Mercurial > hg > Members > tobaru > cbc > CbC_llvm

comparison docs/GarbageCollection.rst @ 3:9ad51c7bc036

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
1st commit. remove git dir and add all files.
author Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
date Wed, 15 May 2013 06:43:32 +0900
parents
children
comparison
equal deleted inserted replaced
-1:000000000000 3:9ad51c7bc036
1 =====================================
2 Accurate Garbage Collection with LLVM
3 =====================================
4
5 .. contents::
6 :local:
7
8 Introduction
9 ============
10
11 Garbage collection is a widely used technique that frees the programmer from
12 having to know the lifetimes of heap objects, making software easier to produce
13 and maintain. Many programming languages rely on garbage collection for
14 automatic memory management. There are two primary forms of garbage collection:
15 conservative and accurate.
16
17 Conservative garbage collection often does not require any special support from
18 either the language or the compiler: it can handle non-type-safe programming
19 languages (such as C/C++) and does not require any special information from the
20 compiler. The `Boehm collector
21 <http://www.hpl.hp.com/personal/Hans_Boehm/gc/>`__ is an example of a
22 state-of-the-art conservative collector.
23
24 Accurate garbage collection requires the ability to identify all pointers in the
25 program at run-time (which requires that the source-language be type-safe in
26 most cases). Identifying pointers at run-time requires compiler support to
27 locate all places that hold live pointer variables at run-time, including the
28 :ref:`processor stack and registers <gcroot>`.
29
30 Conservative garbage collection is attractive because it does not require any
31 special compiler support, but it does have problems. In particular, because the
32 conservative garbage collector cannot *know* that a particular word in the
33 machine is a pointer, it cannot move live objects in the heap (preventing the
34 use of compacting and generational GC algorithms) and it can occasionally suffer
35 from memory leaks due to integer values that happen to point to objects in the
36 program. In addition, some aggressive compiler transformations can break
37 conservative garbage collectors (though these seem rare in practice).
38
39 Accurate garbage collectors do not suffer from any of these problems, but they
40 can suffer from degraded scalar optimization of the program. In particular,
41 because the runtime must be able to identify and update all pointers active in
42 the program, some optimizations are less effective. In practice, however, the
43 locality and performance benefits of using aggressive garbage collection
44 techniques dominates any low-level losses.
45
46 This document describes the mechanisms and interfaces provided by LLVM to
47 support accurate garbage collection.
48
49 Goals and non-goals
50 -------------------
51
52 LLVM's intermediate representation provides :ref:`garbage collection intrinsics
53 <gc_intrinsics>` that offer support for a broad class of collector models. For
54 instance, the intrinsics permit:
55
56 * semi-space collectors
57
58 * mark-sweep collectors
59
60 * generational collectors
61
62 * reference counting
63
64 * incremental collectors
65
66 * concurrent collectors
67
68 * cooperative collectors
69
70 We hope that the primitive support built into the LLVM IR is sufficient to
71 support a broad class of garbage collected languages including Scheme, ML, Java,
72 C#, Perl, Python, Lua, Ruby, other scripting languages, and more.
73
74 However, LLVM does not itself provide a garbage collector --- this should be
75 part of your language's runtime library. LLVM provides a framework for compile
76 time :ref:`code generation plugins <plugin>`. The role of these plugins is to
77 generate code and data structures which conforms to the *binary interface*
78 specified by the *runtime library*. This is similar to the relationship between
79 LLVM and DWARF debugging info, for example. The difference primarily lies in
80 the lack of an established standard in the domain of garbage collection --- thus
81 the plugins.
82
83 The aspects of the binary interface with which LLVM's GC support is
84 concerned are:
85
86 * Creation of GC-safe points within code where collection is allowed to execute
87 safely.
88
89 * Computation of the stack map. For each safe point in the code, object
90 references within the stack frame must be identified so that the collector may
91 traverse and perhaps update them.
92
93 * Write barriers when storing object references to the heap. These are commonly
94 used to optimize incremental scans in generational collectors.
95
96 * Emission of read barriers when loading object references. These are useful
97 for interoperating with concurrent collectors.
98
99 There are additional areas that LLVM does not directly address:
100
101 * Registration of global roots with the runtime.
102
103 * Registration of stack map entries with the runtime.
104
105 * The functions used by the program to allocate memory, trigger a collection,
106 etc.
107
108 * Computation or compilation of type maps, or registration of them with the
109 runtime. These are used to crawl the heap for object references.
110
111 In general, LLVM's support for GC does not include features which can be
112 adequately addressed with other features of the IR and does not specify a
113 particular binary interface. On the plus side, this means that you should be
114 able to integrate LLVM with an existing runtime. On the other hand, it leaves a
115 lot of work for the developer of a novel language. However, it's easy to get
116 started quickly and scale up to a more sophisticated implementation as your
117 compiler matures.
118
119 Getting started
120 ===============
121
122 Using a GC with LLVM implies many things, for example:
123
124 * Write a runtime library or find an existing one which implements a GC heap.
125
126 #. Implement a memory allocator.
