xref: /freebsd/contrib/llvm-project/llvm/lib/Transforms/Scalar/PlaceSafepoints.cpp (revision 2e3507c25e42292b45a5482e116d278f5515d04d)
1 //===- PlaceSafepoints.cpp - Place GC Safepoints --------------------------===//
2 //
3 // Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
4 // See https://llvm.org/LICENSE.txt for license information.
5 // SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
6 //
7 //===----------------------------------------------------------------------===//
8 //
9 // Place garbage collection safepoints at appropriate locations in the IR. This
10 // does not make relocation semantics or variable liveness explicit.  That's
11 // done by RewriteStatepointsForGC.
12 //
13 // Terminology:
14 // - A call is said to be "parseable" if there is a stack map generated for the
15 // return PC of the call.  A runtime can determine where values listed in the
16 // deopt arguments and (after RewriteStatepointsForGC) gc arguments are located
17 // on the stack when the code is suspended inside such a call.  Every parse
18 // point is represented by a call wrapped in an gc.statepoint intrinsic.
19 // - A "poll" is an explicit check in the generated code to determine if the
20 // runtime needs the generated code to cooperate by calling a helper routine
21 // and thus suspending its execution at a known state. The call to the helper
22 // routine will be parseable.  The (gc & runtime specific) logic of a poll is
23 // assumed to be provided in a function of the name "gc.safepoint_poll".
24 //
25 // We aim to insert polls such that running code can quickly be brought to a
26 // well defined state for inspection by the collector.  In the current
27 // implementation, this is done via the insertion of poll sites at method entry
28 // and the backedge of most loops.  We try to avoid inserting more polls than
29 // are necessary to ensure a finite period between poll sites.  This is not
30 // because the poll itself is expensive in the generated code; it's not.  Polls
31 // do tend to impact the optimizer itself in negative ways; we'd like to avoid
32 // perturbing the optimization of the method as much as we can.
33 //
34 // We also need to make most call sites parseable.  The callee might execute a
35 // poll (or otherwise be inspected by the GC).  If so, the entire stack
36 // (including the suspended frame of the current method) must be parseable.
37 //
38 // This pass will insert:
39 // - Call parse points ("call safepoints") for any call which may need to
40 // reach a safepoint during the execution of the callee function.
41 // - Backedge safepoint polls and entry safepoint polls to ensure that
42 // executing code reaches a safepoint poll in a finite amount of time.
43 //
44 // We do not currently support return statepoints, but adding them would not
45 // be hard.  They are not required for correctness - entry safepoints are an
46 // alternative - but some GCs may prefer them.  Patches welcome.
47 //
48 //===----------------------------------------------------------------------===//
49 
50 #include "llvm/Transforms/Scalar/PlaceSafepoints.h"
51 #include "llvm/InitializePasses.h"
52 #include "llvm/Pass.h"
53 
54 #include "llvm/ADT/SetVector.h"
55 #include "llvm/ADT/Statistic.h"
56 #include "llvm/Analysis/CFG.h"
57 #include "llvm/Analysis/LoopInfo.h"
58 #include "llvm/Analysis/ScalarEvolution.h"
59 #include "llvm/Analysis/TargetLibraryInfo.h"
60 #include "llvm/IR/Dominators.h"
61 #include "llvm/IR/IntrinsicInst.h"
62 #include "llvm/IR/LegacyPassManager.h"
63 #include "llvm/IR/Statepoint.h"
64 #include "llvm/Support/CommandLine.h"
65 #include "llvm/Support/Debug.h"
66 #include "llvm/Transforms/Scalar.h"
67 #include "llvm/Transforms/Utils/BasicBlockUtils.h"
68 #include "llvm/Transforms/Utils/Cloning.h"
69 #include "llvm/Transforms/Utils/Local.h"
70 
71 using namespace llvm;
72 
73 #define DEBUG_TYPE "place-safepoints"
74 
75 STATISTIC(NumEntrySafepoints, "Number of entry safepoints inserted");
76 STATISTIC(NumBackedgeSafepoints, "Number of backedge safepoints inserted");
77 
78 STATISTIC(CallInLoop,
79           "Number of loops without safepoints due to calls in loop");
80 STATISTIC(FiniteExecution,
81           "Number of loops without safepoints finite execution");
82 
83 // Ignore opportunities to avoid placing safepoints on backedges, useful for
84 // validation
85 static cl::opt<bool> AllBackedges("spp-all-backedges", cl::Hidden,
86                                   cl::init(false));
87 
88 /// How narrow does the trip count of a loop have to be to have to be considered
89 /// "counted"?  Counted loops do not get safepoints at backedges.
