docs/GarbageCollection.html

   1 <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01//EN"
   2                       "http://www.w3.org/TR/html4/strict.dtd">
   3 <html>
   4 <head>
   5   <meta http-equiv="Content-Type" Content="text/html; charset=UTF-8" >
   6   <title>Accurate Garbage Collection with LLVM</title>
   7   <link rel="stylesheet" href="llvm.css" type="text/css">
   8   <style type="text/css">
   9     .rowhead { text-align: left; background: inherit; }
  10     .indent { padding-left: 1em; }
  11     .optl { color: #BFBFBF; }
  12   </style>
  13 </head>
  14 <body>
  15
  16 <h1>
  17   Accurate Garbage Collection with LLVM
  18 </h1>
  19
  20 <ol>
  21   <li><a href="#introduction">Introduction</a>
  22     <ul>
  23     <li><a href="#feature">Goals and non-goals</a></li>
  24     </ul>
  25   </li>
  26
  27   <li><a href="#quickstart">Getting started</a>
  28     <ul>
  29     <li><a href="#quickstart-compiler">In your compiler</a></li>
  30     <li><a href="#quickstart-runtime">In your runtime library</a></li>
  31     <li><a href="#shadow-stack">About the shadow stack</a></li>
  32     </ul>
  33   </li>
  34
  35   <li><a href="#core">Core support</a>
  36     <ul>
  37     <li><a href="#gcattr">Specifying GC code generation:
  38       <tt>gc "..."</tt></a></li>
  39     <li><a href="#gcroot">Identifying GC roots on the stack:
  40       <tt>llvm.gcroot</tt></a></li>
  41     <li><a href="#barriers">Reading and writing references in the heap</a>
  42       <ul>
  43       <li><a href="#gcwrite">Write barrier: <tt>llvm.gcwrite</tt></a></li>
  44       <li><a href="#gcread">Read barrier: <tt>llvm.gcread</tt></a></li>
  45       </ul>
  46     </li>
  47     </ul>
  48   </li>
  49
  50   <li><a href="#plugin">Compiler plugin interface</a>
  51     <ul>
  52     <li><a href="#collector-algos">Overview of available features</a></li>
  53     <li><a href="#stack-map">Computing stack maps</a></li>
  54     <li><a href="#init-roots">Initializing roots to null:
  55       <tt>InitRoots</tt></a></li>
  56     <li><a href="#custom">Custom lowering of intrinsics: <tt>CustomRoots</tt>,
  57       <tt>CustomReadBarriers</tt>, and <tt>CustomWriteBarriers</tt></a></li>
  58     <li><a href="#safe-points">Generating safe points:
  59       <tt>NeededSafePoints</tt></a></li>
  60     <li><a href="#assembly">Emitting assembly code:
  61       <tt>GCMetadataPrinter</tt></a></li>
  62     </ul>
  63   </li>
  64
  65   <li><a href="#runtime-impl">Implementing a collector runtime</a>
  66     <ul>
  67       <li><a href="#gcdescriptors">Tracing GC pointers from heap
  68       objects</a></li>
  69     </ul>
  70   </li>
  71
  72   <li><a href="#references">References</a></li>
  73
  74 </ol>
  75
  76 <div class="doc_author">
  77   <p>Written by <a href="mailto:sabre@nondot.org">Chris Lattner</a> and
  78      Gordon Henriksen</p>
  79 </div>
  80
  81 <!-- *********************************************************************** -->
  82 <h2>
  83   <a name="introduction">Introduction</a>
  84 </h2>
  85 <!-- *********************************************************************** -->
  86
  87 <div>
  88
  89 <p>Garbage collection is a widely used technique that frees the programmer from
  90 having to know the lifetimes of heap objects, making software easier to produce
  91 and maintain. Many programming languages rely on garbage collection for
  92 automatic memory management. There are two primary forms of garbage collection:
  93 conservative and accurate.</p>
  94
  95 <p>Conservative garbage collection often does not require any special support
  96 from either the language or the compiler: it can handle non-type-safe
  97 programming languages (such as C/C++) and does not require any special
  98 information from the compiler. The
  99 <a href="http://www.hpl.hp.com/personal/Hans_Boehm/gc/">Boehm collector</a> is
 100 an example of a state-of-the-art conservative collector.</p>
 101
 102 <p>Accurate garbage collection requires the ability to identify all pointers in
 103 the program at run-time (which requires that the source-language be type-safe in
 104 most cases). Identifying pointers at run-time requires compiler support to
 105 locate all places that hold live pointer variables at run-time, including the
 106 <a href="#gcroot">processor stack and registers</a>.</p>
 107
 108 <p>Conservative garbage collection is attractive because it does not require any
 109 special compiler support, but it does have problems. In particular, because the
 110 conservative garbage collector cannot <i>know</i> that a particular word in the
 111 machine is a pointer, it cannot move live objects in the heap (preventing the
 112 use of compacting and generational GC algorithms) and it can occasionally suffer
 113 from memory leaks due to integer values that happen to point to objects in the
 114 program. In addition, some aggressive compiler transformations can break
 115 conservative garbage collectors (though these seem rare in practice).</p>
 116
 117 <p>Accurate garbage collectors do not suffer from any of these problems, but
 118 they can suffer from degraded scalar optimization of the program. In particular,
 119 because the runtime must be able to identify and update all pointers active in
 120 the program, some optimizations are less effective. In practice, however, the
 121 locality and performance benefits of using aggressive garbage collection
 122 techniques dominates any low-level losses.</p>
 123
 124 <p>This document describes the mechanisms and interfaces provided by LLVM to
 125 support accurate garbage collection.</p>
 126
 127 <!-- ======================================================================= -->
 128 <h3>
 129   <a name="feature">Goals and non-goals</a>
 130 </h3>
 131
 132 <div>
 133
 134 <p>LLVM's intermediate representation provides <a href="#intrinsics">garbage
 135 collection intrinsics</a> that offer support for a broad class of
 136 collector models. For instance, the intrinsics permit:</p>
 137
 138 <ul>
 139   <li>semi-space collectors</li>
 140   <li>mark-sweep collectors</li>
 141   <li>generational collectors</li>
 142   <li>reference counting</li>
 143   <li>incremental collectors</li>
 144   <li>concurrent collectors</li>
 145   <li>cooperative collectors</li>
 146 </ul>
 147
 148 <p>We hope that the primitive support built into the LLVM IR is sufficient to
 149 support a broad class of garbage collected languages including Scheme, ML, Java,
 150 C#, Perl, Python, Lua, Ruby, other scripting languages, and more.</p>
 151
 152 <p>However, LLVM does not itself provide a garbage collector&mdash;this should
 153 be part of your language's runtime library. LLVM provides a framework for
 154 compile time <a href="#plugin">code generation plugins</a>. The role of these
 155 plugins is to generate code and data structures which conforms to the <em>binary
 156 interface</em> specified by the <em>runtime library</em>. This is similar to the
 157 relationship between LLVM and DWARF debugging info, for example. The
 158 difference primarily lies in the lack of an established standard in the domain
 159 of garbage collection&mdash;thus the plugins.</p>
 160
 161 <p>The aspects of the binary interface with which LLVM's GC support is
 162 concerned are:</p>