127
128 #. Design a binary interface for the stack map, used to identify references
129 within a stack frame on the machine stack.\*
130
131 #. Implement a stack crawler to discover functions on the call stack.\*
132
133 #. Implement a registry for global roots.
134
135 #. Design a binary interface for type maps, used to identify references
136 within heap objects.
137
138 #. Implement a collection routine bringing together all of the above.
139
140 * Emit compatible code from your compiler.
141
142 * Initialization in the main function.
143
144 * Use the ``gc "..."`` attribute to enable GC code generation (or
145 ``F.setGC("...")``).
146
147 * Use ``@llvm.gcroot`` to mark stack roots.
148
149 * Use ``@llvm.gcread`` and/or ``@llvm.gcwrite`` to manipulate GC references,
150 if necessary.
151
152 * Allocate memory using the GC allocation routine provided by the runtime
153 library.
154
155 * Generate type maps according to your runtime's binary interface.
156
157 * Write a compiler plugin to interface LLVM with the runtime library.\*
158
159 * Lower ``@llvm.gcread`` and ``@llvm.gcwrite`` to appropriate code
160 sequences.\*
161
162 * Compile LLVM's stack map to the binary form expected by the runtime.
163
164 * Load the plugin into the compiler. Use ``llc -load`` or link the plugin
165 statically with your language's compiler.\*
166
167 * Link program executables with the runtime.
168
169 To help with several of these tasks (those indicated with a \*), LLVM includes a
170 highly portable, built-in ShadowStack code generator. It is compiled into
171 ``llc`` and works even with the interpreter and C backends.
172
173 In your compiler
174 ----------------
175
176 To turn the shadow stack on for your functions, first call:
177
178 .. code-block:: c++
179
180 F.setGC("shadow-stack");
181
182 for each function your compiler emits. Since the shadow stack is built into
183 LLVM, you do not need to load a plugin.
184
185 Your compiler must also use ``@llvm.gcroot`` as documented. Don't forget to
186 create a root for each intermediate value that is generated when evaluating an
187 expression. In ``h(f(), g())``, the result of ``f()`` could easily be collected
188 if evaluating ``g()`` triggers a collection.
189
190 There's no need to use ``@llvm.gcread`` and ``@llvm.gcwrite`` over plain
191 ``load`` and ``store`` for now. You will need them when switching to a more
192 advanced GC.
193
194 In your runtime
195 ---------------
196
197 The shadow stack doesn't imply a memory allocation algorithm. A semispace
198 collector or building atop ``malloc`` are great places to start, and can be
199 implemented with very little code.
200
201 When it comes time to collect, however, your runtime needs to traverse the stack
202 roots, and for this it needs to integrate with the shadow stack. Luckily, doing
203 so is very simple. (This code is heavily commented to help you understand the
204 data structure, but there are only 20 lines of meaningful code.)
205
206 .. code-block:: c++
207
208 /// @brief The map for a single function's stack frame. One of these is
209 /// compiled as constant data into the executable for each function.
210 ///
211 /// Storage of metadata values is elided if the %metadata parameter to
212 /// @llvm.gcroot is null.
213 struct FrameMap {
214 int32_t NumRoots; //< Number of roots in stack frame.
215 int32_t NumMeta; //< Number of metadata entries. May be < NumRoots.
216 const void *Meta[0]; //< Metadata for each root.
217 };
218
219 /// @brief A link in the dynamic shadow stack. One of these is embedded in
220 /// the stack frame of each function on the call stack.
221 struct StackEntry {
222 StackEntry *Next; //< Link to next stack entry (the caller's).
223 const FrameMap *Map; //< Pointer to constant FrameMap.
224 void *Roots[0]; //< Stack roots (in-place array).
225 };
226
227 /// @brief The head of the singly-linked list of StackEntries. Functions push
228 /// and pop onto this in their prologue and epilogue.
229 ///
230 /// Since there is only a global list, this technique is not threadsafe.
231 StackEntry *llvm_gc_root_chain;
232
233 /// @brief Calls Visitor(root, meta) for each GC root on the stack.
234 /// root and meta are exactly the values passed to
235 /// @llvm.gcroot.
236 ///
237 /// Visitor could be a function to recursively mark live objects. Or it
238 /// might copy them to another heap or generation.
239 ///
240 /// @param Visitor A function to invoke for every GC root on the stack.
241 void visitGCRoots(void (*Visitor)(void **Root, const void *Meta)) {
242 for (StackEntry *R = llvm_gc_root_chain; R; R = R->Next) {
243 unsigned i = 0;
244
245 // For roots [0, NumMeta), the metadata pointer is in the FrameMap.
246 for (unsigned e = R->Map->NumMeta; i != e; ++i)
247 Visitor(&R->Roots[i], R->Map->Meta[i]);
248
249 // For roots [NumMeta, NumRoots), the metadata pointer is null.
