blob: dfdc0528455a29710973b1b65d8dadf899c47c8c [file] [log] [blame]
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01//EN"
"http://www.w3.org/TR/html4/strict.dtd">
<html>
<head>
<title>The LLVM Target-Independent Code Generator</title>
<link rel="stylesheet" href="llvm.css" type="text/css">
</head>
<body>
<div class="doc_title">
The LLVM Target-Independent Code Generator
</div>
<ol>
<li><a href="#introduction">Introduction</a>
<ul>
<li><a href="#required">Required components in the code generator</a></li>
<li><a href="#high-level-design">The high-level design of the code generator</a></li>
<li><a href="#tablegen">Using TableGen for target description</a></li>
</ul>
</li>
<li><a href="#targetdesc">Target description classes</a>
<ul>
<li><a href="#targetmachine">The <tt>TargetMachine</tt> class</a></li>
<li><a href="#targetdata">The <tt>TargetData</tt> class</a></li>
<li><a href="#mregisterinfo">The <tt>MRegisterInfo</tt> class</a></li>
<li><a href="#targetinstrinfo">The <tt>TargetInstrInfo</tt> class</a></li>
<li><a href="#targetframeinfo">The <tt>TargetFrameInfo</tt> class</a></li>
<li><a href="#targetjitinfo">The <tt>TargetJITInfo</tt> class</a></li>
</ul>
</li>
<li><a href="#codegendesc">Machine code description classes</a>
<ul>
<li><a href="#machineinstr">The <tt>MachineInstr</tt> class</a></li>
</ul>
</li>
<li><a href="#codegenalgs">Target-independent code generation algorithms</a>
</li>
<li><a href="#targetimpls">Target description implementations</a>
<ul>
<li><a href="#x86">The X86 backend</a></li>
</ul>
</li>
</ol>
<div class="doc_author">
<p>Written by <a href="mailto:sabre@nondot.org">Chris Lattner</a></p>
</div>
<div class="doc_warning">
<p>Warning: This is a work in progress.</p>
</div>
<!-- *********************************************************************** -->
<div class="doc_section">
<a name="introduction">Introduction</a>
</div>
<!-- *********************************************************************** -->
<div class="doc_text">
<p>The LLVM target-independent code generator is a framework that provides a
suite of reusable components for translating the LLVM internal representation to
the machine code for a specified target -- either in assembly form (suitable for
a static compiler) or in binary machine code format (usable for a JIT compiler).
The LLVM target-independent code generator consists of five main components:</p>
<ol>
<li><a href="#targetdesc">Abstract target description</a> interfaces which
capture important properties about various aspects of the machine, independently
of how they will be used. These interfaces are defined in
<tt>include/llvm/Target/</tt>.</li>
<li>Classes used to represent the <a href="#codegendesc">machine code</a> being
generated for a target. These classes are intended to be abstract enough to
represent the machine code for <i>any</i> target machine. These classes are
defined in <tt>include/llvm/CodeGen/</tt>.</li>
<li><a href="#codegenalgs">Target-independent algorithms</a> used to implement
various phases of native code generation (register allocation, scheduling, stack
frame representation, etc). This code lives in <tt>lib/CodeGen/</tt>.</li>
<li><a href="#targetimpls">Implementations of the abstract target description
interfaces</a> for particular targets. These machine descriptions make use of
the components provided by LLVM, and can optionally provide custom
target-specific passes, to build complete code generators for a specific target.
Target descriptions live in <tt>lib/Target/</tt>.</li>
<li><a href="#jit">The target-independent JIT components</a>. The LLVM JIT is
completely target independent (it uses the <tt>TargetJITInfo</tt> structure to
interface for target-specific issues. The code for the target-independent
JIT lives in <tt>lib/ExecutionEngine/JIT</tt>.</li>
</ol>
<p>
Depending on which part of the code generator you are interested in working on,
different pieces of this will be useful to you. In any case, you should be
familiar with the <a href="#targetdesc">target description</a> and <a
href="#codegendesc">machine code representation</a> classes. If you want to add
a backend for a new target, you will need to <a href="#targetimpls">implement the
target description</a> classes for your new target and understand the <a
href="LangRef.html">LLVM code representation</a>. If you are interested in
implementing a new <a href="#codegenalgs">code generation algorithm</a>, it
should only depend on the target-description and machine code representation
classes, ensuring that it is portable.