 163
 164 <ul>
 165   <li>Creation of GC-safe points within code where collection is allowed to
 166       execute safely.</li>
 167   <li>Computation of the stack map. For each safe point in the code, object
 168       references within the stack frame must be identified so that the
 169       collector may traverse and perhaps update them.</li>
 170   <li>Write barriers when storing object references to the heap. These are
 171       commonly used to optimize incremental scans in generational
 172       collectors.</li>
 173   <li>Emission of read barriers when loading object references. These are
 174       useful for interoperating with concurrent collectors.</li>
 175 </ul>
 176
 177 <p>There are additional areas that LLVM does not directly address:</p>
 178
 179 <ul>
 180   <li>Registration of global roots with the runtime.</li>
 181   <li>Registration of stack map entries with the runtime.</li>
 182   <li>The functions used by the program to allocate memory, trigger a
 183       collection, etc.</li>
 184   <li>Computation or compilation of type maps, or registration of them with
 185       the runtime. These are used to crawl the heap for object
 186       references.</li>
 187 </ul>
 188
 189 <p>In general, LLVM's support for GC does not include features which can be
 190 adequately addressed with other features of the IR and does not specify a
 191 particular binary interface. On the plus side, this means that you should be
 192 able to integrate LLVM with an existing runtime. On the other hand, it leaves
 193 a lot of work for the developer of a novel language. However, it's easy to get
 194 started quickly and scale up to a more sophisticated implementation as your
 195 compiler matures.</p>
 196
 197 </div>
 198
 199 </div>
 200
 201 <!-- *********************************************************************** -->
 202 <h2>
 203   <a name="quickstart">Getting started</a>
 204 </h2>
 205 <!-- *********************************************************************** -->
 206
 207 <div>
 208
 209 <p>Using a GC with LLVM implies many things, for example:</p>
 210
 211 <ul>
 212   <li>Write a runtime library or find an existing one which implements a GC
 213       heap.<ol>
 214     <li>Implement a memory allocator.</li>
 215     <li>Design a binary interface for the stack map, used to identify
 216         references within a stack frame on the machine stack.*</li>
 217     <li>Implement a stack crawler to discover functions on the call stack.*</li>
 218     <li>Implement a registry for global roots.</li>
 219     <li>Design a binary interface for type maps, used to identify references
 220         within heap objects.</li>
 221     <li>Implement a collection routine bringing together all of the above.</li>
 222   </ol></li>
 223   <li>Emit compatible code from your compiler.<ul>
 224     <li>Initialization in the main function.</li>
 225     <li>Use the <tt>gc "..."</tt> attribute to enable GC code generation
 226         (or <tt>F.setGC("...")</tt>).</li>
 227     <li>Use <tt>@llvm.gcroot</tt> to mark stack roots.</li>
 228     <li>Use <tt>@llvm.gcread</tt> and/or <tt>@llvm.gcwrite</tt> to
 229         manipulate GC references, if necessary.</li>
 230     <li>Allocate memory using the GC allocation routine provided by the
 231         runtime library.</li>
 232     <li>Generate type maps according to your runtime's binary interface.</li>
 233   </ul></li>
 234   <li>Write a compiler plugin to interface LLVM with the runtime library.*<ul>
 235     <li>Lower <tt>@llvm.gcread</tt> and <tt>@llvm.gcwrite</tt> to appropriate
 236         code sequences.*</li>
 237     <li>Compile LLVM's stack map to the binary form expected by the
 238         runtime.</li>
 239   </ul></li>
 240   <li>Load the plugin into the compiler. Use <tt>llc -load</tt> or link the
 241       plugin statically with your language's compiler.*</li>
 242   <li>Link program executables with the runtime.</li>
 243 </ul>
 244
 245 <p>To help with several of these tasks (those indicated with a *), LLVM
 246 includes a highly portable, built-in ShadowStack code generator. It is compiled
 247 into <tt>llc</tt> and works even with the interpreter and C backends.</p>
 248
 249 <!-- ======================================================================= -->
 250 <h3>
 251   <a name="quickstart-compiler">In your compiler</a>
 252 </h3>
 253
 254 <div>
 255
 256 <p>To turn the shadow stack on for your functions, first call:</p>
 257
 258 <div class="doc_code"><pre
 259 >F.setGC("shadow-stack");</pre></div>
 260
 261 <p>for each function your compiler emits. Since the shadow stack is built into
 262 LLVM, you do not need to load a plugin.</p>
 263
 264 <p>Your compiler must also use <tt>@llvm.gcroot</tt> as documented.
 265 Don't forget to create a root for each intermediate value that is generated
 266 when evaluating an expression. In <tt>h(f(), g())</tt>, the result of
 267 <tt>f()</tt> could easily be collected if evaluating <tt>g()</tt> triggers a
 268 collection.</p>
 269
 270 <p>There's no need to use <tt>@llvm.gcread</tt> and <tt>@llvm.gcwrite</tt> over
 271 plain <tt>load</tt> and <tt>store</tt> for now. You will need them when
 272 switching to a more advanced GC.</p>
 273
 274 </div>
 275
 276 <!-- ======================================================================= -->
 277 <h3>
 278   <a name="quickstart-runtime">In your runtime</a>
 279 </h3>
 280
 281 <div>
 282
 283 <p>The shadow stack doesn't imply a memory allocation algorithm. A semispace
 284 collector or building atop <tt>malloc</tt> are great places to start, and can
 285 be implemented with very little code.</p>
 286
 287 <p>When it comes time to collect, however, your runtime needs to traverse the
 288 stack roots, and for this it needs to integrate with the shadow stack. Luckily,
 289 doing so is very simple. (This code is heavily commented to help you
 290 understand the data structure, but there are only 20 lines of meaningful
 291 code.)</p>
 292
 293 </div>
 294
 295 <div class="doc_code"><pre
 296 >/// @brief The map for a single function's stack frame. One of these is
 297 ///        compiled as constant data into the executable for each function.
 298 ///
 299 /// Storage of metadata values is elided if the %metadata parameter to
 300 /// @llvm.gcroot is null.
 301 struct FrameMap {
 302   int32_t NumRoots;    //&lt; Number of roots in stack frame.
 303   int32_t NumMeta;     //&lt; Number of metadata entries. May be &lt; NumRoots.
 304   const void *Meta[0]; //&lt; Metadata for each root.
 305 };
 306
 307 /// @brief A link in the dynamic shadow stack. One of these is embedded in the
 308 ///        stack frame of each function on the call stack.
 309 struct StackEntry {
 310   StackEntry *Next;    //&lt; Link to next stack entry (the caller's).
 311   const FrameMap *Map; //&lt; Pointer to constant FrameMap.
 312   void *Roots[0];      //&lt; Stack roots (in-place array).
 313 };
 314
 315 /// @brief The head of the singly-linked list of StackEntries. Functions push
 316 ///        and pop onto this in their prologue and epilogue.
 317 ///
 318 /// Since there is only a global list, this technique is not threadsafe.
 319 StackEntry *llvm_gc_root_chain;
 320
 321 /// @brief Calls Visitor(root, meta) for each GC root on the stack.