250 for (unsigned e = R->Map->NumRoots; i != e; ++i)
251 Visitor(&R->Roots[i], NULL);
252 }
253 }
254
255 About the shadow stack
256 ----------------------
257
258 Unlike many GC algorithms which rely on a cooperative code generator to compile
259 stack maps, this algorithm carefully maintains a linked list of stack roots
260 [:ref:`Henderson2002 <henderson02>`]. This so-called "shadow stack" mirrors the
261 machine stack. Maintaining this data structure is slower than using a stack map
262 compiled into the executable as constant data, but has a significant portability
263 advantage because it requires no special support from the target code generator,
264 and does not require tricky platform-specific code to crawl the machine stack.
265
266 The tradeoff for this simplicity and portability is:
267
268 * High overhead per function call.
269
270 * Not thread-safe.
271
272 Still, it's an easy way to get started. After your compiler and runtime are up
273 and running, writing a :ref:`plugin <plugin>` will allow you to take advantage
274 of :ref:`more advanced GC features <collector-algos>` of LLVM in order to
275 improve performance.
276
277 .. _gc_intrinsics:
278
279 IR features
280 ===========
281
282 This section describes the garbage collection facilities provided by the
283 :doc:`LLVM intermediate representation <LangRef>`. The exact behavior of these
284 IR features is specified by the binary interface implemented by a :ref:`code
285 generation plugin <plugin>`, not by this document.
286
287 These facilities are limited to those strictly necessary; they are not intended
288 to be a complete interface to any garbage collector. A program will need to
289 interface with the GC library using the facilities provided by that program.
290
291 Specifying GC code generation: ``gc "..."``
292 -------------------------------------------
293
294 .. code-block:: llvm
295
296 define ty @name(...) gc "name" { ...
297
298 The ``gc`` function attribute is used to specify the desired GC style to the
299 compiler. Its programmatic equivalent is the ``setGC`` method of ``Function``.
300
301 Setting ``gc "name"`` on a function triggers a search for a matching code
302 generation plugin "*name*"; it is that plugin which defines the exact nature of
303 the code generated to support GC. If none is found, the compiler will raise an
304 error.
305
306 Specifying the GC style on a per-function basis allows LLVM to link together
307 programs that use different garbage collection algorithms (or none at all).
308
309 .. _gcroot:
310
311 Identifying GC roots on the stack: ``llvm.gcroot``
312 --------------------------------------------------
313
314 .. code-block:: llvm
315
316 void @llvm.gcroot(i8** %ptrloc, i8* %metadata)
317
318 The ``llvm.gcroot`` intrinsic is used to inform LLVM that a stack variable
319 references an object on the heap and is to be tracked for garbage collection.
320 The exact impact on generated code is specified by a :ref:`compiler plugin
321 <plugin>`. All calls to ``llvm.gcroot`` **must** reside inside the first basic
322 block.
323
324 A compiler which uses mem2reg to raise imperative code using ``alloca`` into SSA
325 form need only add a call to ``@llvm.gcroot`` for those variables which a
326 pointers into the GC heap.
327
328 It is also important to mark intermediate values with ``llvm.gcroot``. For
329 example, consider ``h(f(), g())``. Beware leaking the result of ``f()`` in the
330 case that ``g()`` triggers a collection. Note, that stack variables must be
331 initialized and marked with ``llvm.gcroot`` in function's prologue.
332
333 The first argument **must** be a value referring to an alloca instruction or a
334 bitcast of an alloca. The second contains a pointer to metadata that should be
335 associated with the pointer, and **must** be a constant or global value
336 address. If your target collector uses tags, use a null pointer for metadata.
337
338 The ``%metadata`` argument can be used to avoid requiring heap objects to have
339 'isa' pointers or tag bits. [Appel89_, Goldberg91_, Tolmach94_] If specified,
340 its value will be tracked along with the location of the pointer in the stack
341 frame.
342
343 Consider the following fragment of Java code:
344
345 .. code-block:: java
346
347 {
348 Object X; // A null-initialized reference to an object
349 ...
350 }
351
352 This block (which may be located in the middle of a function or in a loop nest),
353 could be compiled to this LLVM code:
354
355 .. code-block:: llvm
356
357 Entry:
358 ;; In the entry block for the function, allocate the
359 ;; stack space for X, which is an LLVM pointer.
360 %X = alloca %Object*
361
362 ;; Tell LLVM that the stack space is a stack root.
363 ;; Java has type-tags on objects, so we pass null as metadata.
364 %tmp = bitcast %Object** %X to i8**
365 call void @llvm.gcroot(i8** %tmp, i8* null)
366 ...
367
368 ;; "CodeBlock" is the block corresponding to the start
369 ;; of the scope above.
370 CodeBlock:
371 ;; Java null-initializes pointers.
372 store %Object* null, %Object** %X
373
374 ...
375
376 ;; As the pointer goes out of scope, store a null value into
377 ;; it, to indicate that the value is no longer live.
378 store %Object* null, %Object** %X
379 ...
380
381 Reading and writing references in the heap
382 ------------------------------------------
383
384 Some collectors need to be informed when the mutator (the program that needs
385 garbage collection) either reads a pointer from or writes a pointer to a field
386 of a heap object. The code fragments inserted at these points are called *read
387 barriers* and *write barriers*, respectively. The amount of code that needs to
388 be executed is usually quite small and not on the critical path of any
389 computation, so the overall performance impact of the barrier is tolerable.
