Chris Lattner | ce52b7e | 2004-06-01 06:48:00 +0000 | [diff] [blame] | 1 | <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01//EN" |
| 2 | "http://www.w3.org/TR/html4/strict.dtd"> |
| 3 | <html> |
| 4 | <head> |
| 5 | <title>The LLVM Target-Independent Code Generator</title> |
| 6 | <link rel="stylesheet" href="llvm.css" type="text/css"> |
| 7 | </head> |
| 8 | <body> |
| 9 | |
| 10 | <div class="doc_title"> |
| 11 | The LLVM Target-Independent Code Generator |
| 12 | </div> |
| 13 | |
| 14 | <ol> |
| 15 | <li><a href="#introduction">Introduction</a> |
| 16 | <ul> |
| 17 | <li><a href="#required">Required components in the code generator</a></li> |
| 18 | <li><a href="#high-level-design">The high-level design of the code generator</a></li> |
| 19 | <li><a href="#tablegen">Using TableGen for target description</a></li> |
| 20 | </ul> |
| 21 | </li> |
| 22 | <li><a href="#targetdesc">Target description classes</a> |
| 23 | <ul> |
| 24 | <li><a href="#targetmachine">The <tt>TargetMachine</tt> class</a></li> |
| 25 | <li><a href="#targetdata">The <tt>TargetData</tt> class</a></li> |
| 26 | <li><a href="#mregisterinfo">The <tt>MRegisterInfo</tt> class</a></li> |
| 27 | <li><a href="#targetinstrinfo">The <tt>TargetInstrInfo</tt> class</a></li> |
| 28 | <li><a href="#targetframeinfo">The <tt>TargetFrameInfo</tt> class</a></li> |
| 29 | <li><a href="#targetjitinfo">The <tt>TargetJITInfo</tt> class</a></li> |
| 30 | </ul> |
| 31 | </li> |
| 32 | <li><a href="#codegendesc">Machine code description classes</a> |
| 33 | </li> |
| 34 | <li><a href="#codegenalgs">Target-independent code generation algorithms</a> |
| 35 | </li> |
| 36 | <li><a href="#targetimpls">Target description implementations</a> |
| 37 | <ul> |
| 38 | <li><a href="#x86">The X86 backend</a></li> |
Chris Lattner | 10d6800 | 2004-06-01 17:18:11 +0000 | [diff] [blame^] | 39 | </ul> |
Chris Lattner | ce52b7e | 2004-06-01 06:48:00 +0000 | [diff] [blame] | 40 | </li> |
| 41 | |
| 42 | </ol> |
| 43 | |
| 44 | <div class="doc_author"> |
| 45 | <p>Written by <a href="mailto:sabre@nondot.org">Chris Lattner</a></p> |
| 46 | </div> |
| 47 | |
Chris Lattner | 10d6800 | 2004-06-01 17:18:11 +0000 | [diff] [blame^] | 48 | <div class="doc_warning"> |
| 49 | <p>Warning: This is a work in progress.</p> |
| 50 | </div> |
| 51 | |
Chris Lattner | ce52b7e | 2004-06-01 06:48:00 +0000 | [diff] [blame] | 52 | <!-- *********************************************************************** --> |
| 53 | <div class="doc_section"> |
| 54 | <a name="introduction">Introduction</a> |
| 55 | </div> |
| 56 | <!-- *********************************************************************** --> |
| 57 | |
| 58 | <div class="doc_text"> |
| 59 | |
| 60 | <p>The LLVM target-independent code generator is a framework that provides a |
| 61 | suite of reusable components for translating the LLVM internal representation to |
| 62 | the machine code for a specified target -- either in assembly form (suitable for |
| 63 | a static compiler) or in binary machine code format (usable for a JIT compiler). |
| 64 | The LLVM target-independent code generator consists of four main components:</p> |
| 65 | |
| 66 | <ol> |
| 67 | <li><a href="#targetdesc">Abstract target description</a> interfaces which |
| 68 | capture improtant properties about various aspects of the machine independently |
| 69 | of how they will be used. These interfaces are defined in |
| 70 | <tt>include/llvm/Target/</tt>.</li> |
| 71 | |
| 72 | <li>Classes used to represent the <a href="#codegendesc">machine code</a> being |
| 73 | generator for a target. These classes are intended to be abstract enough to |
| 74 | represent the machine code for <i>any</i> target machine. These classes are |
| 75 | defined in <tt>include/llvm/CodeGen/</tt>.</li> |
| 76 | |
| 77 | <li><a href="#codegenalgs">Target-independent algorithms</a> used to implement |
