[llvm] [LV, VP]VP intrinsics support for the Loop Vectorizer + adding new tail-folding mode using EVL. (PR #76172)
Alexey Bataev via llvm-commits
llvm-commits at lists.llvm.org
Mon Mar 4 10:08:19 PST 2024
https://github.com/alexey-bataev updated https://github.com/llvm/llvm-project/pull/76172
>From c1f151ff8dae7882b6828740db407beb00c1d6cb Mon Sep 17 00:00:00 2001
From: Alexey Bataev <a.bataev at outlook.com>
Date: Wed, 28 Feb 2024 08:36:58 -0800
Subject: [PATCH 1/9] [LV, VP]VP intrinsics support for the Loop Vectorizer +
adding new tail-folding mode using EVL.
This patch introduces generating VP intrinsics in the Loop Vectorizer.
Currently the Loop Vectorizer supports vector predication in a very limited capacity via tail-folding and masked load/store/gather/scatter intrinsics. However, this does not let architectures with active vector length predication support take advantage of their capabilities. Architectures with general masked predication support also can only take advantage of predication on memory operations. By having a way for the Loop Vectorizer to generate Vector Predication intrinsics, which (will) provide a target-independent way to model predicated vector instructions, These architectures can make better use of their predication capabilities.
Our first approach (implemented in this patch) builds on top of the existing tail-folding mechanism in the LV (just adds a new tail-folding mode using EVL), but instead of generating masked intrinsics for memory operations it generates VP intrinsics for loads/stores instructions. The patch adds a new VPlanTransforms to replace the wide header predicate compare with EVL and updates codegen for load/stores to use VP store/load with EVL.
Other important part of this approach is how the Explicit Vector Length is computed. (VP intrinsics define this vector length parameter as Explicit Vector Length (EVL)). We use an experimental intrinsic `get_vector_length`, that can be lowered to architecture specific instruction(s) to compute EVL.
Also, added a new recipe to emit instructions for computing EVL. Using VPlan in this way will eventually help build and compare VPlans corresponding to different strategies and alternatives.
Differential Revision: https://reviews.llvm.org/D99750
---
.../llvm/Analysis/TargetTransformInfo.h | 5 +-
.../Target/RISCV/RISCVTargetTransformInfo.cpp | 4 +
.../Target/RISCV/RISCVTargetTransformInfo.h | 16 ++
.../Transforms/Vectorize/LoopVectorize.cpp | 167 ++++++++++++++++-
llvm/lib/Transforms/Vectorize/VPlan.h | 45 +++++
.../Transforms/Vectorize/VPlanAnalysis.cpp | 16 +-
.../lib/Transforms/Vectorize/VPlanRecipes.cpp | 65 +++++++
.../Transforms/Vectorize/VPlanTransforms.cpp | 119 ++++++++++--
.../Transforms/Vectorize/VPlanTransforms.h | 7 +
llvm/lib/Transforms/Vectorize/VPlanValue.h | 1 +
.../Transforms/Vectorize/VPlanVerifier.cpp | 51 ++++++
.../LoopVectorize/RISCV/inloop-reduction.ll | 68 ++++++-
.../RISCV/vectorize-vp-intrinsics.ll | 142 +++++++++++++++
.../RISCV/vplan-vp-intrinsics.ll | 134 ++++++++++++++
.../X86/vectorize-vp-intrinsics.ll | 127 +++++++++++++
.../LoopVectorize/X86/vplan-vp-intrinsics.ll | 89 +++++++++
.../vectorize-vp-intrinsics-gather-scatter.ll | 48 +++++
.../vectorize-vp-intrinsics-interleave.ll | 170 ++++++++++++++++++
.../vectorize-vp-intrinsics-iv32.ll | 85 +++++++++
...ectorize-vp-intrinsics-masked-loadstore.ll | 60 +++++++
.../vectorize-vp-intrinsics-no-masking.ll | 36 ++++
...torize-vp-intrinsics-reverse-load-store.ll | 49 +++++
.../LoopVectorize/vectorize-vp-intrinsics.ll | 98 ++++++++++
.../LoopVectorize/vplan-vp-intrinsics.ll | 37 ++++
24 files changed, 1608 insertions(+), 31 deletions(-)
create mode 100644 llvm/test/Transforms/LoopVectorize/RISCV/vectorize-vp-intrinsics.ll
create mode 100644 llvm/test/Transforms/LoopVectorize/RISCV/vplan-vp-intrinsics.ll
create mode 100644 llvm/test/Transforms/LoopVectorize/X86/vectorize-vp-intrinsics.ll
create mode 100644 llvm/test/Transforms/LoopVectorize/X86/vplan-vp-intrinsics.ll
create mode 100644 llvm/test/Transforms/LoopVectorize/vectorize-vp-intrinsics-gather-scatter.ll
create mode 100644 llvm/test/Transforms/LoopVectorize/vectorize-vp-intrinsics-interleave.ll
create mode 100644 llvm/test/Transforms/LoopVectorize/vectorize-vp-intrinsics-iv32.ll
create mode 100644 llvm/test/Transforms/LoopVectorize/vectorize-vp-intrinsics-masked-loadstore.ll
create mode 100644 llvm/test/Transforms/LoopVectorize/vectorize-vp-intrinsics-no-masking.ll
create mode 100644 llvm/test/Transforms/LoopVectorize/vectorize-vp-intrinsics-reverse-load-store.ll
create mode 100644 llvm/test/Transforms/LoopVectorize/vectorize-vp-intrinsics.ll
create mode 100644 llvm/test/Transforms/LoopVectorize/vplan-vp-intrinsics.ll
diff --git a/llvm/include/llvm/Analysis/TargetTransformInfo.h b/llvm/include/llvm/Analysis/TargetTransformInfo.h
index 58577a6b6eb5c0..5874e89b85b9aa 100644
--- a/llvm/include/llvm/Analysis/TargetTransformInfo.h
+++ b/llvm/include/llvm/Analysis/TargetTransformInfo.h
@@ -190,7 +190,10 @@ enum class TailFoldingStyle {
/// Use predicate to control both data and control flow, but modify
/// the trip count so that a runtime overflow check can be avoided
/// and such that the scalar epilogue loop can always be removed.
- DataAndControlFlowWithoutRuntimeCheck
+ DataAndControlFlowWithoutRuntimeCheck,
+ /// Use predicated EVL instructions for tail-folding.
+ /// Indicates that VP intrinsics should be used if tail-folding is enabled.
+ DataWithEVL,
};
struct TailFoldingInfo {
diff --git a/llvm/lib/Target/RISCV/RISCVTargetTransformInfo.cpp b/llvm/lib/Target/RISCV/RISCVTargetTransformInfo.cpp
index 504970c0a9ac74..bb038510caed60 100644
--- a/llvm/lib/Target/RISCV/RISCVTargetTransformInfo.cpp
+++ b/llvm/lib/Target/RISCV/RISCVTargetTransformInfo.cpp
@@ -237,6 +237,10 @@ RISCVTTIImpl::getIntImmCostIntrin(Intrinsic::ID IID, unsigned Idx,
return TTI::TCC_Free;
}
+bool RISCVTTIImpl::hasActiveVectorLength(unsigned, Type *DataTy, Align) const {
+ return ST->hasVInstructions();
+}
+
TargetTransformInfo::PopcntSupportKind
RISCVTTIImpl::getPopcntSupport(unsigned TyWidth) {
assert(isPowerOf2_32(TyWidth) && "Ty width must be power of 2");
diff --git a/llvm/lib/Target/RISCV/RISCVTargetTransformInfo.h b/llvm/lib/Target/RISCV/RISCVTargetTransformInfo.h
index af36e9d5d5e886..e2ce97aef9c1db 100644
--- a/llvm/lib/Target/RISCV/RISCVTargetTransformInfo.h
+++ b/llvm/lib/Target/RISCV/RISCVTargetTransformInfo.h
@@ -75,6 +75,22 @@ class RISCVTTIImpl : public BasicTTIImplBase<RISCVTTIImpl> {
const APInt &Imm, Type *Ty,
TTI::TargetCostKind CostKind);
+ /// \name Vector Predication Information
+ /// Whether the target supports the %evl parameter of VP intrinsic efficiently
+ /// in hardware, for the given opcode and type/alignment. (see LLVM Language
+ /// Reference - "Vector Predication Intrinsics",
+ /// https://llvm.org/docs/LangRef.html#vector-predication-intrinsics and
+ /// "IR-level VP intrinsics",
+ /// https://llvm.org/docs/Proposals/VectorPredication.html#ir-level-vp-intrinsics).
+ /// \param Opcode the opcode of the instruction checked for predicated version
+ /// support.
+ /// \param DataType the type of the instruction with the \p Opcode checked for
+ /// prediction support.
+ /// \param Alignment the alignment for memory access operation checked for
+ /// predicated version support.
+ bool hasActiveVectorLength(unsigned Opcode, Type *DataType,
+ Align Alignment) const;
+
TargetTransformInfo::PopcntSupportKind getPopcntSupport(unsigned TyWidth);
bool shouldExpandReduction(const IntrinsicInst *II) const;
diff --git a/llvm/lib/Transforms/Vectorize/LoopVectorize.cpp b/llvm/lib/Transforms/Vectorize/LoopVectorize.cpp
index 50a073e890626e..7246144d630c63 100644
--- a/llvm/lib/Transforms/Vectorize/LoopVectorize.cpp
+++ b/llvm/lib/Transforms/Vectorize/LoopVectorize.cpp
@@ -124,6 +124,7 @@
#include "llvm/IR/User.h"
#include "llvm/IR/Value.h"
#include "llvm/IR/ValueHandle.h"
+#include "llvm/IR/VectorBuilder.h"
#include "llvm/IR/Verifier.h"
#include "llvm/Support/Casting.h"
#include "llvm/Support/CommandLine.h"
@@ -248,10 +249,12 @@ static cl::opt<TailFoldingStyle> ForceTailFoldingStyle(
clEnumValN(TailFoldingStyle::DataAndControlFlow, "data-and-control",
"Create lane mask using active.lane.mask intrinsic, and use "
"it for both data and control flow"),
- clEnumValN(
- TailFoldingStyle::DataAndControlFlowWithoutRuntimeCheck,
- "data-and-control-without-rt-check",
- "Similar to data-and-control, but remove the runtime check")));
+ clEnumValN(TailFoldingStyle::DataAndControlFlowWithoutRuntimeCheck,
+ "data-and-control-without-rt-check",
+ "Similar to data-and-control, but remove the runtime check"),
+ clEnumValN(TailFoldingStyle::DataWithEVL, "data-with-evl",
+ "Use predicated EVL instructions for tail folding if the "
+ "target supports vector length predication")));
static cl::opt<bool> MaximizeBandwidth(
"vectorizer-maximize-bandwidth", cl::init(false), cl::Hidden,
@@ -1511,7 +1514,9 @@ class LoopVectorizationCostModel {
/// Selects and saves TailFoldingStyle for 2 options - if IV update may
/// overflow or not.
- void setTailFoldingStyles() {
+ /// \param IsScalableVF true if scalable vector factors enabled.
+ /// \param UserIC User specific interleave count.
+ void setTailFoldingStyles(bool IsScalableVF, unsigned UserIC) {
assert(ChosenTailFoldingStyle.first == TailFoldingStyle::None &&
ChosenTailFoldingStyle.second == TailFoldingStyle::None &&
"Tail folding must not be selected yet.");
@@ -1521,6 +1526,36 @@ class LoopVectorizationCostModel {
if (ForceTailFoldingStyle.getNumOccurrences()) {
ChosenTailFoldingStyle.first = ChosenTailFoldingStyle.second =
ForceTailFoldingStyle;
+ if (ChosenTailFoldingStyle.first == TailFoldingStyle::DataWithEVL) {
+ // FIXME: use actual opcode/data type for analysis here.
+ // FIXME: Investigate opportunity for fixed vector factor.
+ bool EVLIsLegal =
+ IsScalableVF && UserIC <= 1 &&
+ TTI.hasActiveVectorLength(0, nullptr, Align()) &&
+ !EnableVPlanNativePath &&
+ // FIXME: implement support for max safe dependency distance.
+ Legal->isSafeForAnyVectorWidth() &&
+ // FIXME: remove this once reductions are supported.
+ Legal->getReductionVars().empty() &&
+ // FIXME: remove this once vp_reverse is supported.
+ none_of(WideningDecisions,
+ [](const std::pair<std::pair<Instruction *, ElementCount>,
+ std::pair<InstWidening, InstructionCost>>
+ &Data) {
+ return Data.second.first == CM_Widen_Reverse;
+ });
+ if (!EVLIsLegal) {
+ ChosenTailFoldingStyle.first = ChosenTailFoldingStyle.second =
+ TailFoldingStyle::DataWithoutLaneMask;
+ LLVM_DEBUG(dbgs()
+ << "LV: Preference for VP intrinsics indicated. Will "
+ "not try to generate VP Intrinsics since "
+ << (UserIC > 1
+ ? "interleave count specified is greater than 1.\n"
+ : "the target does not support vector length "
+ "predication.\n"));
+ }
+ }
return;
}
@@ -1544,6 +1579,23 @@ class LoopVectorizationCostModel {
return foldTailByMasking() || Legal->blockNeedsPredication(BB);
}
+ /// Returns true if VP intrinsics with explicit vector length support should
+ /// be generated in the tail folded loop.
+ bool useVPIWithVPEVLVectorization() const {
+ return PreferEVL && !EnableVPlanNativePath &&
+ getTailFoldingStyle() == TailFoldingStyle::DataWithEVL &&
+ // FIXME: implement support for max safe dependency distance.
+ Legal->isSafeForAnyVectorWidth() &&
+ // FIXME: remove this once reductions are supported.
+ Legal->getReductionVars().empty() &&
+ // FIXME: remove this once vp_reverse is supported.
+ none_of(
+ WideningDecisions,
+ [](const std::pair<std::pair<Instruction *, ElementCount>,
+ std::pair<InstWidening, InstructionCost>>
+ &Data) { return Data.second.first == CM_Widen_Reverse; });
+ }
+
/// Returns true if the Phi is part of an inloop reduction.
bool isInLoopReduction(PHINode *Phi) const {
return InLoopReductions.contains(Phi);
@@ -1691,6 +1743,10 @@ class LoopVectorizationCostModel {
std::pair<TailFoldingStyle, TailFoldingStyle> ChosenTailFoldingStyle =
std::make_pair(TailFoldingStyle::None, TailFoldingStyle::None);
+ /// Control whether to generate VP intrinsics with explicit-vector-length
+ /// support in vectorized code.