</p>
</div>
<!-- ======================================================================= -->
<div class="doc_subsection">
<a name="required">Required components in the code generator</a>
</div>
<div class="doc_text">
<p>The two pieces of the LLVM code generator are the high-level interface to the
code generator and the set of reusable components that can be used to build
target-specific backends. The two most important interfaces (<a
href="#targetmachine"><tt>TargetMachine</tt></a> and <a
href="#targetdata"><tt>TargetData</tt></a> classes) are the only ones that are
required to be defined for a backend to fit into the LLVM system, but the others
must be defined if the reusable code generator components are going to be
used.</p>
<p>This design has two important implications. The first is that LLVM can
support completely non-traditional code generation targets. For example, the C
backend does not require register allocation, instruction selection, or any of
the other standard components provided by the system. As such, it only
implements these two interfaces, and does its own thing. Another example of a
code generator like this is a (purely hypothetical) backend that converts LLVM
to the GCC RTL form and uses GCC to emit machine code for a target.</p>
<p>This design also implies that it is possible to design and
implement radically different code generators in the LLVM system that do not
make use of any of the built-in components. Doing so is not recommended at all,
but could be required for radically different targets that do not fit into the
LLVM machine description model: programmable FPGAs for example.</p>
<p><b>Important Note:</b> For historical reasons, the LLVM SparcV9 code
generator uses almost entirely different code paths than described in this
document. For this reason, there are some deprecated interfaces (such as
<tt>TargetRegInfo</tt> and <tt>TargetSchedInfo</tt>), which are only used by the
V9 backend and should not be used by any other targets. Also, all code in the
<tt>lib/Target/SparcV9</tt> directory and subdirectories should be considered
deprecated, and should not be used as the basis for future code generator work.
The SparcV9 backend is slowly being merged into the rest of the
target-independent code generators, but this is a low-priority process with no
predictable completion date.</p>
</div>
<!-- ======================================================================= -->
<div class="doc_subsection">
<a name="high-level-design">The high-level design of the code generator</a>
</div>
<div class="doc_text">
<p>The LLVM target-indendent code generator is designed to support efficient and
quality code generation for standard register-based microprocessors. Code
generation in this model is divided into the following stages:</p>
<ol>
<li><b>Instruction Selection</b> - Determining an efficient implementation of the
input LLVM code in the target instruction set. This stage produces the initial
code for the program in the target instruction set, then makes use of virtual
registers in SSA form and physical registers that represent any required
register assignments due to target constraints or calling conventions.</li>
<li><b>SSA-based Machine Code Optimizations</b> - This (optional) stage consists
of a series of machine-code optimizations that operate on the SSA-form produced
by the instruction selector. Optimizations like modulo-scheduling, normal
scheduling, or peephole optimization work here.</li>
<li><b>Register Allocation</b> - The target code is transformed from an infinite
virtual register file in SSA form to the concrete register file used by the
target. This phase introduces spill code and eliminates all virtual register
references from the program.</li>
<li><b>Prolog/Epilog Code Insertion</b> - Once the machine code has been
generated for the function and the amount of stack space required is known (used
for LLVM alloca's and spill slots), the prolog and epilog code for the function
can be inserted and "abstract stack location references" can be eliminated.
