aboutsummaryrefslogtreecommitdiff
path: root/llvm/lib/Analysis
diff options
context:
space:
mode:
Diffstat (limited to 'llvm/lib/Analysis')
-rw-r--r--llvm/lib/Analysis/ConstantFolding.cpp25
-rw-r--r--llvm/lib/Analysis/DXILResource.cpp7
-rw-r--r--llvm/lib/Analysis/DependenceAnalysis.cpp17
-rw-r--r--llvm/lib/Analysis/InstructionSimplify.cpp2
-rw-r--r--llvm/lib/Analysis/MemoryDependenceAnalysis.cpp44
-rw-r--r--llvm/lib/Analysis/ProfileSummaryInfo.cpp14
-rw-r--r--llvm/lib/Analysis/TargetLibraryInfo.cpp28
-rw-r--r--llvm/lib/Analysis/TypeBasedAliasAnalysis.cpp4
-rw-r--r--llvm/lib/Analysis/VectorUtils.cpp30
9 files changed, 109 insertions, 62 deletions
diff --git a/llvm/lib/Analysis/ConstantFolding.cpp b/llvm/lib/Analysis/ConstantFolding.cpp
index e71ba5e..7341dad 100644
--- a/llvm/lib/Analysis/ConstantFolding.cpp
+++ b/llvm/lib/Analysis/ConstantFolding.cpp
@@ -929,12 +929,11 @@ Constant *SymbolicallyEvaluateGEP(const GEPOperator *GEP,
if (!AllConstantInt)
break;
- // TODO: Try to intersect two inrange attributes?
- if (!InRange) {
- InRange = GEP->getInRange();
- if (InRange)
- // Adjust inrange by offset until now.
- InRange = InRange->sextOrTrunc(BitWidth).subtract(Offset);
+ // Adjust inrange offset and intersect inrange attributes
+ if (auto GEPRange = GEP->getInRange()) {
+ auto AdjustedGEPRange = GEPRange->sextOrTrunc(BitWidth).subtract(Offset);
+ InRange =
+ InRange ? InRange->intersectWith(AdjustedGEPRange) : AdjustedGEPRange;
}
Ptr = cast<Constant>(GEP->getOperand(0));
@@ -1374,7 +1373,7 @@ Constant *llvm::FlushFPConstant(Constant *Operand, const Instruction *Inst,
if (ConstantFP *CFP = dyn_cast<ConstantFP>(Operand))
return flushDenormalConstantFP(CFP, Inst, IsOutput);
- if (isa<ConstantAggregateZero, UndefValue, ConstantExpr>(Operand))
+ if (isa<ConstantAggregateZero, UndefValue>(Operand))
return Operand;
Type *Ty = Operand->getType();
@@ -1390,6 +1389,9 @@ Constant *llvm::FlushFPConstant(Constant *Operand, const Instruction *Inst,
Ty = VecTy->getElementType();
}
+ if (isa<ConstantExpr>(Operand))
+ return Operand;
+
if (const auto *CV = dyn_cast<ConstantVector>(Operand)) {
SmallVector<Constant *, 16> NewElts;
for (unsigned i = 0, e = CV->getNumOperands(); i != e; ++i) {
@@ -2004,21 +2006,20 @@ inline bool llvm_fenv_testexcept() {
return false;
}
-static const APFloat FTZPreserveSign(const APFloat &V) {
+static APFloat FTZPreserveSign(const APFloat &V) {
if (V.isDenormal())
return APFloat::getZero(V.getSemantics(), V.isNegative());
return V;
}
-static const APFloat FlushToPositiveZero(const APFloat &V) {
+static APFloat FlushToPositiveZero(const APFloat &V) {
if (V.isDenormal())
return APFloat::getZero(V.getSemantics(), false);
return V;
}
-static const APFloat
-FlushWithDenormKind(const APFloat &V,
- DenormalMode::DenormalModeKind DenormKind) {
+static APFloat FlushWithDenormKind(const APFloat &V,
+ DenormalMode::DenormalModeKind DenormKind) {
assert(DenormKind != DenormalMode::DenormalModeKind::Invalid &&
DenormKind != DenormalMode::DenormalModeKind::Dynamic);
switch (DenormKind) {
diff --git a/llvm/lib/Analysis/DXILResource.cpp b/llvm/lib/Analysis/DXILResource.cpp
index 2da6468..1959ab6 100644
--- a/llvm/lib/Analysis/DXILResource.cpp
+++ b/llvm/lib/Analysis/DXILResource.cpp
@@ -1079,15 +1079,16 @@ void DXILResourceBindingInfo::populate(Module &M, DXILResourceTypeMap &DRTM) {
// add new space
S = &BS->Spaces.emplace_back(B.Space);
- // the space is full - set flag to report overlapping binding later
- if (S->FreeRanges.empty()) {
+ // The space is full - there are no free slots left, or the rest of the
+ // slots are taken by an unbounded array. Set flag to report overlapping
+ // binding later.
