#include "llvm/Transforms/Scalar.h"
#include "llvm/ADT/SetVector.h"
+#include "llvm/Analysis/GlobalsModRef.h"
#include "llvm/Analysis/AssumptionCache.h"
#include "llvm/Analysis/CodeMetrics.h"
#include "llvm/Analysis/InstructionSimplify.h"
///
void getAnalysisUsage(AnalysisUsage &AU) const override {
AU.addRequired<AssumptionCacheTracker>();
+ AU.addRequired<DominatorTreeWrapperPass>();
AU.addRequired<LoopInfoWrapperPass>();
AU.addPreserved<LoopInfoWrapperPass>();
AU.addRequiredID(LoopSimplifyID);
AU.addPreservedID(LoopSimplifyID);
AU.addRequiredID(LCSSAID);
AU.addPreservedID(LCSSAID);
- AU.addRequired<ScalarEvolution>();
- AU.addPreserved<ScalarEvolution>();
+ AU.addRequired<ScalarEvolutionWrapperPass>();
+ AU.addPreserved<ScalarEvolutionWrapperPass>();
AU.addRequired<TargetTransformInfoWrapperPass>();
// FIXME: Loop unroll requires LCSSA. And LCSSA requires dom info.
// If loop unroll does not preserve dom info then LCSSA pass on next
// loop will receive invalid dom info.
// For now, recreate dom info, if loop is unrolled.
AU.addPreserved<DominatorTreeWrapperPass>();
+ AU.addPreserved<GlobalsAAWrapperPass>();
}
// Fill in the UnrollingPreferences parameter with values from the
// total unrolled size. Parameters Threshold and PartialThreshold
// are set to the maximum unrolled size for fully and partially
// unrolled loops respectively.
- void selectThresholds(const Loop *L, bool HasPragma,
+ void selectThresholds(const Loop *L, bool UsePragmaThreshold,
const TargetTransformInfo::UnrollingPreferences &UP,
unsigned &Threshold, unsigned &PartialThreshold,
unsigned &PercentDynamicCostSavedThreshold,
: UP.DynamicCostSavingsDiscount;
if (!UserThreshold &&
+ // FIXME: Use Function::optForSize().
L->getHeader()->getParent()->hasFnAttribute(
Attribute::OptimizeForSize)) {
Threshold = UP.OptSizeThreshold;
PartialThreshold = UP.PartialOptSizeThreshold;
}
- if (HasPragma) {
+ if (UsePragmaThreshold) {
// If the loop has an unrolling pragma, we want to be more
// aggressive with unrolling limits. Set thresholds to at
// least the PragmaTheshold value which is larger than the
INITIALIZE_PASS_BEGIN(LoopUnroll, "loop-unroll", "Unroll loops", false, false)
INITIALIZE_PASS_DEPENDENCY(TargetTransformInfoWrapperPass)
INITIALIZE_PASS_DEPENDENCY(AssumptionCacheTracker)
+INITIALIZE_PASS_DEPENDENCY(DominatorTreeWrapperPass)
INITIALIZE_PASS_DEPENDENCY(LoopInfoWrapperPass)
INITIALIZE_PASS_DEPENDENCY(LoopSimplify)
INITIALIZE_PASS_DEPENDENCY(LCSSA)
-INITIALIZE_PASS_DEPENDENCY(ScalarEvolution)
+INITIALIZE_PASS_DEPENDENCY(ScalarEvolutionWrapperPass)
INITIALIZE_PASS_END(LoopUnroll, "loop-unroll", "Unroll loops", false, false)
Pass *llvm::createLoopUnrollPass(int Threshold, int Count, int AllowPartial,
/// Try to simplify binary operator I.
///
- /// TODO: Probaly it's worth to hoist the code for estimating the
+ /// TODO: Probably it's worth to hoist the code for estimating the
/// simplifications effects to a separate class, since we have a very similar
/// code in InlineCost already.
bool visitBinaryOperator(BinaryOperator &I) {
auto *GV = dyn_cast<GlobalVariable>(AddressIt->second.Base);
// We're only interested in loads that can be completely folded to a
// constant.
- if (!GV || !GV->hasInitializer())
+ if (!GV || !GV->hasDefinitiveInitializer() || !GV->isConstant())
return false;
ConstantDataSequential *CDS =
namespace {
struct EstimatedUnrollCost {
/// \brief The estimated cost after unrolling.
- unsigned UnrolledCost;
+ int UnrolledCost;
/// \brief The estimated dynamic cost of executing the instructions in the
/// rolled form.