 322 ///        root and meta are exactly the values passed to
 323 ///        <tt>@llvm.gcroot</tt>.
 324 ///
 325 /// Visitor could be a function to recursively mark live objects. Or it
 326 /// might copy them to another heap or generation.
 327 ///
 328 /// @param Visitor A function to invoke for every GC root on the stack.
 329 void visitGCRoots(void (*Visitor)(void **Root, const void *Meta)) {
 330   for (StackEntry *R = llvm_gc_root_chain; R; R = R->Next) {
 331     unsigned i = 0;
 332
 333     // For roots [0, NumMeta), the metadata pointer is in the FrameMap.
 334     for (unsigned e = R->Map->NumMeta; i != e; ++i)
 335       Visitor(&amp;R->Roots[i], R->Map->Meta[i]);
 336
 337     // For roots [NumMeta, NumRoots), the metadata pointer is null.
 338     for (unsigned e = R->Map->NumRoots; i != e; ++i)
 339       Visitor(&amp;R->Roots[i], NULL);
 340   }
 341 }</pre></div>
 342
 343 <!-- ======================================================================= -->
 344 <h3>
 345   <a name="shadow-stack">About the shadow stack</a>
 346 </h3>
 347
 348 <div>
 349
 350 <p>Unlike many GC algorithms which rely on a cooperative code generator to
 351 compile stack maps, this algorithm carefully maintains a linked list of stack
 352 roots [<a href="#henderson02">Henderson2002</a>]. This so-called "shadow stack"
 353 mirrors the machine stack. Maintaining this data structure is slower than using
 354 a stack map compiled into the executable as constant data, but has a significant
 355 portability advantage because it requires no special support from the target
 356 code generator, and does not require tricky platform-specific code to crawl
 357 the machine stack.</p>
 358
 359 <p>The tradeoff for this simplicity and portability is:</p>
 360
 361 <ul>
 362   <li>High overhead per function call.</li>
 363   <li>Not thread-safe.</li>
 364 </ul>
 365
 366 <p>Still, it's an easy way to get started. After your compiler and runtime are
 367 up and running, writing a <a href="#plugin">plugin</a> will allow you to take
 368 advantage of <a href="#collector-algos">more advanced GC features</a> of LLVM
 369 in order to improve performance.</p>
 370
 371 </div>
 372
 373 </div>
 374
 375 <!-- *********************************************************************** -->
 376 <h2>
 377   <a name="core">IR features</a><a name="intrinsics"></a>
 378 </h2>
 379 <!-- *********************************************************************** -->
 380
 381 <div>
 382
 383 <p>This section describes the garbage collection facilities provided by the
 384 <a href="LangRef.html">LLVM intermediate representation</a>. The exact behavior
 385 of these IR features is specified by the binary interface implemented by a
 386 <a href="#plugin">code generation plugin</a>, not by this document.</p>
 387
 388 <p>These facilities are limited to those strictly necessary; they are not
 389 intended to be a complete interface to any garbage collector. A program will
 390 need to interface with the GC library using the facilities provided by that
 391 program.</p>
 392
 393 <!-- ======================================================================= -->
 394 <h3>
 395   <a name="gcattr">Specifying GC code generation: <tt>gc "..."</tt></a>
 396 </h3>
 397
 398 <div class="doc_code"><tt>
 399   define <i>ty</i> @<i>name</i>(...) <span style="text-decoration: underline">gc "<i>name</i>"</span> { ...
 400 </tt></div>
 401
 402 <div>
 403
 404 <p>The <tt>gc</tt> function attribute is used to specify the desired GC style
 405 to the compiler. Its programmatic equivalent is the <tt>setGC</tt> method of
 406 <tt>Function</tt>.</p>
 407
 408 <p>Setting <tt>gc "<i>name</i>"</tt> on a function triggers a search for a
 409 matching code generation plugin "<i>name</i>"; it is that plugin which defines
 410 the exact nature of the code generated to support GC. If none is found, the
 411 compiler will raise an error.</p>
 412
 413 <p>Specifying the GC style on a per-function basis allows LLVM to link together
 414 programs that use different garbage collection algorithms (or none at all).</p>
 415
 416 </div>
 417
 418 <!-- ======================================================================= -->
 419 <h3>
 420   <a name="gcroot">Identifying GC roots on the stack: <tt>llvm.gcroot</tt></a>
 421 </h3>
 422
 423 <div class="doc_code"><tt>
 424   void @llvm.gcroot(i8** %ptrloc, i8* %metadata)
 425 </tt></div>
 426
 427 <div>
 428
 429 <p>The <tt>llvm.gcroot</tt> intrinsic is used to inform LLVM that a stack
 430 variable references an object on the heap and is to be tracked for garbage
 431 collection. The exact impact on generated code is specified by a <a
 432 href="#plugin">compiler plugin</a>.</p>
 433
 434 <p>A compiler which uses mem2reg to raise imperative code using <tt>alloca</tt>
 435 into SSA form need only add a call to <tt>@llvm.gcroot</tt> for those variables
 436 which a pointers into the GC heap.</p>
 437
 438 <p>It is also important to mark intermediate values with <tt>llvm.gcroot</tt>.
 439 For example, consider <tt>h(f(), g())</tt>. Beware leaking the result of
 440 <tt>f()</tt> in the case that <tt>g()</tt> triggers a collection.</p>
 441
 442 <p>The first argument <b>must</b> be a value referring to an alloca instruction
 443 or a bitcast of an alloca. The second contains a pointer to metadata that
 444 should be associated with the pointer, and <b>must</b> be a constant or global
 445 value address. If your target collector uses tags, use a null pointer for
 446 metadata.</p>
 447
 448 <p>The <tt>%metadata</tt> argument can be used to avoid requiring heap objects
 449 to have 'isa' pointers or tag bits. [<a href="#appel89">Appel89</a>, <a
 450 href="#goldberg91">Goldberg91</a>, <a href="#tolmach94">Tolmach94</a>] If
 451 specified, its value will be tracked along with the location of the pointer in
 452 the stack frame.</p>
 453
 454 <p>Consider the following fragment of Java code:</p>
 455
 456 <pre>
 457        {
 458          Object X;   // A null-initialized reference to an object
 459          ...
 460        }
 461 </pre>
 462
 463 <p>This block (which may be located in the middle of a function or in a loop
 464 nest), could be compiled to this LLVM code:</p>
 465
 466 <pre>
 467 Entry:
 468    ;; In the entry block for the function, allocate the
 469    ;; stack space for X, which is an LLVM pointer.
 470    %X = alloca %Object*
 471
 472    ;; Tell LLVM that the stack space is a stack root.
 473    ;; Java has type-tags on objects, so we pass null as metadata.
 474    %tmp = bitcast %Object** %X to i8**
 475    call void @llvm.gcroot(i8** %X, i8* null)
 476    ...
 477
 478    ;; "CodeBlock" is the block corresponding to the start
 479    ;;  of the scope above.
 480 CodeBlock:
 481    ;; Java null-initializes pointers.
 482    store %Object* null, %Object** %X
 483
 484    ...
 485
 486    ;; As the pointer goes out of scope, store a null value into
 487    ;; it, to indicate that the value is no longer live.