+ if (S->FreeRanges.empty() || S->FreeRanges.back().UpperBound < UINT32_MAX) {
OverlappingBinding = true;
continue;
}
// adjust the last free range lower bound, split it in two, or remove it
BindingRange &LastFreeRange = S->FreeRanges.back();
- assert(LastFreeRange.UpperBound == UINT32_MAX);
if (LastFreeRange.LowerBound == B.LowerBound) {
if (B.UpperBound < UINT32_MAX)
LastFreeRange.LowerBound = B.UpperBound + 1;
diff --git a/llvm/lib/Analysis/DependenceAnalysis.cpp b/llvm/lib/Analysis/DependenceAnalysis.cpp
index dd9a44b..f1473b2 100644
--- a/llvm/lib/Analysis/DependenceAnalysis.cpp
+++ b/llvm/lib/Analysis/DependenceAnalysis.cpp
@@ -3383,6 +3383,10 @@ bool DependenceInfo::tryDelinearize(Instruction *Src, Instruction *Dst,
SrcSubscripts, DstSubscripts))
return false;
+ assert(isLoopInvariant(SrcBase, SrcLoop) &&
+ isLoopInvariant(DstBase, DstLoop) &&
+ "Expected SrcBase and DstBase to be loop invariant");
+
int Size = SrcSubscripts.size();
LLVM_DEBUG({
dbgs() << "\nSrcSubscripts: ";
@@ -3666,6 +3670,19 @@ DependenceInfo::depends(Instruction *Src, Instruction *Dst,
SCEVUnionPredicate(Assume, *SE));
}
+ // Even if the base pointers are the same, they may not be loop-invariant. It
+ // could lead to incorrect results, as we're analyzing loop-carried
+ // dependencies. Src and Dst can be in different loops, so we need to check
+ // the base pointer is invariant in both loops.
+ Loop *SrcLoop = LI->getLoopFor(Src->getParent());
+ Loop *DstLoop = LI->getLoopFor(Dst->getParent());
+ if (!isLoopInvariant(SrcBase, SrcLoop) ||
+ !isLoopInvariant(DstBase, DstLoop)) {
+ LLVM_DEBUG(dbgs() << "The base pointer is not loop invariant.\n");
+ return std::make_unique<Dependence>(Src, Dst,
+ SCEVUnionPredicate(Assume, *SE));
+ }
+
uint64_t EltSize = SrcLoc.Size.toRaw();
const SCEV *SrcEv = SE->getMinusSCEV(SrcSCEV, SrcBase);
const SCEV *DstEv = SE->getMinusSCEV(DstSCEV, DstBase);
diff --git a/llvm/lib/Analysis/InstructionSimplify.cpp b/llvm/lib/Analysis/InstructionSimplify.cpp
index 82530e7..5907e21 100644
--- a/llvm/lib/Analysis/InstructionSimplify.cpp
+++ b/llvm/lib/Analysis/InstructionSimplify.cpp
@@ -5366,7 +5366,7 @@ static Value *simplifyCastInst(unsigned CastOpc, Value *Op, Type *Ty,
Type *MidTy = CI->getType();
Type *DstTy = Ty;
if (Src->getType() == Ty) {
- auto FirstOp = static_cast<Instruction::CastOps>(CI->getOpcode());
+ auto FirstOp = CI->getOpcode();
auto SecondOp = static_cast<Instruction::CastOps>(CastOpc);
Type *SrcIntPtrTy =
SrcTy->isPtrOrPtrVectorTy() ? Q.DL.getIntPtrType(SrcTy) : nullptr;