390
391 Barriers often require access to the *object pointer* rather than the *derived
392 pointer* (which is a pointer to the field within the object). Accordingly,
393 these intrinsics take both pointers as separate arguments for completeness. In
394 this snippet, ``%object`` is the object pointer, and ``%derived`` is the derived
395 pointer:
396
397 .. code-block:: llvm
398
399 ;; An array type.
400 %class.Array = type { %class.Object, i32, [0 x %class.Object*] }
401 ...
402
403 ;; Load the object pointer from a gcroot.
404 %object = load %class.Array** %object_addr
405
406 ;; Compute the derived pointer.
407 %derived = getelementptr %object, i32 0, i32 2, i32 %n
408
409 LLVM does not enforce this relationship between the object and derived pointer
410 (although a :ref:`plugin <plugin>` might). However, it would be an unusual
411 collector that violated it.
412
413 The use of these intrinsics is naturally optional if the target GC does require
414 the corresponding barrier. Such a GC plugin will replace the intrinsic calls
415 with the corresponding ``load`` or ``store`` instruction if they are used.
416
417 Write barrier: ``llvm.gcwrite``
418 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
419
420 .. code-block:: llvm
421
422 void @llvm.gcwrite(i8* %value, i8* %object, i8** %derived)
423
424 For write barriers, LLVM provides the ``llvm.gcwrite`` intrinsic function. It
425 has exactly the same semantics as a non-volatile ``store`` to the derived
426 pointer (the third argument). The exact code generated is specified by a
427 compiler :ref:`plugin <plugin>`.
428
429 Many important algorithms require write barriers, including generational and
430 concurrent collectors. Additionally, write barriers could be used to implement
431 reference counting.
432
433 Read barrier: ``llvm.gcread``
434 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
435
436 .. code-block:: llvm
437
438 i8* @llvm.gcread(i8* %object, i8** %derived)
439
440 For read barriers, LLVM provides the ``llvm.gcread`` intrinsic function. It has
441 exactly the same semantics as a non-volatile ``load`` from the derived pointer
442 (the second argument). The exact code generated is specified by a
443 :ref:`compiler plugin <plugin>`.
444
445 Read barriers are needed by fewer algorithms than write barriers, and may have a
446 greater performance impact since pointer reads are more frequent than writes.
447
448 .. _plugin:
449
450 Implementing a collector plugin
451 ===============================
452
453 User code specifies which GC code generation to use with the ``gc`` function
454 attribute or, equivalently, with the ``setGC`` method of ``Function``.
455
456 To implement a GC plugin, it is necessary to subclass ``llvm::GCStrategy``,
457 which can be accomplished in a few lines of boilerplate code. LLVM's
458 infrastructure provides access to several important algorithms. For an
459 uncontroversial collector, all that remains may be to compile LLVM's computed
460 stack map to assembly code (using the binary representation expected by the
461 runtime library). This can be accomplished in about 100 lines of code.
462
463 This is not the appropriate place to implement a garbage collected heap or a
464 garbage collector itself. That code should exist in the language's runtime
465 library. The compiler plugin is responsible for generating code which conforms
466 to the binary interface defined by library, most essentially the :ref:`stack map
467 <stack-map>`.
468
469 To subclass ``llvm::GCStrategy`` and register it with the compiler:
470
471 .. code-block:: c++
472
473 // lib/MyGC/MyGC.cpp - Example LLVM GC plugin
474
475 #include "llvm/CodeGen/GCStrategy.h"
476 #include "llvm/CodeGen/GCMetadata.h"
477 #include "llvm/Support/Compiler.h"
478
479 using namespace llvm;
480
481 namespace {
482 class LLVM_LIBRARY_VISIBILITY MyGC : public GCStrategy {
483 public:
484 MyGC() {}
485 };
486
487 GCRegistry::Add<MyGC>
488 X("mygc", "My bespoke garbage collector.");
489 }
490
491 This boilerplate collector does nothing. More specifically:
492
493 * ``llvm.gcread`` calls are replaced with the corresponding ``load``
494 instruction.
495
496 * ``llvm.gcwrite`` calls are replaced with the corresponding ``store``
497 instruction.
498
499 * No safe points are added to the code.
500
501 * The stack map is not compiled into the executable.
502
503 Using the LLVM makefiles (like the `sample project
504 <http://llvm.org/viewvc/llvm-project/llvm/trunk/projects/sample/>`__), this code
505 can be compiled as a plugin using a simple makefile:
506
507 .. code-block:: make
508
509 # lib/MyGC/Makefile
510
511 LEVEL := ../..
512 LIBRARYNAME = MyGC
513 LOADABLE_MODULE = 1
514
515 include $(LEVEL)/Makefile.common
516
517 Once the plugin is compiled, code using it may be compiled using ``llc
518 -load=MyGC.so`` (though MyGC.so may have some other platform-specific
519 extension):
520
521 ::
522
523 $ cat sample.ll
524 define void @f() gc "mygc" {
525 entry:
526 ret void
527 }
528 $ llvm-as < sample.ll | llc -load=MyGC.so
529
530 It is also possible to statically link the collector plugin into tools, such as
531 a language-specific compiler front-end.