+ bool PreferEVL = false;
+
/// A map holding scalar costs for different vectorization factors. The
/// presence of a cost for an instruction in the mapping indicates that the
/// instruction will be scalarized when vectorizing with the associated
@@ -4646,9 +4702,22 @@ LoopVectorizationCostModel::computeMaxVF(ElementCount UserVF, unsigned UserIC) {
// found modulo the vectorization factor is not zero, try to fold the tail
// by masking.
// FIXME: look for a smaller MaxVF that does divide TC rather than masking.
- setTailFoldingStyles();
- if (foldTailByMasking())
+ setTailFoldingStyles(MaxFactors.ScalableVF.isScalable(), UserIC);
+ if (foldTailByMasking()) {
+ if (getTailFoldingStyle() == TailFoldingStyle::DataWithEVL) {
+ LLVM_DEBUG(dbgs() << "LV: Preference for VP intrinsics indicated. Will "
+ "try to generate VP Intrinsics with scalable vector "
+ "factors only.\n");
+ // Tail folded loop using VP intrinsics restricts the VF to be scalable
+ // for now.
+ // TODO: extend it for fixed vectors, if required.
+ assert(MaxFactors.ScalableVF.isScalable() &&
+ "Expected scalable vector factor.");
+
+ MaxFactors.FixedVF = ElementCount::getFixed(1);
+ }
return MaxFactors;
+ }
// If there was a tail-folding hint/switch, but we can't fold the tail by
// masking, fallback to a vectorization with a scalar epilogue.
@@ -5256,6 +5325,10 @@ LoopVectorizationCostModel::selectInterleaveCount(ElementCount VF,
if (!isScalarEpilogueAllowed())
return 1;
+ // Do not interleave if EVL is preferred and no User IC is specified.
+ if (useVPIWithVPEVLVectorization())
+ return 1;
+
// We used the distance for the interleave count.
if (!Legal->isSafeForAnyVectorWidth())
return 1;
@@ -8504,6 +8577,8 @@ void LoopVectorizationPlanner::buildVPlansWithVPRecipes(ElementCount MinVF,
VPlanTransforms::truncateToMinimalBitwidths(
*Plan, CM.getMinimalBitwidths(), PSE.getSE()->getContext());
VPlanTransforms::optimize(*Plan, *PSE.getSE());
+ if (CM.useVPIWithVPEVLVectorization())
+ VPlanTransforms::addExplicitVectorLength(*Plan);
assert(verifyVPlanIsValid(*Plan) && "VPlan is invalid");
VPlans.push_back(std::move(Plan));
}
@@ -9362,6 +9437,52 @@ void VPReplicateRecipe::execute(VPTransformState &State) {
State.ILV->scalarizeInstruction(UI, this, VPIteration(Part, Lane), State);
}
+/// Creates either vp_store or vp_scatter intrinsics calls to represent
+/// predicated store/scatter.
+static Instruction *
+lowerStoreUsingVectorIntrinsics(IRBuilderBase &Builder, Value *Addr,
+ Value *StoredVal, bool IsScatter, Value *Mask,
+ Value *EVLPart, const Align &Alignment) {
+ CallInst *Call;
+ if (IsScatter) {
+ Call = Builder.CreateIntrinsic(Type::getVoidTy(EVLPart->getContext()),
+ Intrinsic::vp_scatter,
+ {StoredVal, Addr, Mask, EVLPart});
+ } else {
+ VectorBuilder VBuilder(Builder);
+ VBuilder.setEVL(EVLPart).setMask(Mask);
+ Call = cast<CallInst>(VBuilder.createVectorInstruction(
+ Instruction::Store, Type::getVoidTy(EVLPart->getContext()),
+ {StoredVal, Addr}));
+ }
+ Call->addParamAttr(
+ 1, Attribute::getWithAlignment(Call->getContext(), Alignment));
+ return Call;
+}
+
+/// Creates either vp_load or vp_gather intrinsics calls to represent
+/// predicated load/gather.
+static Instruction *lowerLoadUsingVectorIntrinsics(IRBuilderBase &Builder,
+ VectorType *DataTy,
+ Value *Addr, bool IsGather,
+ Value *Mask, Value *EVLPart,
+ const Align &Alignment) {
+ CallInst *Call;
+ if (IsGather) {
+ Call = Builder.CreateIntrinsic(DataTy, Intrinsic::vp_gather,
+ {Addr, Mask, EVLPart}, nullptr,
+ "wide.masked.gather");
+ } else {
+ VectorBuilder VBuilder(Builder);
+ VBuilder.setEVL(EVLPart).setMask(Mask);
+ Call = cast<CallInst>(VBuilder.createVectorInstruction(
+ Instruction::Load, DataTy, Addr, "vp.op.load"));
+ }
+ Call->addParamAttr(
+ 0, Attribute::getWithAlignment(Call->getContext(), Alignment));
+ return Call;
+}
+
void VPWidenMemoryInstructionRecipe::execute(VPTransformState &State) {
VPValue *StoredValue = isStore() ? getStoredValue() : nullptr;
@@ -9400,7 +9521,21 @@ void VPWidenMemoryInstructionRecipe::execute(VPTransformState &State) {
for (unsigned Part = 0; Part < State.UF; ++Part) {
Instruction *NewSI = nullptr;
Value *StoredVal = State.get(StoredValue, Part);
- if (CreateGatherScatter) {
+ if (State.EVL) {
+ assert(State.UF == 1 &&
+ "Expected only UF==1 for predicated vectorization.");
+ Value *EVLPart = State.get(State.EVL, Part);
+ // If EVL is not nullptr, then EVL must be a valid value set during plan
+ // creation, possibly default value = whole vector register length. EVL
+ // is created only if TTI prefers predicated vectorization, thus if EVL
+ // is not nullptr it also implies preference for predicated
+ // vectorization.
+ // FIXME: Support reverse store after vp_reverse is added.
+ Value *MaskPart = isMaskRequired ? BlockInMaskParts[Part] : nullptr;
+ NewSI = lowerStoreUsingVectorIntrinsics(
+ Builder, State.get(getAddr(), Part), StoredVal, CreateGatherScatter,
+ MaskPart, EVLPart, Alignment);
+ } else if (CreateGatherScatter) {
Value *MaskPart = isMaskRequired ? BlockInMaskParts[Part] : nullptr;
Value *VectorGep = State.get(getAddr(), Part);
NewSI = Builder.CreateMaskedScatter(StoredVal, VectorGep, Alignment,
@@ -9430,7 +9565,21 @@ void VPWidenMemoryInstructionRecipe::execute(VPTransformState &State) {
State.setDebugLocFrom(LI->getDebugLoc());
for (unsigned Part = 0; Part < State.UF; ++Part) {
Value *NewLI;
- if (CreateGatherScatter) {
+ if (State.EVL) {
+ assert(State.UF == 1 &&
+ "Expected only UF==1 for predicated vectorization.");
+ Value *EVLPart = State.get(State.EVL, Part);
+ // If EVL is not nullptr, then EVL must be a valid value set during plan
+ // creation, possibly default value = whole vector register length. EVL
+ // is created only if TTI prefers predicated vectorization, thus if EVL
+ // is not nullptr it also implies preference for predicated
+ // vectorization.
+ // FIXME: Support reverse loading after vp_reverse is added.
+ Value *MaskPart = isMaskRequired ? BlockInMaskParts[Part] : nullptr;
+ NewLI = lowerLoadUsingVectorIntrinsics(
+ Builder, DataTy, State.get(getAddr(), Part), CreateGatherScatter,
+ MaskPart, EVLPart, Alignment);
+ } else if (CreateGatherScatter) {
Value *MaskPart = isMaskRequired ? BlockInMaskParts[Part] : nullptr;
Value *VectorGep = State.get(getAddr(), Part);
NewLI = Builder.CreateMaskedGather(DataTy, VectorGep, Alignment, MaskPart,
diff --git a/llvm/lib/Transforms/Vectorize/VPlan.h b/llvm/lib/Transforms/Vectorize/VPlan.h
index 16c09a83e777dd..c44b3118f100e5 100644
--- a/llvm/lib/Transforms/Vectorize/VPlan.h
+++ b/llvm/lib/Transforms/Vectorize/VPlan.h
@@ -242,6 +242,12 @@ struct VPTransformState {
ElementCount VF;
unsigned UF;
+ /// If EVL is not nullptr, then EVL must be a valid value set during plan
+ /// transformation, possibly a default value = whole vector register length.
+ /// EVL is created only if TTI prefers predicated vectorization, thus if EVL
+ /// is not nullptr it also implies preference for predicated vectorization.
+ VPValue *EVL = nullptr;
+
/// Hold the indices to generate specific scalar instructions. Null indicates
/// that all instances are to be generated, using either scalar or vector
/// instructions.
@@ -1149,6 +1155,8 @@ class VPInstruction : public VPRecipeWithIRFlags {
SLPLoad,
SLPStore,
ActiveLaneMask,
+ ExplicitVectorLength,
+ ExplicitVectorLengthIVIncrement,
CalculateTripCountMinusVF,
// Increment the canonical IV separately for each unrolled part.
CanonicalIVIncrementForPart,
@@ -2445,6 +2453,43 @@ class VPActiveLaneMaskPHIRecipe : public VPHeaderPHIRecipe {
#endif
};
+/// A recipe for generating the phi node for the current index of elements,
+/// adjusted in accordance with EVL value. It starts at StartIV value and gets
+/// incremented by EVL in each iteration of the vector loop.
+class VPEVLBasedIVPHIRecipe : public VPHeaderPHIRecipe {
+public:
+ VPEVLBasedIVPHIRecipe(VPValue *StartMask, DebugLoc DL)
+ : VPHeaderPHIRecipe(VPDef::VPEVLBasedIVPHISC, nullptr, StartMask, DL) {}
+
+ ~VPEVLBasedIVPHIRecipe() override = default;
+
+ VPEVLBasedIVPHIRecipe *clone() override {
+ return new VPEVLBasedIVPHIRecipe(getOperand(0), getDebugLoc());
+ }
+
+ VP_CLASSOF_IMPL(VPDef::VPEVLBasedIVPHISC)
+
+ static inline bool classof(const VPHeaderPHIRecipe *D) {
+ return D->getVPDefID() == VPDef::VPEVLBasedIVPHISC;
+ }
+
+ /// Generate phi for handling IV based on EVL over iterations correctly.
+ void execute(VPTransformState &State) override;
+
+ /// Returns true if the recipe only uses the first lane of operand \p Op.
+ bool onlyFirstLaneUsed(const VPValue *Op) const override {
+ assert(is_contained(operands(), Op) &&
+ "Op must be an operand of the recipe");
+ return true;
+ }
+
+#if !defined(NDEBUG) || defined(LLVM_ENABLE_DUMP)
+ /// Print the recipe.
+ void print(raw_ostream &O, const Twine &Indent,
+ VPSlotTracker &SlotTracker) const override;
+#endif
+};
+
/// A Recipe for widening the canonical induction variable of the vector loop.
class VPWidenCanonicalIVRecipe : public VPSingleDefRecipe {
public:
diff --git a/llvm/lib/Transforms/Vectorize/VPlanAnalysis.cpp b/llvm/lib/Transforms/Vectorize/VPlanAnalysis.cpp
index f55beac2047c94..9fe1da9da93deb 100644
--- a/llvm/lib/Transforms/Vectorize/VPlanAnalysis.cpp
+++ b/llvm/lib/Transforms/Vectorize/VPlanAnalysis.cpp
@@ -213,14 +213,14 @@ Type *VPTypeAnalysis::inferScalarType(const VPValue *V) {
Type *ResultTy =
TypeSwitch<const VPRecipeBase *, Type *>(V->getDefiningRecipe())
.Case<VPCanonicalIVPHIRecipe, VPFirstOrderRecurrencePHIRecipe,
- VPReductionPHIRecipe, VPWidenPointerInductionRecipe>(
- [this](const auto *R) {
- // Handle header phi recipes, except VPWienIntOrFpInduction
- // which needs special handling due it being possibly truncated.
- // TODO: consider inferring/caching type of siblings, e.g.,
- // backedge value, here and in cases below.
- return inferScalarType(R->getStartValue());
- })
+ VPReductionPHIRecipe, VPWidenPointerInductionRecipe,
+ VPEVLBasedIVPHIRecipe>([this](const auto *R) {
+ // Handle header phi recipes, except VPWienIntOrFpInduction
+ // which needs special handling due it being possibly truncated.
+ // TODO: consider inferring/caching type of siblings, e.g.,
+ // backedge value, here and in cases below.
+ return inferScalarType(R->getStartValue());
+ })
.Case<VPWidenIntOrFpInductionRecipe, VPDerivedIVRecipe>(
[](const auto *R) { return R->getScalarType(); })
.Case<VPPredInstPHIRecipe, VPWidenPHIRecipe, VPScalarIVStepsRecipe,
diff --git a/llvm/lib/Transforms/Vectorize/VPlanRecipes.cpp b/llvm/lib/Transforms/Vectorize/VPlanRecipes.cpp
index 27b72575ddd51a..da3cf5c3e9cb38 100644
--- a/llvm/lib/Transforms/Vectorize/VPlanRecipes.cpp
+++ b/llvm/lib/Transforms/Vectorize/VPlanRecipes.cpp
@@ -350,6 +350,43 @@ Value *VPInstruction::generateInstruction(VPTransformState &State,
Value *Zero = ConstantInt::get(ScalarTC->getType(), 0);
return Builder.CreateSelect(Cmp, Sub, Zero);
}
+ case VPInstruction::ExplicitVectorLength: {
+ // Compute EVL
+ auto GetSetVL = [=](VPTransformState &State, Value *EVL) {
+ assert(EVL->getType()->isIntegerTy() &&
+ "Requested vector length should be an integer.");
+
+ // TODO: Add support for MaxSafeDist for correct loop emission.
+ assert(State.VF.isScalable() && "Expected scalable vector factor.");
+ Value *VFArg = State.Builder.getInt32(State.VF.getKnownMinValue());
+
+ Value *GVL = State.Builder.CreateIntrinsic(
+ State.Builder.getInt32Ty(), Intrinsic::experimental_get_vector_length,
+ {EVL, VFArg, State.Builder.getTrue()});
+ return GVL;
+ };
+ // TODO: Restructure this code with an explicit remainder loop, vsetvli can
+ // be outside of the main loop.
+ assert(Part == 0 && "No unrolling expected for predicated vectorization.");
+ // Compute VTC - IV as the EVL(requested vector length).
+ Value *Index = State.get(getOperand(0), 0);
+ Value *TripCount = State.get(getOperand(1), VPIteration(0, 0));
+ Value *EVL = State.Builder.CreateSub(TripCount, Index);
+ Value *SetVL = GetSetVL(State, EVL);
+ assert(!State.EVL && "multiple EVL recipes");
+ State.EVL = this;
+ return SetVL;
+ }
+ // TODO: remove this once a regular Add VPInstruction is supported.