90 static cl::opt<int> CountedLoopTripWidth("spp-counted-loop-trip-width",
91                                          cl::Hidden, cl::init(32));
92 
93 // If true, split the backedge of a loop when placing the safepoint, otherwise
94 // split the latch block itself.  Both are useful to support for
95 // experimentation, but in practice, it looks like splitting the backedge
96 // optimizes better.
97 static cl::opt<bool> SplitBackedge("spp-split-backedge", cl::Hidden,
98                                    cl::init(false));
99 
100 namespace {
101 /// An analysis pass whose purpose is to identify each of the backedges in
102 /// the function which require a safepoint poll to be inserted.
103 class PlaceBackedgeSafepointsLegacyPass : public FunctionPass {
104 public:
105   static char ID;
106 
107   /// The output of the pass - gives a list of each backedge (described by
108   /// pointing at the branch) which need a poll inserted.
109   std::vector<Instruction *> PollLocations;
110 
111   /// True unless we're running spp-no-calls in which case we need to disable
112   /// the call-dependent placement opts.
113   bool CallSafepointsEnabled;
114 
115   PlaceBackedgeSafepointsLegacyPass(bool CallSafepoints = false)
116       : FunctionPass(ID), CallSafepointsEnabled(CallSafepoints) {
117     initializePlaceBackedgeSafepointsLegacyPassPass(
118         *PassRegistry::getPassRegistry());
119   }
120 
121   bool runOnLoop(Loop *);
122 
123   void runOnLoopAndSubLoops(Loop *L) {
124     // Visit all the subloops
125     for (Loop *I : *L)
126       runOnLoopAndSubLoops(I);
127     runOnLoop(L);
128   }
129 
130   bool runOnFunction(Function &F) override {
131     SE = &getAnalysis<ScalarEvolutionWrapperPass>().getSE();
132     DT = &getAnalysis<DominatorTreeWrapperPass>().getDomTree();
133     LI = &getAnalysis<LoopInfoWrapperPass>().getLoopInfo();
134     TLI = &getAnalysis<TargetLibraryInfoWrapperPass>().getTLI(F);
135     for (Loop *I : *LI) {
136       runOnLoopAndSubLoops(I);
137     }
138     return false;
139   }
140 
141   void getAnalysisUsage(AnalysisUsage &AU) const override {
142     AU.addRequired<DominatorTreeWrapperPass>();
143     AU.addRequired<ScalarEvolutionWrapperPass>();
144     AU.addRequired<LoopInfoWrapperPass>();
145     AU.addRequired<TargetLibraryInfoWrapperPass>();
146     // We no longer modify the IR at all in this pass.  Thus all
147     // analysis are preserved.