532
533 .. _collector-algos:
534
535 Overview of available features
536 ------------------------------
537
538 ``GCStrategy`` provides a range of features through which a plugin may do useful
539 work. Some of these are callbacks, some are algorithms that can be enabled,
540 disabled, or customized. This matrix summarizes the supported (and planned)
541 features and correlates them with the collection techniques which typically
542 require them.
543
544 .. |v| unicode:: 0x2714
545 :trim:
546
547 .. |x| unicode:: 0x2718
548 :trim:
549
550 +------------+------+--------+----------+-------+---------+-------------+----------+------------+
551 | Algorithm | Done | Shadow | refcount | mark- | copying | incremental | threaded | concurrent |
552 | | | stack | | sweep | | | | |
553 +============+======+========+==========+=======+=========+=============+==========+============+
554 | stack map | |v| | | | |x| | |x| | |x| | |x| | |x| |
555 +------------+------+--------+----------+-------+---------+-------------+----------+------------+
556 | initialize | |v| | |x| | |x| | |x| | |x| | |x| | |x| | |x| |
557 | roots | | | | | | | | |
558 +------------+------+--------+----------+-------+---------+-------------+----------+------------+
559 | derived | NO | | | | | | **N**\* | **N**\* |
560 | pointers | | | | | | | | |
561 +------------+------+--------+----------+-------+---------+-------------+----------+------------+
562 | **custom | |v| | | | | | | | |
563 | lowering** | | | | | | | | |
564 +------------+------+--------+----------+-------+---------+-------------+----------+------------+
565 | *gcroot* | |v| | |x| | |x| | | | | | |
566 +------------+------+--------+----------+-------+---------+-------------+----------+------------+
567 | *gcwrite* | |v| | | |x| | | | |x| | | |x| |
568 +------------+------+--------+----------+-------+---------+-------------+----------+------------+
569 | *gcread* | |v| | | | | | | | |x| |
570 +------------+------+--------+----------+-------+---------+-------------+----------+------------+
571 | **safe | | | | | | | | |
572 | points** | | | | | | | | |
573 +------------+------+--------+----------+-------+---------+-------------+----------+------------+
574 | *in | |v| | | | |x| | |x| | |x| | |x| | |x| |
575 | calls* | | | | | | | | |
576 +------------+------+--------+----------+-------+---------+-------------+----------+------------+
577 | *before | |v| | | | | | | |x| | |x| |
578 | calls* | | | | | | | | |
579 +------------+------+--------+----------+-------+---------+-------------+----------+------------+
580 | *for | NO | | | | | | **N** | **N** |
581 | loops* | | | | | | | | |
582 +------------+------+--------+----------+-------+---------+-------------+----------+------------+
583 | *before | |v| | | | | | | |x| | |x| |
584 | escape* | | | | | | | | |
585 +------------+------+--------+----------+-------+---------+-------------+----------+------------+
586 | emit code | NO | | | | | | **N** | **N** |
587 | at safe | | | | | | | | |
588 | points | | | | | | | | |
589 +------------+------+--------+----------+-------+---------+-------------+----------+------------+
590 | **output** | | | | | | | | |
591 +------------+------+--------+----------+-------+---------+-------------+----------+------------+
592 | *assembly* | |v| | | | |x| | |x| | |x| | |x| | |x| |
593 +------------+------+--------+----------+-------+---------+-------------+----------+------------+
594 | *JIT* | NO | | | **?** | **?** | **?** | **?** | **?** |
595 +------------+------+--------+----------+-------+---------+-------------+----------+------------+
596 | *obj* | NO | | | **?** | **?** | **?** | **?** | **?** |
597 +------------+------+--------+----------+-------+---------+-------------+----------+------------+
598 | live | NO | | | **?** | **?** | **?** | **?** | **?** |
599 | analysis | | | | | | | | |
600 +------------+------+--------+----------+-------+---------+-------------+----------+------------+
601 | register | NO | | | **?** | **?** | **?** | **?** | **?** |
602 | map | | | | | | | | |
603 +------------+------+--------+----------+-------+---------+-------------+----------+------------+
604 | \* Derived pointers only pose a hasard to copying collections. |
605 +------------+------+--------+----------+-------+---------+-------------+----------+------------+
606 | **?** denotes a feature which could be utilized if available. |
607 +------------+------+--------+----------+-------+---------+-------------+----------+------------+
608
609 To be clear, the collection techniques above are defined as:
610
611 Shadow Stack
612 The mutator carefully maintains a linked list of stack roots.
613
614 Reference Counting
615 The mutator maintains a reference count for each object and frees an object
616 when its count falls to zero.
617
618 Mark-Sweep
619 When the heap is exhausted, the collector marks reachable objects starting
620 from the roots, then deallocates unreachable objects in a sweep phase.
