail if operand 3 is not @code{const1_rtx}.
Similarly, if the actual or estimated maximum number of iterations is
too large for this instruction, make it fail.

@cindex @code{doloop_begin} instruction pattern
@item @samp{doloop_begin}
Companion instruction to @code{doloop_end} required for machines that
need to perform some initialization, such as loading special registers
used by a low-overhead looping instruction.  If initialization insns do
not always need to be emitted, use a @code{define_expand}
(@pxref{Expander Definitions}) and make it fail.


@cindex @code{canonicalize_funcptr_for_compare} instruction pattern
@item @samp{canonicalize_funcptr_for_compare}
Canonicalize the function pointer in operand 1 and store the result
into operand 0.

Operand 0 is always a @code{reg} and has mode @code{Pmode}; operand 1
may be a @code{reg}, @code{mem}, @code{symbol_ref}, @code{const_int}, etc
and also has mode @code{Pmode}.

Canonicalization of a function pointer usually involves computing
the address of the function which would be called if the function
pointer were used in an indirect call.

Only define this pattern if function pointers on the target machine
can have different values but still call the same function when
used in an indirect call.

@cindex @code{save_stack_block} instruction pattern
@cindex @code{save_stack_function} instruction pattern
@cindex @code{save_stack_nonlocal} instruction pattern
@cindex @code{restore_stack_block} instruction pattern
@cindex @code{restore_stack_function} instruction pattern
@cindex @code{restore_stack_nonlocal} instruction pattern
@item @samp{save_stack_block}
@itemx @samp{save_stack_function}
@itemx @samp{save_stack_nonlocal}
@itemx @samp{restore_stack_block}
@itemx @samp{restore_stack_function}
@itemx @samp{restore_stack_nonlocal}
Most machines save and restore the stack pointer by copying it to or
from an object of mode @code{Pmode}.  Do not define these patterns on
such machines.

Some machines require special handling for stack pointer saves and
restores.  On those machines, define the patterns corresponding to the
non-standard cases by using a @code{define_expand} (@pxref{Expander
Definitions}) that produces the required insns.  The three types of
saves and restores are:

@enumerate
@item
@samp{save_stack_block} saves the stack pointer at the start of a block
that allocates a variable-sized object, and @samp{restore_stack_block}
restores the stack pointer when the block is exited.

@item
@samp{save_stack_function} and @samp{restore_stack_function} do a
similar job for the outermost block of a function and are used when the
function allocates variable-sized objects or calls @code{alloca}.  Only
the epilogue uses the restored stack pointer, allowing a simpler save or
restore sequence on some machines.

@item
@samp{save_stack_nonlocal} is used in functions that contain labels
branched to by nested functions.  It saves the stack pointer in such a
way that the inner function can use @samp{restore_stack_nonlocal} to
restore the stack pointer.  The compiler generates code to restore the
frame and argument pointer registers, but some machines require saving
and restoring additional data such as register window information or
stack backchains.  Place insns in these patterns to save and restore any
such required data.
@end enumerate

When saving the stack pointer, operand 0 is the save area and operand 1
is the stack pointer.  The mode used to allocate the save area defaults
to @code{Pmode} but you can override that choice by defining the
@code{STACK_SAVEAREA_MODE} macro (@pxref{Storage Layout}).  You must
specify an integral mode, or @code{VOIDmode} if no save area is needed
for a particular type of save (either because no save is needed or
because a machine-specific save area can be used).  Operand 0 is the
stack pointer and operand 1 is the save area for restore operations.  If
@samp{save_stack_block} is defined, operand 0 must not be
@code{VOIDmode} since these saves can be arbitrarily nested.

A save area is a @code{mem} that is at a constant offset from
@code{virtual_stack_vars_rtx} when the stack pointer is saved for use by
nonlocal gotos and a @code{reg} in the other two cases.

@cindex @code{allocate_stack} instruction pattern
@item @samp{allocate_stack}
Subtract (or add if @code{STACK_GROWS_DOWNWARD} is undefined) operand 1 from
the stack pointer to create space for dynamically allocated data.

Store the resultant pointer to this space into operand 0.  If you
are allocating space from the main stack, do this by emitting a
move insn to copy @code{virtual_stack_dynamic_rtx} to operand 0.
If you are allocating the space elsewhere, generate code to copy the
location of the space to operand 0.  In the latter case, you must
ensure this space gets freed when the corresponding space on the main
stack is free.

Do not define this pattern if all that must be done is the subtraction.
Some machines require other operations such as stack probes or
maintaining the back chain.  Define this pattern to emit those
operations in addition to updating the stack pointer.

@cindex @code{check_stack} instruction pattern
@item @samp{check_stack}
If stack checking cannot be done on your system by probing the stack with
a load or store instruction (@pxref{Stack Checking}), define this pattern
to perform the needed check and signaling an error if the stack
has overflowed.  The single operand is the location in the stack furthest
from the current stack pointer that you need to validate.  Normally,
on machines where this pattern is needed, you would obtain the stack
limit from a global or thread-specific variable or register.

@cindex @code{nonlocal_goto} instruction pattern
@item @samp{nonlocal_goto}
Emit code to generate a non-local goto, e.g., a jump from one function
to a label in an outer function.  This pattern has four arguments,
each representing a value to be used in the jump.  The first
argument is to be loaded into the frame pointer, the second is
the address to branch to (code to dispatch to the actual label),
the third is the address of a location where the stack is saved,
and the last is the address of the label, to be placed in the
location for the incoming static chain.

On most machines you need not define this pattern, since GCC will
already generate the correct code, which is to load the frame pointer
and static chain, restore the stack (using the
@samp{restore_stack_nonlocal} pattern, if defined), and jump indirectly
to the dispatcher.  You need only define this pattern if this code will
not work on your machine.

@cindex @code{nonlocal_goto_receiver} instruction pattern
@item @samp{nonlocal_goto_receiver}
This pattern, if defined, contains code needed at the target of a
nonlocal goto after the code already generated by GCC@.  You will not
normally need to define this pattern.  A typical reason why you might
need this pattern is if some value, such as a pointer to a global table,
must be restored when the frame pointer is restored.  Note that a nonlocal
goto only occurs within a unit-of-translation, so a global table pointer
that is shared by all functions of a given module need not be restored.
There are no arguments.

@cindex @code{exception_receiver} instruction pattern
@item @samp{exception_receiver}
This pattern, if defined, contains code needed at the site of an
exception handler that isn't needed at the site of a nonlocal goto.  You
will not normally need to define this pattern.  A typical reason why you
might need this pattern is if some value, such as a pointer to a global
table, must be restored after control flow is branched to the handler of
an exception.  There are no arguments.

@cindex @code{builtin_setjmp_setup} instruction pattern
@item @samp{builtin_setjmp_setup}
This pattern, if defined, contains additional code needed to initialize
the @code{jmp_buf}.  You will not normally need to define this pattern.
A typical reason why you might need this pattern is if some value, such
as a pointer to a global table, must be restored.  Though it is
preferred that the pointer value be recalculated if possible (given the
address of a label for instance).  The single argument is a pointer to
the @code{jmp_buf}.  Note that the buffer is five words long and that
the first three are normally used by the generic mechanism.

@cindex @code{builtin_setjmp_receiver} instruction pattern
@item @samp{builtin_setjmp_receiver}
This pattern, if defined, contains code needed at the site of an
built-in setjmp that isn't needed at the site of a nonlocal goto.  You
will not normally need to define this pattern.  A typical reason why you
might need this pattern is if some value, such as a pointer to a global
table, must be restored.  It takes one argument, which is the label
to which builtin_longjmp transfered control; this pattern may be emitted
at a small offset from that label.

@cindex @code{builtin_longjmp} instruction pattern
@item @samp{builtin_longjmp}
This pattern, if defined, performs the entire action of the longjmp.
You will not normally need to define this pattern unless you also define
@code{builtin_setjmp_setup}.  The single argument is a pointer to the
@code{jmp_buf}.

@cindex @code{eh_return} instruction pattern
@item @samp{eh_return}
This pattern, if defined, affects the way @code{__builtin_eh_return},
and thence the call frame exception handling library routines, are
built.  It is intended to handle non-trivial actions needed along
the abnormal return path.

The address of the exception handler to which the function should return
is passed as operand to this pattern.  It will normally need to copied by
the pattern to some special register or memory location.
If the pattern needs to determine the location of the target call
frame in order to do so, it may use @code{EH_RETURN_STACKADJ_RTX},
if defined; it will have already been assigned.

If this pattern is not defined, the default action will be to simply
copy the return address to @code{EH_RETURN_HANDLER_RTX}.  Either
that macro or this pattern needs to be defined if call frame exception
handling is to be used.

@cindex @code{prologue} instruction pattern
@anchor{prologue instruction pattern}
@item @samp{prologue}
This pattern, if defined, emits RTL for entry to a function.  The function
entry is responsible for setting up the stack frame, initializing the frame
pointer register, saving callee saved registers, etc.

Using a prologue pattern is generally preferred over defining
@code{TARGET_ASM_FUNCTION_PROLOGUE} to emit assembly code for the prologue.

The @code{prologue} pattern is particularly useful for targets which perform
instruction scheduling.

@cindex @code{epilogue} instruction pattern
@anchor{epilogue instruction pattern}
@item @samp{epilogue}
This pattern emits RTL for exit from a function.  The function
exit is responsible for deallocating the stack frame, restoring callee saved
registers and emitting the return instruction.

Using an epilogue pattern is generally preferred over defining
@code{TARGET_ASM_FUNCTION_EPILOGUE} to emit assembly code for the epilogue.

The @code{epilogue} pattern is particularly useful for targets which perform
instruction scheduling or which have delay slots for their return instruction.

@cindex @code{sibcall_epilogue} instruction pattern
@item @samp{sibcall_epilogue}
This pattern, if defined, emits RTL for exit from a function without the final
branch back to the calling function.  This pattern will be emitted before any
sibling call (aka tail call) sites.

The @code{sibcall_epilogue} pattern must not clobber any arguments used for
parameter passing or any stack slots for arguments passed to the current
function.

@cindex @code{trap} instruction pattern
@item @samp{trap}
This pattern, if defined, signals an error, typically by causing some
kind of signal to be raised.  Among other places, it is used by the Java
front end to signal `invalid array index' exceptions.

@cindex @code{conditional_trap} instruction pattern
@item @samp{conditional_trap}
Conditional trap instruction.  Operand 0 is a piece of RTL which
performs a comparison.  Operand 1 is the trap code, an integer.

A typical @code{conditional_trap} pattern looks like

@smallexample
(define_insn "conditional_trap"
  [(trap_if (match_operator 0 "trap_operator"
             [(cc0) (const_int 0)])
            (match_operand 1 "const_int_operand" "i"))]
  ""
  "@dots{}")
@end smallexample

@cindex @code{prefetch} instruction pattern
@item @samp{prefetch}

This pattern, if defined, emits code for a non-faulting data prefetch
instruction.  Operand 0 is the address of the memory to prefetch.  Operand 1
is a constant 1 if the prefetch is preparing for a write to the memory
address, or a constant 0 otherwise.  Operand 2 is the expected degree of
temporal locality of the data and is a value between 0 and 3, inclusive; 0
means that the data has no temporal locality, so it need not be left in the
cache after the access; 3 means that the data has a high degree of temporal
locality and should be left in all levels of cache possible;  1 and 2 mean,
respectively, a low or moderate degree of temporal locality.

Targets that do not support write prefetches or locality hints can ignore
the values of operands 1 and 2.

@cindex @code{blockage} instruction pattern
@item @samp{blockage}

This pattern defines a pseudo insn that prevents the instruction
scheduler from moving instructions across the boundary defined by the
blockage insn.  Normally an UNSPEC_VOLATILE pattern.

@cindex @code{memory_barrier} instruction pattern
@item @samp{memory_barrier}

If the target memory model is not fully synchronous, then this pattern
should be defined to an instruction that orders both loads and stores
before the instruction with respect to loads and stores after the instruction.
This pattern has no operands.

@cindex @code{sync_compare_and_swap@var{mode}} instruction pattern
@item @samp{sync_compare_and_swap@var{mode}}

This pattern, if defined, emits code for an atomic compare-and-swap
operation.  Operand 1 is the memory on which the atomic operation is
performed.  Operand 2 is the ``old'' value to be compared against the
current contents of the memory location.  Operand 3 is the ``new'' value
to store in the memory if the compare succeeds.  Operand 0 is the result
of the operation; it should contain the contents of the memory
before the operation.  If the compare succeeds, this should obviously be
a copy of operand 2.

This pattern must show that both operand 0 and operand 1 are modified.

This pattern must issue any memory barrier instructions such that all
memory operations before the atomic operation occur before the atomic
operation and all memory operations after the atomic operation occur
after the atomic operation.

@cindex @code{sync_compare_and_swap_cc@var{mode}} instruction pattern
@item @samp{sync_compare_and_swap_cc@var{mode}}

This pattern is just like @code{sync_compare_and_swap@var{mode}}, except
it should act as if compare part of the compare-and-swap were issued via
@code{cmp@var{m}}.  This comparison will only be used with @code{EQ} and
@code{NE} branches and @code{setcc} operations.

Some targets do expose the success or failure of the compare-and-swap
operation via the status flags.  Ideally we wouldn't need a separate
named pattern in order to take advantage of this, but the combine pass
does not handle patterns with multiple sets, which is required by
definition for @code{sync_compare_and_swap@var{mode}}.

@cindex @code{sync_add@var{mode}} instruction pattern
@cindex @code{sync_sub@var{mode}} instruction pattern
@cindex @code{sync_ior@var{mode}} instruction pattern
@cindex @code{sync_and@var{mode}} instruction pattern
@cindex @code{sync_xor@var{mode}} instruction pattern
@cindex @code{sync_nand@var{mode}} instruction pattern
@item @samp{sync_add@var{mode}}, @samp{sync_sub@var{mode}}
@itemx @samp{sync_ior@var{mode}}, @samp{sync_and@var{mode}}
@itemx @samp{sync_xor@var{mode}}, @samp{sync_nand@var{mode}}

These patterns emit code for an atomic operation on memory.
Operand 0 is the memory on which the atomic operation is performed.
Operand 1 is the second operand to the binary operator.

The ``nand'' operation is @code{~op0 & op1}.

This pattern must issue any memory barrier instructions such that all
memory operations before the atomic operation occur before the atomic
operation and all memory operations after the atomic operation occur
after the atomic operation.

If these patterns are not defined, the operation will be constructed
from a compare-and-swap operation, if defined.