 488    store %Object* null, %Object** %X
 489    ...
 490 </pre>
 491
 492 </div>
 493
 494 <!-- ======================================================================= -->
 495 <h3>
 496   <a name="barriers">Reading and writing references in the heap</a>
 497 </h3>
 498
 499 <div>
 500
 501 <p>Some collectors need to be informed when the mutator (the program that needs
 502 garbage collection) either reads a pointer from or writes a pointer to a field
 503 of a heap object. The code fragments inserted at these points are called
 504 <em>read barriers</em> and <em>write barriers</em>, respectively. The amount of
 505 code that needs to be executed is usually quite small and not on the critical
 506 path of any computation, so the overall performance impact of the barrier is
 507 tolerable.</p>
 508
 509 <p>Barriers often require access to the <em>object pointer</em> rather than the
 510 <em>derived pointer</em> (which is a pointer to the field within the
 511 object). Accordingly, these intrinsics take both pointers as separate arguments
 512 for completeness. In this snippet, <tt>%object</tt> is the object pointer, and
 513 <tt>%derived</tt> is the derived pointer:</p>
 514
 515 <blockquote><pre>
 516     ;; An array type.
 517     %class.Array = type { %class.Object, i32, [0 x %class.Object*] }
 518     ...
 519
 520     ;; Load the object pointer from a gcroot.
 521     %object = load %class.Array** %object_addr
 522
 523     ;; Compute the derived pointer.
 524     %derived = getelementptr %object, i32 0, i32 2, i32 %n</pre></blockquote>
 525
 526 <p>LLVM does not enforce this relationship between the object and derived
 527 pointer (although a <a href="#plugin">plugin</a> might). However, it would be
 528 an unusual collector that violated it.</p>
 529
 530 <p>The use of these intrinsics is naturally optional if the target GC does
 531 require the corresponding barrier. Such a GC plugin will replace the intrinsic
 532 calls with the corresponding <tt>load</tt> or <tt>store</tt> instruction if they
 533 are used.</p>
 534
 535 <!-- ======================================================================= -->
 536 <h4>
 537   <a name="gcwrite">Write barrier: <tt>llvm.gcwrite</tt></a>
 538 </h4>
 539
 540 <div class="doc_code"><tt>
 541 void @llvm.gcwrite(i8* %value, i8* %object, i8** %derived)
 542 </tt></div>
 543
 544 <div>
 545
 546 <p>For write barriers, LLVM provides the <tt>llvm.gcwrite</tt> intrinsic
 547 function. It has exactly the same semantics as a non-volatile <tt>store</tt> to
 548 the derived pointer (the third argument). The exact code generated is specified
 549 by a <a href="#plugin">compiler plugin</a>.</p>
 550
 551 <p>Many important algorithms require write barriers, including generational
 552 and concurrent collectors. Additionally, write barriers could be used to
 553 implement reference counting.</p>
 554
 555 </div>
 556
 557 <!-- ======================================================================= -->
 558 <h4>
 559   <a name="gcread">Read barrier: <tt>llvm.gcread</tt></a>
 560 </h4>
 561
 562 <div class="doc_code"><tt>
 563 i8* @llvm.gcread(i8* %object, i8** %derived)<br>
 564 </tt></div>
 565
 566 <div>
 567
 568 <p>For read barriers, LLVM provides the <tt>llvm.gcread</tt> intrinsic function.
 569 It has exactly the same semantics as a non-volatile <tt>load</tt> from the
 570 derived pointer (the second argument). The exact code generated is specified by
 571 a <a href="#plugin">compiler plugin</a>.</p>
 572
 573 <p>Read barriers are needed by fewer algorithms than write barriers, and may
 574 have a greater performance impact since pointer reads are more frequent than
 575 writes.</p>
 576
 577 </div>
 578
 579 </div>
 580
 581 </div>
 582
 583 <!-- *********************************************************************** -->
 584 <h2>
 585   <a name="plugin">Implementing a collector plugin</a>
 586 </h2>
 587 <!-- *********************************************************************** -->
 588
 589 <div>
 590
 591 <p>User code specifies which GC code generation to use with the <tt>gc</tt>
 592 function attribute or, equivalently, with the <tt>setGC</tt> method of
 593 <tt>Function</tt>.</p>
 594
 595 <p>To implement a GC plugin, it is necessary to subclass
 596 <tt>llvm::GCStrategy</tt>, which can be accomplished in a few lines of
 597 boilerplate code. LLVM's infrastructure provides access to several important
 598 algorithms. For an uncontroversial collector, all that remains may be to
 599 compile LLVM's computed stack map to assembly code (using the binary
 600 representation expected by the runtime library). This can be accomplished in
 601 about 100 lines of code.</p>
 602
 603 <p>This is not the appropriate place to implement a garbage collected heap or a
 604 garbage collector itself. That code should exist in the language's runtime
 605 library. The compiler plugin is responsible for generating code which
 606 conforms to the binary interface defined by library, most essentially the
 607 <a href="#stack-map">stack map</a>.</p>
 608
 609 <p>To subclass <tt>llvm::GCStrategy</tt> and register it with the compiler:</p>
 610
 611 <blockquote><pre>// lib/MyGC/MyGC.cpp - Example LLVM GC plugin
 612
 613 #include "llvm/CodeGen/GCStrategy.h"
 614 #include "llvm/CodeGen/GCMetadata.h"
 615 #include "llvm/Support/Compiler.h"
 616
 617 using namespace llvm;
 618
 619 namespace {
 620   class LLVM_LIBRARY_VISIBILITY MyGC : public GCStrategy {
 621   public:
 622     MyGC() {}
 623   };
 624
 625   GCRegistry::Add&lt;MyGC&gt;
 626   X("mygc", "My bespoke garbage collector.");
 627 }</pre></blockquote>
 628
 629 <p>This boilerplate collector does nothing. More specifically:</p>
 630
 631 <ul>
 632   <li><tt>llvm.gcread</tt> calls are replaced with the corresponding
 633       <tt>load</tt> instruction.</li>
 634   <li><tt>llvm.gcwrite</tt> calls are replaced with the corresponding
 635       <tt>store</tt> instruction.</li>
 636   <li>No safe points are added to the code.</li>
 637   <li>The stack map is not compiled into the executable.</li>
 638 </ul>
 639
 640 <p>Using the LLVM makefiles (like the <a
 641 href="http://llvm.org/viewvc/llvm-project/llvm/trunk/projects/sample/">sample