+ case VPInstruction::ExplicitVectorLengthIVIncrement: {
+ assert(Part == 0 && "Expected unroll factor 1 for VP vectorization.");
+ Value *Phi = State.get(getOperand(0), VPIteration(Part, 0));
+ Value *EVL = State.get(getOperand(1), VPIteration(Part, 0));
+ assert(EVL->getType() == Phi->getType() &&
+ "EVL and Phi must have the same type.");
+ return Builder.CreateAdd(Phi, EVL, Name, hasNoUnsignedWrap(),
+ hasNoSignedWrap());
+ }
case VPInstruction::CanonicalIVIncrementForPart: {
auto *IV = State.get(getOperand(0), VPIteration(0, 0));
if (Part == 0)
@@ -570,6 +607,12 @@ void VPInstruction::print(raw_ostream &O, const Twine &Indent,
case VPInstruction::ActiveLaneMask:
O << "active lane mask";
break;
+ case VPInstruction::ExplicitVectorLength:
+ O << "EXPLICIT-VECTOR-LENGTH";
+ break;
+ case VPInstruction::ExplicitVectorLengthIVIncrement:
+ O << "EXPLICIT-VECTOR-LENGTH +";
+ break;
case VPInstruction::FirstOrderRecurrenceSplice:
O << "first-order splice";
break;
@@ -1913,3 +1956,25 @@ void VPActiveLaneMaskPHIRecipe::print(raw_ostream &O, const Twine &Indent,
printOperands(O, SlotTracker);
}
#endif
+
+void VPEVLBasedIVPHIRecipe::execute(VPTransformState &State) {
+ BasicBlock *VectorPH = State.CFG.getPreheaderBBFor(this);
+ assert(State.UF == 1 && "Expected unroll factor 1 for VP vectorization.");
+ Value *Start = State.get(getOperand(0), VPIteration(0, 0));
+ PHINode *EntryPart =
+ State.Builder.CreatePHI(Start->getType(), 2, "evl.based.iv");
+ EntryPart->addIncoming(Start, VectorPH);
+ EntryPart->setDebugLoc(getDebugLoc());
+ State.set(this, EntryPart, 0);
+}
+
+#if !defined(NDEBUG) || defined(LLVM_ENABLE_DUMP)
+void VPEVLBasedIVPHIRecipe::print(raw_ostream &O, const Twine &Indent,
+ VPSlotTracker &SlotTracker) const {
+ O << Indent << "EXPLICIT-VECTOR-LENGTH-BASED-IV-PHI ";
+
+ printAsOperand(O, SlotTracker);
+ O << " = phi ";
+ printOperands(O, SlotTracker);
+}
+#endif
diff --git a/llvm/lib/Transforms/Vectorize/VPlanTransforms.cpp b/llvm/lib/Transforms/Vectorize/VPlanTransforms.cpp
index 9d6deb802e2090..328888c122b15d 100644
--- a/llvm/lib/Transforms/Vectorize/VPlanTransforms.cpp
+++ b/llvm/lib/Transforms/Vectorize/VPlanTransforms.cpp
@@ -1172,6 +1172,45 @@ static VPActiveLaneMaskPHIRecipe *addVPLaneMaskPhiAndUpdateExitBranch(
return LaneMaskPhi;
}
+/// Replaces (ICMP_ULE, WideCanonicalIV, backedge-taken-count) pattern using
+/// the given idiom \p Idiom.
+static void replaceHeaderPredicateWithIdiom(
+ VPlan &Plan, VPValue &Idiom,
+ function_ref<bool(VPUser &, unsigned)> Cond = {}) {
+ auto *FoundWidenCanonicalIVUser =
+ find_if(Plan.getCanonicalIV()->users(),
+ [](VPUser *U) { return isa<VPWidenCanonicalIVRecipe>(U); });
+ if (FoundWidenCanonicalIVUser == Plan.getCanonicalIV()->users().end())
+ return;
+ auto *WideCanonicalIV =
+ cast<VPWidenCanonicalIVRecipe>(*FoundWidenCanonicalIVUser);
+ // Walk users of WideCanonicalIV and replace all compares of the form
+ // (ICMP_ULE, WideCanonicalIV, backedge-taken-count) with
+ // the given idiom VPValue.
+ VPValue *BTC = Plan.getOrCreateBackedgeTakenCount();
+ for (VPUser *U : SmallVector<VPUser *>(WideCanonicalIV->users())) {
+ auto *CompareToReplace = dyn_cast<VPInstruction>(U);
+ if (!CompareToReplace ||
+ CompareToReplace->getOpcode() != Instruction::ICmp ||
+ CompareToReplace->getPredicate() != CmpInst::ICMP_ULE ||
+ CompareToReplace->getOperand(1) != BTC)
+ continue;
+
+ assert(CompareToReplace->getOperand(0) == WideCanonicalIV &&
+ "WidenCanonicalIV must be the first operand of the compare");
+ if (Cond) {
+ CompareToReplace->replaceUsesWithIf(&Idiom, Cond);
+ if (!CompareToReplace->getNumUsers())
+ CompareToReplace->eraseFromParent();
+ } else {
+ CompareToReplace->replaceAllUsesWith(&Idiom);
+ CompareToReplace->eraseFromParent();
+ }
+ }
+ if (!WideCanonicalIV->getNumUsers())
+ WideCanonicalIV->eraseFromParent();
+}
+
void VPlanTransforms::addActiveLaneMask(
VPlan &Plan, bool UseActiveLaneMaskForControlFlow,
bool DataAndControlFlowWithoutRuntimeCheck) {
@@ -1201,20 +1240,76 @@ void VPlanTransforms::addActiveLaneMask(
// Walk users of WideCanonicalIV and replace all compares of the form
// (ICMP_ULE, WideCanonicalIV, backedge-taken-count) with an
// active-lane-mask.
- VPValue *BTC = Plan.getOrCreateBackedgeTakenCount();
- for (VPUser *U : SmallVector<VPUser *>(WideCanonicalIV->users())) {
- auto *CompareToReplace = dyn_cast<VPInstruction>(U);
- if (!CompareToReplace ||
- CompareToReplace->getOpcode() != Instruction::ICmp ||
- CompareToReplace->getPredicate() != CmpInst::ICMP_ULE ||
- CompareToReplace->getOperand(1) != BTC)
- continue;
+ replaceHeaderPredicateWithIdiom(Plan, *LaneMask);
+}
- assert(CompareToReplace->getOperand(0) == WideCanonicalIV &&
- "WidenCanonicalIV must be the first operand of the compare");
- CompareToReplace->replaceAllUsesWith(LaneMask);
- CompareToReplace->eraseFromParent();
+// Add a VPEVLBasedIVPHIRecipe and related recipes to \p Plan and
+// replaces all uses except the canonical IV increment of VPCanonicalIVPHIRecipe
+// with a VPEVLBasedIVPHIRecipe. VPCanonicalIVPHIRecipe is used only
+// for loop iterations counting after this transformation.
+//
+// The function uses the following definitions:
+// %StartV is the canonical induction start value.
+//
+// The function adds the following recipes:
+//
+// vector.ph:
+// ...
+//
+// vector.body:
+// ...
+// %P = EXPLICIT-VECTOR-LENGTH-BASED-IV-PHI [ %StartV, %vector.ph ], [ %NextEVL,
+// %vector.body ]
+// %EVL = EXPLICIT-VECTOR-LENGTH %P, original TC
+// ...
+// %NextEVL = EXPLICIT-VECTOR-LENGTH + %P, %EVL
+// ...
+//
+void VPlanTransforms::addExplicitVectorLength(VPlan &Plan) {
+ VPBasicBlock *Header = Plan.getVectorLoopRegion()->getEntryBasicBlock();
+ auto *CanonicalIVPHI = Plan.getCanonicalIV();
+ VPValue *StartV = CanonicalIVPHI->getStartValue();
+
+ // Walk users of WideCanonicalIV and replace all compares of the form
+ // (ICMP_ULE, WideCanonicalIV, backedge-taken-count) with an
+ // all-true-mask.
+ Value *TrueMask =
+ ConstantInt::getTrue(CanonicalIVPHI->getScalarType()->getContext());
+ VPValue *VPTrueMask = Plan.getVPValueOrAddLiveIn(TrueMask);
+ replaceHeaderPredicateWithIdiom(Plan, *VPTrueMask, [](VPUser &U, unsigned) {
+ return isa<VPWidenMemoryInstructionRecipe>(U);
+ });
+ // Now create the ExplicitVectorLengthPhi recipe in the main loop.
+ auto *EVLPhi = new VPEVLBasedIVPHIRecipe(StartV, DebugLoc());
+ EVLPhi->insertAfter(CanonicalIVPHI);
+ auto *VPEVL = new VPInstruction(VPInstruction::ExplicitVectorLength,
+ {EVLPhi, Plan.getTripCount()});
+ VPEVL->insertBefore(*Header, Header->getFirstNonPhi());
+
+ auto *CanonicalIVIncrement =
+ cast<VPInstruction>(CanonicalIVPHI->getBackedgeValue());
+ VPSingleDefRecipe *OpVPEVL = VPEVL;
+ if (CanonicalIVPHI->getScalarType() !=
+ IntegerType::get(CanonicalIVPHI->getScalarType()->getContext(),
+ /*NumBits=*/32)) {
+ OpVPEVL = new VPScalarCastRecipe(Instruction::ZExt, OpVPEVL,
+ CanonicalIVPHI->getScalarType());
+ OpVPEVL->insertBefore(CanonicalIVIncrement);
}
+ auto *NextEVLIV = new VPInstruction(
+ VPInstruction::ExplicitVectorLengthIVIncrement, {OpVPEVL, EVLPhi},
+ {CanonicalIVIncrement->hasNoUnsignedWrap(),
+ CanonicalIVIncrement->hasNoSignedWrap()},
+ CanonicalIVIncrement->getDebugLoc(), "index.evl.next");
+ NextEVLIV->insertBefore(CanonicalIVIncrement);
+ EVLPhi->addOperand(NextEVLIV);
+
+ // Replace all uses of VPCanonicalIVPHIRecipe by
+ // VPEVLBasedIVPHIRecipe except for VPInstruction::CanonicalIVIncrement.
+ CanonicalIVPHI->replaceAllUsesWith(EVLPhi);
+ CanonicalIVIncrement->setOperand(0, CanonicalIVPHI);
+ // TODO: support unroll factor > 1.
+ Plan.setUF(1);
}
void VPlanTransforms::dropPoisonGeneratingRecipes(
diff --git a/llvm/lib/Transforms/Vectorize/VPlanTransforms.h b/llvm/lib/Transforms/Vectorize/VPlanTransforms.h
index ff83c3f083b093..0cbc70713d9c10 100644
--- a/llvm/lib/Transforms/Vectorize/VPlanTransforms.h
+++ b/llvm/lib/Transforms/Vectorize/VPlanTransforms.h
@@ -98,6 +98,13 @@ struct VPlanTransforms {
/// VPlan directly.
static void dropPoisonGeneratingRecipes(
VPlan &Plan, function_ref<bool(BasicBlock *)> BlockNeedsPredication);
+
+ /// Add a VPEVLBasedIVPHIRecipe and related recipes to \p Plan and
+ /// replaces all uses except the canonical IV increment of
+ /// VPCanonicalIVPHIRecipe with a VPEVLBasedIVPHIRecipe.
+ /// VPCanonicalIVPHIRecipe is only used to control the loop after
+ /// this transformation.
+ static void addExplicitVectorLength(VPlan &Plan);
};
} // namespace llvm
diff --git a/llvm/lib/Transforms/Vectorize/VPlanValue.h b/llvm/lib/Transforms/Vectorize/VPlanValue.h
index 1d2c17e91b7ab9..8b221d30e5254c 100644
--- a/llvm/lib/Transforms/Vectorize/VPlanValue.h
+++ b/llvm/lib/Transforms/Vectorize/VPlanValue.h
@@ -368,6 +368,7 @@ class VPDef {
// VPHeaderPHIRecipe need to be kept together.
VPCanonicalIVPHISC,
VPActiveLaneMaskPHISC,
+ VPEVLBasedIVPHISC,
VPFirstOrderRecurrencePHISC,
VPWidenIntOrFpInductionSC,
VPWidenPointerInductionSC,
diff --git a/llvm/lib/Transforms/Vectorize/VPlanVerifier.cpp b/llvm/lib/Transforms/Vectorize/VPlanVerifier.cpp
index 7ebdb914fb852f..364e9cda4fa044 100644
--- a/llvm/lib/Transforms/Vectorize/VPlanVerifier.cpp
+++ b/llvm/lib/Transforms/Vectorize/VPlanVerifier.cpp
@@ -92,7 +92,58 @@ static bool verifyVPBasicBlock(const VPBasicBlock *VPBB,
for (const VPRecipeBase &R : *VPBB)
RecipeNumbering[&R] = Cnt++;
+ // Check if EVL recipes exist only in Entry block and only once.