148     AU.setPreservesAll();
149   }
150 
151 private:
152   ScalarEvolution *SE = nullptr;
153   DominatorTree *DT = nullptr;
154   LoopInfo *LI = nullptr;
155   TargetLibraryInfo *TLI = nullptr;
156 };
157 } // namespace
158 
159 static cl::opt<bool> NoEntry("spp-no-entry", cl::Hidden, cl::init(false));
160 static cl::opt<bool> NoCall("spp-no-call", cl::Hidden, cl::init(false));
161 static cl::opt<bool> NoBackedge("spp-no-backedge", cl::Hidden, cl::init(false));
162 
163 char PlaceBackedgeSafepointsLegacyPass::ID = 0;
164 
165 INITIALIZE_PASS_BEGIN(PlaceBackedgeSafepointsLegacyPass,
166                       "place-backedge-safepoints-impl",
167                       "Place Backedge Safepoints", false, false)
168 INITIALIZE_PASS_DEPENDENCY(ScalarEvolutionWrapperPass)
169 INITIALIZE_PASS_DEPENDENCY(DominatorTreeWrapperPass)
170 INITIALIZE_PASS_DEPENDENCY(LoopInfoWrapperPass)
171 INITIALIZE_PASS_END(PlaceBackedgeSafepointsLegacyPass,
172                     "place-backedge-safepoints-impl",
173                     "Place Backedge Safepoints", false, false)
174 
175 static bool containsUnconditionalCallSafepoint(Loop *L, BasicBlock *Header,
176                                                BasicBlock *Pred,
177                                                DominatorTree &DT,
178                                                const TargetLibraryInfo &TLI);
179 
180 static bool mustBeFiniteCountedLoop(Loop *L, ScalarEvolution *SE,
181                                     BasicBlock *Pred);
182 
183 static Instruction *findLocationForEntrySafepoint(Function &F,
184                                                   DominatorTree &DT);
185 
186 static bool isGCSafepointPoll(Function &F);
187 static bool shouldRewriteFunction(Function &F);
188 static bool enableEntrySafepoints(Function &F);
189 static bool enableBackedgeSafepoints(Function &F);
190 static bool enableCallSafepoints(Function &F);
191 
192 static void
193 InsertSafepointPoll(Instruction *InsertBefore,
194                     std::vector<CallBase *> &ParsePointsNeeded /*rval*/,
195                     const TargetLibraryInfo &TLI);
196 
197 bool PlaceBackedgeSafepointsLegacyPass::runOnLoop(Loop *L) {
198   // Loop through all loop latches (branches controlling backedges).  We need
199   // to place a safepoint on every backedge (potentially).
200   // Note: In common usage, there will be only one edge due to LoopSimplify
201   // having run sometime earlier in the pipeline, but this code must be correct
202   // w.r.t. loops with multiple backedges.
203   BasicBlock *Header = L->getHeader();
204   SmallVector<BasicBlock *, 16> LoopLatches;
205   L->getLoopLatches(LoopLatches);
206   for (BasicBlock *Pred : LoopLatches) {
207     assert(L->contains(Pred));
208 
209     // Make a policy decision about whether this loop needs a safepoint or
210     // not.  Note that this is about unburdening the optimizer in loops, not
211     // avoiding the runtime cost of the actual safepoint.
212     if (!AllBackedges) {
213       if (mustBeFiniteCountedLoop(L, SE, Pred)) {
214         LLVM_DEBUG(dbgs() << "skipping safepoint placement in finite loop\n");
215         FiniteExecution++;
216         continue;
217       }
218       if (CallSafepointsEnabled &&
219           containsUnconditionalCallSafepoint(L, Header, Pred, *DT, *TLI)) {
220         // Note: This is only semantically legal since we won't do any further
221         // IPO or inlining before the actual call insertion..  If we hadn't, we
222         // might latter loose this call safepoint.
223         LLVM_DEBUG(
224             dbgs()
225             << "skipping safepoint placement due to unconditional call\n");
226         CallInLoop++;
227         continue;
228       }
229     }
230 
231     // TODO: We can create an inner loop which runs a finite number of
232     // iterations with an outer loop which contains a safepoint.  This would
233     // not help runtime performance that much, but it might help our ability to
234     // optimize the inner loop.