 642 project</a>), this code can be compiled as a plugin using a simple
 643 makefile:</p>
 644
 645 <blockquote><pre
 646 ># lib/MyGC/Makefile
 647
 648 LEVEL := ../..
 649 LIBRARYNAME = <var>MyGC</var>
 650 LOADABLE_MODULE = 1
 651
 652 include $(LEVEL)/Makefile.common</pre></blockquote>
 653
 654 <p>Once the plugin is compiled, code using it may be compiled using <tt>llc
 655 -load=<var>MyGC.so</var></tt> (though <var>MyGC.so</var> may have some other
 656 platform-specific extension):</p>
 657
 658 <blockquote><pre
 659 >$ cat sample.ll
 660 define void @f() gc "mygc" {
 661 entry:
 662         ret void
 663 }
 664 $ llvm-as &lt; sample.ll | llc -load=MyGC.so</pre></blockquote>
 665
 666 <p>It is also possible to statically link the collector plugin into tools, such
 667 as a language-specific compiler front-end.</p>
 668
 669 <!-- ======================================================================= -->
 670 <h3>
 671   <a name="collector-algos">Overview of available features</a>
 672 </h3>
 673
 674 <div>
 675
 676 <p><tt>GCStrategy</tt> provides a range of features through which a plugin
 677 may do useful work. Some of these are callbacks, some are algorithms that can
 678 be enabled, disabled, or customized. This matrix summarizes the supported (and
 679 planned) features and correlates them with the collection techniques which
 680 typically require them.</p>
 681
 682 <table>
 683   <tr>
 684     <th>Algorithm</th>
 685     <th>Done</th>
 686     <th>shadow stack</th>
 687     <th>refcount</th>
 688     <th>mark-sweep</th>
 689     <th>copying</th>
 690     <th>incremental</th>
 691     <th>threaded</th>
 692     <th>concurrent</th>
 693   </tr>
 694   <tr>
 695     <th class="rowhead"><a href="#stack-map">stack map</a></th>
 696     <td>&#10004;</td>
 697     <td></td>
 698     <td></td>
 699     <td>&#10008;</td>
 700     <td>&#10008;</td>
 701     <td>&#10008;</td>
 702     <td>&#10008;</td>
 703     <td>&#10008;</td>
 704   </tr>
 705   <tr>
 706     <th class="rowhead"><a href="#init-roots">initialize roots</a></th>
 707     <td>&#10004;</td>
 708     <td>&#10008;</td>
 709     <td>&#10008;</td>
 710     <td>&#10008;</td>
 711     <td>&#10008;</td>
 712     <td>&#10008;</td>
 713     <td>&#10008;</td>
 714     <td>&#10008;</td>
 715   </tr>
 716   <tr class="doc_warning">
 717     <th class="rowhead">derived pointers</th>
 718     <td>NO</td>
 719     <td></td>
 720     <td></td>
 721     <td></td>
 722     <td></td>
 723     <td></td>
 724     <td>&#10008;*</td>
 725     <td>&#10008;*</td>
 726   </tr>
 727   <tr>
 728     <th class="rowhead"><em><a href="#custom">custom lowering</a></em></th>
 729     <td>&#10004;</td>
 730     <th></th>
 731     <th></th>
 732     <th></th>
 733     <th></th>
 734     <th></th>
 735     <th></th>
 736     <th></th>
 737   </tr>
 738   <tr>
 739     <th class="rowhead indent">gcroot</th>
 740     <td>&#10004;</td>
 741     <td>&#10008;</td>
 742     <td>&#10008;</td>
 743     <td></td>
 744     <td></td>
 745     <td></td>
 746     <td></td>
 747     <td></td>
 748   </tr>
 749   <tr>
 750     <th class="rowhead indent">gcwrite</th>
 751     <td>&#10004;</td>
 752     <td></td>
 753     <td>&#10008;</td>
 754     <td></td>
 755     <td></td>
 756     <td>&#10008;</td>
 757     <td></td>
 758     <td>&#10008;</td>
 759   </tr>
 760   <tr>
 761     <th class="rowhead indent">gcread</th>
 762     <td>&#10004;</td>
 763     <td></td>
 764     <td></td>
 765     <td></td>
 766     <td></td>
 767     <td></td>
 768     <td></td>
 769     <td>&#10008;</td>
 770   </tr>
 771   <tr>
 772     <th class="rowhead"><em><a href="#safe-points">safe points</a></em></th>
 773     <td></td>
 774     <th></th>
 775     <th></th>
 776     <th></th>
 777     <th></th>
 778     <th></th>
 779     <th></th>
 780     <th></th>
 781   </tr>
 782   <tr>
 783     <th class="rowhead indent">in calls</th>
 784     <td>&#10004;</td>
 785     <td></td>
 786     <td></td>
 787     <td>&#10008;</td>
 788     <td>&#10008;</td>
 789     <td>&#10008;</td>
 790     <td>&#10008;</td>
 791     <td>&#10008;</td>
 792   </tr>
 793   <tr>
 794     <th class="rowhead indent">before calls</th>
 795     <td>&#10004;</td>
 796     <td></td>
 797     <td></td>
 798     <td></td>
 799     <td></td>
 800     <td></td>
 801     <td>&#10008;</td>
 802     <td>&#10008;</td>
 803   </tr>
 804   <tr class="doc_warning">
 805     <th class="rowhead indent">for loops</th>
 806     <td>NO</td>
 807     <td></td>
 808     <td></td>
 809     <td></td>
 810     <td></td>
 811     <td></td>
 812     <td>&#10008;</td>
 813     <td>&#10008;</td>
 814   </tr>
 815   <tr>
 816     <th class="rowhead indent">before escape</th>
 817     <td>&#10004;</td>
 818     <td></td>
 819     <td></td>
 820     <td></td>
 821     <td></td>
 822     <td></td>
 823     <td>&#10008;</td>
 824     <td>&#10008;</td>
 825   </tr>
 826   <tr class="doc_warning">
 827     <th class="rowhead">emit code at safe points</th>
 828     <td>NO</td>
 829     <td></td>
 830     <td></td>
 831     <td></td>
 832     <td></td>
 833     <td></td>
 834     <td>&#10008;</td>
 835     <td>&#10008;</td>
 836   </tr>
 837   <tr>
 838     <th class="rowhead"><em>output</em></th>
 839     <td></td>
 840     <th></th>
 841     <th></th>
 842     <th></th>
 843     <th></th>
 844     <th></th>
 845     <th></th>
 846     <th></th>
 847   </tr>
 848   <tr>
 849     <th class="rowhead indent"><a href="#assembly">assembly</a></th>
 850     <td>&#10004;</td>
 851     <td></td>
 852     <td></td>
 853     <td>&#10008;</td>
 854     <td>&#10008;</td>
 855     <td>&#10008;</td>
 856     <td>&#10008;</td>
 857     <td>&#10008;</td>
 858   </tr>
 859   <tr class="doc_warning">
 860     <th class="rowhead indent">JIT</th>
 861     <td>NO</td>
 862     <td></td>
 863     <td></td>
 864     <td class="optl">&#10008;</td>
 865     <td class="optl">&#10008;</td>
 866     <td class="optl">&#10008;</td>
 867     <td class="optl">&#10008;</td>
 868     <td class="optl">&#10008;</td>
 869   </tr>
 870   <tr class="doc_warning">
 871     <th class="rowhead indent">obj</th>
 872     <td>NO</td>
 873     <td></td>
 874     <td></td>
 875     <td class="optl">&#10008;</td>
 876     <td class="optl">&#10008;</td>
 877     <td class="optl">&#10008;</td>
 878     <td class="optl">&#10008;</td>
 879     <td class="optl">&#10008;</td>
 880   </tr>
 881   <tr class="doc_warning">
 882     <th class="rowhead">live analysis</th>
 883     <td>NO</td>
 884     <td></td>
 885     <td></td>
 886     <td class="optl">&#10008;</td>
 887     <td class="optl">&#10008;</td>
 888     <td class="optl">&#10008;</td>
 889     <td class="optl">&#10008;</td>
 890     <td class="optl">&#10008;</td>
 891   </tr>
 892   <tr class="doc_warning">
 893     <th class="rowhead">register map</th>
 894     <td>NO</td>
 895     <td></td>
 896     <td></td>
 897     <td class="optl">&#10008;</td>
 898     <td class="optl">&#10008;</td>
 899     <td class="optl">&#10008;</td>
 900     <td class="optl">&#10008;</td>
 901     <td class="optl">&#10008;</td>
 902   </tr>
 903   <tr>
 904     <td colspan="10">
 905       <div><span class="doc_warning">*</span> Derived pointers only pose a
 906            hazard to copying collectors.</div>
 907       <div><span class="optl">&#10008;</span> in gray denotes a feature which
 908            could be utilized if available.</div>
 909     </td>
 910   </tr>
 911 </table>
 912