- unsigned RolledDynamicCost;
+ int RolledDynamicCost;
};
}
/// \returns Optional value, holding the RolledDynamicCost and UnrolledCost. If
/// the analysis failed (no benefits expected from the unrolling, or the loop is
/// too big to analyze), the returned value is None.
-Optional<EstimatedUnrollCost>
-analyzeLoopUnrollCost(const Loop *L, unsigned TripCount, ScalarEvolution &SE,
- const TargetTransformInfo &TTI,
- unsigned MaxUnrolledLoopSize) {
+static Optional<EstimatedUnrollCost>
+analyzeLoopUnrollCost(const Loop *L, unsigned TripCount, DominatorTree &DT,
+ ScalarEvolution &SE, const TargetTransformInfo &TTI,
+ int MaxUnrolledLoopSize) {
// We want to be able to scale offsets by the trip count and add more offsets
// to them without checking for overflows, and we already don't want to
// analyze *massive* trip counts, so we force the max to be reasonably small.
SmallSetVector<BasicBlock *, 16> BBWorklist;
DenseMap<Value *, Constant *> SimplifiedValues;
+ SmallVector<std::pair<Value *, Constant *>, 4> SimplifiedInputValues;
// The estimated cost of the unrolled form of the loop. We try to estimate
// this by simplifying as much as we can while computing the estimate.
- unsigned UnrolledCost = 0;
+ int UnrolledCost = 0;
// We also track the estimated dynamic (that is, actually executed) cost in
// the rolled form. This helps identify cases when the savings from unrolling
// aren't just exposing dead control flows, but actual reduced dynamic
// instructions due to the simplifications which we expect to occur after
// unrolling.
- unsigned RolledDynamicCost = 0;
+ int RolledDynamicCost = 0;
+
+ // Ensure that we don't violate the loop structure invariants relied on by
+ // this analysis.
+ assert(L->isLoopSimplifyForm() && "Must put loop into normal form first.");
+ assert(L->isLCSSAForm(DT) &&
+ "Must have loops in LCSSA form to track live-out values.");
+
+ DEBUG(dbgs() << "Starting LoopUnroll profitability analysis...\n");
// Simulate execution of each iteration of the loop counting instructions,
// which would be simplified.
// Since the same load will take different values on different iterations,
// we literally have to go through all loop's iterations.
for (unsigned Iteration = 0; Iteration < TripCount; ++Iteration) {
+ DEBUG(dbgs() << " Analyzing iteration " << Iteration << "\n");
+
+ // Prepare for the iteration by collecting any simplified entry or backedge
+ // inputs.
+ for (Instruction &I : *L->getHeader()) {
+ auto *PHI = dyn_cast<PHINode>(&I);
+ if (!PHI)
+ break;
+
+ // The loop header PHI nodes must have exactly two input: one from the
+ // loop preheader and one from the loop latch.
+ assert(
+ PHI->getNumIncomingValues() == 2 &&
+ "Must have an incoming value only for the preheader and the latch.");
+
+ Value *V = PHI->getIncomingValueForBlock(
+ Iteration == 0 ? L->getLoopPreheader() : L->getLoopLatch());
+ Constant *C = dyn_cast<Constant>(V);
+ if (Iteration != 0 && !C)
+ C = SimplifiedValues.lookup(V);
+ if (C)
+ SimplifiedInputValues.push_back({PHI, C});
+ }
+
+ // Now clear and re-populate the map for the next iteration.
SimplifiedValues.clear();
+ while (!SimplifiedInputValues.empty())
+ SimplifiedValues.insert(SimplifiedInputValues.pop_back_val());
+
UnrolledInstAnalyzer Analyzer(Iteration, SimplifiedValues, L, SE);
BBWorklist.clear();
// it. We don't change the actual IR, just count optimization
// opportunities.
for (Instruction &I : *BB) {
- unsigned InstCost = TTI.getUserCost(&I);
+ int InstCost = TTI.getUserCost(&I);
// Visit the instruction to analyze its loop cost after unrolling,
// and if the visitor returns false, include this instruction in the
// unrolled cost.
if (!Analyzer.visit(I))
UnrolledCost += InstCost;
+ else {
+ DEBUG(dbgs() << " " << I
+ << " would be simplified if loop is unrolled.\n");
+ (void)0;
+ }
// Also track this instructions expected cost when executing the rolled
// loop form.
RolledDynamicCost += InstCost;
// If unrolled body turns out to be too big, bail out.