235 
236     // Safepoint insertion would involve creating a new basic block (as the
237     // target of the current backedge) which does the safepoint (of all live
238     // variables) and branches to the true header
239     Instruction *Term = Pred->getTerminator();
240 
241     LLVM_DEBUG(dbgs() << "[LSP] terminator instruction: " << *Term);
242 
243     PollLocations.push_back(Term);
244   }
245 
246   return false;
247 }
248 
249 bool PlaceSafepointsPass::runImpl(Function &F, const TargetLibraryInfo &TLI) {
250   if (F.isDeclaration() || F.empty()) {
251     // This is a declaration, nothing to do.  Must exit early to avoid crash in
252     // dom tree calculation
253     return false;
254   }
255 
256   if (isGCSafepointPoll(F)) {
257     // Given we're inlining this inside of safepoint poll insertion, this
258     // doesn't make any sense.  Note that we do make any contained calls
259     // parseable after we inline a poll.
260     return false;
261   }
262 
263   if (!shouldRewriteFunction(F))
264     return false;
265 
266   bool Modified = false;
267 
268   // In various bits below, we rely on the fact that uses are reachable from
269   // defs.  When there are basic blocks unreachable from the entry, dominance
270   // and reachablity queries return non-sensical results.  Thus, we preprocess
271   // the function to ensure these properties hold.
272   Modified |= removeUnreachableBlocks(F);
273 
274   // STEP 1 - Insert the safepoint polling locations.  We do not need to
275   // actually insert parse points yet.  That will be done for all polls and
276   // calls in a single pass.
277 
278   DominatorTree DT;
279   DT.recalculate(F);
280 
281   SmallVector<Instruction *, 16> PollsNeeded;
282   std::vector<CallBase *> ParsePointNeeded;
283 
284   if (enableBackedgeSafepoints(F)) {
285     // Construct a pass manager to run the LoopPass backedge logic.  We
286     // need the pass manager to handle scheduling all the loop passes
287     // appropriately.  Doing this by hand is painful and just not worth messing
288     // with for the moment.
289     legacy::FunctionPassManager FPM(F.getParent());
290     bool CanAssumeCallSafepoints = enableCallSafepoints(F);
291     auto *PBS = new PlaceBackedgeSafepointsLegacyPass(CanAssumeCallSafepoints);
292     FPM.add(PBS);
293     FPM.run(F);
294 
295     // We preserve dominance information when inserting the poll, otherwise
296     // we'd have to recalculate this on every insert
297     DT.recalculate(F);
298 
299     auto &PollLocations = PBS->PollLocations;
300 
301     auto OrderByBBName = [](Instruction *a, Instruction *b) {
302       return a->getParent()->getName() < b->getParent()->getName();
303     };
304     // We need the order of list to be stable so that naming ends up stable
305     // when we split edges.  This makes test cases much easier to write.
306     llvm::sort(PollLocations, OrderByBBName);
307 
308     // We can sometimes end up with duplicate poll locations.  This happens if
309     // a single loop is visited more than once.   The fact this happens seems
310     // wrong, but it does happen for the split-backedge.ll test case.
311     PollLocations.erase(std::unique(PollLocations.begin(), PollLocations.end()),
312                         PollLocations.end());
313 
314     // Insert a poll at each point the analysis pass identified
315     // The poll location must be the terminator of a loop latch block.
316     for (Instruction *Term : PollLocations) {
317       // We are inserting a poll, the function is modified
318       Modified = true;
319 
320       if (SplitBackedge) {
321         // Split the backedge of the loop and insert the poll within that new
322         // basic block.  This creates a loop with two latches per original
323         // latch (which is non-ideal), but this appears to be easier to
324         // optimize in practice than inserting the poll immediately before the
325         // latch test.
326 
327         // Since this is a latch, at least one of the successors must dominate
328         // it. Its possible that we have a) duplicate edges to the same header
329         // and b) edges to distinct loop headers.  We need to insert pools on
330         // each.