 913 <p>To be clear, the collection techniques above are defined as:</p>
 914
 915 <dl>
 916   <dt>Shadow Stack</dt>
 917   <dd>The mutator carefully maintains a linked list of stack roots.</dd>
 918   <dt>Reference Counting</dt>
 919   <dd>The mutator maintains a reference count for each object and frees an
 920       object when its count falls to zero.</dd>
 921   <dt>Mark-Sweep</dt>
 922   <dd>When the heap is exhausted, the collector marks reachable objects starting
 923       from the roots, then deallocates unreachable objects in a sweep
 924       phase.</dd>
 925   <dt>Copying</dt>
 926   <dd>As reachability analysis proceeds, the collector copies objects from one
 927       heap area to another, compacting them in the process. Copying collectors
 928       enable highly efficient "bump pointer" allocation and can improve locality
 929       of reference.</dd>
 930   <dt>Incremental</dt>
 931   <dd>(Including generational collectors.) Incremental collectors generally have
 932       all the properties of a copying collector (regardless of whether the
 933       mature heap is compacting), but bring the added complexity of requiring
 934       write barriers.</dd>
 935   <dt>Threaded</dt>
 936   <dd>Denotes a multithreaded mutator; the collector must still stop the mutator
 937       ("stop the world") before beginning reachability analysis. Stopping a
 938       multithreaded mutator is a complicated problem. It generally requires
 939       highly platform specific code in the runtime, and the production of
 940       carefully designed machine code at safe points.</dd>
 941   <dt>Concurrent</dt>
 942   <dd>In this technique, the mutator and the collector run concurrently, with
 943       the goal of eliminating pause times. In a <em>cooperative</em> collector,
 944       the mutator further aids with collection should a pause occur, allowing
 945       collection to take advantage of multiprocessor hosts. The "stop the world"
 946       problem of threaded collectors is generally still present to a limited
 947       extent. Sophisticated marking algorithms are necessary. Read barriers may
 948       be necessary.</dd>
 949 </dl>
 950
 951 <p>As the matrix indicates, LLVM's garbage collection infrastructure is already
 952 suitable for a wide variety of collectors, but does not currently extend to
 953 multithreaded programs. This will be added in the future as there is
 954 interest.</p>
 955
 956 </div>
 957
 958 <!-- ======================================================================= -->
 959 <h3>
 960   <a name="stack-map">Computing stack maps</a>
 961 </h3>
 962
 963 <div>
 964
 965 <p>LLVM automatically computes a stack map. One of the most important features
 966 of a <tt>GCStrategy</tt> is to compile this information into the executable in
 967 the binary representation expected by the runtime library.</p>
 968
 969 <p>The stack map consists of the location and identity of each GC root in the
 970 each function in the module. For each root:</p>
 971
 972 <ul>
 973   <li><tt>RootNum</tt>: The index of the root.</li>
 974   <li><tt>StackOffset</tt>: The offset of the object relative to the frame
 975       pointer.</li>
 976   <li><tt>RootMetadata</tt>: The value passed as the <tt>%metadata</tt>
 977       parameter to the <a href="#gcroot"><tt>@llvm.gcroot</tt></a> intrinsic.</li>
 978 </ul>
 979
 980 <p>Also, for the function as a whole:</p>
 981
 982 <ul>
 983   <li><tt>getFrameSize()</tt>: The overall size of the function's initial
 984       stack frame, not accounting for any dynamic allocation.</li>
 985   <li><tt>roots_size()</tt>: The count of roots in the function.</li>
 986 </ul>
 987
 988 <p>To access the stack map, use <tt>GCFunctionMetadata::roots_begin()</tt> and
 989 -<tt>end()</tt> from the <tt><a
 990 href="#assembly">GCMetadataPrinter</a></tt>:</p>
 991
 992 <blockquote><pre
 993 >for (iterator I = begin(), E = end(); I != E; ++I) {
 994   GCFunctionInfo *FI = *I;
 995   unsigned FrameSize = FI-&gt;getFrameSize();
 996   size_t RootCount = FI-&gt;roots_size();
 997
 998   for (GCFunctionInfo::roots_iterator RI = FI-&gt;roots_begin(),
 999                                       RE = FI-&gt;roots_end();
1000                                       RI != RE; ++RI) {
1001     int RootNum = RI->Num;
1002     int RootStackOffset = RI->StackOffset;
1003     Constant *RootMetadata = RI->Metadata;
1004   }
1005 }</pre></blockquote>
1006
1007 <p>If the <tt>llvm.gcroot</tt> intrinsic is eliminated before code generation by
1008 a custom lowering pass, LLVM will compute an empty stack map. This may be useful
1009 for collector plugins which implement reference counting or a shadow stack.</p>
1010
1011 </div>
1012
1013
1014 <!-- ======================================================================= -->
1015 <h3>
1016   <a name="init-roots">Initializing roots to null: <tt>InitRoots</tt></a>
1017 </h3>
1018
1019 <div>
1020
1021 <blockquote><pre
1022 >MyGC::MyGC() {
1023   InitRoots = true;
1024 }</pre></blockquote>
1025
1026 <p>When set, LLVM will automatically initialize each root to <tt>null</tt> upon
1027 entry to the function. This prevents the GC's sweep phase from visiting
1028 uninitialized pointers, which will almost certainly cause it to crash. This
1029 initialization occurs before custom lowering, so the two may be used
1030 together.</p>
1031
1032 <p>Since LLVM does not yet compute liveness information, there is no means of
1033 distinguishing an uninitialized stack root from an initialized one. Therefore,
1034 this feature should be used by all GC plugins. It is enabled by default.</p>
1035
1036 </div>
1037
1038
1039 <!-- ======================================================================= -->
1040 <h3>
1041   <a name="custom">Custom lowering of intrinsics: <tt>CustomRoots</tt>,
1042     <tt>CustomReadBarriers</tt>, and <tt>CustomWriteBarriers</tt></a>
1043 </h3>
1044
1045 <div>
1046
1047 <p>For GCs which use barriers or unusual treatment of stack roots, these
1048 flags allow the collector to perform arbitrary transformations of the LLVM
1049 IR:</p>
1050
1051 <blockquote><pre
1052 >class MyGC : public GCStrategy {
1053 public:
1054   MyGC() {
1055     CustomRoots = true;
1056     CustomReadBarriers = true;
1057     CustomWriteBarriers = true;
1058   }
1059
1060   virtual bool initializeCustomLowering(Module &amp;M);
1061   virtual bool performCustomLowering(Function &amp;F);
1062 };</pre></blockquote>
1063
1064 <p>If any of these flags are set, then LLVM suppresses its default lowering for
1065 the corresponding intrinsics and instead calls
1066 <tt>performCustomLowering</tt>.</p>
1067
1068 <p>LLVM's default action for each intrinsic is as follows:</p>
1069
1070 <ul>
1071   <li><tt>llvm.gcroot</tt>: Leave it alone. The code generator must see it
1072                             or the stack map will not be computed.</li>
1073   <li><tt>llvm.gcread</tt>: Substitute a <tt>load</tt> instruction.</li>
1074   <li><tt>llvm.gcwrite</tt>: Substitute a <tt>store</tt> instruction.</li>
1075 </ul>
1076
1077 <p>If <tt>CustomReadBarriers</tt> or <tt>CustomWriteBarriers</tt> are specified,
1078 then <tt>performCustomLowering</tt> <strong>must</strong> eliminate the
1079 corresponding barriers.</p>
1080
1081 <p><tt>performCustomLowering</tt> must comply with the same restrictions as <a
1082 href="WritingAnLLVMPass.html#runOnFunction"><tt
1083 >FunctionPass::runOnFunction</tt></a>.