diff --git a/llvm/lib/Analysis/MemoryDependenceAnalysis.cpp b/llvm/lib/Analysis/MemoryDependenceAnalysis.cpp
index 3aa9909..2b0f212 100644
--- a/llvm/lib/Analysis/MemoryDependenceAnalysis.cpp
+++ b/llvm/lib/Analysis/MemoryDependenceAnalysis.cpp
@@ -983,33 +983,37 @@ MemDepResult MemoryDependenceResults::getNonLocalInfoForBlock(
static void
SortNonLocalDepInfoCache(MemoryDependenceResults::NonLocalDepInfo &Cache,
unsigned NumSortedEntries) {
- switch (Cache.size() - NumSortedEntries) {
- case 0:
- // done, no new entries.
- break;
- case 2: {
- // Two new entries, insert the last one into place.
- NonLocalDepEntry Val = Cache.back();
- Cache.pop_back();
- MemoryDependenceResults::NonLocalDepInfo::iterator Entry =
- std::upper_bound(Cache.begin(), Cache.end() - 1, Val);
- Cache.insert(Entry, Val);
- [[fallthrough]];
+
+ // If only one entry, don't sort.
+ if (Cache.size() < 2)
+ return;
+
+ unsigned s = Cache.size() - NumSortedEntries;
+
+ // If the cache is already sorted, don't sort it again.
+ if (s == 0)
+ return;
+
+ // If no entry is sorted, sort the whole cache.
+ if (NumSortedEntries == 0) {
+ llvm::sort(Cache);
+ return;
}
- case 1:
- // One new entry, Just insert the new value at the appropriate position.
- if (Cache.size() != 1) {
+
+ // If the number of unsorted entires is small and the cache size is big, using
+ // insertion sort is faster. Here use Log2_32 to quickly choose the sort
+ // method.
+ if (s < Log2_32(Cache.size())) {
+ while (s > 0) {
NonLocalDepEntry Val = Cache.back();
Cache.pop_back();
MemoryDependenceResults::NonLocalDepInfo::iterator Entry =
- llvm::upper_bound(Cache, Val);
+ std::upper_bound(Cache.begin(), Cache.end() - s + 1, Val);
Cache.insert(Entry, Val);
+ s--;
}
- break;
- default:
- // Added many values, do a full scale sort.
+ } else {
llvm::sort(Cache);
- break;
}
}
diff --git a/llvm/lib/Analysis/ProfileSummaryInfo.cpp b/llvm/lib/Analysis/ProfileSummaryInfo.cpp
index e8d4e37..f1c3155 100644
--- a/llvm/lib/Analysis/ProfileSummaryInfo.cpp
+++ b/llvm/lib/Analysis/ProfileSummaryInfo.cpp
@@ -121,8 +121,18 @@ void ProfileSummaryInfo::computeThresholds() {
ProfileSummaryBuilder::getHotCountThreshold(DetailedSummary);
ColdCountThreshold =
ProfileSummaryBuilder::getColdCountThreshold(DetailedSummary);
- assert(ColdCountThreshold <= HotCountThreshold &&
- "Cold count threshold cannot exceed hot count threshold!");
+ // When the hot and cold thresholds are identical, we would classify
+ // a count value as both hot and cold since we are doing an inclusive check
+ // (see ::is{Hot|Cold}Count(). To avoid this undesirable overlap, ensure the
+ // thresholds are distinct.