331         SetVector<BasicBlock *> Headers;
332         for (unsigned i = 0; i < Term->getNumSuccessors(); i++) {
333           BasicBlock *Succ = Term->getSuccessor(i);
334           if (DT.dominates(Succ, Term->getParent())) {
335             Headers.insert(Succ);
336           }
337         }
338         assert(!Headers.empty() && "poll location is not a loop latch?");
339 
340         // The split loop structure here is so that we only need to recalculate
341         // the dominator tree once.  Alternatively, we could just keep it up to
342         // date and use a more natural merged loop.
343         SetVector<BasicBlock *> SplitBackedges;
344         for (BasicBlock *Header : Headers) {
345           BasicBlock *NewBB = SplitEdge(Term->getParent(), Header, &DT);
346           PollsNeeded.push_back(NewBB->getTerminator());
347           NumBackedgeSafepoints++;
348         }
349       } else {
350         // Split the latch block itself, right before the terminator.
351         PollsNeeded.push_back(Term);
352         NumBackedgeSafepoints++;
353       }
354     }
355   }
356 
357   if (enableEntrySafepoints(F)) {
358     if (Instruction *Location = findLocationForEntrySafepoint(F, DT)) {
359       PollsNeeded.push_back(Location);
360       Modified = true;
361       NumEntrySafepoints++;
362     }
363     // TODO: else we should assert that there was, in fact, a policy choice to
364     // not insert a entry safepoint poll.
365   }
366 
367   // Now that we've identified all the needed safepoint poll locations, insert
368   // safepoint polls themselves.
369   for (Instruction *PollLocation : PollsNeeded) {
370     std::vector<CallBase *> RuntimeCalls;
371     InsertSafepointPoll(PollLocation, RuntimeCalls, TLI);
372     llvm::append_range(ParsePointNeeded, RuntimeCalls);
373   }
374 
375   return Modified;
376 }
377 
378 PreservedAnalyses PlaceSafepointsPass::run(Function &F,
379                                            FunctionAnalysisManager &AM) {
380   auto &TLI = AM.getResult<TargetLibraryAnalysis>(F);
381 
382   if (!runImpl(F, TLI))
383     return PreservedAnalyses::all();
384 
385   // TODO: can we preserve more?
386   return PreservedAnalyses::none();
387 }
388 
389 static bool needsStatepoint(CallBase *Call, const TargetLibraryInfo &TLI) {
390   if (callsGCLeafFunction(Call, TLI))
391     return false;
392   if (auto *CI = dyn_cast<CallInst>(Call)) {
393     if (CI->isInlineAsm())
394       return false;
395   }
396 
397   return !(isa<GCStatepointInst>(Call) || isa<GCRelocateInst>(Call) ||
398            isa<GCResultInst>(Call));
399 }
400 
401 /// Returns true if this loop is known to contain a call safepoint which
402 /// must unconditionally execute on any iteration of the loop which returns
403 /// to the loop header via an edge from Pred.  Returns a conservative correct
404 /// answer; i.e. false is always valid.
405 static bool containsUnconditionalCallSafepoint(Loop *L, BasicBlock *Header,
406                                                BasicBlock *Pred,
407                                                DominatorTree &DT,
408                                                const TargetLibraryInfo &TLI) {
409   // In general, we're looking for any cut of the graph which ensures
410   // there's a call safepoint along every edge between Header and Pred.
411   // For the moment, we look only for the 'cuts' that consist of a single call
412   // instruction in a block which is dominated by the Header and dominates the
413   // loop latch (Pred) block.  Somewhat surprisingly, walking the entire chain
414   // of such dominating blocks gets substantially more occurrences than just
415   // checking the Pred and Header blocks themselves.  This may be due to the
416   // density of loop exit conditions caused by range and null checks.