This stage is responsible for implementing optimizations like frame-pointer
elimination and stack packing.</li>
<li><b>Late Machine Code Optimizations</b> - Optimizations that operate on
"final" machine code can go here, such as spill code scheduling and peephole
optimizations.</li>
<li><b>Code Emission</b> - The final stage actually outputs the code for
the current function, either in the target assembler format or in machine
code.</li>
</ol>
<p>
The code generator is based on the assumption that the instruction selector will
use an optimal pattern matching selector to create high-quality sequences of
native instructions. Alternative code generator designs based on pattern
expansion and
aggressive iterative peephole optimization are much slower. This design
permits efficient compilation (important for JIT environments) and
aggressive optimization (used when generating code offline) by allowing
components of varying levels of sophisication to be used for any step of
compilation.</p>
<p>
In addition to these stages, target implementations can insert arbitrary
target-specific passes into the flow. For example, the X86 target uses a
special pass to handle the 80x87 floating point stack architecture. Other
targets with unusual requirements can be supported with custom passes as needed.
</p>
</div>
<!-- ======================================================================= -->
<div class="doc_subsection">
<a name="tablegen">Using TableGen for target description</a>
</div>
<div class="doc_text">
<p>The target description classes require a detailed description of the target
architecture. These target descriptions often have a large amount of common
information (e.g., an add instruction is almost identical to a sub instruction).
In order to allow the maximum amount of commonality to be factored out, the LLVM
code generator uses the <a href="TableGenFundamentals.html">TableGen</a> tool to
describe big chunks of the target machine, which allows the use of domain- and
target-specific abstractions to reduce the amount of repetition.
</p>
</div>
<!-- *********************************************************************** -->
<div class="doc_section">
<a name="targetdesc">Target description classes</a>
</div>
<!-- *********************************************************************** -->
<div class="doc_text">
<p>The LLVM target description classes (which are located in the
<tt>include/llvm/Target</tt> directory) provide an abstract description of the
target machine, independent of any particular client. These classes are
designed to capture the <i>abstract</i> properties of the target (such as what
instruction and registers it has), and do not incorporate any particular pieces
of code generation algorithms (these interfaces do not take interference graphs
as inputs or other algorithm-specific data structures).</p>
<p>All of the target description classes (except the <tt><a
href="#targetdata">TargetData</a></tt> class) are designed to be subclassed by
the concrete target implementation, and have virtual methods implemented. To
get to these implementations, the <tt><a
href="#targetmachine">TargetMachine</a></tt> class provides accessors that
should be implemented by the target.</p>
</div>
<!-- ======================================================================= -->
<div class="doc_subsection">
<a name="targetmachine">The <tt>TargetMachine</tt> class</a>
</div>
<div class="doc_text">
<p>The <tt>TargetMachine</tt> class provides virtual methods that are used to
access the target-specific implementations of the various target description
classes (with the <tt>getInstrInfo</tt>, <tt>getRegisterInfo</tt>,
<tt>getFrameInfo</tt>, ... methods). This class is designed to be specialized by
a concrete target implementation (e.g., <tt>X86TargetMachine</tt>) which
implements the various virtual methods. The only required target description
class is the <a href="#targetdata"><tt>TargetData</tt></a> class, but if the
code generator components are to be used, the other interfaces should be
implemented as well.</p>
</div>
<!-- ======================================================================= -->
<div class="doc_subsection">
<a name="targetdata">The <tt>TargetData</tt> class</a>
</div>
<div class="doc_text">
<p>The <tt>TargetData</tt> class is the only required target description class,
and it is the only class that is not extensible (it cannot be derived from). It
specifies information about how the target lays out memory for structures, the
alignment requirements for various data types, the size of pointers in the
target, and whether the target is little- or big-endian.</p>
</div>
<!-- ======================================================================= -->
<div class="doc_subsection">
<a name="mregisterinfo">The <tt>MRegisterInfo</tt> class</a>
</div>
<div class="doc_text">
<p>The <tt>MRegisterInfo</tt> class (which will eventually be renamed to
<tt>TargetRegisterInfo</tt>) is used to describe the register file of the
target and any interactions between the registers.</p>
<p>Registers in the code generator are represented in the code generator by
unsigned numbers. Physical registers (those that actually exist in the target
description) are unique small numbers, and virtual registers are generally
large.</p>
<p>Each register in the processor description has an associated
<tt>MRegisterDesc</tt> entry, which provides a textual name for the register
(used for assembly output and debugging dumps), a set of aliases (used to
indicate that one register overlaps with another), and some flag bits.