+ DenseSet<unsigned> EVLFound;
+ const VPBlockBase *Header = nullptr;
+ const VPBlockBase *Exit = nullptr;
+ const VPlan *Plan = VPBB->getPlan();
+ if (Plan && Plan->getEntry()->getNumSuccessors() == 1) {
+ Header = Plan->getVectorLoopRegion()->getEntry();
+ Exit = Plan->getVectorLoopRegion()->getExiting();
+ }
+ auto CheckEVLRecipiesInsts = [&](const VPRecipeBase *R) {
+ if (isa<VPEVLBasedIVPHIRecipe>(R)) {
+ if (!Header || VPBB != Header) {
+ errs() << "EVL PHI recipe not in entry block!\n";
+ return false;
+ }
+ if (EVLFound.contains(VPDef::VPEVLBasedIVPHISC)) {
+ errs() << "EVL PHI recipe inserted more than once!\n";
+ return false;
+ }
+ EVLFound.insert(VPDef::VPEVLBasedIVPHISC);
+ return true;
+ }
+ auto *RInst = dyn_cast<VPInstruction>(R);
+ if (!RInst)
+ return true;
+ switch (RInst->getOpcode()) {
+ case VPInstruction::ExplicitVectorLength:
+ if (!Header || VPBB != Header) {
+ errs() << "EVL instruction not in entry block!\n";
+ return false;
+ }
+ break;
+ case VPInstruction::ExplicitVectorLengthIVIncrement:
+ if (!Exit || VPBB != Exit) {
+ errs() << "EVL inc instruction not in exit block!\n";
+ return false;
+ }
+ break;
+ default:
+ return true;
+ }
+ if (EVLFound.contains(RInst->getOpcode() + VPDef::VPLastPHISC)) {
+ errs() << "EVL instruction inserted more than once!\n";
+ return false;
+ }
+ EVLFound.insert(RInst->getOpcode() + VPDef::VPLastPHISC);
+ return true;
+ };
+
for (const VPRecipeBase &R : *VPBB) {
+ if (!CheckEVLRecipiesInsts(&R))
+ return false;
for (const VPValue *V : R.definedValues()) {
for (const VPUser *U : V->users()) {
auto *UI = dyn_cast<VPRecipeBase>(U);
diff --git a/llvm/test/Transforms/LoopVectorize/RISCV/inloop-reduction.ll b/llvm/test/Transforms/LoopVectorize/RISCV/inloop-reduction.ll
index 57e1dc9051f4d0..dc2298ed21dfcc 100644
--- a/llvm/test/Transforms/LoopVectorize/RISCV/inloop-reduction.ll
+++ b/llvm/test/Transforms/LoopVectorize/RISCV/inloop-reduction.ll
@@ -1,11 +1,13 @@
; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
; RUN: opt -mtriple riscv64-linux-gnu -mattr=+v,+d -passes=loop-vectorize < %s -S -o - | FileCheck %s -check-prefix=OUTLOOP
; RUN: opt -mtriple riscv64-linux-gnu -mattr=+v,+d -passes=loop-vectorize -prefer-inloop-reductions < %s -S -o - | FileCheck %s -check-prefix=INLOOP
-
+; RUN: opt -passes=loop-vectorize -force-tail-folding-style=data-with-evl -prefer-predicate-over-epilogue=predicate-dont-vectorize -mtriple=riscv64 -mattr=+v -S < %s 2>&1 | FileCheck --check-prefix=IF-EVL %s
target datalayout = "e-m:e-p:64:64-i64:64-i128:128-n64-S128"
target triple = "riscv64"
+; FIXME: inloop reductions are not supported yet with predicated vectorization.
+
define i32 @add_i16_i32(ptr nocapture readonly %x, i32 %n) {
; OUTLOOP-LABEL: @add_i16_i32(
; OUTLOOP-NEXT: entry:
@@ -115,6 +117,70 @@ define i32 @add_i16_i32(ptr nocapture readonly %x, i32 %n) {
; INLOOP-NEXT: [[R_0_LCSSA:%.*]] = phi i32 [ 0, [[ENTRY:%.*]] ], [ [[ADD_LCSSA]], [[FOR_COND_CLEANUP_LOOPEXIT]] ]
; INLOOP-NEXT: ret i32 [[R_0_LCSSA]]
;
+; IF-EVL-LABEL: @add_i16_i32(
+; IF-EVL-NEXT: entry:
+; IF-EVL-NEXT: [[CMP6:%.*]] = icmp sgt i32 [[N:%.*]], 0
+; IF-EVL-NEXT: br i1 [[CMP6]], label [[FOR_BODY_PREHEADER:%.*]], label [[FOR_COND_CLEANUP:%.*]]
+; IF-EVL: for.body.preheader:
+; IF-EVL-NEXT: br i1 false, label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
+; IF-EVL: vector.ph:
+; IF-EVL-NEXT: [[TMP0:%.*]] = call i32 @llvm.vscale.i32()
+; IF-EVL-NEXT: [[TMP1:%.*]] = mul i32 [[TMP0]], 4
+; IF-EVL-NEXT: [[TMP2:%.*]] = call i32 @llvm.vscale.i32()
+; IF-EVL-NEXT: [[TMP3:%.*]] = mul i32 [[TMP2]], 4
+; IF-EVL-NEXT: [[TMP4:%.*]] = sub i32 [[TMP3]], 1
+; IF-EVL-NEXT: [[N_RND_UP:%.*]] = add i32 [[N]], [[TMP4]]
+; IF-EVL-NEXT: [[N_MOD_VF:%.*]] = urem i32 [[N_RND_UP]], [[TMP1]]
+; IF-EVL-NEXT: [[N_VEC:%.*]] = sub i32 [[N_RND_UP]], [[N_MOD_VF]]
+; IF-EVL-NEXT: [[TRIP_COUNT_MINUS_1:%.*]] = sub i32 [[N]], 1
+; IF-EVL-NEXT: [[BROADCAST_SPLATINSERT:%.*]] = insertelement <vscale x 4 x i32> poison, i32 [[TRIP_COUNT_MINUS_1]], i64 0
+; IF-EVL-NEXT: [[BROADCAST_SPLAT:%.*]] = shufflevector <vscale x 4 x i32> [[BROADCAST_SPLATINSERT]], <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
+; IF-EVL-NEXT: [[TMP5:%.*]] = call i32 @llvm.vscale.i32()
+; IF-EVL-NEXT: [[TMP6:%.*]] = mul i32 [[TMP5]], 4
+; IF-EVL-NEXT: br label [[VECTOR_BODY:%.*]]
+; IF-EVL: vector.body:
+; IF-EVL-NEXT: [[INDEX:%.*]] = phi i32 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
+; IF-EVL-NEXT: [[VEC_PHI:%.*]] = phi <vscale x 4 x i32> [ zeroinitializer, [[VECTOR_PH]] ], [ [[TMP14:%.*]], [[VECTOR_BODY]] ]
+; IF-EVL-NEXT: [[TMP7:%.*]] = add i32 [[INDEX]], 0
+; IF-EVL-NEXT: [[BROADCAST_SPLATINSERT1:%.*]] = insertelement <vscale x 4 x i32> poison, i32 [[INDEX]], i64 0
+; IF-EVL-NEXT: [[BROADCAST_SPLAT2:%.*]] = shufflevector <vscale x 4 x i32> [[BROADCAST_SPLATINSERT1]], <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
+; IF-EVL-NEXT: [[TMP8:%.*]] = call <vscale x 4 x i32> @llvm.experimental.stepvector.nxv4i32()
+; IF-EVL-NEXT: [[TMP9:%.*]] = add <vscale x 4 x i32> zeroinitializer, [[TMP8]]
+; IF-EVL-NEXT: [[VEC_IV:%.*]] = add <vscale x 4 x i32> [[BROADCAST_SPLAT2]], [[TMP9]]
+; IF-EVL-NEXT: [[TMP10:%.*]] = icmp ule <vscale x 4 x i32> [[VEC_IV]], [[BROADCAST_SPLAT]]
+; IF-EVL-NEXT: [[TMP11:%.*]] = getelementptr inbounds i16, ptr [[X:%.*]], i32 [[TMP7]]
+; IF-EVL-NEXT: [[TMP12:%.*]] = getelementptr inbounds i16, ptr [[TMP11]], i32 0
+; IF-EVL-NEXT: [[WIDE_MASKED_LOAD:%.*]] = call <vscale x 4 x i16> @llvm.masked.load.nxv4i16.p0(ptr [[TMP12]], i32 2, <vscale x 4 x i1> [[TMP10]], <vscale x 4 x i16> poison)
+; IF-EVL-NEXT: [[TMP13:%.*]] = sext <vscale x 4 x i16> [[WIDE_MASKED_LOAD]] to <vscale x 4 x i32>
+; IF-EVL-NEXT: [[TMP14]] = add <vscale x 4 x i32> [[VEC_PHI]], [[TMP13]]
+; IF-EVL-NEXT: [[TMP15:%.*]] = select <vscale x 4 x i1> [[TMP10]], <vscale x 4 x i32> [[TMP14]], <vscale x 4 x i32> [[VEC_PHI]]
+; IF-EVL-NEXT: [[INDEX_NEXT]] = add i32 [[INDEX]], [[TMP6]]
+; IF-EVL-NEXT: [[TMP16:%.*]] = icmp eq i32 [[INDEX_NEXT]], [[N_VEC]]
+; IF-EVL-NEXT: br i1 [[TMP16]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]]
+; IF-EVL: middle.block:
+; IF-EVL-NEXT: [[TMP17:%.*]] = call i32 @llvm.vector.reduce.add.nxv4i32(<vscale x 4 x i32> [[TMP15]])
+; IF-EVL-NEXT: br i1 true, label [[FOR_COND_CLEANUP_LOOPEXIT:%.*]], label [[SCALAR_PH]]
+; IF-EVL: scalar.ph:
+; IF-EVL-NEXT: [[BC_RESUME_VAL:%.*]] = phi i32 [ [[N_VEC]], [[MIDDLE_BLOCK]] ], [ 0, [[FOR_BODY_PREHEADER]] ]
+; IF-EVL-NEXT: [[BC_MERGE_RDX:%.*]] = phi i32 [ 0, [[FOR_BODY_PREHEADER]] ], [ [[TMP17]], [[MIDDLE_BLOCK]] ]
+; IF-EVL-NEXT: br label [[FOR_BODY:%.*]]
+; IF-EVL: for.body:
+; IF-EVL-NEXT: [[I_08:%.*]] = phi i32 [ [[INC:%.*]], [[FOR_BODY]] ], [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ]
+; IF-EVL-NEXT: [[R_07:%.*]] = phi i32 [ [[ADD:%.*]], [[FOR_BODY]] ], [ [[BC_MERGE_RDX]], [[SCALAR_PH]] ]
+; IF-EVL-NEXT: [[ARRAYIDX:%.*]] = getelementptr inbounds i16, ptr [[X]], i32 [[I_08]]
+; IF-EVL-NEXT: [[TMP18:%.*]] = load i16, ptr [[ARRAYIDX]], align 2
+; IF-EVL-NEXT: [[CONV:%.*]] = sext i16 [[TMP18]] to i32
+; IF-EVL-NEXT: [[ADD]] = add nsw i32 [[R_07]], [[CONV]]
+; IF-EVL-NEXT: [[INC]] = add nuw nsw i32 [[I_08]], 1
+; IF-EVL-NEXT: [[EXITCOND:%.*]] = icmp eq i32 [[INC]], [[N]]
+; IF-EVL-NEXT: br i1 [[EXITCOND]], label [[FOR_COND_CLEANUP_LOOPEXIT]], label [[FOR_BODY]], !llvm.loop [[LOOP3:![0-9]+]]
+; IF-EVL: for.cond.cleanup.loopexit:
+; IF-EVL-NEXT: [[ADD_LCSSA:%.*]] = phi i32 [ [[ADD]], [[FOR_BODY]] ], [ [[TMP17]], [[MIDDLE_BLOCK]] ]
+; IF-EVL-NEXT: br label [[FOR_COND_CLEANUP]]
+; IF-EVL: for.cond.cleanup:
+; IF-EVL-NEXT: [[R_0_LCSSA:%.*]] = phi i32 [ 0, [[ENTRY:%.*]] ], [ [[ADD_LCSSA]], [[FOR_COND_CLEANUP_LOOPEXIT]] ]
+; IF-EVL-NEXT: ret i32 [[R_0_LCSSA]]
+;
entry:
%cmp6 = icmp sgt i32 %n, 0
br i1 %cmp6, label %for.body, label %for.cond.cleanup
diff --git a/llvm/test/Transforms/LoopVectorize/RISCV/vectorize-vp-intrinsics.ll b/llvm/test/Transforms/LoopVectorize/RISCV/vectorize-vp-intrinsics.ll
new file mode 100644
index 00000000000000..60d61f25e9f0ca
--- /dev/null
+++ b/llvm/test/Transforms/LoopVectorize/RISCV/vectorize-vp-intrinsics.ll
@@ -0,0 +1,142 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
+; RUN: opt -passes=loop-vectorize \
+; RUN: -force-tail-folding-style=data-with-evl \
+; RUN: -prefer-predicate-over-epilogue=predicate-dont-vectorize \
+; RUN: -mtriple=riscv64 -mattr=+v -S < %s | FileCheck --check-prefix=IF-EVL %s
+
+; RUN: opt -passes=loop-vectorize \
+; RUN: -force-tail-folding-style=none \
+; RUN: -prefer-predicate-over-epilogue=predicate-dont-vectorize \
+; RUN: -mtriple=riscv64 -mattr=+v -S < %s | FileCheck --check-prefix=NO-VP %s
+
+define void @foo(ptr noalias %a, ptr noalias %b, ptr noalias %c, i64 %N) {
+; IF-EVL-LABEL: @foo(
+; IF-EVL-NEXT: entry:
+; IF-EVL-NEXT: [[TMP0:%.*]] = sub i64 -1, [[N:%.*]]
+; IF-EVL-NEXT: [[TMP1:%.*]] = call i64 @llvm.vscale.i64()
+; IF-EVL-NEXT: [[TMP2:%.*]] = mul i64 [[TMP1]], 4
+; IF-EVL-NEXT: [[TMP3:%.*]] = icmp ult i64 [[TMP0]], [[TMP2]]
+; IF-EVL-NEXT: br i1 [[TMP3]], label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
+; IF-EVL: vector.ph:
+; IF-EVL-NEXT: [[TMP4:%.*]] = call i64 @llvm.vscale.i64()
+; IF-EVL-NEXT: [[TMP5:%.*]] = mul i64 [[TMP4]], 4
+; IF-EVL-NEXT: [[TMP6:%.*]] = call i64 @llvm.vscale.i64()