417   // TODO: structure this as an analysis pass, cache the result for subloops,
418   // avoid dom tree recalculations
419   assert(DT.dominates(Header, Pred) && "loop latch not dominated by header?");
420 
421   BasicBlock *Current = Pred;
422   while (true) {
423     for (Instruction &I : *Current) {
424       if (auto *Call = dyn_cast<CallBase>(&I))
425         // Note: Technically, needing a safepoint isn't quite the right
426         // condition here.  We should instead be checking if the target method
427         // has an
428         // unconditional poll. In practice, this is only a theoretical concern
429         // since we don't have any methods with conditional-only safepoint
430         // polls.
431         if (needsStatepoint(Call, TLI))
432           return true;
433     }
434 
435     if (Current == Header)
436       break;
437     Current = DT.getNode(Current)->getIDom()->getBlock();
438   }
439 
440   return false;
441 }
442 
443 /// Returns true if this loop is known to terminate in a finite number of
444 /// iterations.  Note that this function may return false for a loop which
445 /// does actual terminate in a finite constant number of iterations due to
446 /// conservatism in the analysis.
447 static bool mustBeFiniteCountedLoop(Loop *L, ScalarEvolution *SE,
448                                     BasicBlock *Pred) {
449   // A conservative bound on the loop as a whole.
450   const SCEV *MaxTrips = SE->getConstantMaxBackedgeTakenCount(L);
451   if (!isa<SCEVCouldNotCompute>(MaxTrips) &&
452       SE->getUnsignedRange(MaxTrips).getUnsignedMax().isIntN(
453           CountedLoopTripWidth))
454     return true;
455 
456   // If this is a conditional branch to the header with the alternate path
457   // being outside the loop, we can ask questions about the execution frequency
458   // of the exit block.
459   if (L->isLoopExiting(Pred)) {
460     // This returns an exact expression only.  TODO: We really only need an
461     // upper bound here, but SE doesn't expose that.
462     const SCEV *MaxExec = SE->getExitCount(L, Pred);
463     if (!isa<SCEVCouldNotCompute>(MaxExec) &&
464         SE->getUnsignedRange(MaxExec).getUnsignedMax().isIntN(
465             CountedLoopTripWidth))
466         return true;
467   }
468 
469   return /* not finite */ false;
470 }
471 
472 static void scanOneBB(Instruction *Start, Instruction *End,
473                       std::vector<CallInst *> &Calls,
474                       DenseSet<BasicBlock *> &Seen,
475                       std::vector<BasicBlock *> &Worklist) {
476   for (BasicBlock::iterator BBI(Start), BBE0 = Start->getParent()->end(),
477                                         BBE1 = BasicBlock::iterator(End);
478        BBI != BBE0 && BBI != BBE1; BBI++) {
479     if (CallInst *CI = dyn_cast<CallInst>(&*BBI))
480       Calls.push_back(CI);
481 
482     // FIXME: This code does not handle invokes
483     assert(!isa<InvokeInst>(&*BBI) &&
484            "support for invokes in poll code needed");
485 
486     // Only add the successor blocks if we reach the terminator instruction
487     // without encountering end first
488     if (BBI->isTerminator()) {
489       BasicBlock *BB = BBI->getParent();
490       for (BasicBlock *Succ : successors(BB)) {
491         if (Seen.insert(Succ).second) {
492           Worklist.push_back(Succ);
493         }
494       }
495     }
496   }
497 }
498 
499 static void scanInlinedCode(Instruction *Start, Instruction *End,
500                             std::vector<CallInst *> &Calls,
501                             DenseSet<BasicBlock *> &Seen) {
502   Calls.clear();
503   std::vector<BasicBlock *> Worklist;
504   Seen.insert(Start->getParent());
505   scanOneBB(Start, End, Calls, Seen, Worklist);
506   while (!Worklist.empty()) {
507     BasicBlock *BB = Worklist.back();
508     Worklist.pop_back();
509     scanOneBB(&*BB->begin(), End, Calls, Seen, Worklist);
510   }
511 }
512 
513 /// Returns true if an entry safepoint is not required before this callsite in
514 /// the caller function.