| 78 | various phases of native code generation (register allocation, scheduling, stack |
| 79 | frame representation, etc). This code lives in <tt>lib/CodeGen/</tt>.</li> |
| 80 | |
| 81 | <li><a href="#targetimpls">Implementations of the abstract target description |
| 82 | interfaces</a> for particular targets. These machine descriptions make use of |
| 83 | the components provided by LLVM, and can optionally provide custom |
| 84 | target-specific passes, to build complete code generators for a specific target. |
| 85 | Target descriptions live in <tt>lib/Target/</tt>.</li> |
| 86 | |
| 87 | </ol> |
| 88 | |
| 89 | <p> |
| 90 | Depending on which part of the code generator you are interested in working on, |
| 91 | different pieces of this will be useful to you. In any case, you should be |
| 92 | familiar with the <a href="#targetdesc">target description</a> and <a |
| 93 | href="#codegendesc">machine code representation</a> classes. If you want to add |
| 94 | a backend for a new target, you will need <a href="#targetimpls">implement the |
| 95 | targe description</a> classes for your new target and understand the <a |
| 96 | href="LangRef.html">LLVM code representation</a>. If you are interested in |
| 97 | implementing a new <a href="#codegenalgs">code generation algorithm</a>, it |
| 98 | should only depend on the target-description and machine code representation |
| 99 | classes, ensuring that it is portable. |
| 100 | </p> |
| 101 | |
| 102 | </div> |
| 103 | |
| 104 | <!-- ======================================================================= --> |
| 105 | <div class="doc_subsection"> |
| 106 | <a name="required">Required components in the code generator</a> |
| 107 | </div> |
| 108 | |
| 109 | <div class="doc_text"> |
| 110 | |
| 111 | <p>The two pieces of the LLVM code generator are the high-level interface to the |
| 112 | code generator and the set of reusable components that can be used to build |
| 113 | target-specific backends. The two most important interfaces (<a |
| 114 | href="#targetmachine"><tt>TargetMachine</tt></a> and <a |
| 115 | href="#targetdata"><tt>TargetData</tt></a> classes) are the only ones that are |
| 116 | required to be defined for a backend to fit into the LLVM system, but the others |
| 117 | must be defined if the reusable code generator components are going to be |
| 118 | used.</p> |
| 119 | |
| 120 | <p>This design has two important implications. The first is that LLVM can |
| 121 | support completely non-traditional code generation targets. For example, the C |
| 122 | backend does not require register allocation, instruction selection, or any of |
| 123 | the other standard components provided by the system. As such, it only |
| 124 | implements these two interfaces, and does its own thing. Another example of a |
| 125 | code generator like this is a (purely hypothetical) backend that converts LLVM |
| 126 | to the GCC RTL form and uses GCC to emit machine code for a target.</p> |
| 127 | |
| 128 | <p>The other implication of this design is that it is possible to design and |
| 129 | implement radically different code generators in the LLVM system that do not |
| 130 | make use of any of the built-in components. Doing so is not recommended at all, |
| 131 | but could be required for radically different targets that do not fit into the |
| 132 | LLVM machine description model: programmable FPGAs for example.</p> |
Chris Lattner | ce52b7e | 2004-06-01 06:48:00 +0000 | [diff] [blame] | 133 | </div> |
| 134 | |
| 135 | <!-- ======================================================================= --> |
| 136 | <div class="doc_subsection"> |
Chris Lattner | 10d6800 | 2004-06-01 17:18:11 +0000 | [diff] [blame^] | 137 | <a name="high-level-design">The high-level design of the code generator</a> |
Chris Lattner | ce52b7e | 2004-06-01 06:48:00 +0000 | [diff] [blame] | 138 | </div> |
| 139 | |
| 140 | <div class="doc_text"> |
| 141 | |