@cindex @code{sync_old_add@var{mode}} instruction pattern
@cindex @code{sync_old_sub@var{mode}} instruction pattern
@cindex @code{sync_old_ior@var{mode}} instruction pattern
@cindex @code{sync_old_and@var{mode}} instruction pattern
@cindex @code{sync_old_xor@var{mode}} instruction pattern
@cindex @code{sync_old_nand@var{mode}} instruction pattern
@item @samp{sync_old_add@var{mode}}, @samp{sync_old_sub@var{mode}}
@itemx @samp{sync_old_ior@var{mode}}, @samp{sync_old_and@var{mode}}
@itemx @samp{sync_old_xor@var{mode}}, @samp{sync_old_nand@var{mode}}

These patterns are emit code for an atomic operation on memory,
and return the value that the memory contained before the operation.
Operand 0 is the result value, operand 1 is the memory on which the
atomic operation is performed, and operand 2 is the second operand
to the binary operator.

This pattern must issue any memory barrier instructions such that all
memory operations before the atomic operation occur before the atomic
operation and all memory operations after the atomic operation occur
after the atomic operation.

If these patterns are not defined, the operation will be constructed
from a compare-and-swap operation, if defined.

@cindex @code{sync_new_add@var{mode}} instruction pattern
@cindex @code{sync_new_sub@var{mode}} instruction pattern
@cindex @code{sync_new_ior@var{mode}} instruction pattern
@cindex @code{sync_new_and@var{mode}} instruction pattern
@cindex @code{sync_new_xor@var{mode}} instruction pattern
@cindex @code{sync_new_nand@var{mode}} instruction pattern
@item @samp{sync_new_add@var{mode}}, @samp{sync_new_sub@var{mode}}
@itemx @samp{sync_new_ior@var{mode}}, @samp{sync_new_and@var{mode}}
@itemx @samp{sync_new_xor@var{mode}}, @samp{sync_new_nand@var{mode}}

These patterns are like their @code{sync_old_@var{op}} counterparts,
except that they return the value that exists in the memory location
after the operation, rather than before the operation.

@cindex @code{sync_lock_test_and_set@var{mode}} instruction pattern
@item @samp{sync_lock_test_and_set@var{mode}}

This pattern takes two forms, based on the capabilities of the target.
In either case, operand 0 is the result of the operand, operand 1 is
the memory on which the atomic operation is performed, and operand 2
is the value to set in the lock.

In the ideal case, this operation is an atomic exchange operation, in
which the previous value in memory operand is copied into the result
operand, and the value operand is stored in the memory operand.

For less capable targets, any value operand that is not the constant 1
should be rejected with @code{FAIL}.  In this case the target may use
an atomic test-and-set bit operation.  The result operand should contain
1 if the bit was previously set and 0 if the bit was previously clear.
The true contents of the memory operand are implementation defined.

This pattern must issue any memory barrier instructions such that the
pattern as a whole acts as an acquire barrier, that is all memory
operations after the pattern do not occur until the lock is acquired.

If this pattern is not defined, the operation will be constructed from
a compare-and-swap operation, if defined.

@cindex @code{sync_lock_release@var{mode}} instruction pattern
@item @samp{sync_lock_release@var{mode}}

This pattern, if defined, releases a lock set by
@code{sync_lock_test_and_set@var{mode}}.  Operand 0 is the memory
that contains the lock; operand 1 is the value to store in the lock.

If the target doesn't implement full semantics for
@code{sync_lock_test_and_set@var{mode}}, any value operand which is not
the constant 0 should be rejected with @code{FAIL}, and the true contents
of the memory operand are implementation defined.

This pattern must issue any memory barrier instructions such that the
pattern as a whole acts as a release barrier, that is the lock is
released only after all previous memory operations have completed.

If this pattern is not defined, then a @code{memory_barrier} pattern
will be emitted, followed by a store of the value to the memory operand.

@cindex @code{stack_protect_set} instruction pattern
@item @samp{stack_protect_set}

This pattern, if defined, moves a @code{Pmode} value from the memory
in operand 1 to the memory in operand 0 without leaving the value in
a register afterward.  This is to avoid leaking the value some place
that an attacker might use to rewrite the stack guard slot after
having clobbered it.

If this pattern is not defined, then a plain move pattern is generated.

@cindex @code{stack_protect_test} instruction pattern
@item @samp{stack_protect_test}

This pattern, if defined, compares a @code{Pmode} value from the
memory in operand 1 with the memory in operand 0 without leaving the
value in a register afterward and branches to operand 2 if the values
weren't equal.

If this pattern is not defined, then a plain compare pattern and
conditional branch pattern is used.

@end table

@end ifset
@c Each of the following nodes are wrapped in separate
@c "@ifset INTERNALS" to work around memory limits for the default
@c configuration in older tetex distributions.  Known to not work:
@c tetex-1.0.7, known to work: tetex-2.0.2.
@ifset INTERNALS
@node Pattern Ordering
@section When the Order of Patterns Matters
@cindex Pattern Ordering
@cindex Ordering of Patterns

Sometimes an insn can match more than one instruction pattern.  Then the
pattern that appears first in the machine description is the one used.
Therefore, more specific patterns (patterns that will match fewer things)
and faster instructions (those that will produce better code when they
do match) should usually go first in the description.

In some cases the effect of ordering the patterns can be used to hide
a pattern when it is not valid.  For example, the 68000 has an
instruction for converting a fullword to floating point and another
for converting a byte to floating point.  An instruction converting
an integer to floating point could match either one.  We put the
pattern to convert the fullword first to make sure that one will
be used rather than the other.  (Otherwise a large integer might
be generated as a single-byte immediate quantity, which would not work.)
Instead of using this pattern ordering it would be possible to make the
pattern for convert-a-byte smart enough to deal properly with any
constant value.

@end ifset
@ifset INTERNALS
@node Dependent Patterns
@section Interdependence of Patterns
@cindex Dependent Patterns
@cindex Interdependence of Patterns

Every machine description must have a named pattern for each of the
conditional branch names @samp{b@var{cond}}.  The recognition template
must always have the form

@smallexample
(set (pc)
     (if_then_else (@var{cond} (cc0) (const_int 0))
                   (label_ref (match_operand 0 "" ""))
                   (pc)))
@end smallexample

@noindent
In addition, every machine description must have an anonymous pattern
for each of the possible reverse-conditional branches.  Their templates
look like

@smallexample
(set (pc)
     (if_then_else (@var{cond} (cc0) (const_int 0))
                   (pc)
                   (label_ref (match_operand 0 "" ""))))
@end smallexample

@noindent
They are necessary because jump optimization can turn direct-conditional
branches into reverse-conditional branches.

It is often convenient to use the @code{match_operator} construct to
reduce the number of patterns that must be specified for branches.  For
example,

@smallexample
(define_insn ""
  [(set (pc)
        (if_then_else (match_operator 0 "comparison_operator"
                                      [(cc0) (const_int 0)])
                      (pc)
                      (label_ref (match_operand 1 "" ""))))]
  "@var{condition}"
  "@dots{}")
@end smallexample

In some cases machines support instructions identical except for the
machine mode of one or more operands.  For example, there may be
``sign-extend halfword'' and ``sign-extend byte'' instructions whose
patterns are

@smallexample
(set (match_operand:SI 0 @dots{})
     (extend:SI (match_operand:HI 1 @dots{})))

(set (match_operand:SI 0 @dots{})
     (extend:SI (match_operand:QI 1 @dots{})))
@end smallexample

@noindent
Constant integers do not specify a machine mode, so an instruction to
extend a constant value could match either pattern.  The pattern it
actually will match is the one that appears first in the file.  For correct
results, this must be the one for the widest possible mode (@code{HImode},
here).  If the pattern matches the @code{QImode} instruction, the results
will be incorrect if the constant value does not actually fit that mode.

Such instructions to extend constants are rarely generated because they are
optimized away, but they do occasionally happen in nonoptimized
compilations.

If a constraint in a pattern allows a constant, the reload pass may
replace a register with a constant permitted by the constraint in some
cases.  Similarly for memory references.  Because of this substitution,
you should not provide separate patterns for increment and decrement
instructions.  Instead, they should be generated from the same pattern
that supports register-register add insns by examining the operands and
generating the appropriate machine instruction.

@end ifset
@ifset INTERNALS
@node Jump Patterns
@section Defining Jump Instruction Patterns
@cindex jump instruction patterns
@cindex defining jump instruction patterns

For most machines, GCC assumes that the machine has a condition code.
A comparison insn sets the condition code, recording the results of both
signed and unsigned comparison of the given operands.  A separate branch
insn tests the condition code and branches or not according its value.
The branch insns come in distinct signed and unsigned flavors.  Many
common machines, such as the VAX, the 68000 and the 32000, work this
way.

Some machines have distinct signed and unsigned compare instructions, and
only one set of conditional branch instructions.  The easiest way to handle
these machines is to treat them just like the others until the final stage
where assembly code is written.  At this time, when outputting code for the
compare instruction, peek ahead at the following branch using
@code{next_cc0_user (insn)}.  (The variable @code{insn} refers to the insn
being output, in the output-writing code in an instruction pattern.)  If
the RTL says that is an unsigned branch, output an unsigned compare;
otherwise output a signed compare.  When the branch itself is output, you
can treat signed and unsigned branches identically.

The reason you can do this is that GCC always generates a pair of
consecutive RTL insns, possibly separated by @code{note} insns, one to
set the condition code and one to test it, and keeps the pair inviolate
until the end.

To go with this technique, you must define the machine-description macro
@code{NOTICE_UPDATE_CC} to do @code{CC_STATUS_INIT}; in other words, no
compare instruction is superfluous.

Some machines have compare-and-branch instructions and no condition code.
A similar technique works for them.  When it is time to ``output'' a
compare instruction, record its operands in two static variables.  When
outputting the branch-on-condition-code instruction that follows, actually
output a compare-and-branch instruction that uses the remembered operands.

It also works to define patterns for compare-and-branch instructions.
In optimizing compilation, the pair of compare and branch instructions
will be combined according to these patterns.  But this does not happen
if optimization is not requested.  So you must use one of the solutions
above in addition to any special patterns you define.

In many RISC machines, most instructions do not affect the condition
code and there may not even be a separate condition code register.  On
these machines, the restriction that the definition and use of the
condition code be adjacent insns is not necessary and can prevent
important optimizations.  For example, on the IBM RS/6000, there is a
delay for taken branches unless the condition code register is set three
instructions earlier than the conditional branch.  The instruction
scheduler cannot perform this optimization if it is not permitted to
separate the definition and use of the condition code register.

On these machines, do not use @code{(cc0)}, but instead use a register
to represent the condition code.  If there is a specific condition code
register in the machine, use a hard register.  If the condition code or
comparison result can be placed in any general register, or if there are
multiple condition registers, use a pseudo register.

@findex prev_cc0_setter
@findex next_cc0_user
On some machines, the type of branch instruction generated may depend on
the way the condition code was produced; for example, on the 68k and
SPARC, setting the condition code directly from an add or subtract
instruction does not clear the overflow bit the way that a test
instruction does, so a different branch instruction must be used for
some conditional branches.  For machines that use @code{(cc0)}, the set
and use of the condition code must be adjacent (separated only by
@code{note} insns) allowing flags in @code{cc_status} to be used.
(@xref{Condition Code}.)  Also, the comparison and branch insns can be
located from each other by using the functions @code{prev_cc0_setter}
and @code{next_cc0_user}.

However, this is not true on machines that do not use @code{(cc0)}.  On
those machines, no assumptions can be made about the adjacency of the
compare and branch insns and the above methods cannot be used.  Instead,
we use the machine mode of the condition code register to record
different formats of the condition code register.

Registers used to store the condition code value should have a mode that
is in class @code{MODE_CC}.  Normally, it will be @code{CCmode}.  If
additional modes are required (as for the add example mentioned above in
the SPARC), define them in @file{@var{machine}-modes.def}
(@pxref{Condition Code}).  Also define @code{SELECT_CC_MODE} to choose
a mode given an operand of a compare.

If it is known during RTL generation that a different mode will be
required (for example, if the machine has separate compare instructions
for signed and unsigned quantities, like most IBM processors), they can
be specified at that time.

If the cases that require different modes would be made by instruction
combination, the macro @code{SELECT_CC_MODE} determines which machine
mode should be used for the comparison result.  The patterns should be
written using that mode.  To support the case of the add on the SPARC
discussed above, we have the pattern

@smallexample
(define_insn ""
  [(set (reg:CC_NOOV 0)
        (compare:CC_NOOV
          (plus:SI (match_operand:SI 0 "register_operand" "%r")
                   (match_operand:SI 1 "arith_operand" "rI"))
          (const_int 0)))]
  ""
  "@dots{}")
@end smallexample

The @code{SELECT_CC_MODE} macro on the SPARC returns @code{CC_NOOVmode}
for comparisons whose argument is a @code{plus}.

@end ifset
@ifset INTERNALS
@node Looping Patterns
@section Defining Looping Instruction Patterns
@cindex looping instruction patterns
@cindex defining looping instruction patterns

Some machines have special jump instructions that can be utilized to
make loops more efficient.  A common example is the 68000 @samp{dbra}
instruction which performs a decrement of a register and a branch if the
result was greater than zero.  Other machines, in particular digital
signal processors (DSPs), have special block repeat instructions to
provide low-overhead loop support.  For example, the TI TMS320C3x/C4x
DSPs have a block repeat instruction that loads special registers to
mark the top and end of a loop and to count the number of loop
iterations.  This avoids the need for fetching and executing a
@samp{dbra}-like instruction and avoids pipeline stalls associated with
the jump.

GCC has three special named patterns to support low overhead looping.
They are @samp{decrement_and_branch_until_zero}, @samp{doloop_begin},
and @samp{doloop_end}.  The first pattern,
@samp{decrement_and_branch_until_zero}, is not emitted during RTL
generation but may be emitted during the instruction combination phase.
This requires the assistance of the loop optimizer, using information
collected during strength reduction, to reverse a loop to count down to
zero.  Some targets also require the loop optimizer to add a
@code{REG_NONNEG} note to indicate that the iteration count is always
positive.  This is needed if the target performs a signed loop
termination test.  For example, the 68000 uses a pattern similar to the
following for its @code{dbra} instruction:

@smallexample
@group
(define_insn "decrement_and_branch_until_zero"
  [(set (pc)
        (if_then_else
          (ge (plus:SI (match_operand:SI 0 "general_operand" "+d*am")
                       (const_int -1))
              (const_int 0))
          (label_ref (match_operand 1 "" ""))
          (pc)))
   (set (match_dup 0)
        (plus:SI (match_dup 0)
                 (const_int -1)))]
  "find_reg_note (insn, REG_NONNEG, 0)"
  "@dots{}")
@end group
@end smallexample

Note that since the insn is both a jump insn and has an output, it must
deal with its own reloads, hence the `m' constraints.  Also note that
since this insn is generated by the instruction combination phase
combining two sequential insns together into an implicit parallel insn,
the iteration counter needs to be biased by the same amount as the
decrement operation, in this case @minus{}1.  Note that the following similar
pattern will not be matched by the combiner.