</p>
<p>In addition to the per-register description, the <tt>MRegisterInfo</tt> class
exposes a set of processor specific register classes (instances of the
<tt>TargetRegisterClass</tt> class). Each register class contains sets of
registers that have the same properties (for example, they are all 32-bit
integer registers). Each SSA virtual register created by the instruction
selector has an associated register class. When the register allocator runs, it
replaces virtual registers with a physical register in the set.</p>
<p>
The target-specific implementations of these classes is auto-generated from a <a
href="TableGenFundamentals.html">TableGen</a> description of the register file.
</p>
</div>
<!-- ======================================================================= -->
<div class="doc_subsection">
<a name="targetinstrinfo">The <tt>TargetInstrInfo</tt> class</a>
</div>
<!-- ======================================================================= -->
<div class="doc_subsection">
<a name="targetframeinfo">The <tt>TargetFrameInfo</tt> class</a>
</div>
<!-- ======================================================================= -->
<div class="doc_subsection">
<a name="targetjitinfo">The <tt>TargetJITInfo</tt> class</a>
</div>
<!-- *********************************************************************** -->
<div class="doc_section">
<a name="codegendesc">Machine code description classes</a>
</div>
<!-- *********************************************************************** -->
<div class="doc_text">
<p>
At the high-level, LLVM code is translated to a machine specific representation
formed out of MachineFunction, MachineBasicBlock, and <a
href="#machineinstr"><tt>MachineInstr</tt></a> instances
(defined in include/llvm/CodeGen). This representation is completely target
agnostic, representing instructions in their most abstract form: an opcode and a
series of operands. This representation is designed to support both SSA
representation for machine code, as well as a register allocated, non-SSA form.
</p>
</div>
<!-- ======================================================================= -->
<div class="doc_subsection">
<a name="machineinstr">The <tt>MachineInstr</tt> class</a>
</div>
<div class="doc_text">
<p>Target machine instructions are represented as instances of the
<tt>MachineInstr</tt> class. This class is an extremely abstract way of
representing machine instructions. In particular, all it keeps track of is
an opcode number and some number of operands.</p>
<p>The opcode number is an simple unsigned number that only has meaning to a
specific backend. All of the instructions for a target should be defined in
the <tt>*InstrInfo.td</tt> file for the target, and the opcode enum values
are autogenerated from this description. The <tt>MachineInstr</tt> class does
not have any information about how to intepret the instruction (i.e., what the
semantics of the instruction are): for that you must refer to the
<tt><a href="#targetinstrinfo">TargetInstrInfo</a></tt> class.</p>
<p>The operands of a machine instruction can be of several different types:
they can be a register reference, constant integer, basic block reference, etc.
In addition, a machine operand should be marked as a def or a use of the value
(though only registers are allowed to be defs).</p>
<p>By convention, the LLVM code generator orders instruction operands so that
all register definitions come before the register uses, even on architectures
that are normally printed in other orders. For example, the sparc add
instruction: "<tt>add %i1, %i2, %i3</tt>" adds the "%i1", and "%i2" registers
and stores the result into the "%i3" register. In the LLVM code generator,
the operands should be stored as "<tt>%i3, %i1, %i2</tt>": with the destination
first.</p>
<p>Keeping destination operands at the beginning of the operand list has several
advantages. In particular, the debugging printer will print the instruction
like this:</p>
<pre>
%r3 = add %i1, %i2
</pre>
<p>If the first operand is a def, and it is also easier to <a
href="#buildmi">create instructions</a> whose only def is the first
operand.</p>
</div>
<!-- _______________________________________________________________________ -->
<div class="doc_subsubsection">
<a name="buildmi">Using the <tt>MachineInstrBuilder.h</tt> functions</a>
</div>
<div class="doc_text">
<p>Machine instructions are created by using the <tt>BuildMI</tt> functions,
located in the <tt>include/llvm/CodeGen/MachineInstrBuilder.h</tt> file. The
<tt>BuildMI</tt> functions make it easy to build arbitrary machine
instructions. Usage of the <tt>BuildMI</tt> functions look like this:
</p>
<pre>
// Create a 'DestReg = mov 42' (rendered in X86 assembly as 'mov DestReg, 42')
// instruction. The '1' specifies how many operands will be added.
