• Home
  • Line#
  • Scopes#
  • Navigate#
  • Raw
  • Download
1Inlining
2========
3
4There are several options that control which calls the analyzer will consider for
5inlining. The major one is -analyzer-ipa:
6
7  -analyzer-ipa=none - All inlining is disabled. This is the only mode available
8     in LLVM 3.1 and earlier and in Xcode 4.3 and earlier.
9
10  -analyzer-ipa=basic-inlining - Turns on inlining for C functions, C++ static
11     member functions, and blocks -- essentially, the calls that behave like
12     simple C function calls. This is essentially the mode used in Xcode 4.4.
13
14  -analyzer-ipa=inlining - Turns on inlining when we can confidently find the
15    function/method body corresponding to the call. (C functions, static
16    functions, devirtualized C++ methods, Objective-C class methods, Objective-C
17    instance methods when ExprEngine is confident about the dynamic type of the
18    instance).
19
20  -analyzer-ipa=dynamic - Inline instance methods for which the type is
21   determined at runtime and we are not 100% sure that our type info is
22   correct. For virtual calls, inline the most plausible definition.
23
24  -analyzer-ipa=dynamic-bifurcate - Same as -analyzer-ipa=dynamic, but the path
25   is split. We inline on one branch and do not inline on the other. This mode
26   does not drop the coverage in cases when the parent class has code that is
27   only exercised when some of its methods are overridden.
28
29Currently, -analyzer-ipa=dynamic-bifurcate is the default mode.
30
31While -analyzer-ipa determines in general how aggressively the analyzer will try to
32inline functions, several additional options control which types of functions can
33inlined, in an all-or-nothing way. These options use the analyzer's configuration
34table, so they are all specified as follows:
35
36    -analyzer-config OPTION=VALUE
37
38### c++-inlining ###
39
40This option controls which C++ member functions may be inlined.
41
42    -analyzer-config c++-inlining=[none | methods | constructors | destructors]
43
44Each of these modes implies that all the previous member function kinds will be
45inlined as well; it doesn't make sense to inline destructors without inlining
46constructors, for example.
47
48The default c++-inlining mode is 'methods', meaning only regular member
49functions and overloaded operators will be inlined. Note that no C++ member
50functions will be inlined under -analyzer-ipa=none or
51-analyzer-ipa=basic-inlining.
52
53### c++-template-inlining ###
54
55This option controls whether C++ templated functions may be inlined.
56
57    -analyzer-config c++-template-inlining=[true | false]
58
59Currently, template functions are considered for inlining by default.
60
61The motivation behind this option is that very generic code can be a source
62of false positives, either by considering paths that the caller considers
63impossible (by some unstated precondition), or by inlining some but not all
64of a deep implementation of a function.
65
66### c++-stdlib-inlining ###
67
68This option controls whether functions from the C++ standard library, including
69methods of the container classes in the Standard Template Library, should be
70considered for inlining.
71
72    -analyzer-config c++-template-inlining=[true | false]
73
74Currently, C++ standard library functions are NOT considered for inlining by default.
75
76The standard library functions and the STL in particular are used ubiquitously
77enough that our tolerance for false positives is even lower here. A false
78positive due to poor modeling of the STL leads to a poor user experience, since
79most users would not be comfortable adding assertions to system headers in order
80to silence analyzer warnings.
81
82
83Basics of Implementation
84-----------------------
85
86The low-level mechanism of inlining a function is handled in
87ExprEngine::inlineCall and ExprEngine::processCallExit.
88
89If the conditions are right for inlining, a CallEnter node is created and added
90to the analysis work list. The CallEnter node marks the change to a new
91LocationContext representing the called function, and its state includes the
92contents of the new stack frame. When the CallEnter node is actually processed,
93its single successor will be a edge to the first CFG block in the function.
94
95Exiting an inlined function is a bit more work, fortunately broken up into
96reasonable steps:
97
981. The CoreEngine realizes we're at the end of an inlined call and generates a
99   CallExitBegin node.
100
1012. ExprEngine takes over (in processCallExit) and finds the return value of the
102   function, if it has one. This is bound to the expression that triggered the
103   call. (In the case of calls without origin expressions, such as destructors,
104   this step is skipped.)
105
1063. Dead symbols and bindings are cleaned out from the state, including any local
107   bindings.
108
1094. A CallExitEnd node is generated, which marks the transition back to the
110   caller's LocationContext.
111
1125. Custom post-call checks are processed and the final nodes are pushed back
113   onto the work list, so that evaluation of the caller can continue.
114
115Retry Without Inlining
116----------------------
117
118In some cases, we would like to retry analysis without inlining a particular
119call.
120
121Currently, we use this technique to recover coverage in case we stop
122analyzing a path due to exceeding the maximum block count inside an inlined
123function.
124
125When this situation is detected, we walk up the path to find the first node
126before inlining was started and enqueue it on the WorkList with a special
127ReplayWithoutInlining bit added to it (ExprEngine::replayWithoutInlining).  The
128path is then re-analyzed from that point without inlining that particular call.