@smallexample
@group
(define_insn "decrement_and_branch_until_zero"
  [(set (pc)
        (if_then_else
          (ge (match_operand:SI 0 "general_operand" "+d*am")
              (const_int 1))
          (label_ref (match_operand 1 "" ""))
          (pc)))
   (set (match_dup 0)
        (plus:SI (match_dup 0)
                 (const_int -1)))]
  "find_reg_note (insn, REG_NONNEG, 0)"
  "@dots{}")
@end group
@end smallexample

The other two special looping patterns, @samp{doloop_begin} and
@samp{doloop_end}, are emitted by the loop optimizer for certain
well-behaved loops with a finite number of loop iterations using
information collected during strength reduction.

The @samp{doloop_end} pattern describes the actual looping instruction
(or the implicit looping operation) and the @samp{doloop_begin} pattern
is an optional companion pattern that can be used for initialization
needed for some low-overhead looping instructions.

Note that some machines require the actual looping instruction to be
emitted at the top of the loop (e.g., the TMS320C3x/C4x DSPs).  Emitting
the true RTL for a looping instruction at the top of the loop can cause
problems with flow analysis.  So instead, a dummy @code{doloop} insn is
emitted at the end of the loop.  The machine dependent reorg pass checks
for the presence of this @code{doloop} insn and then searches back to
the top of the loop, where it inserts the true looping insn (provided
there are no instructions in the loop which would cause problems).  Any
additional labels can be emitted at this point.  In addition, if the
desired special iteration counter register was not allocated, this
machine dependent reorg pass could emit a traditional compare and jump
instruction pair.

The essential difference between the
@samp{decrement_and_branch_until_zero} and the @samp{doloop_end}
patterns is that the loop optimizer allocates an additional pseudo
register for the latter as an iteration counter.  This pseudo register
cannot be used within the loop (i.e., general induction variables cannot
be derived from it), however, in many cases the loop induction variable
may become redundant and removed by the flow pass.


@end ifset
@ifset INTERNALS
@node Insn Canonicalizations
@section Canonicalization of Instructions
@cindex canonicalization of instructions
@cindex insn canonicalization

There are often cases where multiple RTL expressions could represent an
operation performed by a single machine instruction.  This situation is
most commonly encountered with logical, branch, and multiply-accumulate
instructions.  In such cases, the compiler attempts to convert these
multiple RTL expressions into a single canonical form to reduce the
number of insn patterns required.

In addition to algebraic simplifications, following canonicalizations
are performed:

@itemize @bullet
@item
For commutative and comparison operators, a constant is always made the
second operand.  If a machine only supports a constant as the second
operand, only patterns that match a constant in the second operand need
be supplied.

@item
For associative operators, a sequence of operators will always chain
to the left; for instance, only the left operand of an integer @code{plus}
can itself be a @code{plus}.  @code{and}, @code{ior}, @code{xor},
@code{plus}, @code{mult}, @code{smin}, @code{smax}, @code{umin}, and
@code{umax} are associative when applied to integers, and sometimes to
floating-point.

@item
@cindex @code{neg}, canonicalization of
@cindex @code{not}, canonicalization of
@cindex @code{mult}, canonicalization of
@cindex @code{plus}, canonicalization of
@cindex @code{minus}, canonicalization of
For these operators, if only one operand is a @code{neg}, @code{not},
@code{mult}, @code{plus}, or @code{minus} expression, it will be the
first operand.

@item
In combinations of @code{neg}, @code{mult}, @code{plus}, and
@code{minus}, the @code{neg} operations (if any) will be moved inside
the operations as far as possible.  For instance,
@code{(neg (mult A B))} is canonicalized as @code{(mult (neg A) B)}, but
@code{(plus (mult (neg A) B) C)} is canonicalized as
@code{(minus A (mult B C))}.

@cindex @code{compare}, canonicalization of
@item
For the @code{compare} operator, a constant is always the second operand
on machines where @code{cc0} is used (@pxref{Jump Patterns}).  On other
machines, there are rare cases where the compiler might want to construct
a @code{compare} with a constant as the first operand.  However, these
cases are not common enough for it to be worthwhile to provide a pattern
matching a constant as the first operand unless the machine actually has
such an instruction.

An operand of @code{neg}, @code{not}, @code{mult}, @code{plus}, or
@code{minus} is made the first operand under the same conditions as
above.

@item
@code{(minus @var{x} (const_int @var{n}))} is converted to
@code{(plus @var{x} (const_int @var{-n}))}.

@item
Within address computations (i.e., inside @code{mem}), a left shift is
converted into the appropriate multiplication by a power of two.

@cindex @code{ior}, canonicalization of
@cindex @code{and}, canonicalization of
@cindex De Morgan's law
@item
De Morgan's Law is used to move bitwise negation inside a bitwise
logical-and or logical-or operation.  If this results in only one
operand being a @code{not} expression, it will be the first one.

A machine that has an instruction that performs a bitwise logical-and of one
operand with the bitwise negation of the other should specify the pattern
for that instruction as

@smallexample
(define_insn ""
  [(set (match_operand:@var{m} 0 @dots{})
        (and:@var{m} (not:@var{m} (match_operand:@var{m} 1 @dots{}))
                     (match_operand:@var{m} 2 @dots{})))]
  "@dots{}"
  "@dots{}")
@end smallexample

@noindent
Similarly, a pattern for a ``NAND'' instruction should be written

@smallexample
(define_insn ""
  [(set (match_operand:@var{m} 0 @dots{})
        (ior:@var{m} (not:@var{m} (match_operand:@var{m} 1 @dots{}))
                     (not:@var{m} (match_operand:@var{m} 2 @dots{}))))]
  "@dots{}"
  "@dots{}")
@end smallexample

In both cases, it is not necessary to include patterns for the many
logically equivalent RTL expressions.

@cindex @code{xor}, canonicalization of
@item
The only possible RTL expressions involving both bitwise exclusive-or
and bitwise negation are @code{(xor:@var{m} @var{x} @var{y})}
and @code{(not:@var{m} (xor:@var{m} @var{x} @var{y}))}.

@item
The sum of three items, one of which is a constant, will only appear in
the form

@smallexample
(plus:@var{m} (plus:@var{m} @var{x} @var{y}) @var{constant})
@end smallexample

@item
On machines that do not use @code{cc0},
@code{(compare @var{x} (const_int 0))} will be converted to
@var{x}.

@cindex @code{zero_extract}, canonicalization of
@cindex @code{sign_extract}, canonicalization of
@item
Equality comparisons of a group of bits (usually a single bit) with zero
will be written using @code{zero_extract} rather than the equivalent
@code{and} or @code{sign_extract} operations.

@end itemize

Further canonicalization rules are defined in the function
@code{commutative_operand_precedence} in @file{gcc/rtlanal.c}.

@end ifset
@ifset INTERNALS
@node Expander Definitions
@section Defining RTL Sequences for Code Generation
@cindex expander definitions
@cindex code generation RTL sequences
@cindex defining RTL sequences for code generation

On some target machines, some standard pattern names for RTL generation
cannot be handled with single insn, but a sequence of RTL insns can
represent them.  For these target machines, you can write a
@code{define_expand} to specify how to generate the sequence of RTL@.

@findex define_expand
A @code{define_expand} is an RTL expression that looks almost like a
@code{define_insn}; but, unlike the latter, a @code{define_expand} is used
only for RTL generation and it can produce more than one RTL insn.

A @code{define_expand} RTX has four operands:

@itemize @bullet
@item
The name.  Each @code{define_expand} must have a name, since the only
use for it is to refer to it by name.

@item
The RTL template.  This is a vector of RTL expressions representing
a sequence of separate instructions.  Unlike @code{define_insn}, there
is no implicit surrounding @code{PARALLEL}.

@item
The condition, a string containing a C expression.  This expression is
used to express how the availability of this pattern depends on
subclasses of target machine, selected by command-line options when GCC
is run.  This is just like the condition of a @code{define_insn} that
has a standard name.  Therefore, the condition (if present) may not
depend on the data in the insn being matched, but only the
target-machine-type flags.  The compiler needs to test these conditions
during initialization in order to learn exactly which named instructions
are available in a particular run.

@item
The preparation statements, a string containing zero or more C
statements which are to be executed before RTL code is generated from
the RTL template.

Usually these statements prepare temporary registers for use as
internal operands in the RTL template, but they can also generate RTL
insns directly by calling routines such as @code{emit_insn}, etc.
Any such insns precede the ones that come from the RTL template.
@end itemize

Every RTL insn emitted by a @code{define_expand} must match some
@code{define_insn} in the machine description.  Otherwise, the compiler
will crash when trying to generate code for the insn or trying to optimize
it.

The RTL template, in addition to controlling generation of RTL insns,
also describes the operands that need to be specified when this pattern
is used.  In particular, it gives a predicate for each operand.

A true operand, which needs to be specified in order to generate RTL from
the pattern, should be described with a @code{match_operand} in its first
occurrence in the RTL template.  This enters information on the operand's
predicate into the tables that record such things.  GCC uses the
information to preload the operand into a register if that is required for
valid RTL code.  If the operand is referred to more than once, subsequent
references should use @code{match_dup}.

The RTL template may also refer to internal ``operands'' which are
temporary registers or labels used only within the sequence made by the
@code{define_expand}.  Internal operands are substituted into the RTL
template with @code{match_dup}, never with @code{match_operand}.  The
values of the internal operands are not passed in as arguments by the
compiler when it requests use of this pattern.  Instead, they are computed
within the pattern, in the preparation statements.  These statements
compute the values and store them into the appropriate elements of
@code{operands} so that @code{match_dup} can find them.

There are two special macros defined for use in the preparation statements:
@code{DONE} and @code{FAIL}.  Use them with a following semicolon,
as a statement.

@table @code

@findex DONE
@item DONE
Use the @code{DONE} macro to end RTL generation for the pattern.  The
only RTL insns resulting from the pattern on this occasion will be
those already emitted by explicit calls to @code{emit_insn} within the
preparation statements; the RTL template will not be generated.

@findex FAIL
@item FAIL
Make the pattern fail on this occasion.  When a pattern fails, it means
that the pattern was not truly available.  The calling routines in the
compiler will try other strategies for code generation using other patterns.

Failure is currently supported only for binary (addition, multiplication,
shifting, etc.) and bit-field (@code{extv}, @code{extzv}, and @code{insv})
operations.
@end table

If the preparation falls through (invokes neither @code{DONE} nor
@code{FAIL}), then the @code{define_expand} acts like a
@code{define_insn} in that the RTL template is used to generate the
insn.

The RTL template is not used for matching, only for generating the
initial insn list.  If the preparation statement always invokes
@code{DONE} or @code{FAIL}, the RTL template may be reduced to a simple
list of operands, such as this example:

@smallexample
@group
(define_expand "addsi3"
  [(match_operand:SI 0 "register_operand" "")
   (match_operand:SI 1 "register_operand" "")
   (match_operand:SI 2 "register_operand" "")]
@end group
@group
  ""
  "
@{
  handle_add (operands[0], operands[1], operands[2]);
  DONE;
@}")
@end group
@end smallexample

Here is an example, the definition of left-shift for the SPUR chip:

@smallexample
@group
(define_expand "ashlsi3"
  [(set (match_operand:SI 0 "register_operand" "")
        (ashift:SI
@end group
@group
          (match_operand:SI 1 "register_operand" "")
          (match_operand:SI 2 "nonmemory_operand" "")))]
  ""
  "
@end group
@end smallexample

@smallexample
@group
@{
  if (GET_CODE (operands[2]) != CONST_INT
      || (unsigned) INTVAL (operands[2]) > 3)
    FAIL;
@}")
@end group
@end smallexample

@noindent
This example uses @code{define_expand} so that it can generate an RTL insn
for shifting when the shift-count is in the supported range of 0 to 3 but
fail in other cases where machine insns aren't available.  When it fails,
the compiler tries another strategy using different patterns (such as, a
library call).

If the compiler were able to handle nontrivial condition-strings in
patterns with names, then it would be possible to use a
@code{define_insn} in that case.  Here is another case (zero-extension
on the 68000) which makes more use of the power of @code{define_expand}:

@smallexample
(define_expand "zero_extendhisi2"
  [(set (match_operand:SI 0 "general_operand" "")
        (const_int 0))
   (set (strict_low_part
          (subreg:HI
            (match_dup 0)
            0))
        (match_operand:HI 1 "general_operand" ""))]
  ""
  "operands[1] = make_safe_from (operands[1], operands[0]);")
@end smallexample

@noindent
@findex make_safe_from
Here two RTL insns are generated, one to clear the entire output operand
and the other to copy the input operand into its low half.  This sequence
is incorrect if the input operand refers to [the old value of] the output
operand, so the preparation statement makes sure this isn't so.  The
function @code{make_safe_from} copies the @code{operands[1]} into a
temporary register if it refers to @code{operands[0]}.  It does this
by emitting another RTL insn.

Finally, a third example shows the use of an internal operand.
Zero-extension on the SPUR chip is done by @code{and}-ing the result
against a halfword mask.  But this mask cannot be represented by a
@code{const_int} because the constant value is too large to be legitimate
on this machine.  So it must be copied into a register with
@code{force_reg} and then the register used in the @code{and}.

@smallexample
(define_expand "zero_extendhisi2"
  [(set (match_operand:SI 0 "register_operand" "")
        (and:SI (subreg:SI
                  (match_operand:HI 1 "register_operand" "")
                  0)
                (match_dup 2)))]
  ""
  "operands[2]
     = force_reg (SImode, GEN_INT (65535)); ")
@end smallexample

@emph{Note:} If the @code{define_expand} is used to serve a
standard binary or unary arithmetic operation or a bit-field operation,
then the last insn it generates must not be a @code{code_label},
@code{barrier} or @code{note}.  It must be an @code{insn},
@code{jump_insn} or @code{call_insn}.  If you don't need a real insn
at the end, emit an insn to copy the result of the operation into
itself.  Such an insn will generate no code, but it can avoid problems
in the compiler.

@end ifset
@ifset INTERNALS
@node Insn Splitting
@section Defining How to Split Instructions
@cindex insn splitting
@cindex instruction splitting
@cindex splitting instructions

There are two cases where you should specify how to split a pattern
into multiple insns.  On machines that have instructions requiring
delay slots (@pxref{Delay Slots}) or that have instructions whose
output is not available for multiple cycles (@pxref{Processor pipeline
description}), the compiler phases that optimize these cases need to
be able to move insns into one-instruction delay slots.  However, some
insns may generate more than one machine instruction.  These insns
cannot be placed into a delay slot.