| 142 | <p>The LLVM target-indendent code generator is designed to support efficient and |
| 143 | quality code generation for standard register-based microprocessors. Code |
| 144 | generation in this model is divided into the following stages:</p> |
| 145 | |
| 146 | <ol> |
| 147 | <li><b>Instruction Selection</b> - Determining a efficient implementation of the |
| 148 | input LLVM code in the target instruction set. This stage produces the initial |
| 149 | code for the program in the target instruction set the makes use of virtual |
| 150 | registers in SSA form and physical registers that represent any required |
| 151 | register assignments due to target constraints or calling conventions.</li> |
| 152 | |
| 153 | <li><b>SSA-based Machine Code Optimizations</b> - This (optional) stage consists |
| 154 | of a series of machine-code optimizations that operate on the SSA-form produced |
| 155 | by the instruction selector. Optimizations like modulo-scheduling, normal |
| 156 | scheduling, or peephole optimization work here.</li> |
| 157 | |
| 158 | <li><b>Register Allocation</b> - The target code is transformed from an infinite |
| 159 | virtual register file in SSA form to the concrete register file used by the |
| 160 | target. This phase introduces spill code and eliminates all virtual register |
| 161 | references from the program.</li> |
| 162 | |
| 163 | <li><b>Prolog/Epilog Code Insertion</b> - Once the machine code has been |
| 164 | generated for the function and the amount of stack space required is known (used |
| 165 | for LLVM alloca's and spill slots), the prolog and epilog code for the function |
| 166 | can be inserted and "abstract stack location references" can be eliminated. |
| 167 | This stage is responsible for implementing optimizations like frame-pointer |
| 168 | elimination and stack packing.</li> |
| 169 | |
| 170 | <li><b>Late Machine Code Optimizations</b> - Optimizations that operate on |
| 171 | "final" machine code can go here, such as spill code scheduling and peephole |
| 172 | optimizations.</li> |
| 173 | |
| 174 | <li><b>Code Emission</b> - The final stage actually outputs the machine code for |
| 175 | the current function, either in the target assembler format or in machine |
| 176 | code.</li> |
| 177 | |
| 178 | </ol> |
| 179 | |
| 180 | <p> |
| 181 | The code generator is based on the assumption that the instruction selector will |
| 182 | use an optimal pattern matching selector to create high-quality sequences of |
| 183 | native code. Alternative code generator designs based on pattern expansion and |
| 184 | aggressive iterative peephole optimization are much slower. This design is |
| 185 | designed to permit efficient compilation (important for JIT environments) and |
| 186 | aggressive optimization (used when generate code offline) by allowing components |
| 187 | of varying levels of sophisication to be used for any step of compilation.</p> |
| 188 | |
| 189 | <p> |
| 190 | In addition to these stages, target implementations can insert arbitrary |
| 191 | target-specific passes into the flow. For example, the X86 target uses a |
| 192 | special pass to handle the 80x87 floating point stack architecture. Other |
| 193 | targets with unusual requirements can be supported with custom passes as needed. |
| 194 | </p> |
| 195 | |
| 196 | </div> |
| 197 | |
| 198 | |
| 199 | <!-- ======================================================================= --> |
| 200 | <div class="doc_subsection"> |
Chris Lattner | 10d6800 | 2004-06-01 17:18:11 +0000 | [diff] [blame^] | 201 | <a name="tablegen">Using TableGen for target description</a> |
Chris Lattner | ce52b7e | 2004-06-01 06:48:00 +0000 | [diff] [blame] | 202 | </div> |
| 203 | |
| 204 | <div class="doc_text"> |
| 205 | |
| 206 | <p>The target description classes require a detailed descriptions of the target |
| 207 | architecture. These target descriptions often have a large amount of common |