- if (UnrolledCost > MaxUnrolledLoopSize)
+ if (UnrolledCost > MaxUnrolledLoopSize) {
+ DEBUG(dbgs() << " Exceeded threshold.. exiting.\n"
+ << " UnrolledCost: " << UnrolledCost
+ << ", MaxUnrolledLoopSize: " << MaxUnrolledLoopSize
+ << "\n");
return None;
+ }
}
TerminatorInst *TI = BB->getTerminator();
if (BI->isConditional()) {
if (Constant *SimpleCond =
SimplifiedValues.lookup(BI->getCondition())) {
- BasicBlock *Succ = BI->getSuccessor(
- cast<ConstantInt>(SimpleCond)->isZero() ? 1 : 0);
+ BasicBlock *Succ = nullptr;
+ // Just take the first successor if condition is undef
+ if (isa<UndefValue>(SimpleCond))
+ Succ = BI->getSuccessor(0);
+ else
+ Succ = BI->getSuccessor(
+ cast<ConstantInt>(SimpleCond)->isZero() ? 1 : 0);
if (L->contains(Succ))
BBWorklist.insert(Succ);
continue;
} else if (SwitchInst *SI = dyn_cast<SwitchInst>(TI)) {
if (Constant *SimpleCond =
SimplifiedValues.lookup(SI->getCondition())) {
- BasicBlock *Succ =
- SI->getSuccessor(cast<ConstantInt>(SimpleCond)->getSExtValue());
+ BasicBlock *Succ = nullptr;
+ // Just take the first successor if condition is undef
+ if (isa<UndefValue>(SimpleCond))
+ Succ = SI->getSuccessor(0);
+ else
+ Succ = SI->findCaseValue(cast<ConstantInt>(SimpleCond))
+ .getCaseSuccessor();
if (L->contains(Succ))
BBWorklist.insert(Succ);
continue;
// If we found no optimization opportunities on the first iteration, we
// won't find them on later ones too.
- if (UnrolledCost == RolledDynamicCost)
+ if (UnrolledCost == RolledDynamicCost) {
+ DEBUG(dbgs() << " No opportunities found.. exiting.\n"
+ << " UnrolledCost: " << UnrolledCost << "\n");
return None;
+ }
}
+ DEBUG(dbgs() << "Analysis finished:\n"
+ << "UnrolledCost: " << UnrolledCost << ", "
+ << "RolledDynamicCost: " << RolledDynamicCost << "\n");
return {{UnrolledCost, RolledDynamicCost}};
}
return GetUnrollMetadataForLoop(L, "llvm.loop.unroll.full");
}
+// Returns true if the loop has an unroll(enable) pragma. This metadata is used
+// for both "#pragma unroll" and "#pragma clang loop unroll(enable)" directives.
+static bool HasUnrollEnablePragma(const Loop *L) {
+ return GetUnrollMetadataForLoop(L, "llvm.loop.unroll.enable");
+}
+
// Returns true if the loop has an unroll(disable) pragma.
static bool HasUnrollDisablePragma(const Loop *L) {
return GetUnrollMetadataForLoop(L, "llvm.loop.unroll.disable");
unsigned Count = UserCount ? CurrentCount : 0;
// If there is no user-specified count, unroll pragmas have the next
- // highest precendence.
+ // highest precedence.
if (Count == 0) {
if (PragmaCount) {
Count = PragmaCount;
Function &F = *L->getHeader()->getParent();
+ auto &DT = getAnalysis<DominatorTreeWrapperPass>().getDomTree();
LoopInfo *LI = &getAnalysis<LoopInfoWrapperPass>().getLoopInfo();
- ScalarEvolution *SE = &getAnalysis<ScalarEvolution>();
+ ScalarEvolution *SE = &getAnalysis<ScalarEvolutionWrapperPass>().getSE();
const TargetTransformInfo &TTI =
getAnalysis<TargetTransformInfoWrapperPass>().getTTI(F);
auto &AC = getAnalysis<AssumptionCacheTracker>().getAssumptionCache(F);
return false;
}
bool PragmaFullUnroll = HasUnrollFullPragma(L);
+ bool PragmaEnableUnroll = HasUnrollEnablePragma(L);
unsigned PragmaCount = UnrollCountPragmaValue(L);
- bool HasPragma = PragmaFullUnroll || PragmaCount > 0;
+ bool HasPragma = PragmaFullUnroll || PragmaEnableUnroll || PragmaCount > 0;
TargetTransformInfo::UnrollingPreferences UP;
getUnrollingPreferences(L, TTI, UP);
unsigned Threshold, PartialThreshold;
unsigned PercentDynamicCostSavedThreshold;
unsigned DynamicCostSavingsDiscount;
- selectThresholds(L, HasPragma, UP, Threshold, PartialThreshold,
+ // Only use the high pragma threshold when we have a target unroll factor such
+ // as with "#pragma unroll N" or a pragma indicating full unrolling and the
+ // trip count is known. Otherwise we rely on the standard threshold to
+ // heuristically select a reasonable unroll count.