Often you can rewrite the single insn as a list of individual insns,
each corresponding to one machine instruction.  The disadvantage of
doing so is that it will cause the compilation to be slower and require
more space.  If the resulting insns are too complex, it may also
suppress some optimizations.  The compiler splits the insn if there is a
reason to believe that it might improve instruction or delay slot
scheduling.

The insn combiner phase also splits putative insns.  If three insns are
merged into one insn with a complex expression that cannot be matched by
some @code{define_insn} pattern, the combiner phase attempts to split
the complex pattern into two insns that are recognized.  Usually it can
break the complex pattern into two patterns by splitting out some
subexpression.  However, in some other cases, such as performing an
addition of a large constant in two insns on a RISC machine, the way to
split the addition into two insns is machine-dependent.

@findex define_split
The @code{define_split} definition tells the compiler how to split a
complex insn into several simpler insns.  It looks like this:

@smallexample
(define_split
  [@var{insn-pattern}]
  "@var{condition}"
  [@var{new-insn-pattern-1}
   @var{new-insn-pattern-2}
   @dots{}]
  "@var{preparation-statements}")
@end smallexample

@var{insn-pattern} is a pattern that needs to be split and
@var{condition} is the final condition to be tested, as in a
@code{define_insn}.  When an insn matching @var{insn-pattern} and
satisfying @var{condition} is found, it is replaced in the insn list
with the insns given by @var{new-insn-pattern-1},
@var{new-insn-pattern-2}, etc.

The @var{preparation-statements} are similar to those statements that
are specified for @code{define_expand} (@pxref{Expander Definitions})
and are executed before the new RTL is generated to prepare for the
generated code or emit some insns whose pattern is not fixed.  Unlike
those in @code{define_expand}, however, these statements must not
generate any new pseudo-registers.  Once reload has completed, they also
must not allocate any space in the stack frame.

Patterns are matched against @var{insn-pattern} in two different
circumstances.  If an insn needs to be split for delay slot scheduling
or insn scheduling, the insn is already known to be valid, which means
that it must have been matched by some @code{define_insn} and, if
@code{reload_completed} is nonzero, is known to satisfy the constraints
of that @code{define_insn}.  In that case, the new insn patterns must
also be insns that are matched by some @code{define_insn} and, if
@code{reload_completed} is nonzero, must also satisfy the constraints
of those definitions.

As an example of this usage of @code{define_split}, consider the following
example from @file{a29k.md}, which splits a @code{sign_extend} from
@code{HImode} to @code{SImode} into a pair of shift insns:

@smallexample
(define_split
  [(set (match_operand:SI 0 "gen_reg_operand" "")
        (sign_extend:SI (match_operand:HI 1 "gen_reg_operand" "")))]
  ""
  [(set (match_dup 0)
        (ashift:SI (match_dup 1)
                   (const_int 16)))
   (set (match_dup 0)
        (ashiftrt:SI (match_dup 0)
                     (const_int 16)))]
  "
@{ operands[1] = gen_lowpart (SImode, operands[1]); @}")
@end smallexample

When the combiner phase tries to split an insn pattern, it is always the
case that the pattern is @emph{not} matched by any @code{define_insn}.
The combiner pass first tries to split a single @code{set} expression
and then the same @code{set} expression inside a @code{parallel}, but
followed by a @code{clobber} of a pseudo-reg to use as a scratch
register.  In these cases, the combiner expects exactly two new insn
patterns to be generated.  It will verify that these patterns match some
@code{define_insn} definitions, so you need not do this test in the
@code{define_split} (of course, there is no point in writing a
@code{define_split} that will never produce insns that match).

Here is an example of this use of @code{define_split}, taken from
@file{rs6000.md}:

@smallexample
(define_split
  [(set (match_operand:SI 0 "gen_reg_operand" "")
        (plus:SI (match_operand:SI 1 "gen_reg_operand" "")
                 (match_operand:SI 2 "non_add_cint_operand" "")))]
  ""
  [(set (match_dup 0) (plus:SI (match_dup 1) (match_dup 3)))
   (set (match_dup 0) (plus:SI (match_dup 0) (match_dup 4)))]
"
@{
  int low = INTVAL (operands[2]) & 0xffff;
  int high = (unsigned) INTVAL (operands[2]) >> 16;

  if (low & 0x8000)
    high++, low |= 0xffff0000;

  operands[3] = GEN_INT (high << 16);
  operands[4] = GEN_INT (low);
@}")
@end smallexample

Here the predicate @code{non_add_cint_operand} matches any
@code{const_int} that is @emph{not} a valid operand of a single add
insn.  The add with the smaller displacement is written so that it
can be substituted into the address of a subsequent operation.

An example that uses a scratch register, from the same file, generates
an equality comparison of a register and a large constant:

@smallexample
(define_split
  [(set (match_operand:CC 0 "cc_reg_operand" "")
        (compare:CC (match_operand:SI 1 "gen_reg_operand" "")
                    (match_operand:SI 2 "non_short_cint_operand" "")))
   (clobber (match_operand:SI 3 "gen_reg_operand" ""))]
  "find_single_use (operands[0], insn, 0)
   && (GET_CODE (*find_single_use (operands[0], insn, 0)) == EQ
       || GET_CODE (*find_single_use (operands[0], insn, 0)) == NE)"
  [(set (match_dup 3) (xor:SI (match_dup 1) (match_dup 4)))
   (set (match_dup 0) (compare:CC (match_dup 3) (match_dup 5)))]
  "
@{
  /* @r{Get the constant we are comparing against, C, and see what it
     looks like sign-extended to 16 bits.  Then see what constant
     could be XOR'ed with C to get the sign-extended value.}  */

  int c = INTVAL (operands[2]);
  int sextc = (c << 16) >> 16;
  int xorv = c ^ sextc;

  operands[4] = GEN_INT (xorv);
  operands[5] = GEN_INT (sextc);
@}")
@end smallexample

To avoid confusion, don't write a single @code{define_split} that
accepts some insns that match some @code{define_insn} as well as some
insns that don't.  Instead, write two separate @code{define_split}
definitions, one for the insns that are valid and one for the insns that
are not valid.

The splitter is allowed to split jump instructions into sequence of
jumps or create new jumps in while splitting non-jump instructions.  As
the central flowgraph and branch prediction information needs to be updated,
several restriction apply.

Splitting of jump instruction into sequence that over by another jump
instruction is always valid, as compiler expect identical behavior of new
jump.  When new sequence contains multiple jump instructions or new labels,
more assistance is needed.  Splitter is required to create only unconditional
jumps, or simple conditional jump instructions.  Additionally it must attach a
@code{REG_BR_PROB} note to each conditional jump.  A global variable
@code{split_branch_probability} holds the probability of the original branch in case
it was an simple conditional jump, @minus{}1 otherwise.  To simplify
recomputing of edge frequencies, the new sequence is required to have only
forward jumps to the newly created labels.

@findex define_insn_and_split
For the common case where the pattern of a define_split exactly matches the
pattern of a define_insn, use @code{define_insn_and_split}.  It looks like
this:

@smallexample
(define_insn_and_split
  [@var{insn-pattern}]
  "@var{condition}"
  "@var{output-template}"
  "@var{split-condition}"
  [@var{new-insn-pattern-1}
   @var{new-insn-pattern-2}
   @dots{}]
  "@var{preparation-statements}"
  [@var{insn-attributes}])

@end smallexample

@var{insn-pattern}, @var{condition}, @var{output-template}, and
@var{insn-attributes} are used as in @code{define_insn}.  The
@var{new-insn-pattern} vector and the @var{preparation-statements} are used as
in a @code{define_split}.  The @var{split-condition} is also used as in
@code{define_split}, with the additional behavior that if the condition starts
with @samp{&&}, the condition used for the split will be the constructed as a
logical ``and'' of the split condition with the insn condition.  For example,
from i386.md:

@smallexample
(define_insn_and_split "zero_extendhisi2_and"
  [(set (match_operand:SI 0 "register_operand" "=r")
     (zero_extend:SI (match_operand:HI 1 "register_operand" "0")))
   (clobber (reg:CC 17))]
  "TARGET_ZERO_EXTEND_WITH_AND && !optimize_size"
  "#"
  "&& reload_completed"
  [(parallel [(set (match_dup 0)
                   (and:SI (match_dup 0) (const_int 65535)))
              (clobber (reg:CC 17))])]
  ""
  [(set_attr "type" "alu1")])

@end smallexample

In this case, the actual split condition will be
@samp{TARGET_ZERO_EXTEND_WITH_AND && !optimize_size && reload_completed}.

The @code{define_insn_and_split} construction provides exactly the same
functionality as two separate @code{define_insn} and @code{define_split}
patterns.  It exists for compactness, and as a maintenance tool to prevent
having to ensure the two patterns' templates match.

@end ifset
@ifset INTERNALS
@node Including Patterns
@section Including Patterns in Machine Descriptions.
@cindex insn includes

@findex include
The @code{include} pattern tells the compiler tools where to
look for patterns that are in files other than in the file
@file{.md}.  This is used only at build time and there is no preprocessing allowed.

It looks like:

@smallexample

(include
  @var{pathname})
@end smallexample

For example:

@smallexample

(include "filestuff")

@end smallexample

Where @var{pathname} is a string that specifies the location of the file,
specifies the include file to be in @file{gcc/config/target/filestuff}.  The
directory @file{gcc/config/target} is regarded as the default directory.


Machine descriptions may be split up into smaller more manageable subsections
and placed into subdirectories.

By specifying:

@smallexample

(include "BOGUS/filestuff")

@end smallexample

the include file is specified to be in @file{gcc/config/@var{target}/BOGUS/filestuff}.

Specifying an absolute path for the include file such as;
@smallexample

(include "/u2/BOGUS/filestuff")

@end smallexample
is permitted but is not encouraged.

@subsection RTL Generation Tool Options for Directory Search
@cindex directory options .md
@cindex options, directory search
@cindex search options

The @option{-I@var{dir}} option specifies directories to search for machine descriptions.
For example:

@smallexample

genrecog -I/p1/abc/proc1 -I/p2/abcd/pro2 target.md

@end smallexample


Add the directory @var{dir} to the head of the list of directories to be
searched for header files.  This can be used to override a system machine definition
file, substituting your own version, since these directories are
searched before the default machine description file directories.  If you use more than
one @option{-I} option, the directories are scanned in left-to-right
order; the standard default directory come after.


@end ifset
@ifset INTERNALS
@node Peephole Definitions
@section Machine-Specific Peephole Optimizers
@cindex peephole optimizer definitions
@cindex defining peephole optimizers

In addition to instruction patterns the @file{md} file may contain
definitions of machine-specific peephole optimizations.

The combiner does not notice certain peephole optimizations when the data
flow in the program does not suggest that it should try them.  For example,
sometimes two consecutive insns related in purpose can be combined even
though the second one does not appear to use a register computed in the
first one.  A machine-specific peephole optimizer can detect such
opportunities.

There are two forms of peephole definitions that may be used.  The
original @code{define_peephole} is run at assembly output time to
match insns and substitute assembly text.  Use of @code{define_peephole}
is deprecated.

A newer @code{define_peephole2} matches insns and substitutes new
insns.  The @code{peephole2} pass is run after register allocation
but before scheduling, which may result in much better code for
targets that do scheduling.

@menu
* define_peephole::     RTL to Text Peephole Optimizers
* define_peephole2::    RTL to RTL Peephole Optimizers
@end menu

@end ifset
@ifset INTERNALS
@node define_peephole
@subsection RTL to Text Peephole Optimizers
@findex define_peephole

@need 1000
A definition looks like this:

@smallexample
(define_peephole
  [@var{insn-pattern-1}
   @var{insn-pattern-2}
   @dots{}]
  "@var{condition}"
  "@var{template}"
  "@var{optional-insn-attributes}")
@end smallexample

@noindent
The last string operand may be omitted if you are not using any
machine-specific information in this machine description.  If present,
it must obey the same rules as in a @code{define_insn}.

In this skeleton, @var{insn-pattern-1} and so on are patterns to match
consecutive insns.  The optimization applies to a sequence of insns when
@var{insn-pattern-1} matches the first one, @var{insn-pattern-2} matches
the next, and so on.

Each of the insns matched by a peephole must also match a
@code{define_insn}.  Peepholes are checked only at the last stage just
before code generation, and only optionally.  Therefore, any insn which
would match a peephole but no @code{define_insn} will cause a crash in code
generation in an unoptimized compilation, or at various optimization
stages.

The operands of the insns are matched with @code{match_operands},
@code{match_operator}, and @code{match_dup}, as usual.  What is not
usual is that the operand numbers apply to all the insn patterns in the
definition.  So, you can check for identical operands in two insns by
using @code{match_operand} in one insn and @code{match_dup} in the
other.

The operand constraints used in @code{match_operand} patterns do not have
any direct effect on the applicability of the peephole, but they will
be validated afterward, so make sure your constraints are general enough
to apply whenever the peephole matches.  If the peephole matches
but the constraints are not satisfied, the compiler will crash.

It is safe to omit constraints in all the operands of the peephole; or
you can write constraints which serve as a double-check on the criteria
previously tested.

Once a sequence of insns matches the patterns, the @var{condition} is
checked.  This is a C expression which makes the final decision whether to
perform the optimization (we do so if the expression is nonzero).  If
@var{condition} is omitted (in other words, the string is empty) then the
optimization is applied to every sequence of insns that matches the
patterns.

The defined peephole optimizations are applied after register allocation
is complete.  Therefore, the peephole definition can check which
operands have ended up in which kinds of registers, just by looking at
the operands.

@findex prev_active_insn
The way to refer to the operands in @var{condition} is to write
@code{operands[@var{i}]} for operand number @var{i} (as matched by
@code{(match_operand @var{i} @dots{})}).  Use the variable @code{insn}
to refer to the last of the insns being matched; use
@code{prev_active_insn} to find the preceding insns.

@findex dead_or_set_p
When optimizing computations with intermediate results, you can use
@var{condition} to match only when the intermediate results are not used
elsewhere.  Use the C expression @code{dead_or_set_p (@var{insn},
@var{op})}, where @var{insn} is the insn in which you expect the value
to be used for the last time (from the value of @code{insn}, together
with use of @code{prev_nonnote_insn}), and @var{op} is the intermediate
value (from @code{operands[@var{i}]}).

Applying the optimization means replacing the sequence of insns with one
new insn.  The @var{template} controls ultimate output of assembler code
for this combined insn.  It works exactly like the template of a
@code{define_insn}.  Operand numbers in this template are the same ones
used in matching the original sequence of insns.

The result of a defined peephole optimizer does not need to match any of
the insn patterns in the machine description; it does not even have an
opportunity to match them.  The peephole optimizer definition itself serves
as the insn pattern to control how the insn is output.