+ if (HotCountThreshold == ColdCountThreshold) {
+ if (ColdCountThreshold > 0)
+ (*ColdCountThreshold)--;
+ else
+ (*HotCountThreshold)++;
+ }
+ assert(ColdCountThreshold < HotCountThreshold &&
+ "Cold count threshold should be less than hot count threshold!");
if (!hasPartialSampleProfile() || !ScalePartialSampleProfileWorkingSetSize) {
HasHugeWorkingSetSize =
HotEntry.NumCounts > ProfileSummaryHugeWorkingSetSizeThreshold;
diff --git a/llvm/lib/Analysis/TargetLibraryInfo.cpp b/llvm/lib/Analysis/TargetLibraryInfo.cpp
index e475be2..6e92766 100644
--- a/llvm/lib/Analysis/TargetLibraryInfo.cpp
+++ b/llvm/lib/Analysis/TargetLibraryInfo.cpp
@@ -875,6 +875,34 @@ static void initializeLibCalls(TargetLibraryInfoImpl &TLI, const Triple &T,
TLI.setUnavailable(LibFunc_toascii);
}
+ if (T.isOSFreeBSD()) {
+ TLI.setAvailable(LibFunc_dunder_strtok_r);
+ TLI.setAvailable(LibFunc_memalign);
+ TLI.setAvailable(LibFunc_fputc_unlocked);
+ TLI.setAvailable(LibFunc_fputs_unlocked);
+ TLI.setAvailable(LibFunc_fread_unlocked);
+ TLI.setAvailable(LibFunc_fwrite_unlocked);
+ TLI.setAvailable(LibFunc_getc_unlocked);
+ TLI.setAvailable(LibFunc_getchar_unlocked);
+ TLI.setAvailable(LibFunc_putc_unlocked);
+ TLI.setAvailable(LibFunc_putchar_unlocked);
+
+ TLI.setUnavailable(LibFunc___kmpc_alloc_shared);
+ TLI.setUnavailable(LibFunc___kmpc_free_shared);
+ TLI.setUnavailable(LibFunc_dunder_strndup);
+ TLI.setUnavailable(LibFunc_memccpy_chk);
+ TLI.setUnavailable(LibFunc_strlen_chk);
+ TLI.setUnavailable(LibFunc_fmaximum_num);
+ TLI.setUnavailable(LibFunc_fmaximum_numf);
+ TLI.setUnavailable(LibFunc_fmaximum_numl);
+ TLI.setUnavailable(LibFunc_fminimum_num);
+ TLI.setUnavailable(LibFunc_fminimum_numf);
+ TLI.setUnavailable(LibFunc_fminimum_numl);
+ TLI.setUnavailable(LibFunc_roundeven);
+ TLI.setUnavailable(LibFunc_roundevenf);
+ TLI.setUnavailable(LibFunc_roundevenl);
+ }
+
// As currently implemented in clang, NVPTX code has no standard library to
// speak of. Headers provide a standard-ish library implementation, but many
// of the signatures are wrong -- for example, many libm functions are not
diff --git a/llvm/lib/Analysis/TypeBasedAliasAnalysis.cpp b/llvm/lib/Analysis/TypeBasedAliasAnalysis.cpp
index c871070..7025b83 100644
--- a/llvm/lib/Analysis/TypeBasedAliasAnalysis.cpp
+++ b/llvm/lib/Analysis/TypeBasedAliasAnalysis.cpp
@@ -525,6 +525,8 @@ AAMDNodes AAMDNodes::merge(const AAMDNodes &Other) const {
Result.TBAAStruct = nullptr;
Result.Scope = MDNode::getMostGenericAliasScope(Scope, Other.Scope);
Result.NoAlias = MDNode::intersect(NoAlias, Other.NoAlias);
+ Result.NoAliasAddrSpace = MDNode::getMostGenericNoaliasAddrspace(
+ NoAliasAddrSpace, Other.NoAliasAddrSpace);
return Result;
}
@@ -533,6 +535,8 @@ AAMDNodes AAMDNodes::concat(const AAMDNodes &Other) const {