515 static bool doesNotRequireEntrySafepointBefore(CallBase *Call) {
516   if (IntrinsicInst *II = dyn_cast<IntrinsicInst>(Call)) {
517     switch (II->getIntrinsicID()) {
518     case Intrinsic::experimental_gc_statepoint:
519     case Intrinsic::experimental_patchpoint_void:
520     case Intrinsic::experimental_patchpoint_i64:
521       // The can wrap an actual call which may grow the stack by an unbounded
522       // amount or run forever.
523       return false;
524     default:
525       // Most LLVM intrinsics are things which do not expand to actual calls, or
526       // at least if they do, are leaf functions that cause only finite stack
527       // growth.  In particular, the optimizer likes to form things like memsets
528       // out of stores in the original IR.  Another important example is
529       // llvm.localescape which must occur in the entry block.  Inserting a
530       // safepoint before it is not legal since it could push the localescape
531       // out of the entry block.
532       return true;
533     }
534   }
535   return false;
536 }
537 
538 static Instruction *findLocationForEntrySafepoint(Function &F,
539                                                   DominatorTree &DT) {
540 
541   // Conceptually, this poll needs to be on method entry, but in
542   // practice, we place it as late in the entry block as possible.  We
543   // can place it as late as we want as long as it dominates all calls
544   // that can grow the stack.  This, combined with backedge polls,
545   // give us all the progress guarantees we need.
546 
547   // hasNextInstruction and nextInstruction are used to iterate
548   // through a "straight line" execution sequence.
549 
550   auto HasNextInstruction = [](Instruction *I) {
551     if (!I->isTerminator())
552       return true;
553 
554     BasicBlock *nextBB = I->getParent()->getUniqueSuccessor();
555     return nextBB && (nextBB->getUniquePredecessor() != nullptr);
556   };
557 
558   auto NextInstruction = [&](Instruction *I) {
559     assert(HasNextInstruction(I) &&
560            "first check if there is a next instruction!");
561 
562     if (I->isTerminator())
563       return &I->getParent()->getUniqueSuccessor()->front();
564     return &*++I->getIterator();
565   };
566 
567   Instruction *Cursor = nullptr;
568   for (Cursor = &F.getEntryBlock().front(); HasNextInstruction(Cursor);
569        Cursor = NextInstruction(Cursor)) {
570 
571     // We need to ensure a safepoint poll occurs before any 'real' call.  The
572     // easiest way to ensure finite execution between safepoints in the face of
573     // recursive and mutually recursive functions is to enforce that each take
574     // a safepoint.  Additionally, we need to ensure a poll before any call
575     // which can grow the stack by an unbounded amount.  This isn't required
576     // for GC semantics per se, but is a common requirement for languages
577     // which detect stack overflow via guard pages and then throw exceptions.
578     if (auto *Call = dyn_cast<CallBase>(Cursor)) {
579       if (doesNotRequireEntrySafepointBefore(Call))
580         continue;
581       break;
582     }
583   }
584 
585   assert((HasNextInstruction(Cursor) || Cursor->isTerminator()) &&
586          "either we stopped because of a call, or because of terminator");
587 
588   return Cursor;
589 }
590 
591 const char GCSafepointPollName[] = "gc.safepoint_poll";
592 
593 static bool isGCSafepointPoll(Function &F) {
594   return F.getName().equals(GCSafepointPollName);
595 }
596 
597 /// Returns true if this function should be rewritten to include safepoint
598 /// polls and parseable call sites.  The main point of this function is to be
599 /// an extension point for custom logic.