129
130Deciding When to Inline
131-----------------------
132
133In general, the analyzer attempts to inline as much as possible, since it
134provides a better summary of what actually happens in the program.  There are
135some cases, however, where the analyzer chooses not to inline:
136
137- If there is no definition available for the called function or method.  In
138  this case, there is no opportunity to inline.
139
140- If the CFG cannot be constructed for a called function, or the liveness
141  cannot be computed.  These are prerequisites for analyzing a function body,
142  with or without inlining.
143
144- If the LocationContext chain for a given ExplodedNode reaches a maximum cutoff
145  depth.  This prevents unbounded analysis due to infinite recursion, but also
146  serves as a useful cutoff for performance reasons.
147
148- If the function is variadic.  This is not a hard limitation, but an engineering
149  limitation.
150
151  Tracked by: <rdar://problem/12147064> Support inlining of variadic functions
152
153- In C++, constructors are not inlined unless the destructor call will be
154  processed by the ExprEngine. Thus, if the CFG was built without nodes for
155  implicit destructors, or if the destructors for the given object are not
156  represented in the CFG, the constructor will not be inlined. (As an exception,
157  constructors for objects with trivial constructors can still be inlined.)
158  See "C++ Caveats" below.
159
160- In C++, ExprEngine does not inline custom implementations of operator 'new'
161  or operator 'delete', nor does it inline the constructors and destructors
162  associated with these. See "C++ Caveats" below.
163
164- Calls resulting in "dynamic dispatch" are specially handled.  See more below.
165
166- The FunctionSummaries map stores additional information about declarations,
167  some of which is collected at runtime based on previous analyses.
168  We do not inline functions which were not profitable to inline in a different
169  context (for example, if the maximum block count was exceeded; see
170  "Retry Without Inlining").
171
172
173Dynamic Calls and Devirtualization
174----------------------------------
175
176"Dynamic" calls are those that are resolved at runtime, such as C++ virtual
177method calls and Objective-C message sends. Due to the path-sensitive nature of
178the analysis, the analyzer may be able to reason about the dynamic type of the
179object whose method is being called and thus "devirtualize" the call.
180
181This path-sensitive devirtualization occurs when the analyzer can determine what
182method would actually be called at runtime.  This is possible when the type
183information is constrained enough for a simulated C++/Objective-C object that
184the analyzer can make such a decision.
185
186 == DynamicTypeInfo ==
187
188As the analyzer analyzes a path, it may accrue information to refine the
189knowledge about the type of an object.  This can then be used to make better
190decisions about the target method of a call.
191
192Such type information is tracked as DynamicTypeInfo.  This is path-sensitive
193data that is stored in ProgramState, which defines a mapping from MemRegions to
194an (optional) DynamicTypeInfo.
195
196If no DynamicTypeInfo has been explicitly set for a MemRegion, it will be lazily
197inferred from the region's type or associated symbol. Information from symbolic
198regions is weaker than from true typed regions.
199
200  EXAMPLE: A C++ object declared "A obj" is known to have the class 'A', but a
201           reference "A &ref" may dynamically be a subclass of 'A'.
202
203The DynamicTypePropagation checker gathers and propagates DynamicTypeInfo,
204updating it as information is observed along a path that can refine that type
205information for a region.
206
207  WARNING: Not all of the existing analyzer code has been retrofitted to use
208           DynamicTypeInfo, nor is it universally appropriate. In particular,
209           DynamicTypeInfo always applies to a region with all casts stripped
210           off, but sometimes the information provided by casts can be useful.
211
212
213 == RuntimeDefinition ==
214
215The basis of devirtualization is CallEvent's getRuntimeDefinition() method,
216which returns a RuntimeDefinition object.  When asked to provide a definition,
217the CallEvents for dynamic calls will use the DynamicTypeInfo in their
218ProgramState to attempt to devirtualize the call.  In the case of no dynamic
219dispatch, or perfectly constrained devirtualization, the resulting
220RuntimeDefinition contains a Decl corresponding to the definition of the called
221function, and RuntimeDefinition::mayHaveOtherDefinitions will return FALSE.
222
223In the case of dynamic dispatch where our information is not perfect, CallEvent
224can make a guess, but RuntimeDefinition::mayHaveOtherDefinitions will return
225TRUE. The RuntimeDefinition object will then also include a MemRegion
226corresponding to the object being called (i.e., the "receiver" in Objective-C
227parlance), which ExprEngine uses to decide whether or not the call should be
228inlined.
229
230 == Inlining Dynamic Calls ==
231
232The -analyzer-ipa option has five different modes: none, basic-inlining,
233inlining, dynamic, and dynamic-bifurcate. Under -analyzer-ipa=dynamic, all
234dynamic calls are inlined, whether we are certain or not that this will actually
235be the definition used at runtime. Under -analyzer-ipa=inlining, only
236"near-perfect" devirtualized calls are inlined*, and other dynamic calls are
237evaluated conservatively (as if no definition were available).