MachineInstr *MI = BuildMI(X86::MOV32ri, 1, DestReg).addImm(42);
// Create the same instr, but insert it at the end of a basic block.
MachineBasicBlock &amp;MBB = ...
BuildMI(MBB, X86::MOV32ri, 1, DestReg).addImm(42);
// Create the same instr, but insert it before a specified iterator point.
MachineBasicBlock::iterator MBBI = ...
BuildMI(MBB, MBBI, X86::MOV32ri, 1, DestReg).addImm(42);
// Create a 'cmp Reg, 0' instruction, no destination reg.
MI = BuildMI(X86::CMP32ri, 2).addReg(Reg).addImm(0);
// Create an 'sahf' instruction which takes no operands and stores nothing.
MI = BuildMI(X86::SAHF, 0);
// Create a self looping branch instruction.
BuildMI(MBB, X86::JNE, 1).addMBB(&amp;MBB);
</pre>
<p>
The key thing to remember with the <tt>BuildMI</tt> functions is that you have
to specify the number of operands that the machine instruction will take
(allowing efficient memory allocation). Also, if operands default to be uses
of values, not definitions. If you need to add a definition operand (other
than the optional destination register), you must explicitly mark it as such.
</p>
</div>
<!-- _______________________________________________________________________ -->
<div class="doc_subsubsection">
<a name="fixedregs">Fixed (aka preassigned) registers</a>
</div>
<div class="doc_text">
<p>One important issue that the code generator needs to be aware of is the
presence of fixed registers. In particular, there are often places in the
instruction stream where the register allocator <em>must</em> arrange for a
particular value to be in a particular register. This can occur due to
limitations in the instruction set (e.g., the X86 can only do a 32-bit divide
with the <tt>EAX</tt>/<tt>EDX</tt> registers), or external factors like calling
conventions. In any case, the instruction selector should emit code that
copies a virtual register into or out of a physical register when needed.</p>
<p>For example, consider this simple LLVM example:</p>
<pre>
int %test(int %X, int %Y) {
%Z = div int %X, %Y
ret int %Z
}
</pre>
<p>The X86 instruction selector produces this machine code for the div
and ret (use
"<tt>llc X.bc -march=x86 -print-machineinstrs</tt>" to get this):</p>
<pre>
;; Start of div
%EAX = mov %reg1024 ;; Copy X (in reg1024) into EAX
%reg1027 = sar %reg1024, 31
%EDX = mov %reg1027 ;; Sign extend X into EDX
idiv %reg1025 ;; Divide by Y (in reg1025)
%reg1026 = mov %EAX ;; Read the result (Z) out of EAX
;; Start of ret
%EAX = mov %reg1026 ;; 32-bit return value goes in EAX
ret
</pre>
<p>By the end of code generation, the register allocator has coallesced
the registers and deleted the resultant identity moves, producing the
following code:</p>
<pre>
;; X is in EAX, Y is in ECX
mov %EAX, %EDX
sar %EDX, 31
idiv %ECX
ret
</pre>
<p>This approach is extremely general (if it can handle the X86 architecture,
it can handle anything!) and allows all of the target specific
knowledge about the instruction stream to be isolated in the instruction
selector. Note that physical registers should have a short lifetime for good
code generation, and all physical registers are assumed dead on entry and
exit of basic blocks (before register allocation). Thus if you need a value
to be live across basic block boundaries, it <em>must</em> live in a virtual
register.</p>
</div>
<!-- _______________________________________________________________________ -->
<div class="doc_subsubsection">
<a name="ssa">Machine code SSA form</a>
</div>
<div class="doc_text">
<p><tt>MachineInstr</tt>'s are initially instruction selected in SSA-form, and