600 static bool shouldRewriteFunction(Function &F) {
601   // TODO: This should check the GCStrategy
602   if (F.hasGC()) {
603     const auto &FunctionGCName = F.getGC();
604     const StringRef StatepointExampleName("statepoint-example");
605     const StringRef CoreCLRName("coreclr");
606     return (StatepointExampleName == FunctionGCName) ||
607            (CoreCLRName == FunctionGCName);
608   } else
609     return false;
610 }
611 
612 // TODO: These should become properties of the GCStrategy, possibly with
613 // command line overrides.
614 static bool enableEntrySafepoints(Function &F) { return !NoEntry; }
615 static bool enableBackedgeSafepoints(Function &F) { return !NoBackedge; }
616 static bool enableCallSafepoints(Function &F) { return !NoCall; }
617 
618 // Insert a safepoint poll immediately before the given instruction.  Does
619 // not handle the parsability of state at the runtime call, that's the
620 // callers job.
621 static void
622 InsertSafepointPoll(Instruction *InsertBefore,
623                     std::vector<CallBase *> &ParsePointsNeeded /*rval*/,
624                     const TargetLibraryInfo &TLI) {
625   BasicBlock *OrigBB = InsertBefore->getParent();
626   Module *M = InsertBefore->getModule();
627   assert(M && "must be part of a module");
628 
629   // Inline the safepoint poll implementation - this will get all the branch,
630   // control flow, etc..  Most importantly, it will introduce the actual slow
631   // path call - where we need to insert a safepoint (parsepoint).
632 
633   auto *F = M->getFunction(GCSafepointPollName);
634   assert(F && "gc.safepoint_poll function is missing");
635   assert(F->getValueType() ==
636          FunctionType::get(Type::getVoidTy(M->getContext()), false) &&
637          "gc.safepoint_poll declared with wrong type");
638   assert(!F->empty() && "gc.safepoint_poll must be a non-empty function");
639   CallInst *PollCall = CallInst::Create(F, "", InsertBefore);
640 
641   // Record some information about the call site we're replacing
642   BasicBlock::iterator Before(PollCall), After(PollCall);
643   bool IsBegin = false;
644   if (Before == OrigBB->begin())
645     IsBegin = true;
646   else
647     Before--;
648 
649   After++;
650   assert(After != OrigBB->end() && "must have successor");
651 
652   // Do the actual inlining
653   InlineFunctionInfo IFI;
654   bool InlineStatus = InlineFunction(*PollCall, IFI).isSuccess();
655   assert(InlineStatus && "inline must succeed");
656   (void)InlineStatus; // suppress warning in release-asserts
657 
658   // Check post-conditions
659   assert(IFI.StaticAllocas.empty() && "can't have allocs");
660 
661   std::vector<CallInst *> Calls; // new calls
662   DenseSet<BasicBlock *> BBs;    // new BBs + insertee
663 
664   // Include only the newly inserted instructions, Note: begin may not be valid
665   // if we inserted to the beginning of the basic block
666   BasicBlock::iterator Start = IsBegin ? OrigBB->begin() : std::next(Before);
667 
668   // If your poll function includes an unreachable at the end, that's not
669   // valid.  Bugpoint likes to create this, so check for it.
670   assert(isPotentiallyReachable(&*Start, &*After) &&
671          "malformed poll function");
672 
673   scanInlinedCode(&*Start, &*After, Calls, BBs);
674   assert(!Calls.empty() && "slow path not found for safepoint poll");
675 
676   // Record the fact we need a parsable state at the runtime call contained in
677   // the poll function.  This is required so that the runtime knows how to
678   // parse the last frame when we actually take  the safepoint (i.e. execute
679   // the slow path)
680   assert(ParsePointsNeeded.empty());
681   for (auto *CI : Calls) {
682     // No safepoint needed or wanted
683     if (!needsStatepoint(CI, TLI))
684       continue;
685 
686     // These are likely runtime calls.  Should we assert that via calling
687     // convention or something?
688     ParsePointsNeeded.push_back(CI);
689   }
690   assert(ParsePointsNeeded.size() <= Calls.size());
691 }
692