621
622 Copying
623 As reachability analysis proceeds, the collector copies objects from one heap
624 area to another, compacting them in the process. Copying collectors enable
625 highly efficient "bump pointer" allocation and can improve locality of
626 reference.
627
628 Incremental
629 (Including generational collectors.) Incremental collectors generally have all
630 the properties of a copying collector (regardless of whether the mature heap
631 is compacting), but bring the added complexity of requiring write barriers.
632
633 Threaded
634 Denotes a multithreaded mutator; the collector must still stop the mutator
635 ("stop the world") before beginning reachability analysis. Stopping a
636 multithreaded mutator is a complicated problem. It generally requires highly
637 platform specific code in the runtime, and the production of carefully
638 designed machine code at safe points.
639
640 Concurrent
641 In this technique, the mutator and the collector run concurrently, with the
642 goal of eliminating pause times. In a *cooperative* collector, the mutator
643 further aids with collection should a pause occur, allowing collection to take
644 advantage of multiprocessor hosts. The "stop the world" problem of threaded
645 collectors is generally still present to a limited extent. Sophisticated
646 marking algorithms are necessary. Read barriers may be necessary.
647
648 As the matrix indicates, LLVM's garbage collection infrastructure is already
649 suitable for a wide variety of collectors, but does not currently extend to
650 multithreaded programs. This will be added in the future as there is
651 interest.
652
653 .. _stack-map:
654
655 Computing stack maps
656 --------------------
657
658 LLVM automatically computes a stack map. One of the most important features
659 of a ``GCStrategy`` is to compile this information into the executable in
660 the binary representation expected by the runtime library.
661
662 The stack map consists of the location and identity of each GC root in the
663 each function in the module. For each root:
664
665 * ``RootNum``: The index of the root.
666
667 * ``StackOffset``: The offset of the object relative to the frame pointer.
668
669 * ``RootMetadata``: The value passed as the ``%metadata`` parameter to the
670 ``@llvm.gcroot`` intrinsic.
671
672 Also, for the function as a whole:
673
674 * ``getFrameSize()``: The overall size of the function's initial stack frame,
675 not accounting for any dynamic allocation.
676
677 * ``roots_size()``: The count of roots in the function.
678
679 To access the stack map, use ``GCFunctionMetadata::roots_begin()`` and
680 -``end()`` from the :ref:`GCMetadataPrinter <assembly>`:
681
682 .. code-block:: c++
683
684 for (iterator I = begin(), E = end(); I != E; ++I) {
685 GCFunctionInfo *FI = *I;
686 unsigned FrameSize = FI->getFrameSize();
687 size_t RootCount = FI->roots_size();
688
689 for (GCFunctionInfo::roots_iterator RI = FI->roots_begin(),
690 RE = FI->roots_end();
691 RI != RE; ++RI) {
692 int RootNum = RI->Num;
693 int RootStackOffset = RI->StackOffset;
694 Constant *RootMetadata = RI->Metadata;
695 }
696 }
697
698 If the ``llvm.gcroot`` intrinsic is eliminated before code generation by a
699 custom lowering pass, LLVM will compute an empty stack map. This may be useful
700 for collector plugins which implement reference counting or a shadow stack.
701
702 .. _init-roots:
703
704 Initializing roots to null: ``InitRoots``
705 -----------------------------------------
706
707 .. code-block:: c++
708
709 MyGC::MyGC() {
710 InitRoots = true;
711 }
712
713 When set, LLVM will automatically initialize each root to ``null`` upon entry to
714 the function. This prevents the GC's sweep phase from visiting uninitialized
715 pointers, which will almost certainly cause it to crash. This initialization
716 occurs before custom lowering, so the two may be used together.
717
718 Since LLVM does not yet compute liveness information, there is no means of
719 distinguishing an uninitialized stack root from an initialized one. Therefore,
720 this feature should be used by all GC plugins. It is enabled by default.
721
722 Custom lowering of intrinsics: ``CustomRoots``, ``CustomReadBarriers``, and ``CustomWriteBarriers``
723 ---------------------------------------------------------------------------------------------------
724
725 For GCs which use barriers or unusual treatment of stack roots, these flags
726 allow the collector to perform arbitrary transformations of the LLVM IR:
727
728 .. code-block:: c++
729
730 class MyGC : public GCStrategy {
731 public:
732 MyGC() {
733 CustomRoots = true;
734 CustomReadBarriers = true;
735 CustomWriteBarriers = true;
736 }
737
738 virtual bool initializeCustomLowering(Module &M);
739 virtual bool performCustomLowering(Function &F);
740 };
741
742 If any of these flags are set, then LLVM suppresses its default lowering for the
743 corresponding intrinsics and instead calls ``performCustomLowering``.
744
745 LLVM's default action for each intrinsic is as follows:
746
747 * ``llvm.gcroot``: Leave it alone. The code generator must see it or the stack
748 map will not be computed.
749
750 * ``llvm.gcread``: Substitute a ``load`` instruction.
751
752 * ``llvm.gcwrite``: Substitute a ``store`` instruction.