+; IF-EVL-NEXT: [[TMP7:%.*]] = mul i64 [[TMP6]], 4
+; IF-EVL-NEXT: [[TMP8:%.*]] = sub i64 [[TMP7]], 1
+; IF-EVL-NEXT: [[N_RND_UP:%.*]] = add i64 [[N]], [[TMP8]]
+; IF-EVL-NEXT: [[N_MOD_VF:%.*]] = urem i64 [[N_RND_UP]], [[TMP5]]
+; IF-EVL-NEXT: [[N_VEC:%.*]] = sub i64 [[N_RND_UP]], [[N_MOD_VF]]
+; IF-EVL-NEXT: [[TMP9:%.*]] = call i64 @llvm.vscale.i64()
+; IF-EVL-NEXT: [[TMP10:%.*]] = mul i64 [[TMP9]], 4
+; IF-EVL-NEXT: br label [[VECTOR_BODY:%.*]]
+; IF-EVL: vector.body:
+; IF-EVL-NEXT: [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
+; IF-EVL-NEXT: [[EVL_BASED_IV:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_EVL_NEXT:%.*]], [[VECTOR_BODY]] ]
+; IF-EVL-NEXT: [[TMP11:%.*]] = sub i64 [[N]], [[EVL_BASED_IV]]
+; IF-EVL-NEXT: [[TMP12:%.*]] = call i32 @llvm.experimental.get.vector.length.i64(i64 [[TMP11]], i32 4, i1 true)
+; IF-EVL-NEXT: [[TMP13:%.*]] = add i64 [[EVL_BASED_IV]], 0
+; IF-EVL-NEXT: [[TMP14:%.*]] = getelementptr inbounds i32, ptr [[B:%.*]], i64 [[TMP13]]
+; IF-EVL-NEXT: [[TMP15:%.*]] = getelementptr inbounds i32, ptr [[TMP14]], i32 0
+; IF-EVL-NEXT: [[VP_OP_LOAD:%.*]] = call <vscale x 4 x i32> @llvm.vp.load.nxv4i32.p0(ptr align 4 [[TMP15]], <vscale x 4 x i1> shufflevector (<vscale x 4 x i1> insertelement (<vscale x 4 x i1> poison, i1 true, i64 0), <vscale x 4 x i1> poison, <vscale x 4 x i32> zeroinitializer), i32 [[TMP12]])
+; IF-EVL-NEXT: [[TMP16:%.*]] = getelementptr inbounds i32, ptr [[C:%.*]], i64 [[TMP13]]
+; IF-EVL-NEXT: [[TMP17:%.*]] = getelementptr inbounds i32, ptr [[TMP16]], i32 0
+; IF-EVL-NEXT: [[VP_OP_LOAD1:%.*]] = call <vscale x 4 x i32> @llvm.vp.load.nxv4i32.p0(ptr align 4 [[TMP17]], <vscale x 4 x i1> shufflevector (<vscale x 4 x i1> insertelement (<vscale x 4 x i1> poison, i1 true, i64 0), <vscale x 4 x i1> poison, <vscale x 4 x i32> zeroinitializer), i32 [[TMP12]])
+; IF-EVL-NEXT: [[TMP18:%.*]] = add nsw <vscale x 4 x i32> [[VP_OP_LOAD1]], [[VP_OP_LOAD]]
+; IF-EVL-NEXT: [[TMP19:%.*]] = getelementptr inbounds i32, ptr [[A:%.*]], i64 [[TMP13]]
+; IF-EVL-NEXT: [[TMP20:%.*]] = getelementptr inbounds i32, ptr [[TMP19]], i32 0
+; IF-EVL-NEXT: call void @llvm.vp.store.nxv4i32.p0(<vscale x 4 x i32> [[TMP18]], ptr align 4 [[TMP20]], <vscale x 4 x i1> shufflevector (<vscale x 4 x i1> insertelement (<vscale x 4 x i1> poison, i1 true, i64 0), <vscale x 4 x i1> poison, <vscale x 4 x i32> zeroinitializer), i32 [[TMP12]])
+; IF-EVL-NEXT: [[TMP21:%.*]] = zext i32 [[TMP12]] to i64
+; IF-EVL-NEXT: [[INDEX_EVL_NEXT]] = add i64 [[TMP21]], [[EVL_BASED_IV]]
+; IF-EVL-NEXT: [[INDEX_NEXT]] = add i64 [[INDEX]], [[TMP10]]
+; IF-EVL-NEXT: [[TMP22:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
+; IF-EVL-NEXT: br i1 [[TMP22]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]]
+; IF-EVL: middle.block:
+; IF-EVL-NEXT: br i1 true, label [[FOR_COND_CLEANUP:%.*]], label [[SCALAR_PH]]
+; IF-EVL: scalar.ph:
+; IF-EVL-NEXT: [[BC_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ]
+; IF-EVL-NEXT: br label [[FOR_BODY:%.*]]
+; IF-EVL: for.body:
+; IF-EVL-NEXT: [[IV:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[IV_NEXT:%.*]], [[FOR_BODY]] ]
+; IF-EVL-NEXT: [[ARRAYIDX:%.*]] = getelementptr inbounds i32, ptr [[B]], i64 [[IV]]
+; IF-EVL-NEXT: [[TMP23:%.*]] = load i32, ptr [[ARRAYIDX]], align 4
+; IF-EVL-NEXT: [[ARRAYIDX2:%.*]] = getelementptr inbounds i32, ptr [[C]], i64 [[IV]]
+; IF-EVL-NEXT: [[TMP24:%.*]] = load i32, ptr [[ARRAYIDX2]], align 4
+; IF-EVL-NEXT: [[ADD:%.*]] = add nsw i32 [[TMP24]], [[TMP23]]
+; IF-EVL-NEXT: [[ARRAYIDX4:%.*]] = getelementptr inbounds i32, ptr [[A]], i64 [[IV]]
+; IF-EVL-NEXT: store i32 [[ADD]], ptr [[ARRAYIDX4]], align 4
+; IF-EVL-NEXT: [[IV_NEXT]] = add nuw nsw i64 [[IV]], 1
+; IF-EVL-NEXT: [[EXITCOND_NOT:%.*]] = icmp eq i64 [[IV_NEXT]], [[N]]
+; IF-EVL-NEXT: br i1 [[EXITCOND_NOT]], label [[FOR_COND_CLEANUP]], label [[FOR_BODY]], !llvm.loop [[LOOP3:![0-9]+]]
+; IF-EVL: for.cond.cleanup:
+; IF-EVL-NEXT: ret void
+;
+; NO-VP-LABEL: @foo(
+; NO-VP-NEXT: entry:
+; NO-VP-NEXT: [[TMP0:%.*]] = call i64 @llvm.vscale.i64()
+; NO-VP-NEXT: [[TMP1:%.*]] = mul i64 [[TMP0]], 4
+; NO-VP-NEXT: [[MIN_ITERS_CHECK:%.*]] = icmp ult i64 [[N:%.*]], [[TMP1]]
+; NO-VP-NEXT: br i1 [[MIN_ITERS_CHECK]], label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
+; NO-VP: vector.ph:
+; NO-VP-NEXT: [[TMP2:%.*]] = call i64 @llvm.vscale.i64()
+; NO-VP-NEXT: [[TMP3:%.*]] = mul i64 [[TMP2]], 4
+; NO-VP-NEXT: [[N_MOD_VF:%.*]] = urem i64 [[N]], [[TMP3]]
+; NO-VP-NEXT: [[N_VEC:%.*]] = sub i64 [[N]], [[N_MOD_VF]]
+; NO-VP-NEXT: [[TMP4:%.*]] = call i64 @llvm.vscale.i64()
+; NO-VP-NEXT: [[TMP5:%.*]] = mul i64 [[TMP4]], 4
+; NO-VP-NEXT: br label [[VECTOR_BODY:%.*]]
+; NO-VP: vector.body:
+; NO-VP-NEXT: [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
+; NO-VP-NEXT: [[TMP6:%.*]] = add i64 [[INDEX]], 0
+; NO-VP-NEXT: [[TMP7:%.*]] = getelementptr inbounds i32, ptr [[B:%.*]], i64 [[TMP6]]
+; NO-VP-NEXT: [[TMP8:%.*]] = getelementptr inbounds i32, ptr [[TMP7]], i32 0
+; NO-VP-NEXT: [[WIDE_LOAD:%.*]] = load <vscale x 4 x i32>, ptr [[TMP8]], align 4
+; NO-VP-NEXT: [[TMP9:%.*]] = getelementptr inbounds i32, ptr [[C:%.*]], i64 [[TMP6]]
+; NO-VP-NEXT: [[TMP10:%.*]] = getelementptr inbounds i32, ptr [[TMP9]], i32 0
+; NO-VP-NEXT: [[WIDE_LOAD1:%.*]] = load <vscale x 4 x i32>, ptr [[TMP10]], align 4
+; NO-VP-NEXT: [[TMP11:%.*]] = add nsw <vscale x 4 x i32> [[WIDE_LOAD1]], [[WIDE_LOAD]]
+; NO-VP-NEXT: [[TMP12:%.*]] = getelementptr inbounds i32, ptr [[A:%.*]], i64 [[TMP6]]
+; NO-VP-NEXT: [[TMP13:%.*]] = getelementptr inbounds i32, ptr [[TMP12]], i32 0
+; NO-VP-NEXT: store <vscale x 4 x i32> [[TMP11]], ptr [[TMP13]], align 4
+; NO-VP-NEXT: [[INDEX_NEXT]] = add nuw i64 [[INDEX]], [[TMP5]]
+; NO-VP-NEXT: [[TMP14:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
+; NO-VP-NEXT: br i1 [[TMP14]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]]
+; NO-VP: middle.block:
+; NO-VP-NEXT: [[CMP_N:%.*]] = icmp eq i64 [[N]], [[N_VEC]]
+; NO-VP-NEXT: br i1 [[CMP_N]], label [[FOR_COND_CLEANUP:%.*]], label [[SCALAR_PH]]
+; NO-VP: scalar.ph:
+; NO-VP-NEXT: [[BC_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ]
+; NO-VP-NEXT: br label [[FOR_BODY:%.*]]
+; NO-VP: for.body:
+; NO-VP-NEXT: [[IV:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[IV_NEXT:%.*]], [[FOR_BODY]] ]
+; NO-VP-NEXT: [[ARRAYIDX:%.*]] = getelementptr inbounds i32, ptr [[B]], i64 [[IV]]
+; NO-VP-NEXT: [[TMP15:%.*]] = load i32, ptr [[ARRAYIDX]], align 4
+; NO-VP-NEXT: [[ARRAYIDX2:%.*]] = getelementptr inbounds i32, ptr [[C]], i64 [[IV]]
+; NO-VP-NEXT: [[TMP16:%.*]] = load i32, ptr [[ARRAYIDX2]], align 4
+; NO-VP-NEXT: [[ADD:%.*]] = add nsw i32 [[TMP16]], [[TMP15]]
+; NO-VP-NEXT: [[ARRAYIDX4:%.*]] = getelementptr inbounds i32, ptr [[A]], i64 [[IV]]
+; NO-VP-NEXT: store i32 [[ADD]], ptr [[ARRAYIDX4]], align 4
+; NO-VP-NEXT: [[IV_NEXT]] = add nuw nsw i64 [[IV]], 1
+; NO-VP-NEXT: [[EXITCOND_NOT:%.*]] = icmp eq i64 [[IV_NEXT]], [[N]]
+; NO-VP-NEXT: br i1 [[EXITCOND_NOT]], label [[FOR_COND_CLEANUP]], label [[FOR_BODY]], !llvm.loop [[LOOP3:![0-9]+]]
+; NO-VP: for.cond.cleanup:
+; NO-VP-NEXT: ret void
+;
+entry:
+ br label %for.body
+
+for.body:
+ %iv = phi i64 [ 0, %entry ], [ %iv.next, %for.body ]
+ %arrayidx = getelementptr inbounds i32, ptr %b, i64 %iv
+ %0 = load i32, ptr %arrayidx, align 4
+ %arrayidx2 = getelementptr inbounds i32, ptr %c, i64 %iv
+ %1 = load i32, ptr %arrayidx2, align 4
+ %add = add nsw i32 %1, %0
+ %arrayidx4 = getelementptr inbounds i32, ptr %a, i64 %iv
+ store i32 %add, ptr %arrayidx4, align 4
+ %iv.next = add nuw nsw i64 %iv, 1
+ %exitcond.not = icmp eq i64 %iv.next, %N
+ br i1 %exitcond.not, label %for.cond.cleanup, label %for.body
+
+for.cond.cleanup:
+ ret void
+}
diff --git a/llvm/test/Transforms/LoopVectorize/RISCV/vplan-vp-intrinsics.ll b/llvm/test/Transforms/LoopVectorize/RISCV/vplan-vp-intrinsics.ll
new file mode 100644
index 00000000000000..9f03d945aceadb
--- /dev/null
+++ b/llvm/test/Transforms/LoopVectorize/RISCV/vplan-vp-intrinsics.ll
@@ -0,0 +1,134 @@
+; REQUIRES: asserts
+
+; RUN: opt -passes=loop-vectorize -debug-only=loop-vectorize \
+; RUN: -force-tail-folding-style=data-with-evl \
+; RUN: -prefer-predicate-over-epilogue=predicate-dont-vectorize \
+; RUN: -mtriple=riscv64 -mattr=+v -riscv-v-vector-bits-max=128 -disable-output < %s 2>&1 | FileCheck --check-prefixes=IF-EVL,CHECK %s
+
+; RUN: opt -passes=loop-vectorize -debug-only=loop-vectorize \
+; RUN: -force-tail-folding-style=none \
+; RUN: -prefer-predicate-over-epilogue=predicate-dont-vectorize \
+; RUN: -mtriple=riscv64 -mattr=+v -riscv-v-vector-bits-max=128 -disable-output < %s 2>&1 | FileCheck --check-prefixes=NO-VP,CHECK %s
+
+define void @foo(ptr noalias %a, ptr noalias %b, ptr noalias %c, i64 %N) {
+; IF-EVL: VPlan 'Initial VPlan for VF={vscale x 1,vscale x 2,vscale x 4},UF={1}' {
+; IF-EVL-NEXT: Live-in vp<[[VFUF:%[0-9]+]]> = VF * UF
+; IF-EVL-NEXT: Live-in vp<[[VTC:%[0-9]+]]> = vector-trip-count
+; IF-EVL-NEXT: Live-in ir<%N> = original trip-count
+; IF-EVL-EMPTY:
+; IF-EVL: vector.ph:
+; IF-EVL-NEXT: Successor(s): vector loop
+; IF-EVL-EMPTY:
+; IF-EVL-NEXT: <x1> vector loop: {
+; IF-EVL-NEXT: vector.body:
+; IF-EVL-NEXT: EMIT vp<[[IV:%[0-9]+]]> = CANONICAL-INDUCTION
+; IF-EVL-NEXT: EXPLICIT-VECTOR-LENGTH-BASED-IV-PHI vp<[[EVL_PHI:%[0-9]+]]> = phi ir<0>, vp<[[IV_NEXT:%[0-9]+]]>