1084 Likewise, <tt>initializeCustomLowering</tt> has the same semantics as <a
1085 href="WritingAnLLVMPass.html#doInitialization_mod"><tt
1086 >Pass::doInitialization(Module&amp;)</tt></a>.</p>
1087
1088 <p>The following can be used as a template:</p>
1089
1090 <blockquote><pre
1091 >#include "llvm/Module.h"
1092 #include "llvm/IntrinsicInst.h"
1093
1094 bool MyGC::initializeCustomLowering(Module &amp;M) {
1095   return false;
1096 }
1097
1098 bool MyGC::performCustomLowering(Function &amp;F) {
1099   bool MadeChange = false;
1100
1101   for (Function::iterator BB = F.begin(), E = F.end(); BB != E; ++BB)
1102     for (BasicBlock::iterator II = BB-&gt;begin(), E = BB-&gt;end(); II != E; )
1103       if (IntrinsicInst *CI = dyn_cast&lt;IntrinsicInst&gt;(II++))
1104         if (Function *F = CI-&gt;getCalledFunction())
1105           switch (F-&gt;getIntrinsicID()) {
1106           case Intrinsic::gcwrite:
1107             // Handle llvm.gcwrite.
1108             CI-&gt;eraseFromParent();
1109             MadeChange = true;
1110             break;
1111           case Intrinsic::gcread:
1112             // Handle llvm.gcread.
1113             CI-&gt;eraseFromParent();
1114             MadeChange = true;
1115             break;
1116           case Intrinsic::gcroot:
1117             // Handle llvm.gcroot.
1118             CI-&gt;eraseFromParent();
1119             MadeChange = true;
1120             break;
1121           }
1122
1123   return MadeChange;
1124 }</pre></blockquote>
1125
1126 </div>
1127
1128
1129 <!-- ======================================================================= -->
1130 <h3>
1131   <a name="safe-points">Generating safe points: <tt>NeededSafePoints</tt></a>
1132 </h3>
1133
1134 <div>
1135
1136 <p>LLVM can compute four kinds of safe points:</p>
1137
1138 <blockquote><pre
1139 >namespace GC {
1140   /// PointKind - The type of a collector-safe point.
1141   ///
1142   enum PointKind {
1143     Loop,    //&lt; Instr is a loop (backwards branch).
1144     Return,  //&lt; Instr is a return instruction.
1145     PreCall, //&lt; Instr is a call instruction.
1146     PostCall //&lt; Instr is the return address of a call.
1147   };
1148 }</pre></blockquote>
1149
1150 <p>A collector can request any combination of the four by setting the
1151 <tt>NeededSafePoints</tt> mask:</p>
1152
1153 <blockquote><pre
1154 >MyGC::MyGC() {
1155   NeededSafePoints = 1 &lt;&lt; GC::Loop
1156                    | 1 &lt;&lt; GC::Return
1157                    | 1 &lt;&lt; GC::PreCall
1158                    | 1 &lt;&lt; GC::PostCall;
1159 }</pre></blockquote>
1160
1161 <p>It can then use the following routines to access safe points.</p>
1162
1163 <blockquote><pre
1164 >for (iterator I = begin(), E = end(); I != E; ++I) {
1165   GCFunctionInfo *MD = *I;
1166   size_t PointCount = MD-&gt;size();
1167
1168   for (GCFunctionInfo::iterator PI = MD-&gt;begin(),
1169                                 PE = MD-&gt;end(); PI != PE; ++PI) {
1170     GC::PointKind PointKind = PI-&gt;Kind;
1171     unsigned PointNum = PI-&gt;Num;
1172   }
1173 }
1174 </pre></blockquote>
1175
1176 <p>Almost every collector requires <tt>PostCall</tt> safe points, since these
1177 correspond to the moments when the function is suspended during a call to a
1178 subroutine.</p>
1179
1180 <p>Threaded programs generally require <tt>Loop</tt> safe points to guarantee
1181 that the application will reach a safe point within a bounded amount of time,
1182 even if it is executing a long-running loop which contains no function
1183 calls.</p>
1184
1185 <p>Threaded collectors may also require <tt>Return</tt> and <tt>PreCall</tt>
1186 safe points to implement "stop the world" techniques using self-modifying code,
1187 where it is important that the program not exit the function without reaching a
1188 safe point (because only the topmost function has been patched).</p>
1189
1190 </div>
1191
1192
1193 <!-- ======================================================================= -->
1194 <h3>
1195   <a name="assembly">Emitting assembly code: <tt>GCMetadataPrinter</tt></a>
1196 </h3>
1197
1198 <div>
1199
1200 <p>LLVM allows a plugin to print arbitrary assembly code before and after the
1201 rest of a module's assembly code. At the end of the module, the GC can compile
1202 the LLVM stack map into assembly code. (At the beginning, this information is not
1203 yet computed.)</p>
1204
1205 <p>Since AsmWriter and CodeGen are separate components of LLVM, a separate
1206 abstract base class and registry is provided for printing assembly code, the
1207 <tt>GCMetadaPrinter</tt> and <tt>GCMetadataPrinterRegistry</tt>. The AsmWriter
1208 will look for such a subclass if the <tt>GCStrategy</tt> sets
1209 <tt>UsesMetadata</tt>:</p>
1210
1211 <blockquote><pre
1212 >MyGC::MyGC() {
1213   UsesMetadata = true;
1214 }</pre></blockquote>
1215
1216 <p>This separation allows JIT-only clients to be smaller.</p>
1217
1218 <p>Note that LLVM does not currently have analogous APIs to support code
1219 generation in the JIT, nor using the object writers.</p>
1220
1221 <blockquote><pre
1222 >// lib/MyGC/MyGCPrinter.cpp - Example LLVM GC printer
1223
1224 #include "llvm/CodeGen/GCMetadataPrinter.h"
1225 #include "llvm/Support/Compiler.h"
1226
1227 using namespace llvm;
1228
1229 namespace {
1230   class LLVM_LIBRARY_VISIBILITY MyGCPrinter : public GCMetadataPrinter {
1231   public:
1232     virtual void beginAssembly(std::ostream &amp;OS, AsmPrinter &amp;AP,
1233                                const TargetAsmInfo &amp;TAI);
1234
1235     virtual void finishAssembly(std::ostream &amp;OS, AsmPrinter &amp;AP,
1236                                 const TargetAsmInfo &amp;TAI);
1237   };
1238
1239   GCMetadataPrinterRegistry::Add&lt;MyGCPrinter&gt;
1240   X("mygc", "My bespoke garbage collector.");
1241 }</pre></blockquote>
1242
1243 <p>The collector should use <tt>AsmPrinter</tt> and <tt>TargetAsmInfo</tt> to
1244 print portable assembly code to the <tt>std::ostream</tt>. The collector itself
1245 contains the stack map for the entire module, and may access the
1246 <tt>GCFunctionInfo</tt> using its own <tt>begin()</tt> and <tt>end()</tt>
1247 methods. Here's a realistic example:</p>
1248
1249 <blockquote><pre
1250 >#include "llvm/CodeGen/AsmPrinter.h"
1251 #include "llvm/Function.h"
1252 #include "llvm/Target/TargetMachine.h"
1253 #include "llvm/Target/TargetData.h"
1254 #include "llvm/Target/TargetAsmInfo.h"
1255
1256 void MyGCPrinter::beginAssembly(std::ostream &amp;OS, AsmPrinter &amp;AP,
1257                                 const TargetAsmInfo &amp;TAI) {
1258   // Nothing to do.