| 208 | information (e.g., an add instruction is almost identical to a sub instruction). |
| 209 | In order to allow the maximum amount of commonality to be factored out, the LLVM |
| 210 | code generator uses the <a href="TableGenFundamentals.html">TableGen</a> tool to |
| 211 | allow |
| 212 | </p> |
| 213 | |
| 214 | </div> |
| 215 | |
| 216 | <!-- *********************************************************************** --> |
| 217 | <div class="doc_section"> |
| 218 | <a name="targetdesc">Target description classes</a> |
| 219 | </div> |
| 220 | <!-- *********************************************************************** --> |
| 221 | |
| 222 | <div class="doc_text"> |
| 223 | |
| 224 | <p>The LLVM target description classes (which are located in the |
| 225 | <tt>include/llvm/Target</tt> directory) provide an abstract description of the |
| 226 | target machine, independent of any particular client. These classes are |
| 227 | designed to capture the <i>abstract</i> properties of the target (such as what |
| 228 | instruction and registers it has), and do not incorporate any particular pieces |
| 229 | of code generation algorithms (these interfaces do not take interference graphs |
| 230 | as inputs or other algorithm-specific data structures).</p> |
| 231 | |
| 232 | <p>All of the target description classes (except the <tt><a |
| 233 | href="#targetdata">TargetData</a></tt> class) are designed to be subclassed by |
| 234 | the concrete target implementation, and have virtual methods implemented. To |
| 235 | get to these implementations, <tt><a |
| 236 | href="#targetmachine">TargetMachine</a></tt> class provides accessors that |
| 237 | should be implemented by the target.</p> |
| 238 | |
| 239 | </div> |
| 240 | |
| 241 | <!-- ======================================================================= --> |
| 242 | <div class="doc_subsection"> |
| 243 | <a name="targetmachine">The <tt>TargetMachine</tt> class</a> |
| 244 | </div> |
| 245 | |
| 246 | <div class="doc_text"> |
| 247 | |
| 248 | <p>The <tt>TargetMachine</tt> class provides virtual methods that are used to |
| 249 | access the target-specific implementations of the various target description |
| 250 | classes (with the <tt>getInstrInfo</tt>, <tt>getRegisterInfo</tt>, |
| 251 | <tt>getFrameInfo</tt>, ... methods). This class is designed to be subclassed by |
| 252 | a concrete target implementation (e.g., <tt>X86TargetMachine</tt>) which |
| 253 | implements the various virtual methods. The only required target description |
| 254 | class is the <a href="#targetdata"><tt>TargetData</tt></a> class, but if the |
| 255 | code generator components are to be used, the other interfaces should be |
| 256 | implemented as well.</p> |
| 257 | |
| 258 | </div> |
| 259 | |
| 260 | |
| 261 | <!-- ======================================================================= --> |
| 262 | <div class="doc_subsection"> |
| 263 | <a name="targetdata">The <tt>TargetData</tt> class</a> |
| 264 | </div> |
| 265 | |
| 266 | <div class="doc_text"> |
| 267 | |
| 268 | <p>The <tt>TargetData</tt> class is the only required target description class, |
| 269 | and it is the only class that is not extensible (it cannot be derived from). It |
| 270 | specifies information about how the target lays out memory for structures, the |
| 271 | alignment requirements for various data types, the size of pointers in the |
| 272 | target, and whether the target is little- or big-endian.</p> |
| 273 | |
| 274 | </div> |
| 275 | |
| 276 | |
| 277 | <!-- ======================================================================= --> |
| 278 | <div class="doc_subsection"> |
Chris Lattner | 10d6800 | 2004-06-01 17:18:11 +0000 | [diff] [blame^] | 279 | <a name="mregisterinfo">The <tt>MRegisterInfo</tt> class</a> |
Chris Lattner | ce52b7e | 2004-06-01 06:48:00 +0000 | [diff] [blame] | 280 | </div> |