+; IF-EVL-NEXT: EMIT vp<[[EVL:%.+]]> = EXPLICIT-VECTOR-LENGTH vp<[[EVL_PHI]]>, ir<%N>
+; IF-EVL-NEXT: vp<[[ST:%[0-9]+]]> = SCALAR-STEPS vp<[[EVL_PHI]]>, ir<1>
+; IF-EVL-NEXT: CLONE ir<[[GEP1:%.+]]> = getelementptr inbounds ir<%b>, vp<[[ST]]>
+; IF-EVL-NEXT: vp<[[PTR1:%[0-9]+]]> = vector-pointer ir<[[GEP1]]>
+; IF-EVL-NEXT: WIDEN ir<[[LD1:%.+]]> = load vp<[[PTR1]]>, ir<true>
+; IF-EVL-NEXT: CLONE ir<[[GEP2:%.+]]> = getelementptr inbounds ir<%c>, vp<[[ST]]>
+; IF-EVL-NEXT: vp<[[PTR2:%[0-9]+]]> = vector-pointer ir<[[GEP2]]>
+; IF-EVL-NEXT: WIDEN ir<[[LD2:%.+]]> = load vp<[[PTR2]]>, ir<true>
+; IF-EVL-NEXT: WIDEN ir<[[ADD:%.+]]> = add nsw ir<[[LD2]]>, ir<[[LD1]]>
+; IF-EVL-NEXT: CLONE ir<[[GEP3:%.+]]> = getelementptr inbounds ir<%a>, vp<[[ST]]>
+; IF-EVL-NEXT: vp<[[PTR3:%[0-9]+]]> = vector-pointer ir<[[GEP3]]>
+; IF-EVL-NEXT: WIDEN store vp<[[PTR3]]>, ir<[[ADD]]>, ir<true>
+; IF-EVL-NEXT: SCALAR-CAST vp<[[CAST:%[0-9]+]]> = zext vp<[[EVL]]> to i64
+; IF-EVL-NEXT: EMIT vp<[[IV_NEXT]]> = EXPLICIT-VECTOR-LENGTH + vp<[[CAST]]>, vp<[[EVL_PHI]]>
+; IF-EVL-NEXT: EMIT vp<[[IV_NEXT_EXIT:%[0-9]+]]> = add vp<[[IV]]>, vp<[[VFUF]]>
+; IF-EVL-NEXT: EMIT branch-on-count vp<[[IV_NEXT_EXIT]]>, vp<[[VTC]]>
+; IF-EVL-NEXT: No successors
+; IF-EVL-NEXT: }
+
+; NO-VP: VPlan 'Initial VPlan for VF={vscale x 1,vscale x 2,vscale x 4},UF>=1' {
+; NO-VP-NEXT: Live-in vp<[[VFUF:%[0-9]+]]> = VF * UF
+; NO-VP-NEXT: Live-in vp<[[VTC:%[0-9]+]]> = vector-trip-count
+; NO-VP-NEXT: Live-in ir<%N> = original trip-count
+; NO-VP-EMPTY:
+; NO-VP: vector.ph:
+; NO-VP-NEXT: Successor(s): vector loop
+; NO-VP-EMPTY:
+; NO-VP-NEXT: <x1> vector loop: {
+; NO-VP-NEXT: vector.body:
+; NO-VP-NEXT: EMIT vp<[[IV:%[0-9]+]]> = CANONICAL-INDUCTION
+; NO-VP-NEXT: vp<[[ST:%[0-9]+]]> = SCALAR-STEPS vp<[[IV]]>, ir<1>
+; NO-VP-NEXT: CLONE ir<[[GEP1:%.+]]> = getelementptr inbounds ir<%b>, vp<[[ST]]>
+; NO-VP-NEXT: vp<[[PTR1:%[0-9]+]]> = vector-pointer ir<[[GEP1]]>
+; NO-VP-NEXT: WIDEN ir<[[LD1:%.+]]> = load vp<[[PTR1]]>
+; NO-VP-NEXT: CLONE ir<[[GEP2:%.+]]> = getelementptr inbounds ir<%c>, vp<[[ST]]>
+; NO-VP-NEXT: vp<[[PTR2:%[0-9]+]]> = vector-pointer ir<[[GEP2]]>
+; NO-VP-NEXT: WIDEN ir<[[LD2:%.+]]> = load vp<[[PTR2]]>
+; NO-VP-NEXT: WIDEN ir<[[ADD:%.+]]> = add nsw ir<[[LD2]]>, ir<[[LD1]]>
+; NO-VP-NEXT: CLONE ir<[[GEP3:%.+]]> = getelementptr inbounds ir<%a>, vp<[[ST]]>
+; NO-VP-NEXT: vp<[[PTR3:%[0-9]+]]> = vector-pointer ir<[[GEP3]]>
+; NO-VP-NEXT: WIDEN store vp<[[PTR3]]>, ir<[[ADD]]>
+; NO-VP-NEXT: EMIT vp<[[IV_NEXT:%[0-9]+]]> = add nuw vp<[[IV]]>, vp<[[VFUF]]>
+; NO-VP-NEXT: EMIT branch-on-count vp<[[IV_NEXT]]>, vp<[[VTC]]>
+; NO-VP-NEXT: No successors
+; NO-VP-NEXT: }
+
+entry:
+ br label %for.body
+
+for.body:
+ %iv = phi i64 [ 0, %entry ], [ %iv.next, %for.body ]
+ %arrayidx = getelementptr inbounds i32, ptr %b, i64 %iv
+ %0 = load i32, ptr %arrayidx, align 4
+ %arrayidx2 = getelementptr inbounds i32, ptr %c, i64 %iv
+ %1 = load i32, ptr %arrayidx2, align 4
+ %add = add nsw i32 %1, %0
+ %arrayidx4 = getelementptr inbounds i32, ptr %a, i64 %iv
+ store i32 %add, ptr %arrayidx4, align 4
+ %iv.next = add nuw nsw i64 %iv, 1
+ %exitcond.not = icmp eq i64 %iv.next, %N
+ br i1 %exitcond.not, label %for.cond.cleanup, label %for.body
+
+for.cond.cleanup:
+ ret void
+}
+
+define void @safe_dep(ptr %p) {
+; CHECK: VPlan 'Initial VPlan for VF={vscale x 1,vscale x 2},UF>=1' {
+; CHECK-NEXT: Live-in vp<[[VFUF:%[0-9]+]]> = VF * UF
+; CHECK-NEXT: Live-in vp<[[VTC:%[0-9]+]]> = vector-trip-count
+; CHECK-NEXT: Live-in ir<512> = original trip-count
+; CHECK-EMPTY:
+; CHECK: vector.ph:
+; CHECK-NEXT: Successor(s): vector loop
+; CHECK-EMPTY:
+; CHECK-NEXT: <x1> vector loop: {
+; CHECK-NEXT: vector.body:
+; CHECK-NEXT: EMIT vp<[[IV:%[0-9]+]]> = CANONICAL-INDUCTION
+; CHECK-NEXT: vp<[[ST:%[0-9]+]]> = SCALAR-STEPS vp<[[IV]]>, ir<1>
+; CHECK-NEXT: CLONE ir<[[GEP1:%.+]]> = getelementptr ir<%p>, vp<[[ST]]>
+; CHECK-NEXT: vp<[[PTR1:%[0-9]+]]> = vector-pointer ir<[[GEP1]]>
+; CHECK-NEXT: WIDEN ir<[[V:%.+]]> = load vp<[[PTR1]]>
+; CHECK-NEXT: CLONE ir<[[OFFSET:.+]]> = add vp<[[ST]]>, ir<100>
+; CHECK-NEXT: CLONE ir<[[GEP2:%.+]]> = getelementptr ir<%p>, ir<[[OFFSET]]>
+; CHECK-NEXT: vp<[[PTR2:%[0-9]+]]> = vector-pointer ir<[[GEP2]]>
+; CHECK-NEXT: WIDEN store vp<[[PTR2]]>, ir<[[V]]>
+; CHECK-NEXT: EMIT vp<[[IV_NEXT:%[0-9]+]]> = add nuw vp<[[IV]]>, vp<[[VFUF]]>
+; CHECK-NEXT: EMIT branch-on-count vp<[[IV_NEXT]]>, vp<[[VTC]]>
+; CHECK-NEXT: No successors
+; CHECK-NEXT: }
+
+entry:
+ br label %loop
+
+loop:
+ %iv = phi i64 [0, %entry], [%iv.next, %loop]
+ %a1 = getelementptr i64, ptr %p, i64 %iv
+ %v = load i64, ptr %a1, align 32
+ %offset = add i64 %iv, 100
+ %a2 = getelementptr i64, ptr %p, i64 %offset
+ store i64 %v, ptr %a2, align 32
+ %iv.next = add i64 %iv, 1
+ %cmp = icmp ne i64 %iv, 511
+ br i1 %cmp, label %loop, label %exit
+
+exit:
+ ret void
+}
+
diff --git a/llvm/test/Transforms/LoopVectorize/X86/vectorize-vp-intrinsics.ll b/llvm/test/Transforms/LoopVectorize/X86/vectorize-vp-intrinsics.ll
new file mode 100644
index 00000000000000..cc9711f869d5ef
--- /dev/null
+++ b/llvm/test/Transforms/LoopVectorize/X86/vectorize-vp-intrinsics.ll
@@ -0,0 +1,127 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
+; RUN: opt -passes=loop-vectorize \
+; RUN: -force-tail-folding-style=data-with-evl \
+; RUN: -prefer-predicate-over-epilogue=predicate-dont-vectorize \
+; RUN: -mtriple=x86_64 -mattr=+avx512f -S < %s 2>&1 | FileCheck --check-prefix=IF-EVL %s
+
+; RUN: opt -passes=loop-vectorize \
+; RUN: -force-tail-folding-style=none \
+; RUN: -prefer-predicate-over-epilogue=predicate-dont-vectorize \
+; RUN: -mtriple=x86_64 -mattr=+avx512f -S < %s 2>&1 | FileCheck --check-prefix=NO-VP %s
+
+define void @foo(ptr noalias %a, ptr noalias %b, ptr noalias %c, i64 %N) {
+; IF-EVL-LABEL: @foo(
+; IF-EVL-NEXT: entry:
+; IF-EVL-NEXT: br i1 false, label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
+; IF-EVL: vector.ph:
+; IF-EVL-NEXT: [[N_RND_UP:%.*]] = add i64 [[N:%.*]], 15
+; IF-EVL-NEXT: [[N_MOD_VF:%.*]] = urem i64 [[N_RND_UP]], 16
+; IF-EVL-NEXT: [[N_VEC:%.*]] = sub i64 [[N_RND_UP]], [[N_MOD_VF]]
+; IF-EVL-NEXT: [[TRIP_COUNT_MINUS_1:%.*]] = sub i64 [[N]], 1
+; IF-EVL-NEXT: [[BROADCAST_SPLATINSERT:%.*]] = insertelement <16 x i64> poison, i64 [[TRIP_COUNT_MINUS_1]], i64 0
+; IF-EVL-NEXT: [[BROADCAST_SPLAT:%.*]] = shufflevector <16 x i64> [[BROADCAST_SPLATINSERT]], <16 x i64> poison, <16 x i32> zeroinitializer
+; IF-EVL-NEXT: br label [[VECTOR_BODY:%.*]]
+; IF-EVL: vector.body:
+; IF-EVL-NEXT: [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
+; IF-EVL-NEXT: [[TMP0:%.*]] = add i64 [[INDEX]], 0
+; IF-EVL-NEXT: [[BROADCAST_SPLATINSERT1:%.*]] = insertelement <16 x i64> poison, i64 [[INDEX]], i64 0
+; IF-EVL-NEXT: [[BROADCAST_SPLAT2:%.*]] = shufflevector <16 x i64> [[BROADCAST_SPLATINSERT1]], <16 x i64> poison, <16 x i32> zeroinitializer
+; IF-EVL-NEXT: [[VEC_IV:%.*]] = add <16 x i64> [[BROADCAST_SPLAT2]], <i64 0, i64 1, i64 2, i64 3, i64 4, i64 5, i64 6, i64 7, i64 8, i64 9, i64 10, i64 11, i64 12, i64 13, i64 14, i64 15>
+; IF-EVL-NEXT: [[TMP1:%.*]] = icmp ule <16 x i64> [[VEC_IV]], [[BROADCAST_SPLAT]]
+; IF-EVL-NEXT: [[TMP2:%.*]] = getelementptr inbounds i32, ptr [[B:%.*]], i64 [[TMP0]]
+; IF-EVL-NEXT: [[TMP3:%.*]] = getelementptr inbounds i32, ptr [[TMP2]], i32 0
+; IF-EVL-NEXT: [[WIDE_MASKED_LOAD:%.*]] = call <16 x i32> @llvm.masked.load.v16i32.p0(ptr [[TMP3]], i32 4, <16 x i1> [[TMP1]], <16 x i32> poison)
+; IF-EVL-NEXT: [[TMP4:%.*]] = getelementptr inbounds i32, ptr [[C:%.*]], i64 [[TMP0]]
+; IF-EVL-NEXT: [[TMP5:%.*]] = getelementptr inbounds i32, ptr [[TMP4]], i32 0
+; IF-EVL-NEXT: [[WIDE_MASKED_LOAD3:%.*]] = call <16 x i32> @llvm.masked.load.v16i32.p0(ptr [[TMP5]], i32 4, <16 x i1> [[TMP1]], <16 x i32> poison)
+; IF-EVL-NEXT: [[TMP6:%.*]] = add nsw <16 x i32> [[WIDE_MASKED_LOAD3]], [[WIDE_MASKED_LOAD]]
+; IF-EVL-NEXT: [[TMP7:%.*]] = getelementptr inbounds i32, ptr [[A:%.*]], i64 [[TMP0]]
+; IF-EVL-NEXT: [[TMP8:%.*]] = getelementptr inbounds i32, ptr [[TMP7]], i32 0
+; IF-EVL-NEXT: call void @llvm.masked.store.v16i32.p0(<16 x i32> [[TMP6]], ptr [[TMP8]], i32 4, <16 x i1> [[TMP1]])
+; IF-EVL-NEXT: [[INDEX_NEXT]] = add i64 [[INDEX]], 16
+; IF-EVL-NEXT: [[TMP9:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
+; IF-EVL-NEXT: br i1 [[TMP9]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]]
+; IF-EVL: middle.block:
+; IF-EVL-NEXT: br i1 true, label [[FOR_COND_CLEANUP:%.*]], label [[SCALAR_PH]]
+; IF-EVL: scalar.ph:
+; IF-EVL-NEXT: [[BC_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ]
+; IF-EVL-NEXT: br label [[FOR_BODY:%.*]]
+; IF-EVL: for.body:
+; IF-EVL-NEXT: [[IV:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[IV_NEXT:%.*]], [[FOR_BODY]] ]
+; IF-EVL-NEXT: [[ARRAYIDX:%.*]] = getelementptr inbounds i32, ptr [[B]], i64 [[IV]]
+; IF-EVL-NEXT: [[TMP10:%.*]] = load i32, ptr [[ARRAYIDX]], align 4
+; IF-EVL-NEXT: [[ARRAYIDX2:%.*]] = getelementptr inbounds i32, ptr [[C]], i64 [[IV]]
+; IF-EVL-NEXT: [[TMP11:%.*]] = load i32, ptr [[ARRAYIDX2]], align 4
+; IF-EVL-NEXT: [[ADD:%.*]] = add nsw i32 [[TMP11]], [[TMP10]]
+; IF-EVL-NEXT: [[ARRAYIDX4:%.*]] = getelementptr inbounds i32, ptr [[A]], i64 [[IV]]
+; IF-EVL-NEXT: store i32 [[ADD]], ptr [[ARRAYIDX4]], align 4