Result.TBAA = Result.TBAAStruct = nullptr;
Result.Scope = MDNode::getMostGenericAliasScope(Scope, Other.Scope);
Result.NoAlias = MDNode::intersect(NoAlias, Other.NoAlias);
+ Result.NoAliasAddrSpace = MDNode::getMostGenericNoaliasAddrspace(
+ NoAliasAddrSpace, Other.NoAliasAddrSpace);
return Result;
}
diff --git a/llvm/lib/Analysis/VectorUtils.cpp b/llvm/lib/Analysis/VectorUtils.cpp
index 1b3da59..e9cf2ee 100644
--- a/llvm/lib/Analysis/VectorUtils.cpp
+++ b/llvm/lib/Analysis/VectorUtils.cpp
@@ -81,6 +81,7 @@ bool llvm::isTriviallyVectorizable(Intrinsic::ID ID) {
case Intrinsic::exp:
case Intrinsic::exp10:
case Intrinsic::exp2:
+ case Intrinsic::ldexp:
case Intrinsic::log:
case Intrinsic::log10:
case Intrinsic::log2:
@@ -108,6 +109,8 @@ bool llvm::isTriviallyVectorizable(Intrinsic::ID ID) {
case Intrinsic::canonicalize:
case Intrinsic::fptosi_sat:
case Intrinsic::fptoui_sat:
+ case Intrinsic::lround:
+ case Intrinsic::llround:
case Intrinsic::lrint:
case Intrinsic::llrint:
case Intrinsic::ucmp:
@@ -189,6 +192,8 @@ bool llvm::isVectorIntrinsicWithOverloadTypeAtArg(
switch (ID) {
case Intrinsic::fptosi_sat:
case Intrinsic::fptoui_sat:
+ case Intrinsic::lround:
+ case Intrinsic::llround:
case Intrinsic::lrint:
case Intrinsic::llrint:
case Intrinsic::vp_lrint:
@@ -203,6 +208,7 @@ bool llvm::isVectorIntrinsicWithOverloadTypeAtArg(
case Intrinsic::vp_is_fpclass:
return OpdIdx == 0;
case Intrinsic::powi:
+ case Intrinsic::ldexp:
return OpdIdx == -1 || OpdIdx == 1;
default:
return OpdIdx == -1;
@@ -240,30 +246,6 @@ Intrinsic::ID llvm::getVectorIntrinsicIDForCall(const CallInst *CI,
return Intrinsic::not_intrinsic;
}
-struct InterleaveIntrinsic {
- Intrinsic::ID Interleave, Deinterleave;
-};
-
-static InterleaveIntrinsic InterleaveIntrinsics[] = {
- {Intrinsic::vector_interleave2, Intrinsic::vector_deinterleave2},
- {Intrinsic::vector_interleave3, Intrinsic::vector_deinterleave3},
- {Intrinsic::vector_interleave4, Intrinsic::vector_deinterleave4},
- {Intrinsic::vector_interleave5, Intrinsic::vector_deinterleave5},
- {Intrinsic::vector_interleave6, Intrinsic::vector_deinterleave6},
- {Intrinsic::vector_interleave7, Intrinsic::vector_deinterleave7},
- {Intrinsic::vector_interleave8, Intrinsic::vector_deinterleave8},
-};
-
-Intrinsic::ID llvm::getInterleaveIntrinsicID(unsigned Factor) {
- assert(Factor >= 2 && Factor <= 8 && "Unexpected factor");
- return InterleaveIntrinsics[Factor - 2].Interleave;
-}
-
-Intrinsic::ID llvm::getDeinterleaveIntrinsicID(unsigned Factor) {
- assert(Factor >= 2 && Factor <= 8 && "Unexpected factor");
- return InterleaveIntrinsics[Factor - 2].Deinterleave;
-}
-
unsigned llvm::getInterleaveIntrinsicFactor(Intrinsic::ID ID) {
switch (ID) {
case Intrinsic::vector_interleave2:
generated by cgit v1.1 at 2025-07-30 07:10:50 +0000