Defined peephole optimizers are run as assembler code is being output,
so the insns they produce are never combined or rearranged in any way.

Here is an example, taken from the 68000 machine description:

@smallexample
(define_peephole
  [(set (reg:SI 15) (plus:SI (reg:SI 15) (const_int 4)))
   (set (match_operand:DF 0 "register_operand" "=f")
        (match_operand:DF 1 "register_operand" "ad"))]
  "FP_REG_P (operands[0]) && ! FP_REG_P (operands[1])"
@{
  rtx xoperands[2];
  xoperands[1] = gen_rtx_REG (SImode, REGNO (operands[1]) + 1);
#ifdef MOTOROLA
  output_asm_insn ("move.l %1,(sp)", xoperands);
  output_asm_insn ("move.l %1,-(sp)", operands);
  return "fmove.d (sp)+,%0";
#else
  output_asm_insn ("movel %1,sp@@", xoperands);
  output_asm_insn ("movel %1,sp@@-", operands);
  return "fmoved sp@@+,%0";
#endif
@})
@end smallexample

@need 1000
The effect of this optimization is to change

@smallexample
@group
jbsr _foobar
addql #4,sp
movel d1,sp@@-
movel d0,sp@@-
fmoved sp@@+,fp0
@end group
@end smallexample

@noindent
into

@smallexample
@group
jbsr _foobar
movel d1,sp@@
movel d0,sp@@-
fmoved sp@@+,fp0
@end group
@end smallexample

@ignore
@findex CC_REVERSED
If a peephole matches a sequence including one or more jump insns, you must
take account of the flags such as @code{CC_REVERSED} which specify that the
condition codes are represented in an unusual manner.  The compiler
automatically alters any ordinary conditional jumps which occur in such
situations, but the compiler cannot alter jumps which have been replaced by
peephole optimizations.  So it is up to you to alter the assembler code
that the peephole produces.  Supply C code to write the assembler output,
and in this C code check the condition code status flags and change the
assembler code as appropriate.
@end ignore

@var{insn-pattern-1} and so on look @emph{almost} like the second
operand of @code{define_insn}.  There is one important difference: the
second operand of @code{define_insn} consists of one or more RTX's
enclosed in square brackets.  Usually, there is only one: then the same
action can be written as an element of a @code{define_peephole}.  But
when there are multiple actions in a @code{define_insn}, they are
implicitly enclosed in a @code{parallel}.  Then you must explicitly
write the @code{parallel}, and the square brackets within it, in the
@code{define_peephole}.  Thus, if an insn pattern looks like this,

@smallexample
(define_insn "divmodsi4"
  [(set (match_operand:SI 0 "general_operand" "=d")
        (div:SI (match_operand:SI 1 "general_operand" "0")
                (match_operand:SI 2 "general_operand" "dmsK")))
   (set (match_operand:SI 3 "general_operand" "=d")
        (mod:SI (match_dup 1) (match_dup 2)))]
  "TARGET_68020"
  "divsl%.l %2,%3:%0")
@end smallexample

@noindent
then the way to mention this insn in a peephole is as follows:

@smallexample
(define_peephole
  [@dots{}
   (parallel
    [(set (match_operand:SI 0 "general_operand" "=d")
          (div:SI (match_operand:SI 1 "general_operand" "0")
                  (match_operand:SI 2 "general_operand" "dmsK")))
     (set (match_operand:SI 3 "general_operand" "=d")
          (mod:SI (match_dup 1) (match_dup 2)))])
   @dots{}]
  @dots{})
@end smallexample

@end ifset
@ifset INTERNALS
@node define_peephole2
@subsection RTL to RTL Peephole Optimizers
@findex define_peephole2

The @code{define_peephole2} definition tells the compiler how to
substitute one sequence of instructions for another sequence,
what additional scratch registers may be needed and what their
lifetimes must be.

@smallexample
(define_peephole2
  [@var{insn-pattern-1}
   @var{insn-pattern-2}
   @dots{}]
  "@var{condition}"
  [@var{new-insn-pattern-1}
   @var{new-insn-pattern-2}
   @dots{}]
  "@var{preparation-statements}")
@end smallexample

The definition is almost identical to @code{define_split}
(@pxref{Insn Splitting}) except that the pattern to match is not a
single instruction, but a sequence of instructions.

It is possible to request additional scratch registers for use in the
output template.  If appropriate registers are not free, the pattern
will simply not match.

@findex match_scratch
@findex match_dup
Scratch registers are requested with a @code{match_scratch} pattern at
the top level of the input pattern.  The allocated register (initially) will
be dead at the point requested within the original sequence.  If the scratch
is used at more than a single point, a @code{match_dup} pattern at the
top level of the input pattern marks the last position in the input sequence
at which the register must be available.

Here is an example from the IA-32 machine description:

@smallexample
(define_peephole2
  [(match_scratch:SI 2 "r")
   (parallel [(set (match_operand:SI 0 "register_operand" "")
                   (match_operator:SI 3 "arith_or_logical_operator"
                     [(match_dup 0)
                      (match_operand:SI 1 "memory_operand" "")]))
              (clobber (reg:CC 17))])]
  "! optimize_size && ! TARGET_READ_MODIFY"
  [(set (match_dup 2) (match_dup 1))
   (parallel [(set (match_dup 0)
                   (match_op_dup 3 [(match_dup 0) (match_dup 2)]))
              (clobber (reg:CC 17))])]
  "")
@end smallexample

@noindent
This pattern tries to split a load from its use in the hopes that we'll be
able to schedule around the memory load latency.  It allocates a single
@code{SImode} register of class @code{GENERAL_REGS} (@code{"r"}) that needs
to be live only at the point just before the arithmetic.

A real example requiring extended scratch lifetimes is harder to come by,
so here's a silly made-up example:

@smallexample
(define_peephole2
  [(match_scratch:SI 4 "r")
   (set (match_operand:SI 0 "" "") (match_operand:SI 1 "" ""))
   (set (match_operand:SI 2 "" "") (match_dup 1))
   (match_dup 4)
   (set (match_operand:SI 3 "" "") (match_dup 1))]
  "/* @r{determine 1 does not overlap 0 and 2} */"
  [(set (match_dup 4) (match_dup 1))
   (set (match_dup 0) (match_dup 4))
   (set (match_dup 2) (match_dup 4))]
   (set (match_dup 3) (match_dup 4))]
  "")
@end smallexample

@noindent
If we had not added the @code{(match_dup 4)} in the middle of the input
sequence, it might have been the case that the register we chose at the
beginning of the sequence is killed by the first or second @code{set}.

@end ifset
@ifset INTERNALS
@node Insn Attributes
@section Instruction Attributes
@cindex insn attributes
@cindex instruction attributes

In addition to describing the instruction supported by the target machine,
the @file{md} file also defines a group of @dfn{attributes} and a set of
values for each.  Every generated insn is assigned a value for each attribute.
One possible attribute would be the effect that the insn has on the machine's
condition code.  This attribute can then be used by @code{NOTICE_UPDATE_CC}
to track the condition codes.

@menu
* Defining Attributes:: Specifying attributes and their values.
* Expressions::         Valid expressions for attribute values.
* Tagging Insns::       Assigning attribute values to insns.
* Attr Example::        An example of assigning attributes.
* Insn Lengths::        Computing the length of insns.
* Constant Attributes:: Defining attributes that are constant.
* Delay Slots::         Defining delay slots required for a machine.
* Processor pipeline description:: Specifying information for insn scheduling.
@end menu

@end ifset
@ifset INTERNALS
@node Defining Attributes
@subsection Defining Attributes and their Values
@cindex defining attributes and their values
@cindex attributes, defining

@findex define_attr
The @code{define_attr} expression is used to define each attribute required
by the target machine.  It looks like:

@smallexample
(define_attr @var{name} @var{list-of-values} @var{default})
@end smallexample

@var{name} is a string specifying the name of the attribute being defined.

@var{list-of-values} is either a string that specifies a comma-separated
list of values that can be assigned to the attribute, or a null string to
indicate that the attribute takes numeric values.

@var{default} is an attribute expression that gives the value of this
attribute for insns that match patterns whose definition does not include
an explicit value for this attribute.  @xref{Attr Example}, for more
information on the handling of defaults.  @xref{Constant Attributes},
for information on attributes that do not depend on any particular insn.

@findex insn-attr.h
For each defined attribute, a number of definitions are written to the
@file{insn-attr.h} file.  For cases where an explicit set of values is
specified for an attribute, the following are defined:

@itemize @bullet
@item
A @samp{#define} is written for the symbol @samp{HAVE_ATTR_@var{name}}.

@item
An enumerated class is defined for @samp{attr_@var{name}} with
elements of the form @samp{@var{upper-name}_@var{upper-value}} where
the attribute name and value are first converted to uppercase.

@item
A function @samp{get_attr_@var{name}} is defined that is passed an insn and
returns the attribute value for that insn.
@end itemize

For example, if the following is present in the @file{md} file:

@smallexample
(define_attr "type" "branch,fp,load,store,arith" @dots{})
@end smallexample

@noindent
the following lines will be written to the file @file{insn-attr.h}.

@smallexample
#define HAVE_ATTR_type
enum attr_type @{TYPE_BRANCH, TYPE_FP, TYPE_LOAD,
                 TYPE_STORE, TYPE_ARITH@};
extern enum attr_type get_attr_type ();
@end smallexample

If the attribute takes numeric values, no @code{enum} type will be
defined and the function to obtain the attribute's value will return
@code{int}.

@end ifset
@ifset INTERNALS
@node Expressions
@subsection Attribute Expressions
@cindex attribute expressions

RTL expressions used to define attributes use the codes described above
plus a few specific to attribute definitions, to be discussed below.
Attribute value expressions must have one of the following forms:

@table @code
@cindex @code{const_int} and attributes
@item (const_int @var{i})
The integer @var{i} specifies the value of a numeric attribute.  @var{i}
must be non-negative.

The value of a numeric attribute can be specified either with a
@code{const_int}, or as an integer represented as a string in
@code{const_string}, @code{eq_attr} (see below), @code{attr},
@code{symbol_ref}, simple arithmetic expressions, and @code{set_attr}
overrides on specific instructions (@pxref{Tagging Insns}).

@cindex @code{const_string} and attributes
@item (const_string @var{value})
The string @var{value} specifies a constant attribute value.
If @var{value} is specified as @samp{"*"}, it means that the default value of
the attribute is to be used for the insn containing this expression.
@samp{"*"} obviously cannot be used in the @var{default} expression
of a @code{define_attr}.

If the attribute whose value is being specified is numeric, @var{value}
must be a string containing a non-negative integer (normally
@code{const_int} would be used in this case).  Otherwise, it must
contain one of the valid values for the attribute.

@cindex @code{if_then_else} and attributes
@item (if_then_else @var{test} @var{true-value} @var{false-value})
@var{test} specifies an attribute test, whose format is defined below.
The value of this expression is @var{true-value} if @var{test} is true,
otherwise it is @var{false-value}.

@cindex @code{cond} and attributes
@item (cond [@var{test1} @var{value1} @dots{}] @var{default})
The first operand of this expression is a vector containing an even
number of expressions and consisting of pairs of @var{test} and @var{value}
expressions.  The value of the @code{cond} expression is that of the
@var{value} corresponding to the first true @var{test} expression.  If
none of the @var{test} expressions are true, the value of the @code{cond}
expression is that of the @var{default} expression.
@end table

@var{test} expressions can have one of the following forms:

@table @code
@cindex @code{const_int} and attribute tests
@item (const_int @var{i})
This test is true if @var{i} is nonzero and false otherwise.

@cindex @code{not} and attributes
@cindex @code{ior} and attributes
@cindex @code{and} and attributes
@item (not @var{test})
@itemx (ior @var{test1} @var{test2})
@itemx (and @var{test1} @var{test2})
These tests are true if the indicated logical function is true.

@cindex @code{match_operand} and attributes
@item (match_operand:@var{m} @var{n} @var{pred} @var{constraints})
This test is true if operand @var{n} of the insn whose attribute value
is being determined has mode @var{m} (this part of the test is ignored
if @var{m} is @code{VOIDmode}) and the function specified by the string
@var{pred} returns a nonzero value when passed operand @var{n} and mode
@var{m} (this part of the test is ignored if @var{pred} is the null
string).

The @var{constraints} operand is ignored and should be the null string.

@cindex @code{le} and attributes
@cindex @code{leu} and attributes
@cindex @code{lt} and attributes
@cindex @code{gt} and attributes
@cindex @code{gtu} and attributes
@cindex @code{ge} and attributes
@cindex @code{geu} and attributes
@cindex @code{ne} and attributes
@cindex @code{eq} and attributes
@cindex @code{plus} and attributes
@cindex @code{minus} and attributes
@cindex @code{mult} and attributes
@cindex @code{div} and attributes
@cindex @code{mod} and attributes
@cindex @code{abs} and attributes
@cindex @code{neg} and attributes
@cindex @code{ashift} and attributes
@cindex @code{lshiftrt} and attributes
@cindex @code{ashiftrt} and attributes
@item (le @var{arith1} @var{arith2})
@itemx (leu @var{arith1} @var{arith2})
@itemx (lt @var{arith1} @var{arith2})
@itemx (ltu @var{arith1} @var{arith2})
@itemx (gt @var{arith1} @var{arith2})
@itemx (gtu @var{arith1} @var{arith2})
@itemx (ge @var{arith1} @var{arith2})
@itemx (geu @var{arith1} @var{arith2})
@itemx (ne @var{arith1} @var{arith2})
@itemx (eq @var{arith1} @var{arith2})
These tests are true if the indicated comparison of the two arithmetic
expressions is true.  Arithmetic expressions are formed with
@code{plus}, @code{minus}, @code{mult}, @code{div}, @code{mod},
@code{abs}, @code{neg}, @code{and}, @code{ior}, @code{xor}, @code{not},
@code{ashift}, @code{lshiftrt}, and @code{ashiftrt} expressions.

@findex get_attr
@code{const_int} and @code{symbol_ref} are always valid terms (@pxref{Insn
Lengths},for additional forms).  @code{symbol_ref} is a string
denoting a C expression that yields an @code{int} when evaluated by the
@samp{get_attr_@dots{}} routine.  It should normally be a global
variable.

@findex eq_attr
@item (eq_attr @var{name} @var{value})
@var{name} is a string specifying the name of an attribute.

@var{value} is a string that is either a valid value for attribute
@var{name}, a comma-separated list of values, or @samp{!} followed by a
value or list.  If @var{value} does not begin with a @samp{!}, this
test is true if the value of the @var{name} attribute of the current
insn is in the list specified by @var{value}.  If @var{value} begins
with a @samp{!}, this test is true if the attribute's value is
@emph{not} in the specified list.