+; IF-EVL-NEXT: [[IV_NEXT]] = add nuw nsw i64 [[IV]], 1
+; IF-EVL-NEXT: [[EXITCOND_NOT:%.*]] = icmp eq i64 [[IV_NEXT]], [[N]]
+; IF-EVL-NEXT: br i1 [[EXITCOND_NOT]], label [[FOR_COND_CLEANUP]], label [[FOR_BODY]], !llvm.loop [[LOOP3:![0-9]+]]
+; IF-EVL: for.cond.cleanup:
+; IF-EVL-NEXT: ret void
+;
+; NO-VP-LABEL: @foo(
+; NO-VP-NEXT: entry:
+; NO-VP-NEXT: [[MIN_ITERS_CHECK:%.*]] = icmp ult i64 [[N:%.*]], 16
+; NO-VP-NEXT: br i1 [[MIN_ITERS_CHECK]], label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
+; NO-VP: vector.ph:
+; NO-VP-NEXT: [[N_MOD_VF:%.*]] = urem i64 [[N]], 16
+; NO-VP-NEXT: [[N_VEC:%.*]] = sub i64 [[N]], [[N_MOD_VF]]
+; NO-VP-NEXT: br label [[VECTOR_BODY:%.*]]
+; NO-VP: vector.body:
+; NO-VP-NEXT: [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
+; NO-VP-NEXT: [[TMP0:%.*]] = add i64 [[INDEX]], 0
+; NO-VP-NEXT: [[TMP1:%.*]] = getelementptr inbounds i32, ptr [[B:%.*]], i64 [[TMP0]]
+; NO-VP-NEXT: [[TMP2:%.*]] = getelementptr inbounds i32, ptr [[TMP1]], i32 0
+; NO-VP-NEXT: [[WIDE_LOAD:%.*]] = load <16 x i32>, ptr [[TMP2]], align 4
+; NO-VP-NEXT: [[TMP3:%.*]] = getelementptr inbounds i32, ptr [[C:%.*]], i64 [[TMP0]]
+; NO-VP-NEXT: [[TMP4:%.*]] = getelementptr inbounds i32, ptr [[TMP3]], i32 0
+; NO-VP-NEXT: [[WIDE_LOAD1:%.*]] = load <16 x i32>, ptr [[TMP4]], align 4
+; NO-VP-NEXT: [[TMP5:%.*]] = add nsw <16 x i32> [[WIDE_LOAD1]], [[WIDE_LOAD]]
+; NO-VP-NEXT: [[TMP6:%.*]] = getelementptr inbounds i32, ptr [[A:%.*]], i64 [[TMP0]]
+; NO-VP-NEXT: [[TMP7:%.*]] = getelementptr inbounds i32, ptr [[TMP6]], i32 0
+; NO-VP-NEXT: store <16 x i32> [[TMP5]], ptr [[TMP7]], align 4
+; NO-VP-NEXT: [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 16
+; NO-VP-NEXT: [[TMP8:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
+; NO-VP-NEXT: br i1 [[TMP8]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]]
+; NO-VP: middle.block:
+; NO-VP-NEXT: [[CMP_N:%.*]] = icmp eq i64 [[N]], [[N_VEC]]
+; NO-VP-NEXT: br i1 [[CMP_N]], label [[FOR_COND_CLEANUP:%.*]], label [[SCALAR_PH]]
+; NO-VP: scalar.ph:
+; NO-VP-NEXT: [[BC_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ]
+; NO-VP-NEXT: br label [[FOR_BODY:%.*]]
+; NO-VP: for.body:
+; NO-VP-NEXT: [[IV:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[IV_NEXT:%.*]], [[FOR_BODY]] ]
+; NO-VP-NEXT: [[ARRAYIDX:%.*]] = getelementptr inbounds i32, ptr [[B]], i64 [[IV]]
+; NO-VP-NEXT: [[TMP9:%.*]] = load i32, ptr [[ARRAYIDX]], align 4
+; NO-VP-NEXT: [[ARRAYIDX2:%.*]] = getelementptr inbounds i32, ptr [[C]], i64 [[IV]]
+; NO-VP-NEXT: [[TMP10:%.*]] = load i32, ptr [[ARRAYIDX2]], align 4
+; NO-VP-NEXT: [[ADD:%.*]] = add nsw i32 [[TMP10]], [[TMP9]]
+; NO-VP-NEXT: [[ARRAYIDX4:%.*]] = getelementptr inbounds i32, ptr [[A]], i64 [[IV]]
+; NO-VP-NEXT: store i32 [[ADD]], ptr [[ARRAYIDX4]], align 4
+; NO-VP-NEXT: [[IV_NEXT]] = add nuw nsw i64 [[IV]], 1
+; NO-VP-NEXT: [[EXITCOND_NOT:%.*]] = icmp eq i64 [[IV_NEXT]], [[N]]
+; NO-VP-NEXT: br i1 [[EXITCOND_NOT]], label [[FOR_COND_CLEANUP]], label [[FOR_BODY]], !llvm.loop [[LOOP3:![0-9]+]]
+; NO-VP: for.cond.cleanup:
+; NO-VP-NEXT: ret void
+;
+entry:
+ br label %for.body
+
+for.body:
+ %iv = phi i64 [ 0, %entry ], [ %iv.next, %for.body ]
+ %arrayidx = getelementptr inbounds i32, ptr %b, i64 %iv
+ %0 = load i32, ptr %arrayidx, align 4
+ %arrayidx2 = getelementptr inbounds i32, ptr %c, i64 %iv
+ %1 = load i32, ptr %arrayidx2, align 4
+ %add = add nsw i32 %1, %0
+ %arrayidx4 = getelementptr inbounds i32, ptr %a, i64 %iv
+ store i32 %add, ptr %arrayidx4, align 4
+ %iv.next = add nuw nsw i64 %iv, 1
+ %exitcond.not = icmp eq i64 %iv.next, %N
+ br i1 %exitcond.not, label %for.cond.cleanup, label %for.body
+
+for.cond.cleanup:
+ ret void
+}
diff --git a/llvm/test/Transforms/LoopVectorize/X86/vplan-vp-intrinsics.ll b/llvm/test/Transforms/LoopVectorize/X86/vplan-vp-intrinsics.ll
new file mode 100644
index 00000000000000..395339faaa60d6
--- /dev/null
+++ b/llvm/test/Transforms/LoopVectorize/X86/vplan-vp-intrinsics.ll
@@ -0,0 +1,89 @@
+; REQUIRES: asserts
+
+; RUN: opt -passes=loop-vectorize -debug-only=loop-vectorize -force-vector-width=4 \
+; RUN: -force-tail-folding-style=data-with-evl \
+; RUN: -prefer-predicate-over-epilogue=predicate-dont-vectorize \
+; RUN: -mtriple=x86_64 -mattr=+avx512f -disable-output < %s 2>&1 | FileCheck --check-prefix=IF-EVL %s
+
+; RUN: opt -passes=loop-vectorize -debug-only=loop-vectorize -force-vector-width=4 \
+; RUN: -force-tail-folding-style=none \
+; RUN: -prefer-predicate-over-epilogue=predicate-dont-vectorize \
+; RUN: -mtriple=x86_64 -mattr=+avx512f -disable-output < %s 2>&1 | FileCheck --check-prefix=NO-VP %s
+
+define void @foo(ptr noalias %a, ptr noalias %b, ptr noalias %c, i64 %N) {
+; IF-EVL: VPlan 'Initial VPlan for VF={4},UF>=1' {
+; IF-EVL-NEXT: Live-in vp<[[VFUF:%[0-9]+]]> = VF * UF
+; IF-EVL-NEXT: Live-in vp<[[VTC:%[0-9]+]]> = vector-trip-count
+; IF-EVL-NEXT: Live-in vp<[[BETC:%[0-9]+]]> = backedge-taken count
+; IF-EVL-NEXT: Live-in ir<%N> = original trip-count
+; IF-EVL-EMPTY:
+; IF-EVL: vector.ph:
+; IF-EVL-NEXT: Successor(s): vector loop
+; IF-EVL-EMPTY:
+; IF-EVL-NEXT: <x1> vector loop: {
+; IF-EVL-NEXT: vector.body:
+; IF-EVL-NEXT: EMIT vp<[[IV:%[0-9]+]]> = CANONICAL-INDUCTION
+; IF-EVL-NEXT: vp<[[ST:%[0-9]+]]> = SCALAR-STEPS vp<[[IV]]>, ir<1>
+; IF-EVL-NEXT: EMIT vp<[[VIV:%[0-9]+]]> = WIDEN-CANONICAL-INDUCTION vp<[[IV]]>
+; IF-EVL-NEXT: EMIT vp<[[MASK:%[0-9]+]]> = icmp ule vp<[[VIV]]>, vp<[[BETC]]>
+; IF-EVL-NEXT: CLONE ir<[[GEP1:%.+]]> = getelementptr inbounds ir<%b>, vp<[[ST]]>
+; IF-EVL-NEXT: vp<[[PTR1:%[0-9]+]]> = vector-pointer ir<[[GEP1]]>
+; IF-EVL-NEXT: WIDEN ir<[[LD1:%.+]]> = load vp<[[PTR1]]>, vp<[[MASK]]>
+; IF-EVL-NEXT: CLONE ir<[[GEP2:%.+]]> = getelementptr inbounds ir<%c>, vp<[[ST]]>
+; IF-EVL-NEXT: vp<[[PTR2:%[0-9]+]]> = vector-pointer ir<[[GEP2]]>
+; IF-EVL-NEXT: WIDEN ir<[[LD2:%.+]]> = load vp<[[PTR2]]>, vp<[[MASK]]>
+; IF-EVL-NEXT: WIDEN ir<[[ADD:%.+]]> = add nsw ir<[[LD2]]>, ir<[[LD1]]>
+; IF-EVL-NEXT: CLONE ir<[[GEP3:%.+]]> = getelementptr inbounds ir<%a>, vp<[[ST]]>
+; IF-EVL-NEXT: vp<[[PTR3:%[0-9]+]]> = vector-pointer ir<[[GEP3]]>
+; IF-EVL-NEXT: WIDEN store vp<[[PTR3]]>, ir<[[ADD]]>, vp<[[MASK]]>
+; IF-EVL-NEXT: EMIT vp<[[IV_NEXT:%[0-9]+]]> = add vp<[[IV]]>, vp<[[VFUF]]>
+; IF-EVL-NEXT: EMIT branch-on-count vp<[[IV_NEXT]]>, vp<[[VTC]]>
+; IF-EVL-NEXT: No successors
+; IF-EVL-NEXT: }
+
+; NO-VP: VPlan 'Initial VPlan for VF={4},UF>=1' {
+; NO-VP-NEXT: Live-in vp<[[VFUF:%[0-9]+]]> = VF * UF
+; NO-VP-NEXT: Live-in vp<[[VTC:%[0-9]+]]> = vector-trip-count
+; NO-VP-NEXT: Live-in ir<%N> = original trip-count
+; NO-VP-EMPTY:
+; NO-VP: vector.ph:
+; NO-VP-NEXT: Successor(s): vector loop
+; NO-VP-EMPTY:
+; NO-VP-NEXT: <x1> vector loop: {
+; NO-VP-NEXT: vector.body:
+; NO-VP-NEXT: EMIT vp<[[IV:%[0-9]+]]> = CANONICAL-INDUCTION
+; NO-VP-NEXT: vp<[[ST:%[0-9]+]]> = SCALAR-STEPS vp<[[IV]]>, ir<1>
+; NO-VP-NEXT: CLONE ir<[[GEP1:%.+]]> = getelementptr inbounds ir<%b>, vp<[[ST]]>
+; NO-VP-NEXT: vp<[[PTR1:%[0-9]+]]> = vector-pointer ir<[[GEP1]]>
+; NO-VP-NEXT: WIDEN ir<[[LD1:%.+]]> = load vp<[[PTR1]]>
+; NO-VP-NEXT: CLONE ir<[[GEP2:%.+]]> = getelementptr inbounds ir<%c>, vp<[[ST]]>
+; NO-VP-NEXT: vp<[[PTR2:%[0-9]+]]> = vector-pointer ir<[[GEP2]]>
+; NO-VP-NEXT: WIDEN ir<[[LD2:%.+]]> = load vp<[[PTR2]]>
+; NO-VP-NEXT: WIDEN ir<[[ADD:%.+]]> = add nsw ir<[[LD2]]>, ir<[[LD1]]>
+; NO-VP-NEXT: CLONE ir<[[GEP3:%.+]]> = getelementptr inbounds ir<%a>, vp<[[ST]]>
+; NO-VP-NEXT: vp<[[PTR3:%[0-9]+]]> = vector-pointer ir<[[GEP3]]>
+; NO-VP-NEXT: WIDEN store vp<[[PTR3]]>, ir<[[ADD]]>
+; NO-VP-NEXT: EMIT vp<[[IV_NEXT:%[0-9]+]]> = add nuw vp<[[IV]]>, vp<[[VFUF]]>
+; NO-VP-NEXT: EMIT branch-on-count vp<[[IV_NEXT]]>, vp<[[VTC]]>
+; NO-VP-NEXT: No successors
+; NO-VP-NEXT: }
+
+entry:
+ br label %for.body
+
+for.body:
+ %iv = phi i64 [ 0, %entry ], [ %iv.next, %for.body ]
+ %arrayidx = getelementptr inbounds i32, ptr %b, i64 %iv
+ %0 = load i32, ptr %arrayidx, align 4
+ %arrayidx2 = getelementptr inbounds i32, ptr %c, i64 %iv
+ %1 = load i32, ptr %arrayidx2, align 4
+ %add = add nsw i32 %1, %0
+ %arrayidx4 = getelementptr inbounds i32, ptr %a, i64 %iv
+ store i32 %add, ptr %arrayidx4, align 4
+ %iv.next = add nuw nsw i64 %iv, 1
+ %exitcond.not = icmp eq i64 %iv.next, %N
+ br i1 %exitcond.not, label %for.cond.cleanup, label %for.body
+
+for.cond.cleanup:
+ ret void
+}
diff --git a/llvm/test/Transforms/LoopVectorize/vectorize-vp-intrinsics-gather-scatter.ll b/llvm/test/Transforms/LoopVectorize/vectorize-vp-intrinsics-gather-scatter.ll
new file mode 100644
index 00000000000000..98d2938523ce44
--- /dev/null
+++ b/llvm/test/Transforms/LoopVectorize/vectorize-vp-intrinsics-gather-scatter.ll
@@ -0,0 +1,48 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
+; RUN: opt -passes=loop-vectorize \
+; RUN: -force-tail-folding-style=data-with-evl \
+; RUN: -prefer-predicate-over-epilogue=predicate-dont-vectorize \
+; RUN: -force-target-supports-scalable-vectors -scalable-vectorization=on -S < %s | FileCheck %s
+
+; RUN: opt -passes=loop-vectorize \
+; RUN: -force-tail-folding-style=none \
+; RUN: -prefer-predicate-over-epilogue=predicate-dont-vectorize \
+; RUN: -force-target-supports-scalable-vectors -scalable-vectorization=on -S < %s | FileCheck %s
+
+; The target does not support predicated vectorization.