1259 }
1260
1261 void MyGCPrinter::finishAssembly(std::ostream &amp;OS, AsmPrinter &amp;AP,
1262                                  const TargetAsmInfo &amp;TAI) {
1263   // Set up for emitting addresses.
1264   const char *AddressDirective;
1265   int AddressAlignLog;
1266   if (AP.TM.getTargetData()->getPointerSize() == sizeof(int32_t)) {
1267     AddressDirective = TAI.getData32bitsDirective();
1268     AddressAlignLog = 2;
1269   } else {
1270     AddressDirective = TAI.getData64bitsDirective();
1271     AddressAlignLog = 3;
1272   }
1273
1274   // Put this in the data section.
1275   AP.SwitchToDataSection(TAI.getDataSection());
1276
1277   // For each function...
1278   for (iterator FI = begin(), FE = end(); FI != FE; ++FI) {
1279     GCFunctionInfo &amp;MD = **FI;
1280
1281     // Emit this data structure:
1282     //
1283     // struct {
1284     //   int32_t PointCount;
1285     //   struct {
1286     //     void *SafePointAddress;
1287     //     int32_t LiveCount;
1288     //     int32_t LiveOffsets[LiveCount];
1289     //   } Points[PointCount];
1290     // } __gcmap_&lt;FUNCTIONNAME&gt;;
1291
1292     // Align to address width.
1293     AP.EmitAlignment(AddressAlignLog);
1294
1295     // Emit the symbol by which the stack map entry can be found.
1296     std::string Symbol;
1297     Symbol += TAI.getGlobalPrefix();
1298     Symbol += "__gcmap_";
1299     Symbol += MD.getFunction().getName();
1300     if (const char *GlobalDirective = TAI.getGlobalDirective())
1301       OS &lt;&lt; GlobalDirective &lt;&lt; Symbol &lt;&lt; "\n";
1302     OS &lt;&lt; TAI.getGlobalPrefix() &lt;&lt; Symbol &lt;&lt; ":\n";
1303
1304     // Emit PointCount.
1305     AP.EmitInt32(MD.size());
1306     AP.EOL("safe point count");
1307
1308     // And each safe point...
1309     for (GCFunctionInfo::iterator PI = MD.begin(),
1310                                      PE = MD.end(); PI != PE; ++PI) {
1311       // Align to address width.
1312       AP.EmitAlignment(AddressAlignLog);
1313
1314       // Emit the address of the safe point.
1315       OS &lt;&lt; AddressDirective
1316          &lt;&lt; TAI.getPrivateGlobalPrefix() &lt;&lt; "label" &lt;&lt; PI-&gt;Num;
1317       AP.EOL("safe point address");
1318
1319       // Emit the stack frame size.
1320       AP.EmitInt32(MD.getFrameSize());
1321       AP.EOL("stack frame size");
1322
1323       // Emit the number of live roots in the function.
1324       AP.EmitInt32(MD.live_size(PI));
1325       AP.EOL("live root count");
1326
1327       // And for each live root...
1328       for (GCFunctionInfo::live_iterator LI = MD.live_begin(PI),
1329                                             LE = MD.live_end(PI);
1330                                             LI != LE; ++LI) {
1331         // Print its offset within the stack frame.
1332         AP.EmitInt32(LI-&gt;StackOffset);
1333         AP.EOL("stack offset");
1334       }
1335     }
1336   }
1337 }
1338 </pre></blockquote>
1339
1340 </div>
1341
1342 </div>
1343
1344 <!-- *********************************************************************** -->
1345 <h2>
1346   <a name="references">References</a>
1347 </h2>
1348 <!-- *********************************************************************** -->
1349
1350 <div>
1351
1352 <p><a name="appel89">[Appel89]</a> Runtime Tags Aren't Necessary. Andrew
1353 W. Appel. Lisp and Symbolic Computation 19(7):703-705, July 1989.</p>
1354
1355 <p><a name="goldberg91">[Goldberg91]</a> Tag-free garbage collection for
1356 strongly typed programming languages. Benjamin Goldberg. ACM SIGPLAN
1357 PLDI'91.</p>
1358
1359 <p><a name="tolmach94">[Tolmach94]</a> Tag-free garbage collection using
1360 explicit type parameters. Andrew Tolmach. Proceedings of the 1994 ACM
1361 conference on LISP and functional programming.</p>
1362
1363 <p><a name="henderson02">[Henderson2002]</a> <a
1364 href="http://citeseer.ist.psu.edu/henderson02accurate.html">
1365 Accurate Garbage Collection in an Uncooperative Environment</a>.
1366 Fergus Henderson. International Symposium on Memory Management 2002.</p>
1367
1368 </div>
1369
1370
1371 <!-- *********************************************************************** -->
1372
1373 <hr>
1374 <address>
1375   <a href="http://jigsaw.w3.org/css-validator/check/referer"><img
1376   src="http://jigsaw.w3.org/css-validator/images/vcss-blue" alt="Valid CSS"></a>
1377   <a href="http://validator.w3.org/check/referer"><img
1378   src="http://www.w3.org/Icons/valid-html401-blue" alt="Valid HTML 4.01"></a>
1379
1380   <a href="mailto:sabre@nondot.org">Chris Lattner</a><br>
1381   <a href="http://llvm.org/">LLVM Compiler Infrastructure</a><br>
1382   Last modified: $Date$
1383 </address>
1384
1385 </body>
1386 </html>