For example,

@smallexample
(eq_attr "type" "load,store")
@end smallexample

@noindent
is equivalent to

@smallexample
(ior (eq_attr "type" "load") (eq_attr "type" "store"))
@end smallexample

If @var{name} specifies an attribute of @samp{alternative}, it refers to the
value of the compiler variable @code{which_alternative}
(@pxref{Output Statement}) and the values must be small integers.  For
example,

@smallexample
(eq_attr "alternative" "2,3")
@end smallexample

@noindent
is equivalent to

@smallexample
(ior (eq (symbol_ref "which_alternative") (const_int 2))
     (eq (symbol_ref "which_alternative") (const_int 3)))
@end smallexample

Note that, for most attributes, an @code{eq_attr} test is simplified in cases
where the value of the attribute being tested is known for all insns matching
a particular pattern.  This is by far the most common case.

@findex attr_flag
@item (attr_flag @var{name})
The value of an @code{attr_flag} expression is true if the flag
specified by @var{name} is true for the @code{insn} currently being
scheduled.

@var{name} is a string specifying one of a fixed set of flags to test.
Test the flags @code{forward} and @code{backward} to determine the
direction of a conditional branch.  Test the flags @code{very_likely},
@code{likely}, @code{very_unlikely}, and @code{unlikely} to determine
if a conditional branch is expected to be taken.

If the @code{very_likely} flag is true, then the @code{likely} flag is also
true.  Likewise for the @code{very_unlikely} and @code{unlikely} flags.

This example describes a conditional branch delay slot which
can be nullified for forward branches that are taken (annul-true) or
for backward branches which are not taken (annul-false).

@smallexample
(define_delay (eq_attr "type" "cbranch")
  [(eq_attr "in_branch_delay" "true")
   (and (eq_attr "in_branch_delay" "true")
        (attr_flag "forward"))
   (and (eq_attr "in_branch_delay" "true")
        (attr_flag "backward"))])
@end smallexample

The @code{forward} and @code{backward} flags are false if the current
@code{insn} being scheduled is not a conditional branch.

The @code{very_likely} and @code{likely} flags are true if the
@code{insn} being scheduled is not a conditional branch.
The @code{very_unlikely} and @code{unlikely} flags are false if the
@code{insn} being scheduled is not a conditional branch.

@code{attr_flag} is only used during delay slot scheduling and has no
meaning to other passes of the compiler.

@findex attr
@item (attr @var{name})
The value of another attribute is returned.  This is most useful
for numeric attributes, as @code{eq_attr} and @code{attr_flag}
produce more efficient code for non-numeric attributes.
@end table

@end ifset
@ifset INTERNALS
@node Tagging Insns
@subsection Assigning Attribute Values to Insns
@cindex tagging insns
@cindex assigning attribute values to insns

The value assigned to an attribute of an insn is primarily determined by
which pattern is matched by that insn (or which @code{define_peephole}
generated it).  Every @code{define_insn} and @code{define_peephole} can
have an optional last argument to specify the values of attributes for
matching insns.  The value of any attribute not specified in a particular
insn is set to the default value for that attribute, as specified in its
@code{define_attr}.  Extensive use of default values for attributes
permits the specification of the values for only one or two attributes
in the definition of most insn patterns, as seen in the example in the
next section.

The optional last argument of @code{define_insn} and
@code{define_peephole} is a vector of expressions, each of which defines
the value for a single attribute.  The most general way of assigning an
attribute's value is to use a @code{set} expression whose first operand is an
@code{attr} expression giving the name of the attribute being set.  The
second operand of the @code{set} is an attribute expression
(@pxref{Expressions}) giving the value of the attribute.

When the attribute value depends on the @samp{alternative} attribute
(i.e., which is the applicable alternative in the constraint of the
insn), the @code{set_attr_alternative} expression can be used.  It
allows the specification of a vector of attribute expressions, one for
each alternative.

@findex set_attr
When the generality of arbitrary attribute expressions is not required,
the simpler @code{set_attr} expression can be used, which allows
specifying a string giving either a single attribute value or a list
of attribute values, one for each alternative.

The form of each of the above specifications is shown below.  In each case,
@var{name} is a string specifying the attribute to be set.

@table @code
@item (set_attr @var{name} @var{value-string})
@var{value-string} is either a string giving the desired attribute value,
or a string containing a comma-separated list giving the values for
succeeding alternatives.  The number of elements must match the number
of alternatives in the constraint of the insn pattern.

Note that it may be useful to specify @samp{*} for some alternative, in
which case the attribute will assume its default value for insns matching
that alternative.

@findex set_attr_alternative
@item (set_attr_alternative @var{name} [@var{value1} @var{value2} @dots{}])
Depending on the alternative of the insn, the value will be one of the
specified values.  This is a shorthand for using a @code{cond} with
tests on the @samp{alternative} attribute.

@findex attr
@item (set (attr @var{name}) @var{value})
The first operand of this @code{set} must be the special RTL expression
@code{attr}, whose sole operand is a string giving the name of the
attribute being set.  @var{value} is the value of the attribute.
@end table

The following shows three different ways of representing the same
attribute value specification:

@smallexample
(set_attr "type" "load,store,arith")

(set_attr_alternative "type"
                      [(const_string "load") (const_string "store")
                       (const_string "arith")])

(set (attr "type")
     (cond [(eq_attr "alternative" "1") (const_string "load")
            (eq_attr "alternative" "2") (const_string "store")]
           (const_string "arith")))
@end smallexample

@need 1000
@findex define_asm_attributes
The @code{define_asm_attributes} expression provides a mechanism to
specify the attributes assigned to insns produced from an @code{asm}
statement.  It has the form:

@smallexample
(define_asm_attributes [@var{attr-sets}])
@end smallexample

@noindent
where @var{attr-sets} is specified the same as for both the
@code{define_insn} and the @code{define_peephole} expressions.

These values will typically be the ``worst case'' attribute values.  For
example, they might indicate that the condition code will be clobbered.

A specification for a @code{length} attribute is handled specially.  The
way to compute the length of an @code{asm} insn is to multiply the
length specified in the expression @code{define_asm_attributes} by the
number of machine instructions specified in the @code{asm} statement,
determined by counting the number of semicolons and newlines in the
string.  Therefore, the value of the @code{length} attribute specified
in a @code{define_asm_attributes} should be the maximum possible length
of a single machine instruction.

@end ifset
@ifset INTERNALS
@node Attr Example
@subsection Example of Attribute Specifications
@cindex attribute specifications example
@cindex attribute specifications

The judicious use of defaulting is important in the efficient use of
insn attributes.  Typically, insns are divided into @dfn{types} and an
attribute, customarily called @code{type}, is used to represent this
value.  This attribute is normally used only to define the default value
for other attributes.  An example will clarify this usage.

Assume we have a RISC machine with a condition code and in which only
full-word operations are performed in registers.  Let us assume that we
can divide all insns into loads, stores, (integer) arithmetic
operations, floating point operations, and branches.

Here we will concern ourselves with determining the effect of an insn on
the condition code and will limit ourselves to the following possible
effects:  The condition code can be set unpredictably (clobbered), not
be changed, be set to agree with the results of the operation, or only
changed if the item previously set into the condition code has been
modified.

Here is part of a sample @file{md} file for such a machine:

@smallexample
(define_attr "type" "load,store,arith,fp,branch" (const_string "arith"))

(define_attr "cc" "clobber,unchanged,set,change0"
             (cond [(eq_attr "type" "load")
                        (const_string "change0")
                    (eq_attr "type" "store,branch")
                        (const_string "unchanged")
                    (eq_attr "type" "arith")
                        (if_then_else (match_operand:SI 0 "" "")
                                      (const_string "set")
                                      (const_string "clobber"))]
                   (const_string "clobber")))

(define_insn ""
  [(set (match_operand:SI 0 "general_operand" "=r,r,m")
        (match_operand:SI 1 "general_operand" "r,m,r"))]
  ""
  "@@
   move %0,%1
   load %0,%1
   store %0,%1"
  [(set_attr "type" "arith,load,store")])
@end smallexample

Note that we assume in the above example that arithmetic operations
performed on quantities smaller than a machine word clobber the condition
code since they will set the condition code to a value corresponding to the
full-word result.

@end ifset
@ifset INTERNALS
@node Insn Lengths
@subsection Computing the Length of an Insn
@cindex insn lengths, computing
@cindex computing the length of an insn

For many machines, multiple types of branch instructions are provided, each
for different length branch displacements.  In most cases, the assembler
will choose the correct instruction to use.  However, when the assembler
cannot do so, GCC can when a special attribute, the @code{length}
attribute, is defined.  This attribute must be defined to have numeric
values by specifying a null string in its @code{define_attr}.

In the case of the @code{length} attribute, two additional forms of
arithmetic terms are allowed in test expressions:

@table @code
@cindex @code{match_dup} and attributes
@item (match_dup @var{n})
This refers to the address of operand @var{n} of the current insn, which
must be a @code{label_ref}.

@cindex @code{pc} and attributes
@item (pc)
This refers to the address of the @emph{current} insn.  It might have
been more consistent with other usage to make this the address of the
@emph{next} insn but this would be confusing because the length of the
current insn is to be computed.
@end table

@cindex @code{addr_vec}, length of
@cindex @code{addr_diff_vec}, length of
For normal insns, the length will be determined by value of the
@code{length} attribute.  In the case of @code{addr_vec} and
@code{addr_diff_vec} insn patterns, the length is computed as
the number of vectors multiplied by the size of each vector.

Lengths are measured in addressable storage units (bytes).

The following macros can be used to refine the length computation:

@table @code
@findex ADJUST_INSN_LENGTH
@item ADJUST_INSN_LENGTH (@var{insn}, @var{length})
If defined, modifies the length assigned to instruction @var{insn} as a
function of the context in which it is used.  @var{length} is an lvalue
that contains the initially computed length of the insn and should be
updated with the correct length of the insn.

This macro will normally not be required.  A case in which it is
required is the ROMP@.  On this machine, the size of an @code{addr_vec}
insn must be increased by two to compensate for the fact that alignment
may be required.
@end table

@findex get_attr_length
The routine that returns @code{get_attr_length} (the value of the
@code{length} attribute) can be used by the output routine to
determine the form of the branch instruction to be written, as the
example below illustrates.

As an example of the specification of variable-length branches, consider
the IBM 360.  If we adopt the convention that a register will be set to
the starting address of a function, we can jump to labels within 4k of
the start using a four-byte instruction.  Otherwise, we need a six-byte
sequence to load the address from memory and then branch to it.

On such a machine, a pattern for a branch instruction might be specified
as follows:

@smallexample
(define_insn "jump"
  [(set (pc)
        (label_ref (match_operand 0 "" "")))]
  ""
@{
   return (get_attr_length (insn) == 4
           ? "b %l0" : "l r15,=a(%l0); br r15");
@}
  [(set (attr "length")
        (if_then_else (lt (match_dup 0) (const_int 4096))
                      (const_int 4)
                      (const_int 6)))])
@end smallexample

@end ifset
@ifset INTERNALS
@node Constant Attributes
@subsection Constant Attributes
@cindex constant attributes

A special form of @code{define_attr}, where the expression for the
default value is a @code{const} expression, indicates an attribute that
is constant for a given run of the compiler.  Constant attributes may be
used to specify which variety of processor is used.  For example,

@smallexample
(define_attr "cpu" "m88100,m88110,m88000"
 (const
  (cond [(symbol_ref "TARGET_88100") (const_string "m88100")
         (symbol_ref "TARGET_88110") (const_string "m88110")]
        (const_string "m88000"))))

(define_attr "memory" "fast,slow"
 (const
  (if_then_else (symbol_ref "TARGET_FAST_MEM")
                (const_string "fast")
                (const_string "slow"))))
@end smallexample

The routine generated for constant attributes has no parameters as it
does not depend on any particular insn.  RTL expressions used to define
the value of a constant attribute may use the @code{symbol_ref} form,
but may not use either the @code{match_operand} form or @code{eq_attr}
forms involving insn attributes.

@end ifset
@ifset INTERNALS
@node Delay Slots
@subsection Delay Slot Scheduling
@cindex delay slots, defining

The insn attribute mechanism can be used to specify the requirements for
delay slots, if any, on a target machine.  An instruction is said to
require a @dfn{delay slot} if some instructions that are physically
after the instruction are executed as if they were located before it.
Classic examples are branch and call instructions, which often execute
the following instruction before the branch or call is performed.

On some machines, conditional branch instructions can optionally
@dfn{annul} instructions in the delay slot.  This means that the
instruction will not be executed for certain branch outcomes.  Both
instructions that annul if the branch is true and instructions that
annul if the branch is false are supported.

Delay slot scheduling differs from instruction scheduling in that
determining whether an instruction needs a delay slot is dependent only
on the type of instruction being generated, not on data flow between the
instructions.  See the next section for a discussion of data-dependent
instruction scheduling.

@findex define_delay
The requirement of an insn needing one or more delay slots is indicated
via the @code{define_delay} expression.  It has the following form:

@smallexample
(define_delay @var{test}
              [@var{delay-1} @var{annul-true-1} @var{annul-false-1}
               @var{delay-2} @var{annul-true-2} @var{annul-false-2}
               @dots{}])
@end smallexample

@var{test} is an attribute test that indicates whether this
@code{define_delay} applies to a particular insn.  If so, the number of
required delay slots is determined by the length of the vector specified
as the second argument.  An insn placed in delay slot @var{n} must
satisfy attribute test @var{delay-n}.  @var{annul-true-n} is an
attribute test that specifies which insns may be annulled if the branch
is true.  Similarly, @var{annul-false-n} specifies which insns in the
delay slot may be annulled if the branch is false.  If annulling is not
supported for that delay slot, @code{(nil)} should be coded.

For example, in the common case where branch and call insns require
a single delay slot, which may contain any insn other than a branch or
call, the following would be placed in the @file{md} file:

@smallexample
(define_delay (eq_attr "type" "branch,call")
              [(eq_attr "type" "!branch,call") (nil) (nil)])
@end smallexample

Multiple @code{define_delay} expressions may be specified.  In this
case, each such expression specifies different delay slot requirements
and there must be no insn for which tests in two @code{define_delay}
expressions are both true.