+ bool UsePragmaThreshold =
+ PragmaCount > 0 ||
+ ((PragmaFullUnroll || PragmaEnableUnroll) && TripCount != 0);
+
+ selectThresholds(L, UsePragmaThreshold, UP, Threshold, PartialThreshold,
PercentDynamicCostSavedThreshold,
DynamicCostSavingsDiscount);
// The loop isn't that small, but we still can fully unroll it if that
// helps to remove a significant number of instructions.
// To check that, run additional analysis on the loop.
- if (Optional<EstimatedUnrollCost> Cost = analyzeLoopUnrollCost(
- L, TripCount, *SE, TTI, Threshold + DynamicCostSavingsDiscount))
+ if (Optional<EstimatedUnrollCost> Cost =
+ analyzeLoopUnrollCost(L, TripCount, DT, *SE, TTI,
+ Threshold + DynamicCostSavingsDiscount))
if (canUnrollCompletely(L, Threshold, PercentDynamicCostSavedThreshold,
DynamicCostSavingsDiscount, Cost->UnrolledCost,
Cost->RolledDynamicCost)) {
// Reduce count based on the type of unrolling and the threshold values.
unsigned OriginalCount = Count;
- bool AllowRuntime =
- (PragmaCount > 0) || (UserRuntime ? CurrentRuntime : UP.Runtime);
+ bool AllowRuntime = PragmaEnableUnroll || (PragmaCount > 0) ||
+ (UserRuntime ? CurrentRuntime : UP.Runtime);
// Don't unroll a runtime trip count loop with unroll full pragma.
if (HasRuntimeUnrollDisablePragma(L) || PragmaFullUnroll) {
AllowRuntime = false;
}
if (Unrolling == Partial) {
- bool AllowPartial = UserAllowPartial ? CurrentAllowPartial : UP.Partial;
+ bool AllowPartial = PragmaEnableUnroll ||
+ (UserAllowPartial ? CurrentAllowPartial : UP.Partial);
if (!AllowPartial && !CountSetExplicitly) {
DEBUG(dbgs() << " will not try to unroll partially because "
<< "-unroll-allow-partial not given\n");
DebugLoc LoopLoc = L->getStartLoc();
Function *F = Header->getParent();
LLVMContext &Ctx = F->getContext();
- if (PragmaFullUnroll && PragmaCount == 0) {
- if (TripCount && Count != TripCount) {
- emitOptimizationRemarkMissed(
- Ctx, DEBUG_TYPE, *F, LoopLoc,
- "Unable to fully unroll loop as directed by unroll(full) pragma "
- "because unrolled size is too large.");
- } else if (!TripCount) {
- emitOptimizationRemarkMissed(
- Ctx, DEBUG_TYPE, *F, LoopLoc,
- "Unable to fully unroll loop as directed by unroll(full) pragma "
- "because loop has a runtime trip count.");
- }
- } else if (PragmaCount > 0 && Count != OriginalCount) {
+ if ((PragmaCount > 0) && Count != OriginalCount) {
emitOptimizationRemarkMissed(
Ctx, DEBUG_TYPE, *F, LoopLoc,
"Unable to unroll loop the number of times directed by "
"unroll_count pragma because unrolled size is too large.");
+ } else if (PragmaFullUnroll && !TripCount) {
+ emitOptimizationRemarkMissed(
+ Ctx, DEBUG_TYPE, *F, LoopLoc,
+ "Unable to fully unroll loop as directed by unroll(full) pragma "
+ "because loop has a runtime trip count.");
+ } else if (PragmaEnableUnroll && Count != TripCount && Count < 2) {
+ emitOptimizationRemarkMissed(
+ Ctx, DEBUG_TYPE, *F, LoopLoc,
+ "Unable to unroll loop as directed by unroll(enable) pragma because "
+ "unrolled size is too large.");
+ } else if ((PragmaFullUnroll || PragmaEnableUnroll) && TripCount &&
+ Count != TripCount) {
+ emitOptimizationRemarkMissed(
+ Ctx, DEBUG_TYPE, *F, LoopLoc,
+ "Unable to fully unroll loop as directed by unroll pragma because "
+ "unrolled size is too large.");
}
}