753
754 If ``CustomReadBarriers`` or ``CustomWriteBarriers`` are specified, then
755 ``performCustomLowering`` **must** eliminate the corresponding barriers.
756
757 ``performCustomLowering`` must comply with the same restrictions as
758 :ref:`FunctionPass::runOnFunction <writing-an-llvm-pass-runOnFunction>`
759 Likewise, ``initializeCustomLowering`` has the same semantics as
760 :ref:`Pass::doInitialization(Module&)
761 <writing-an-llvm-pass-doInitialization-mod>`
762
763 The following can be used as a template:
764
765 .. code-block:: c++
766
767 #include "llvm/Module.h"
768 #include "llvm/IntrinsicInst.h"
769
770 bool MyGC::initializeCustomLowering(Module &M) {
771 return false;
772 }
773
774 bool MyGC::performCustomLowering(Function &F) {
775 bool MadeChange = false;
776
777 for (Function::iterator BB = F.begin(), E = F.end(); BB != E; ++BB)
778 for (BasicBlock::iterator II = BB->begin(), E = BB->end(); II != E; )
779 if (IntrinsicInst *CI = dyn_cast<IntrinsicInst>(II++))
780 if (Function *F = CI->getCalledFunction())
781 switch (F->getIntrinsicID()) {
782 case Intrinsic::gcwrite:
783 // Handle llvm.gcwrite.
784 CI->eraseFromParent();
785 MadeChange = true;
786 break;
787 case Intrinsic::gcread:
788 // Handle llvm.gcread.
789 CI->eraseFromParent();
790 MadeChange = true;
791 break;
792 case Intrinsic::gcroot:
793 // Handle llvm.gcroot.
794 CI->eraseFromParent();
795 MadeChange = true;
796 break;
797 }
798
799 return MadeChange;
800 }
801
802 .. _safe-points:
803
804 Generating safe points: ``NeededSafePoints``
805 --------------------------------------------
806
807 LLVM can compute four kinds of safe points:
808
809 .. code-block:: c++
810
811 namespace GC {
812 /// PointKind - The type of a collector-safe point.
813 ///
814 enum PointKind {
815 Loop, //< Instr is a loop (backwards branch).
816 Return, //< Instr is a return instruction.
817 PreCall, //< Instr is a call instruction.
818 PostCall //< Instr is the return address of a call.
819 };
820 }
821
822 A collector can request any combination of the four by setting the
823 ``NeededSafePoints`` mask:
824
825 .. code-block:: c++
826
827 MyGC::MyGC() {
828 NeededSafePoints = 1 << GC::Loop
829 | 1 << GC::Return
830 | 1 << GC::PreCall
831 | 1 << GC::PostCall;
832 }
833
834 It can then use the following routines to access safe points.
835
836 .. code-block:: c++
837
838 for (iterator I = begin(), E = end(); I != E; ++I) {
839 GCFunctionInfo *MD = *I;
840 size_t PointCount = MD->size();
841
842 for (GCFunctionInfo::iterator PI = MD->begin(),
843 PE = MD->end(); PI != PE; ++PI) {
844 GC::PointKind PointKind = PI->Kind;
845 unsigned PointNum = PI->Num;
846 }
847 }
848
849 Almost every collector requires ``PostCall`` safe points, since these correspond
850 to the moments when the function is suspended during a call to a subroutine.
851
852 Threaded programs generally require ``Loop`` safe points to guarantee that the
853 application will reach a safe point within a bounded amount of time, even if it
854 is executing a long-running loop which contains no function calls.
855
856 Threaded collectors may also require ``Return`` and ``PreCall`` safe points to
857 implement "stop the world" techniques using self-modifying code, where it is
858 important that the program not exit the function without reaching a safe point
859 (because only the topmost function has been patched).
860
861 .. _assembly:
862
863 Emitting assembly code: ``GCMetadataPrinter``
864 ---------------------------------------------
865
866 LLVM allows a plugin to print arbitrary assembly code before and after the rest
867 of a module's assembly code. At the end of the module, the GC can compile the
868 LLVM stack map into assembly code. (At the beginning, this information is not
869 yet computed.)
870
871 Since AsmWriter and CodeGen are separate components of LLVM, a separate abstract
872 base class and registry is provided for printing assembly code, the
873 ``GCMetadaPrinter`` and ``GCMetadataPrinterRegistry``. The AsmWriter will look
874 for such a subclass if the ``GCStrategy`` sets ``UsesMetadata``:
875
876 .. code-block:: c++
877
878 MyGC::MyGC() {
879 UsesMetadata = true;
880 }
881
882 This separation allows JIT-only clients to be smaller.
883
884 Note that LLVM does not currently have analogous APIs to support code generation
885 in the JIT, nor using the object writers.