For example, if we have a machine that requires one delay slot for branches
but two for calls,  no delay slot can contain a branch or call insn,
and any valid insn in the delay slot for the branch can be annulled if the
branch is true, we might represent this as follows:

@smallexample
(define_delay (eq_attr "type" "branch")
   [(eq_attr "type" "!branch,call")
    (eq_attr "type" "!branch,call")
    (nil)])

(define_delay (eq_attr "type" "call")
              [(eq_attr "type" "!branch,call") (nil) (nil)
               (eq_attr "type" "!branch,call") (nil) (nil)])
@end smallexample
@c the above is *still* too long.  --mew 4feb93

@end ifset
@ifset INTERNALS
@node Processor pipeline description
@subsection Specifying processor pipeline description
@cindex processor pipeline description
@cindex processor functional units
@cindex instruction latency time
@cindex interlock delays
@cindex data dependence delays
@cindex reservation delays
@cindex pipeline hazard recognizer
@cindex automaton based pipeline description
@cindex regular expressions
@cindex deterministic finite state automaton
@cindex automaton based scheduler
@cindex RISC
@cindex VLIW

To achieve better performance, most modern processors
(super-pipelined, superscalar @acronym{RISC}, and @acronym{VLIW}
processors) have many @dfn{functional units} on which several
instructions can be executed simultaneously.  An instruction starts
execution if its issue conditions are satisfied.  If not, the
instruction is stalled until its conditions are satisfied.  Such
@dfn{interlock (pipeline) delay} causes interruption of the fetching
of successor instructions (or demands nop instructions, e.g.@: for some
MIPS processors).

There are two major kinds of interlock delays in modern processors.
The first one is a data dependence delay determining @dfn{instruction
latency time}.  The instruction execution is not started until all
source data have been evaluated by prior instructions (there are more
complex cases when the instruction execution starts even when the data
are not available but will be ready in given time after the
instruction execution start).  Taking the data dependence delays into
account is simple.  The data dependence (true, output, and
anti-dependence) delay between two instructions is given by a
constant.  In most cases this approach is adequate.  The second kind
of interlock delays is a reservation delay.  The reservation delay
means that two instructions under execution will be in need of shared
processors resources, i.e.@: buses, internal registers, and/or
functional units, which are reserved for some time.  Taking this kind
of delay into account is complex especially for modern @acronym{RISC}
processors.

The task of exploiting more processor parallelism is solved by an
instruction scheduler.  For a better solution to this problem, the
instruction scheduler has to have an adequate description of the
processor parallelism (or @dfn{pipeline description}).  GCC
machine descriptions describe processor parallelism and functional
unit reservations for groups of instructions with the aid of
@dfn{regular expressions}.

The GCC instruction scheduler uses a @dfn{pipeline hazard recognizer} to
figure out the possibility of the instruction issue by the processor
on a given simulated processor cycle.  The pipeline hazard recognizer is
automatically generated from the processor pipeline description.  The
pipeline hazard recognizer generated from the machine description
is based on a deterministic finite state automaton (@acronym{DFA}):
the instruction issue is possible if there is a transition from one
automaton state to another one.  This algorithm is very fast, and
furthermore, its speed is not dependent on processor
complexity@footnote{However, the size of the automaton depends on
  processor complexity.  To limit this effect, machine descriptions
  can split orthogonal parts of the machine description among several
  automata: but then, since each of these must be stepped independently,
  this does cause a small decrease in the algorithm's performance.}.

@cindex automaton based pipeline description
The rest of this section describes the directives that constitute
an automaton-based processor pipeline description.  The order of
these constructions within the machine description file is not
important.

@findex define_automaton
@cindex pipeline hazard recognizer
The following optional construction describes names of automata
generated and used for the pipeline hazards recognition.  Sometimes
the generated finite state automaton used by the pipeline hazard
recognizer is large.  If we use more than one automaton and bind functional
units to the automata, the total size of the automata is usually
less than the size of the single automaton.  If there is no one such
construction, only one finite state automaton is generated.

@smallexample
(define_automaton @var{automata-names})
@end smallexample

@var{automata-names} is a string giving names of the automata.  The
names are separated by commas.  All the automata should have unique names.
The automaton name is used in the constructions @code{define_cpu_unit} and
@code{define_query_cpu_unit}.

@findex define_cpu_unit
@cindex processor functional units
Each processor functional unit used in the description of instruction
reservations should be described by the following construction.

@smallexample
(define_cpu_unit @var{unit-names} [@var{automaton-name}])
@end smallexample

@var{unit-names} is a string giving the names of the functional units
separated by commas.  Don't use name @samp{nothing}, it is reserved
for other goals.

@var{automaton-name} is a string giving the name of the automaton with
which the unit is bound.  The automaton should be described in
construction @code{define_automaton}.  You should give
@dfn{automaton-name}, if there is a defined automaton.

The assignment of units to automata are constrained by the uses of the
units in insn reservations.  The most important constraint is: if a
unit reservation is present on a particular cycle of an alternative
for an insn reservation, then some unit from the same automaton must
be present on the same cycle for the other alternatives of the insn
reservation.  The rest of the constraints are mentioned in the
description of the subsequent constructions.

@findex define_query_cpu_unit
@cindex querying function unit reservations
The following construction describes CPU functional units analogously
to @code{define_cpu_unit}.  The reservation of such units can be
queried for an automaton state.  The instruction scheduler never
queries reservation of functional units for given automaton state.  So
as a rule, you don't need this construction.  This construction could
be used for future code generation goals (e.g.@: to generate
@acronym{VLIW} insn templates).

@smallexample
(define_query_cpu_unit @var{unit-names} [@var{automaton-name}])
@end smallexample

@var{unit-names} is a string giving names of the functional units
separated by commas.

@var{automaton-name} is a string giving the name of the automaton with
which the unit is bound.

@findex define_insn_reservation
@cindex instruction latency time
@cindex regular expressions
@cindex data bypass
The following construction is the major one to describe pipeline
characteristics of an instruction.

@smallexample
(define_insn_reservation @var{insn-name} @var{default_latency}
                         @var{condition} @var{regexp})
@end smallexample

@var{default_latency} is a number giving latency time of the
instruction.  There is an important difference between the old
description and the automaton based pipeline description.  The latency
time is used for all dependencies when we use the old description.  In
the automaton based pipeline description, the given latency time is only
used for true dependencies.  The cost of anti-dependencies is always
zero and the cost of output dependencies is the difference between
latency times of the producing and consuming insns (if the difference
is negative, the cost is considered to be zero).  You can always
change the default costs for any description by using the target hook
@code{TARGET_SCHED_ADJUST_COST} (@pxref{Scheduling}).

@var{insn-name} is a string giving the internal name of the insn.  The
internal names are used in constructions @code{define_bypass} and in
the automaton description file generated for debugging.  The internal
name has nothing in common with the names in @code{define_insn}.  It is a
good practice to use insn classes described in the processor manual.

@var{condition} defines what RTL insns are described by this
construction.  You should remember that you will be in trouble if
@var{condition} for two or more different
@code{define_insn_reservation} constructions is TRUE for an insn.  In
this case what reservation will be used for the insn is not defined.
Such cases are not checked during generation of the pipeline hazards
recognizer because in general recognizing that two conditions may have
the same value is quite difficult (especially if the conditions
contain @code{symbol_ref}).  It is also not checked during the
pipeline hazard recognizer work because it would slow down the
recognizer considerably.

@var{regexp} is a string describing the reservation of the cpu's functional
units by the instruction.  The reservations are described by a regular
expression according to the following syntax:

@smallexample
       regexp = regexp "," oneof
              | oneof

       oneof = oneof "|" allof
             | allof

       allof = allof "+" repeat
             | repeat

       repeat = element "*" number
              | element

       element = cpu_function_unit_name
               | reservation_name
               | result_name
               | "nothing"
               | "(" regexp ")"
@end smallexample

@itemize @bullet
@item
@samp{,} is used for describing the start of the next cycle in
the reservation.

@item
@samp{|} is used for describing a reservation described by the first
regular expression @strong{or} a reservation described by the second
regular expression @strong{or} etc.

@item
@samp{+} is used for describing a reservation described by the first
regular expression @strong{and} a reservation described by the
second regular expression @strong{and} etc.

@item
@samp{*} is used for convenience and simply means a sequence in which
the regular expression are repeated @var{number} times with cycle
advancing (see @samp{,}).

@item
@samp{cpu_function_unit_name} denotes reservation of the named
functional unit.

@item
@samp{reservation_name} --- see description of construction
@samp{define_reservation}.

@item
@samp{nothing} denotes no unit reservations.
@end itemize

@findex define_reservation
Sometimes unit reservations for different insns contain common parts.
In such case, you can simplify the pipeline description by describing
the common part by the following construction

@smallexample
(define_reservation @var{reservation-name} @var{regexp})
@end smallexample

@var{reservation-name} is a string giving name of @var{regexp}.
Functional unit names and reservation names are in the same name
space.  So the reservation names should be different from the
functional unit names and can not be the reserved name @samp{nothing}.

@findex define_bypass
@cindex instruction latency time
@cindex data bypass
The following construction is used to describe exceptions in the
latency time for given instruction pair.  This is so called bypasses.

@smallexample
(define_bypass @var{number} @var{out_insn_names} @var{in_insn_names}
               [@var{guard}])
@end smallexample

@var{number} defines when the result generated by the instructions
given in string @var{out_insn_names} will be ready for the
instructions given in string @var{in_insn_names}.  The instructions in
the string are separated by commas.

@var{guard} is an optional string giving the name of a C function which
defines an additional guard for the bypass.  The function will get the
two insns as parameters.  If the function returns zero the bypass will
be ignored for this case.  The additional guard is necessary to
recognize complicated bypasses, e.g.@: when the consumer is only an address
of insn @samp{store} (not a stored value).

@findex exclusion_set
@findex presence_set
@findex final_presence_set
@findex absence_set
@findex final_absence_set
@cindex VLIW
@cindex RISC
The following five constructions are usually used to describe
@acronym{VLIW} processors, or more precisely, to describe a placement
of small instructions into @acronym{VLIW} instruction slots.  They
can be used for @acronym{RISC} processors, too.

@smallexample
(exclusion_set @var{unit-names} @var{unit-names})
(presence_set @var{unit-names} @var{patterns})
(final_presence_set @var{unit-names} @var{patterns})
(absence_set @var{unit-names} @var{patterns})
(final_absence_set @var{unit-names} @var{patterns})
@end smallexample

@var{unit-names} is a string giving names of functional units
separated by commas.

@var{patterns} is a string giving patterns of functional units
separated by comma.  Currently pattern is one unit or units
separated by white-spaces.

The first construction (@samp{exclusion_set}) means that each
functional unit in the first string can not be reserved simultaneously
with a unit whose name is in the second string and vice versa.  For
example, the construction is useful for describing processors
(e.g.@: some SPARC processors) with a fully pipelined floating point
functional unit which can execute simultaneously only single floating
point insns or only double floating point insns.

The second construction (@samp{presence_set}) means that each
functional unit in the first string can not be reserved unless at
least one of pattern of units whose names are in the second string is
reserved.  This is an asymmetric relation.  For example, it is useful
for description that @acronym{VLIW} @samp{slot1} is reserved after
@samp{slot0} reservation.  We could describe it by the following
construction

@smallexample
(presence_set "slot1" "slot0")
@end smallexample

Or @samp{slot1} is reserved only after @samp{slot0} and unit @samp{b0}
reservation.  In this case we could write

@smallexample
(presence_set "slot1" "slot0 b0")
@end smallexample

The third construction (@samp{final_presence_set}) is analogous to
@samp{presence_set}.  The difference between them is when checking is
done.  When an instruction is issued in given automaton state
reflecting all current and planned unit reservations, the automaton
state is changed.  The first state is a source state, the second one
is a result state.  Checking for @samp{presence_set} is done on the
source state reservation, checking for @samp{final_presence_set} is
done on the result reservation.  This construction is useful to
describe a reservation which is actually two subsequent reservations.
For example, if we use

@smallexample
(presence_set "slot1" "slot0")
@end smallexample

the following insn will be never issued (because @samp{slot1} requires
@samp{slot0} which is absent in the source state).

@smallexample
(define_reservation "insn_and_nop" "slot0 + slot1")
@end smallexample

but it can be issued if we use analogous @samp{final_presence_set}.

The forth construction (@samp{absence_set}) means that each functional
unit in the first string can be reserved only if each pattern of units
whose names are in the second string is not reserved.  This is an
asymmetric relation (actually @samp{exclusion_set} is analogous to
this one but it is symmetric).  For example it might be useful in a 
@acronym{VLIW} description to say that @samp{slot0} cannot be reserved
after either @samp{slot1} or @samp{slot2} have been reserved.  This
can be described as:

@smallexample
(absence_set "slot0" "slot1, slot2")
@end smallexample

Or @samp{slot2} can not be reserved if @samp{slot0} and unit @samp{b0}
are reserved or @samp{slot1} and unit @samp{b1} are reserved.  In
this case we could write

@smallexample
(absence_set "slot2" "slot0 b0, slot1 b1")
@end smallexample

All functional units mentioned in a set should belong to the same
automaton.

The last construction (@samp{final_absence_set}) is analogous to
@samp{absence_set} but checking is done on the result (state)
reservation.  See comments for @samp{final_presence_set}.

@findex automata_option
@cindex deterministic finite state automaton
@cindex nondeterministic finite state automaton
@cindex finite state automaton minimization
You can control the generator of the pipeline hazard recognizer with
the following construction.

@smallexample
(automata_option @var{options})
@end smallexample

@var{options} is a string giving options which affect the generated
code.  Currently there are the following options:

@itemize @bullet
@item
@dfn{no-minimization} makes no minimization of the automaton.  This is
only worth to do when we are debugging the description and need to
look more accurately at reservations of states.

@item
@dfn{time} means printing time statistics about the generation of
automata.

@item
@dfn{stats} means printing statistics about the generated automata
such as the number of DFA states, NDFA states and arcs.

@item
@dfn{v} means a generation of the file describing the result automata.
The file has suffix @samp{.dfa} and can be used for the description
verification and debugging.

@item
@dfn{w} means a generation of warning instead of error for
non-critical errors.

@item
@dfn{ndfa} makes nondeterministic finite state automata.  This affects
the treatment of operator @samp{|} in the regular expressions.  The
usual treatment of the operator is to try the first alternative and,
if the reservation is not possible, the second alternative.  The
nondeterministic treatment means trying all alternatives, some of them
may be rejected by reservations in the subsequent insns.

@item
@dfn{progress} means output of a progress bar showing how many states
were generated so far for automaton being processed.  This is useful
during debugging a @acronym{DFA} description.  If you see too many
generated states, you could interrupt the generator of the pipeline
hazard recognizer and try to figure out a reason for generation of the
huge automaton.
@end itemize

As an example, consider a superscalar @acronym{RISC} machine which can
issue three insns (two integer insns and one floating point insn) on
the cycle but can finish only two insns.  To describe this, we define
the following functional units.