| 281 | |
| 282 | <div class="doc_text"> |
| 283 | |
| 284 | <p>The <tt>MRegisterInfo</tt> class (which will eventually be renamed to |
| 285 | <tt>TargetRegisterInfo</tt>) is used to describe the register file of the |
| 286 | target and any interactions between the registers.</p> |
| 287 | |
| 288 | <p>Registers in the code generator are represented in the code generator by |
| 289 | unsigned numbers. Physical registers (those that actually exist in the target |
| 290 | description) are unique small numbers, and virtual registers are generally |
| 291 | large.</p> |
| 292 | |
| 293 | <p>Each register in the processor description has an associated |
| 294 | <tt>MRegisterDesc</tt> entry, which provides a textual name for the register |
| 295 | (used for assembly output and debugging dumps), a set of aliases (used to |
| 296 | indicate that one register overlaps with another), and some flag bits. |
| 297 | </p> |
| 298 | |
| 299 | <p>In addition to the per-register description, the <tt>MRegisterInfo</tt> class |
| 300 | exposes a set of processor specific register classes (instances of the |
| 301 | <tt>TargetRegisterClass</tt> class). Each register class contains sets of |
| 302 | registers that have the same properties (for example, they are all 32-bit |
| 303 | integer registers). Each SSA virtual register created by the instruction |
| 304 | selector has an associated register class. When the register allocator runs, it |
| 305 | replaces virtual registers with a physical register in the set.</p> |
| 306 | |
| 307 | <p> |
| 308 | The target-specific implementations of these classes is auto-generated from a <a |
| 309 | href="TableGenFundamentals.html">TableGen</a> description of the register file. |
| 310 | </p> |
| 311 | |
| 312 | </div> |
| 313 | |
| 314 | <!-- ======================================================================= --> |
| 315 | <div class="doc_subsection"> |
Chris Lattner | 10d6800 | 2004-06-01 17:18:11 +0000 | [diff] [blame^] | 316 | <a name="targetinstrinfo">The <tt>TargetInstrInfo</tt> class</a> |
Chris Lattner | ce52b7e | 2004-06-01 06:48:00 +0000 | [diff] [blame] | 317 | </div> |
| 318 | |
| 319 | <!-- ======================================================================= --> |
| 320 | <div class="doc_subsection"> |
Chris Lattner | 10d6800 | 2004-06-01 17:18:11 +0000 | [diff] [blame^] | 321 | <a name="targetframeinfo">The <tt>TargetFrameInfo</tt> class</a> |
Chris Lattner | ce52b7e | 2004-06-01 06:48:00 +0000 | [diff] [blame] | 322 | </div> |
| 323 | |
| 324 | <!-- ======================================================================= --> |
| 325 | <div class="doc_subsection"> |
Chris Lattner | 10d6800 | 2004-06-01 17:18:11 +0000 | [diff] [blame^] | 326 | <a name="targetjitinfo">The <tt>TargetJITInfo</tt> class</a> |
Chris Lattner | ce52b7e | 2004-06-01 06:48:00 +0000 | [diff] [blame] | 327 | </div> |
| 328 | |
| 329 | <!-- *********************************************************************** --> |
| 330 | <div class="doc_section"> |
| 331 | <a name="codegendesc">Machine code description classes</a> |
| 332 | </div> |
| 333 | <!-- *********************************************************************** --> |
| 334 | |
| 335 | |
| 336 | |
| 337 | <!-- *********************************************************************** --> |
| 338 | <hr> |
| 339 | <address> |
| 340 | <a href="http://jigsaw.w3.org/css-validator/check/referer"><img |
| 341 | src="http://jigsaw.w3.org/css-validator/images/vcss" alt="Valid CSS!"></a> |
| 342 | <a href="http://validator.w3.org/check/referer"><img |
| 343 | src="http://www.w3.org/Icons/valid-html401" alt="Valid HTML 4.01!" /></a> |
| 344 | |
| 345 | <a href="mailto:sabre@nondot.org">Chris Lattner</a><br> |
| 346 | <a href="http://llvm.cs.uiuc.edu">The LLVM Compiler Infrastructure</a><br> |
| 347 | Last modified: $Date$ |
| 348 | </address> |
| 349 | |
| 350 | </body> |
| 351 | </html> |