886
887 .. code-block:: c++
888
889 // lib/MyGC/MyGCPrinter.cpp - Example LLVM GC printer
890
891 #include "llvm/CodeGen/GCMetadataPrinter.h"
892 #include "llvm/Support/Compiler.h"
893
894 using namespace llvm;
895
896 namespace {
897 class LLVM_LIBRARY_VISIBILITY MyGCPrinter : public GCMetadataPrinter {
898 public:
899 virtual void beginAssembly(std::ostream &OS, AsmPrinter &AP,
900 const TargetAsmInfo &TAI);
901
902 virtual void finishAssembly(std::ostream &OS, AsmPrinter &AP,
903 const TargetAsmInfo &TAI);
904 };
905
906 GCMetadataPrinterRegistry::Add<MyGCPrinter>
907 X("mygc", "My bespoke garbage collector.");
908 }
909
910 The collector should use ``AsmPrinter`` and ``TargetAsmInfo`` to print portable
911 assembly code to the ``std::ostream``. The collector itself contains the stack
912 map for the entire module, and may access the ``GCFunctionInfo`` using its own
913 ``begin()`` and ``end()`` methods. Here's a realistic example:
914
915 .. code-block:: c++
916
917 #include "llvm/CodeGen/AsmPrinter.h"
918 #include "llvm/Function.h"
919 #include "llvm/Target/TargetMachine.h"
920 #include "llvm/DataLayout.h"
921 #include "llvm/Target/TargetAsmInfo.h"
922
923 void MyGCPrinter::beginAssembly(std::ostream &OS, AsmPrinter &AP,
924 const TargetAsmInfo &TAI) {
925 // Nothing to do.
926 }
927
928 void MyGCPrinter::finishAssembly(std::ostream &OS, AsmPrinter &AP,
929 const TargetAsmInfo &TAI) {
930 // Set up for emitting addresses.
931 const char *AddressDirective;
932 int AddressAlignLog;
933 if (AP.TM.getDataLayout()->getPointerSize() == sizeof(int32_t)) {
934 AddressDirective = TAI.getData32bitsDirective();
935 AddressAlignLog = 2;
936 } else {
937 AddressDirective = TAI.getData64bitsDirective();
938 AddressAlignLog = 3;
939 }
940
941 // Put this in the data section.
942 AP.SwitchToDataSection(TAI.getDataSection());
943
944 // For each function...
945 for (iterator FI = begin(), FE = end(); FI != FE; ++FI) {
946 GCFunctionInfo &MD = **FI;
947
948 // Emit this data structure:
949 //
950 // struct {
951 // int32_t PointCount;
952 // struct {
953 // void *SafePointAddress;
954 // int32_t LiveCount;
955 // int32_t LiveOffsets[LiveCount];
956 // } Points[PointCount];
957 // } __gcmap_<FUNCTIONNAME>;
958
959 // Align to address width.
960 AP.EmitAlignment(AddressAlignLog);
961
962 // Emit the symbol by which the stack map entry can be found.
963 std::string Symbol;
964 Symbol += TAI.getGlobalPrefix();
965 Symbol += "__gcmap_";
966 Symbol += MD.getFunction().getName();
967 if (const char *GlobalDirective = TAI.getGlobalDirective())
968 OS << GlobalDirective << Symbol << "\n";
969 OS << TAI.getGlobalPrefix() << Symbol << ":\n";
970
971 // Emit PointCount.
972 AP.EmitInt32(MD.size());
973 AP.EOL("safe point count");
974
975 // And each safe point...
976 for (GCFunctionInfo::iterator PI = MD.begin(),
977 PE = MD.end(); PI != PE; ++PI) {
978 // Align to address width.
979 AP.EmitAlignment(AddressAlignLog);
980
981 // Emit the address of the safe point.
982 OS << AddressDirective
983 << TAI.getPrivateGlobalPrefix() << "label" << PI->Num;
984 AP.EOL("safe point address");
985
986 // Emit the stack frame size.
987 AP.EmitInt32(MD.getFrameSize());
988 AP.EOL("stack frame size");
989
990 // Emit the number of live roots in the function.
991 AP.EmitInt32(MD.live_size(PI));
992 AP.EOL("live root count");
993
994 // And for each live root...
995 for (GCFunctionInfo::live_iterator LI = MD.live_begin(PI),
996 LE = MD.live_end(PI);
997 LI != LE; ++LI) {
998 // Print its offset within the stack frame.
999 AP.EmitInt32(LI->StackOffset);
1000 AP.EOL("stack offset");
1001 }
1002 }
1003 }
1004 }
1005
1006 References
1007 ==========
1008
1009 .. _appel89:
1010
1011 [Appel89] Runtime Tags Aren't Necessary. Andrew W. Appel. Lisp and Symbolic
1012 Computation 19(7):703-705, July 1989.
1013
1014 .. _goldberg91:
1015
1016 [Goldberg91] Tag-free garbage collection for strongly typed programming
1017 languages. Benjamin Goldberg. ACM SIGPLAN PLDI'91.
1018
1019 .. _tolmach94:
1020
1021 [Tolmach94] Tag-free garbage collection using explicit type parameters. Andrew
1022 Tolmach. Proceedings of the 1994 ACM conference on LISP and functional
1023 programming.
1024
1025 .. _henderson02:
1026
1027 [Henderson2002] `Accurate Garbage Collection in an Uncooperative Environment
1028 <http://citeseer.ist.psu.edu/henderson02accurate.html>`__
1029