are maintained in SSA-form until register allocation happens. For the most
part, this is trivially simple since LLVM is already in SSA form: LLVM PHI nodes
become machine code PHI nodes, and virtual registers are only allowed to have a
single definition.</p>
<p>After register allocation, machine code is no longer in SSA-form, as there
are no virtual registers left in the code.</p>
</div>
<!-- *********************************************************************** -->
<div class="doc_section">
<a name="targetimpls">Target description implementations</a>
</div>
<!-- *********************************************************************** -->
<div class="doc_text">
<p>This section of the document explains any features or design decisions that
are specific to the code generator for a particular target.</p>
</div>
<!-- ======================================================================= -->
<div class="doc_subsection">
<a name="x86">The X86 backend</a>
</div>
<div class="doc_text">
<p>
The X86 code generator lives in the <tt>lib/Target/X86</tt> directory. This
code generator currently targets a generic P6-like processor. As such, it
produces a few P6-and-above instructions (like conditional moves), but it does
not make use of newer features like MMX or SSE. In the future, the X86 backend
will have subtarget support added for specific processor families and
implementations.</p>
</div>
<!-- _______________________________________________________________________ -->
<div class="doc_subsubsection">
<a name="x86_memory">Representing X86 addressing modes in MachineInstrs</a>
</div>
<div class="doc_text">
<p>
The x86 has a very, uhm, flexible, way of accessing memory. It is capable of
forming memory addresses of the following expression directly in integer
instructions (which use ModR/M addressing):</p>
<pre>
Base+[1,2,4,8]*IndexReg+Disp32
</pre>
<p>Wow, that's crazy. In order to represent this, LLVM tracks no less that 4
operands for each memory operand of this form. This means that the "load" form
of 'mov' has the following "Operands" in this order:</p>
<pre>
Index: 0 | 1 2 3 4
Meaning: DestReg, | BaseReg, Scale, IndexReg, Displacement
OperandTy: VirtReg, | VirtReg, UnsImm, VirtReg, SignExtImm
</pre>
<p>Stores and all other instructions treat the four memory operands in the same
way, in the same order.</p>
</div>
<!-- _______________________________________________________________________ -->
<div class="doc_subsubsection">
<a name="x86_names">Instruction naming</a>
</div>
<div class="doc_text">
<p>
An instruction name consists of the base name, a default operand size
followed by a character per operand with an optional special size. For
example:</p>
<p>
<tt>ADD8rr</tt> -&gt; add, 8-bit register, 8-bit register<br>
<tt>IMUL16rmi</tt> -&gt; imul, 16-bit register, 16-bit memory, 16-bit immediate<br>
<tt>IMUL16rmi8</tt> -&gt; imul, 16-bit register, 16-bit memory, 8-bit immediate<br>
<tt>MOVSX32rm16</tt> -&gt; movsx, 32-bit register, 16-bit memory
</p>
</div>
<!-- *********************************************************************** -->
<hr>
<address>
<a href="http://jigsaw.w3.org/css-validator/check/referer"><img
src="http://jigsaw.w3.org/css-validator/images/vcss" alt="Valid CSS!"></a>
<a href="http://validator.w3.org/check/referer"><img
src="http://www.w3.org/Icons/valid-html401" alt="Valid HTML 4.01!" /></a>
<a href="mailto:sabre@nondot.org">Chris Lattner</a><br>
<a href="http://llvm.cs.uiuc.edu">The LLVM Compiler Infrastructure</a><br>
Last modified: $Date$
</address>
</body>
</html>