@smallexample
(define_cpu_unit "i0_pipeline, i1_pipeline, f_pipeline")
(define_cpu_unit "port0, port1")
@end smallexample

All simple integer insns can be executed in any integer pipeline and
their result is ready in two cycles.  The simple integer insns are
issued into the first pipeline unless it is reserved, otherwise they
are issued into the second pipeline.  Integer division and
multiplication insns can be executed only in the second integer
pipeline and their results are ready correspondingly in 8 and 4
cycles.  The integer division is not pipelined, i.e.@: the subsequent
integer division insn can not be issued until the current division
insn finished.  Floating point insns are fully pipelined and their
results are ready in 3 cycles.  Where the result of a floating point
insn is used by an integer insn, an additional delay of one cycle is
incurred.  To describe all of this we could specify

@smallexample
(define_cpu_unit "div")

(define_insn_reservation "simple" 2 (eq_attr "type" "int")
                         "(i0_pipeline | i1_pipeline), (port0 | port1)")

(define_insn_reservation "mult" 4 (eq_attr "type" "mult")
                         "i1_pipeline, nothing*2, (port0 | port1)")

(define_insn_reservation "div" 8 (eq_attr "type" "div")
                         "i1_pipeline, div*7, div + (port0 | port1)")

(define_insn_reservation "float" 3 (eq_attr "type" "float")
                         "f_pipeline, nothing, (port0 | port1))

(define_bypass 4 "float" "simple,mult,div")
@end smallexample

To simplify the description we could describe the following reservation

@smallexample
(define_reservation "finish" "port0|port1")
@end smallexample

and use it in all @code{define_insn_reservation} as in the following
construction

@smallexample
(define_insn_reservation "simple" 2 (eq_attr "type" "int")
                         "(i0_pipeline | i1_pipeline), finish")
@end smallexample


@end ifset
@ifset INTERNALS
@node Conditional Execution
@section Conditional Execution
@cindex conditional execution
@cindex predication

A number of architectures provide for some form of conditional
execution, or predication.  The hallmark of this feature is the
ability to nullify most of the instructions in the instruction set.
When the instruction set is large and not entirely symmetric, it
can be quite tedious to describe these forms directly in the
@file{.md} file.  An alternative is the @code{define_cond_exec} template.

@findex define_cond_exec
@smallexample
(define_cond_exec
  [@var{predicate-pattern}]
  "@var{condition}"
  "@var{output-template}")
@end smallexample

@var{predicate-pattern} is the condition that must be true for the
insn to be executed at runtime and should match a relational operator.
One can use @code{match_operator} to match several relational operators
at once.  Any @code{match_operand} operands must have no more than one
alternative.

@var{condition} is a C expression that must be true for the generated
pattern to match.

@findex current_insn_predicate
@var{output-template} is a string similar to the @code{define_insn}
output template (@pxref{Output Template}), except that the @samp{*}
and @samp{@@} special cases do not apply.  This is only useful if the
assembly text for the predicate is a simple prefix to the main insn.
In order to handle the general case, there is a global variable
@code{current_insn_predicate} that will contain the entire predicate
if the current insn is predicated, and will otherwise be @code{NULL}.

When @code{define_cond_exec} is used, an implicit reference to
the @code{predicable} instruction attribute is made.
@xref{Insn Attributes}.  This attribute must be boolean (i.e.@: have
exactly two elements in its @var{list-of-values}).  Further, it must
not be used with complex expressions.  That is, the default and all
uses in the insns must be a simple constant, not dependent on the
alternative or anything else.

For each @code{define_insn} for which the @code{predicable}
attribute is true, a new @code{define_insn} pattern will be
generated that matches a predicated version of the instruction.
For example,

@smallexample
(define_insn "addsi"
  [(set (match_operand:SI 0 "register_operand" "r")
        (plus:SI (match_operand:SI 1 "register_operand" "r")
                 (match_operand:SI 2 "register_operand" "r")))]
  "@var{test1}"
  "add %2,%1,%0")

(define_cond_exec
  [(ne (match_operand:CC 0 "register_operand" "c")
       (const_int 0))]
  "@var{test2}"
  "(%0)")
@end smallexample

@noindent
generates a new pattern

@smallexample
(define_insn ""
  [(cond_exec
     (ne (match_operand:CC 3 "register_operand" "c") (const_int 0))
     (set (match_operand:SI 0 "register_operand" "r")
          (plus:SI (match_operand:SI 1 "register_operand" "r")
                   (match_operand:SI 2 "register_operand" "r"))))]
  "(@var{test2}) && (@var{test1})"
  "(%3) add %2,%1,%0")
@end smallexample

@end ifset
@ifset INTERNALS
@node Constant Definitions
@section Constant Definitions
@cindex constant definitions
@findex define_constants

Using literal constants inside instruction patterns reduces legibility and
can be a maintenance problem.

To overcome this problem, you may use the @code{define_constants}
expression.  It contains a vector of name-value pairs.  From that
point on, wherever any of the names appears in the MD file, it is as
if the corresponding value had been written instead.  You may use
@code{define_constants} multiple times; each appearance adds more
constants to the table.  It is an error to redefine a constant with
a different value.

To come back to the a29k load multiple example, instead of

@smallexample
(define_insn ""
  [(match_parallel 0 "load_multiple_operation"
     [(set (match_operand:SI 1 "gpc_reg_operand" "=r")
           (match_operand:SI 2 "memory_operand" "m"))
      (use (reg:SI 179))
      (clobber (reg:SI 179))])]
  ""
  "loadm 0,0,%1,%2")
@end smallexample

You could write:

@smallexample
(define_constants [
    (R_BP 177)
    (R_FC 178)
    (R_CR 179)
    (R_Q  180)
])

(define_insn ""
  [(match_parallel 0 "load_multiple_operation"
     [(set (match_operand:SI 1 "gpc_reg_operand" "=r")
           (match_operand:SI 2 "memory_operand" "m"))
      (use (reg:SI R_CR))
      (clobber (reg:SI R_CR))])]
  ""
  "loadm 0,0,%1,%2")
@end smallexample

The constants that are defined with a define_constant are also output
in the insn-codes.h header file as #defines.
@end ifset
@ifset INTERNALS
@node Macros
@section Macros
@cindex macros in @file{.md} files

Ports often need to define similar patterns for more than one machine
mode or for more than one rtx code.  GCC provides some simple macro
facilities to make this process easier.

@menu
* Mode Macros::         Generating variations of patterns for different modes.
* Code Macros::         Doing the same for codes.
@end menu

@node Mode Macros
@subsection Mode Macros
@cindex mode macros in @file{.md} files

Ports often need to define similar patterns for two or more different modes.
For example:

@itemize @bullet
@item
If a processor has hardware support for both single and double
floating-point arithmetic, the @code{SFmode} patterns tend to be
very similar to the @code{DFmode} ones.

@item
If a port uses @code{SImode} pointers in one configuration and
@code{DImode} pointers in another, it will usually have very similar
@code{SImode} and @code{DImode} patterns for manipulating pointers.
@end itemize

Mode macros allow several patterns to be instantiated from one
@file{.md} file template.  They can be used with any type of
rtx-based construct, such as a @code{define_insn},
@code{define_split}, or @code{define_peephole2}.

@menu
* Defining Mode Macros:: Defining a new mode macro.
* Substitutions::        Combining mode macros with substitutions
* Examples::             Examples
@end menu

@node Defining Mode Macros
@subsubsection Defining Mode Macros
@findex define_mode_macro

The syntax for defining a mode macro is:

@smallexample
(define_mode_macro @var{name} [(@var{mode1} "@var{cond1}") ... (@var{moden} "@var{condn}")])
@end smallexample

This allows subsequent @file{.md} file constructs to use the mode suffix
@code{:@var{name}}.  Every construct that does so will be expanded
@var{n} times, once with every use of @code{:@var{name}} replaced by
@code{:@var{mode1}}, once with every use replaced by @code{:@var{mode2}},
and so on.  In the expansion for a particular @var{modei}, every
C condition will also require that @var{condi} be true.

For example:

@smallexample
(define_mode_macro P [(SI "Pmode == SImode") (DI "Pmode == DImode")])
@end smallexample

defines a new mode suffix @code{:P}.  Every construct that uses
@code{:P} will be expanded twice, once with every @code{:P} replaced
by @code{:SI} and once with every @code{:P} replaced by @code{:DI}.
The @code{:SI} version will only apply if @code{Pmode == SImode} and
the @code{:DI} version will only apply if @code{Pmode == DImode}.

As with other @file{.md} conditions, an empty string is treated
as ``always true''.  @code{(@var{mode} "")} can also be abbreviated
to @code{@var{mode}}.  For example:

@smallexample
(define_mode_macro GPR [SI (DI "TARGET_64BIT")])
@end smallexample

means that the @code{:DI} expansion only applies if @code{TARGET_64BIT}
but that the @code{:SI} expansion has no such constraint.

Macros are applied in the order they are defined.  This can be
significant if two macros are used in a construct that requires
substitutions.  @xref{Substitutions}.

@node Substitutions
@subsubsection Substitution in Mode Macros
@findex define_mode_attr

If an @file{.md} file construct uses mode macros, each version of the
construct will often need slightly different strings or modes.  For
example:

@itemize @bullet
@item
When a @code{define_expand} defines several @code{add@var{m}3} patterns
(@pxref{Standard Names}), each expander will need to use the
appropriate mode name for @var{m}.

@item
When a @code{define_insn} defines several instruction patterns,
each instruction will often use a different assembler mnemonic.

@item
When a @code{define_insn} requires operands with different modes,
using a macro for one of the operand modes usually requires a specific
mode for the other operand(s).
@end itemize

GCC supports such variations through a system of ``mode attributes''.
There are two standard attributes: @code{mode}, which is the name of
the mode in lower case, and @code{MODE}, which is the same thing in
upper case.  You can define other attributes using:

@smallexample
(define_mode_attr @var{name} [(@var{mode1} "@var{value1}") ... (@var{moden} "@var{valuen}")])
@end smallexample

where @var{name} is the name of the attribute and @var{valuei}
is the value associated with @var{modei}.

When GCC replaces some @var{:macro} with @var{:mode}, it will scan
each string and mode in the pattern for sequences of the form
@code{<@var{macro}:@var{attr}>}, where @var{attr} is the name of a
mode attribute.  If the attribute is defined for @var{mode}, the whole
@code{<...>} sequence will be replaced by the appropriate attribute
value.

For example, suppose an @file{.md} file has:

@smallexample
(define_mode_macro P [(SI "Pmode == SImode") (DI "Pmode == DImode")])
(define_mode_attr load [(SI "lw") (DI "ld")])
@end smallexample

If one of the patterns that uses @code{:P} contains the string
@code{"<P:load>\t%0,%1"}, the @code{SI} version of that pattern
will use @code{"lw\t%0,%1"} and the @code{DI} version will use
@code{"ld\t%0,%1"}.

Here is an example of using an attribute for a mode:

@smallexample
(define_mode_macro LONG [SI DI])
(define_mode_attr SHORT [(SI "HI") (DI "SI")])
(define_insn ...
  (sign_extend:LONG (match_operand:<LONG:SHORT> ...)) ...)
@end smallexample

The @code{@var{macro}:} prefix may be omitted, in which case the
substitution will be attempted for every macro expansion.

@node Examples
@subsubsection Mode Macro Examples

Here is an example from the MIPS port.  It defines the following
modes and attributes (among others):

@smallexample
(define_mode_macro GPR [SI (DI "TARGET_64BIT")])
(define_mode_attr d [(SI "") (DI "d")])
@end smallexample

and uses the following template to define both @code{subsi3}
and @code{subdi3}:

@smallexample
(define_insn "sub<mode>3"
  [(set (match_operand:GPR 0 "register_operand" "=d")
        (minus:GPR (match_operand:GPR 1 "register_operand" "d")
                   (match_operand:GPR 2 "register_operand" "d")))]
  ""
  "<d>subu\t%0,%1,%2"
  [(set_attr "type" "arith")
   (set_attr "mode" "<MODE>")])
@end smallexample

This is exactly equivalent to:

@smallexample
(define_insn "subsi3"
  [(set (match_operand:SI 0 "register_operand" "=d")
        (minus:SI (match_operand:SI 1 "register_operand" "d")
                  (match_operand:SI 2 "register_operand" "d")))]
  ""
  "subu\t%0,%1,%2"
  [(set_attr "type" "arith")
   (set_attr "mode" "SI")])

(define_insn "subdi3"
  [(set (match_operand:DI 0 "register_operand" "=d")
        (minus:DI (match_operand:DI 1 "register_operand" "d")
                  (match_operand:DI 2 "register_operand" "d")))]
  ""
  "dsubu\t%0,%1,%2"
  [(set_attr "type" "arith")
   (set_attr "mode" "DI")])
@end smallexample

@node Code Macros
@subsection Code Macros
@cindex code macros in @file{.md} files
@findex define_code_macro
@findex define_code_attr

Code macros operate in a similar way to mode macros.  @xref{Mode Macros}.

The construct:

@smallexample
(define_code_macro @var{name} [(@var{code1} "@var{cond1}") ... (@var{coden} "@var{condn}")])
@end smallexample

defines a pseudo rtx code @var{name} that can be instantiated as
@var{codei} if condition @var{condi} is true.  Each @var{codei}
must have the same rtx format.  @xref{RTL Classes}.

As with mode macros, each pattern that uses @var{name} will be
expanded @var{n} times, once with all uses of @var{name} replaced by
@var{code1}, once with all uses replaced by @var{code2}, and so on.
@xref{Defining Mode Macros}.

It is possible to define attributes for codes as well as for modes.
There are two standard code attributes: @code{code}, the name of the
code in lower case, and @code{CODE}, the name of the code in upper case.
Other attributes are defined using:

@smallexample
(define_code_attr @var{name} [(@var{code1} "@var{value1}") ... (@var{coden} "@var{valuen}")])
@end smallexample

Here's an example of code macros in action, taken from the MIPS port:

@smallexample
(define_code_macro any_cond [unordered ordered unlt unge uneq ltgt unle ungt
                             eq ne gt ge lt le gtu geu ltu leu])

(define_expand "b<code>"
  [(set (pc)
        (if_then_else (any_cond:CC (cc0)
                                   (const_int 0))
                      (label_ref (match_operand 0 ""))
                      (pc)))]
  ""
@{
  gen_conditional_branch (operands, <CODE>);
  DONE;
@})
@end smallexample

This is equivalent to:

@smallexample
(define_expand "bunordered"
  [(set (pc)
        (if_then_else (unordered:CC (cc0)
                                    (const_int 0))
                      (label_ref (match_operand 0 ""))
                      (pc)))]
  ""
@{
  gen_conditional_branch (operands, UNORDERED);
  DONE;
@})

(define_expand "bordered"
  [(set (pc)
        (if_then_else (ordered:CC (cc0)
                                  (const_int 0))
                      (label_ref (match_operand 0 ""))
                      (pc)))]
  ""
@{
  gen_conditional_branch (operands, ORDERED);
  DONE;
@})

...
@end smallexample

@end ifset