238
239* Currently, no Objective-C messages are not inlined under
240  -analyzer-ipa=inlining, even if we are reasonably confident of the type of the
241  receiver. We plan to enable this once we have tested our heuristics more
242  thoroughly.
243
244The last option, -analyzer-ipa=dynamic-bifurcate, behaves similarly to
245"dynamic", but performs a conservative invalidation in the general virtual case
246in *addition* to inlining. The details of this are discussed below.
247
248As stated above, -analyzer-ipa=basic-inlining does not inline any C++ member
249functions or Objective-C method calls, even if they are non-virtual or can be
250safely devirtualized.
251
252
253Bifurcation
254-----------
255
256ExprEngine::BifurcateCall implements the -analyzer-ipa=dynamic-bifurcate
257mode.
258
259When a call is made on an object with imprecise dynamic type information
260(RuntimeDefinition::mayHaveOtherDefinitions() evaluates to TRUE), ExprEngine
261bifurcates the path and marks the object's region (retrieved from the
262RuntimeDefinition object) with a path-sensitive "mode" in the ProgramState.
263
264Currently, there are 2 modes:
265
266 DynamicDispatchModeInlined - Models the case where the dynamic type information
267   of the receiver (MemoryRegion) is assumed to be perfectly constrained so
268   that a given definition of a method is expected to be the code actually
269   called. When this mode is set, ExprEngine uses the Decl from
270   RuntimeDefinition to inline any dynamically dispatched call sent to this
271   receiver because the function definition is considered to be fully resolved.
272
273 DynamicDispatchModeConservative - Models the case where the dynamic type
274   information is assumed to be incorrect, for example, implies that the method
275   definition is overriden in a subclass. In such cases, ExprEngine does not
276   inline the methods sent to the receiver (MemoryRegion), even if a candidate
277   definition is available. This mode is conservative about simulating the
278   effects of a call.
279
280Going forward along the symbolic execution path, ExprEngine consults the mode
281of the receiver's MemRegion to make decisions on whether the calls should be
282inlined or not, which ensures that there is at most one split per region.
283
284At a high level, "bifurcation mode" allows for increased semantic coverage in
285cases where the parent method contains code which is only executed when the
286class is subclassed. The disadvantages of this mode are a (considerable?)
287performance hit and the possibility of false positives on the path where the
288conservative mode is used.
289
290Objective-C Message Heuristics
291------------------------------
292
293ExprEngine relies on a set of heuristics to partition the set of Objective-C
294method calls into those that require bifurcation and those that do not. Below
295are the cases when the DynamicTypeInfo of the object is considered precise
296(cannot be a subclass):
297
298 - If the object was created with +alloc or +new and initialized with an -init
299   method.
300
301 - If the calls are property accesses using dot syntax. This is based on the
302   assumption that children rarely override properties, or do so in an
303   essentially compatible way.
304
305 - If the class interface is declared inside the main source file. In this case
306   it is unlikely that it will be subclassed.
307
308 - If the method is not declared outside of main source file, either by the
309   receiver's class or by any superclasses.
310
311C++ Caveats
312--------------------
313
314C++11 [class.cdtor]p4 describes how the vtable of an object is modified as it is
315being constructed or destructed; that is, the type of the object depends on
316which base constructors have been completed. This is tracked using
317DynamicTypeInfo in the DynamicTypePropagation checker.
318
319There are several limitations in the current implementation:
320
321- Temporaries are poorly modeled right now because we're not confident in the
322  placement of their destructors in the CFG. We currently won't inline their
323  constructors unless the destructor is trivial, and don't process their
324  destructors at all, not even to invalidate the region.
325
326- 'new' is poorly modeled due to some nasty CFG/design issues.  This is tracked
327  in PR12014.  'delete' is not modeled at all.
328
329- Arrays of objects are modeled very poorly right now.  ExprEngine currently
330  only simulates the first constructor and first destructor. Because of this,
331  ExprEngine does not inline any constructors or destructors for arrays.
332
333
334CallEvent
335=========
336
337A CallEvent represents a specific call to a function, method, or other body of
338code. It is path-sensitive, containing both the current state (ProgramStateRef)
339and stack space (LocationContext), and provides uniform access to the argument
340values and return type of a call, no matter how the call is written in the
341source or what sort of code body is being invoked.
342
343  NOTE: For those familiar with Cocoa, CallEvent is roughly equivalent to
344        NSInvocation.
345
346CallEvent should be used whenever there is logic dealing with function calls
347that does not care how the call occurred.
348
349Examples include checking that arguments satisfy preconditions (such as
350__attribute__((nonnull))), and attempting to inline a call.
351
352CallEvents are reference-counted objects managed by a CallEventManager. While
353there is no inherent issue with persisting them (say, in a ProgramState's GDM),
354they are intended for short-lived use, and can be recreated from CFGElements or
355non-top-level StackFrameContexts fairly easily.
356