+define void @gather_scatter(ptr noalias %in, ptr noalias %out, ptr noalias %index, i64 %n) {
+; CHECK-LABEL: @gather_scatter(
+; CHECK-NEXT: entry:
+; CHECK-NEXT: br label [[FOR_BODY:%.*]]
+; CHECK: for.body:
+; CHECK-NEXT: [[INDVARS_IV:%.*]] = phi i64 [ 0, [[ENTRY:%.*]] ], [ [[INDVARS_IV_NEXT:%.*]], [[FOR_BODY]] ]
+; CHECK-NEXT: [[ARRAYIDX3:%.*]] = getelementptr inbounds i32, ptr [[INDEX:%.*]], i64 [[INDVARS_IV]]
+; CHECK-NEXT: [[TMP0:%.*]] = load i64, ptr [[ARRAYIDX3]], align 8
+; CHECK-NEXT: [[ARRAYIDX5:%.*]] = getelementptr inbounds float, ptr [[IN:%.*]], i64 [[TMP0]]
+; CHECK-NEXT: [[TMP1:%.*]] = load float, ptr [[ARRAYIDX5]], align 4
+; CHECK-NEXT: [[ARRAYIDX7:%.*]] = getelementptr inbounds float, ptr [[OUT:%.*]], i64 [[TMP0]]
+; CHECK-NEXT: store float [[TMP1]], ptr [[ARRAYIDX7]], align 4
+; CHECK-NEXT: [[INDVARS_IV_NEXT]] = add nuw nsw i64 [[INDVARS_IV]], 1
+; CHECK-NEXT: [[EXITCOND_NOT:%.*]] = icmp eq i64 [[INDVARS_IV_NEXT]], [[N:%.*]]
+; CHECK-NEXT: br i1 [[EXITCOND_NOT]], label [[FOR_END:%.*]], label [[FOR_BODY]]
+; CHECK: for.end:
+; CHECK-NEXT: ret void
+;
+entry:
+ br label %for.body
+
+for.body:
+ %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
+ %arrayidx3 = getelementptr inbounds i32, ptr %index, i64 %indvars.iv
+ %0 = load i64, ptr %arrayidx3, align 8
+ %arrayidx5 = getelementptr inbounds float, ptr %in, i64 %0
+ %1 = load float, ptr %arrayidx5, align 4
+ %arrayidx7 = getelementptr inbounds float, ptr %out, i64 %0
+ store float %1, ptr %arrayidx7, align 4
+ %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
+ %exitcond.not = icmp eq i64 %indvars.iv.next, %n
+ br i1 %exitcond.not, label %for.end, label %for.body
+
+for.end:
+ ret void
+}
diff --git a/llvm/test/Transforms/LoopVectorize/vectorize-vp-intrinsics-interleave.ll b/llvm/test/Transforms/LoopVectorize/vectorize-vp-intrinsics-interleave.ll
new file mode 100644
index 00000000000000..9d0ae16c265da4
--- /dev/null
+++ b/llvm/test/Transforms/LoopVectorize/vectorize-vp-intrinsics-interleave.ll
@@ -0,0 +1,170 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
+; RUN: opt -passes=loop-vectorize \
+; RUN: -force-tail-folding-style=data-with-evl \
+; RUN: -prefer-predicate-over-epilogue=predicate-dont-vectorize \
+; RUN: -force-target-supports-scalable-vectors -scalable-vectorization=on -S < %s | FileCheck --check-prefix=IF-EVL %s
+
+; RUN: opt -passes=loop-vectorize \
+; RUN: -force-tail-folding-style=none \
+; RUN: -prefer-predicate-over-epilogue=predicate-dont-vectorize \
+; RUN: -force-target-supports-scalable-vectors -scalable-vectorization=on -S < %s | FileCheck --check-prefix=NO-VP %s
+
+; FIXME: interleaved accesses are not supported yet with predicated vectorization.
+define void @interleave(ptr noalias %a, ptr noalias %b, ptr noalias %c, i64 %N) {
+; IF-EVL-LABEL: @interleave(
+; IF-EVL-NEXT: entry:
+; IF-EVL-NEXT: br i1 false, label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
+; IF-EVL: vector.ph:
+; IF-EVL-NEXT: [[N_RND_UP:%.*]] = add i64 [[N:%.*]], 1
+; IF-EVL-NEXT: [[N_MOD_VF:%.*]] = urem i64 [[N_RND_UP]], 2
+; IF-EVL-NEXT: [[N_VEC:%.*]] = sub i64 [[N_RND_UP]], [[N_MOD_VF]]
+; IF-EVL-NEXT: [[TRIP_COUNT_MINUS_1:%.*]] = sub i64 [[N]], 1
+; IF-EVL-NEXT: br label [[VECTOR_BODY:%.*]]
+; IF-EVL: vector.body:
+; IF-EVL-NEXT: [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[PRED_STORE_CONTINUE3:%.*]] ]
+; IF-EVL-NEXT: [[VEC_IV:%.*]] = add i64 [[INDEX]], 0
+; IF-EVL-NEXT: [[VEC_IV1:%.*]] = add i64 [[INDEX]], 1
+; IF-EVL-NEXT: [[TMP0:%.*]] = icmp ule i64 [[VEC_IV]], [[TRIP_COUNT_MINUS_1]]
+; IF-EVL-NEXT: [[TMP1:%.*]] = icmp ule i64 [[VEC_IV1]], [[TRIP_COUNT_MINUS_1]]
+; IF-EVL-NEXT: br i1 [[TMP0]], label [[PRED_STORE_IF:%.*]], label [[PRED_STORE_CONTINUE:%.*]]
+; IF-EVL: pred.store.if:
+; IF-EVL-NEXT: [[TMP2:%.*]] = add i64 [[INDEX]], 0
+; IF-EVL-NEXT: [[TMP3:%.*]] = getelementptr inbounds i32, ptr [[B:%.*]], i64 [[TMP2]]
+; IF-EVL-NEXT: [[TMP4:%.*]] = load i32, ptr [[TMP3]], align 4
+; IF-EVL-NEXT: [[TMP5:%.*]] = getelementptr inbounds i32, ptr [[C:%.*]], i64 [[TMP2]]
+; IF-EVL-NEXT: [[TMP6:%.*]] = load i32, ptr [[TMP5]], align 4
+; IF-EVL-NEXT: [[TMP7:%.*]] = getelementptr inbounds i32, ptr [[A:%.*]], i64 [[TMP2]]
+; IF-EVL-NEXT: [[TMP8:%.*]] = add nsw i32 [[TMP6]], [[TMP4]]
+; IF-EVL-NEXT: store i32 [[TMP8]], ptr [[TMP7]], align 4
+; IF-EVL-NEXT: br label [[PRED_STORE_CONTINUE]]
+; IF-EVL: pred.store.continue:
+; IF-EVL-NEXT: [[TMP9:%.*]] = phi i32 [ poison, [[VECTOR_BODY]] ], [ [[TMP4]], [[PRED_STORE_IF]] ]
+; IF-EVL-NEXT: [[TMP10:%.*]] = phi i32 [ poison, [[VECTOR_BODY]] ], [ [[TMP6]], [[PRED_STORE_IF]] ]
+; IF-EVL-NEXT: br i1 [[TMP1]], label [[PRED_STORE_IF2:%.*]], label [[PRED_STORE_CONTINUE3]]
+; IF-EVL: pred.store.if2:
+; IF-EVL-NEXT: [[TMP11:%.*]] = add i64 [[INDEX]], 1
+; IF-EVL-NEXT: [[TMP12:%.*]] = getelementptr inbounds i32, ptr [[B]], i64 [[TMP11]]
+; IF-EVL-NEXT: [[TMP13:%.*]] = load i32, ptr [[TMP12]], align 4
+; IF-EVL-NEXT: [[TMP14:%.*]] = getelementptr inbounds i32, ptr [[C]], i64 [[TMP11]]
+; IF-EVL-NEXT: [[TMP15:%.*]] = load i32, ptr [[TMP14]], align 4
+; IF-EVL-NEXT: [[TMP16:%.*]] = getelementptr inbounds i32, ptr [[A]], i64 [[TMP11]]
+; IF-EVL-NEXT: [[TMP17:%.*]] = add nsw i32 [[TMP15]], [[TMP13]]
+; IF-EVL-NEXT: store i32 [[TMP17]], ptr [[TMP16]], align 4
+; IF-EVL-NEXT: br label [[PRED_STORE_CONTINUE3]]
+; IF-EVL: pred.store.continue3:
+; IF-EVL-NEXT: [[TMP18:%.*]] = phi i32 [ poison, [[PRED_STORE_CONTINUE]] ], [ [[TMP13]], [[PRED_STORE_IF2]] ]
+; IF-EVL-NEXT: [[TMP19:%.*]] = phi i32 [ poison, [[PRED_STORE_CONTINUE]] ], [ [[TMP15]], [[PRED_STORE_IF2]] ]
+; IF-EVL-NEXT: [[INDEX_NEXT]] = add i64 [[INDEX]], 2
+; IF-EVL-NEXT: [[TMP20:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
+; IF-EVL-NEXT: br i1 [[TMP20]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]]
+; IF-EVL: middle.block:
+; IF-EVL-NEXT: br i1 true, label [[FOR_COND_CLEANUP:%.*]], label [[SCALAR_PH]]
+; IF-EVL: scalar.ph:
+; IF-EVL-NEXT: [[BC_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ]
+; IF-EVL-NEXT: br label [[FOR_BODY:%.*]]
+; IF-EVL: for.body:
+; IF-EVL-NEXT: [[IV:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[IV_NEXT:%.*]], [[FOR_BODY]] ]
+; IF-EVL-NEXT: [[ARRAYIDX:%.*]] = getelementptr inbounds i32, ptr [[B]], i64 [[IV]]
+; IF-EVL-NEXT: [[TMP21:%.*]] = load i32, ptr [[ARRAYIDX]], align 4
+; IF-EVL-NEXT: [[ARRAYIDX2:%.*]] = getelementptr inbounds i32, ptr [[C]], i64 [[IV]]
+; IF-EVL-NEXT: [[TMP22:%.*]] = load i32, ptr [[ARRAYIDX2]], align 4
+; IF-EVL-NEXT: [[ADD:%.*]] = add nsw i32 [[TMP22]], [[TMP21]]
+; IF-EVL-NEXT: [[ARRAYIDX4:%.*]] = getelementptr inbounds i32, ptr [[A]], i64 [[IV]]
+; IF-EVL-NEXT: store i32 [[ADD]], ptr [[ARRAYIDX4]], align 4
+; IF-EVL-NEXT: [[IV_NEXT]] = add nuw nsw i64 [[IV]], 1
+; IF-EVL-NEXT: [[EXITCOND_NOT:%.*]] = icmp eq i64 [[IV_NEXT]], [[N]]
+; IF-EVL-NEXT: br i1 [[EXITCOND_NOT]], label [[FOR_COND_CLEANUP]], label [[FOR_BODY]], !llvm.loop [[LOOP3:![0-9]+]]
+; IF-EVL: for.cond.cleanup:
+; IF-EVL-NEXT: ret void
+;
+; NO-VP-LABEL: @interleave(
+; NO-VP-NEXT: entry:
+; NO-VP-NEXT: [[TMP0:%.*]] = call i64 @llvm.vscale.i64()
+; NO-VP-NEXT: [[TMP1:%.*]] = mul i64 [[TMP0]], 2
+; NO-VP-NEXT: [[MIN_ITERS_CHECK:%.*]] = icmp ult i64 [[N:%.*]], [[TMP1]]
+; NO-VP-NEXT: br i1 [[MIN_ITERS_CHECK]], label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
+; NO-VP: vector.ph:
+; NO-VP-NEXT: [[TMP2:%.*]] = call i64 @llvm.vscale.i64()
+; NO-VP-NEXT: [[TMP3:%.*]] = mul i64 [[TMP2]], 2
+; NO-VP-NEXT: [[N_MOD_VF:%.*]] = urem i64 [[N]], [[TMP3]]
+; NO-VP-NEXT: [[N_VEC:%.*]] = sub i64 [[N]], [[N_MOD_VF]]
+; NO-VP-NEXT: [[TMP4:%.*]] = call i64 @llvm.vscale.i64()
+; NO-VP-NEXT: [[TMP5:%.*]] = mul i64 [[TMP4]], 2
+; NO-VP-NEXT: br label [[VECTOR_BODY:%.*]]
+; NO-VP: vector.body:
+; NO-VP-NEXT: [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
+; NO-VP-NEXT: [[TMP6:%.*]] = add i64 [[INDEX]], 0
+; NO-VP-NEXT: [[TMP7:%.*]] = call i64 @llvm.vscale.i64()
+; NO-VP-NEXT: [[TMP8:%.*]] = add i64 [[TMP7]], 0
+; NO-VP-NEXT: [[TMP9:%.*]] = mul i64 [[TMP8]], 1
+; NO-VP-NEXT: [[TMP10:%.*]] = add i64 [[INDEX]], [[TMP9]]
+; NO-VP-NEXT: [[TMP11:%.*]] = getelementptr inbounds i32, ptr [[B:%.*]], i64 [[TMP6]]
+; NO-VP-NEXT: [[TMP12:%.*]] = getelementptr inbounds i32, ptr [[B]], i64 [[TMP10]]
+; NO-VP-NEXT: [[TMP13:%.*]] = getelementptr inbounds i32, ptr [[TMP11]], i32 0
+; NO-VP-NEXT: [[TMP14:%.*]] = call i64 @llvm.vscale.i64()
+; NO-VP-NEXT: [[TMP15:%.*]] = getelementptr inbounds i32, ptr [[TMP11]], i64 [[TMP14]]
+; NO-VP-NEXT: [[WIDE_LOAD:%.*]] = load <vscale x 1 x i32>, ptr [[TMP13]], align 4
+; NO-VP-NEXT: [[WIDE_LOAD1:%.*]] = load <vscale x 1 x i32>, ptr [[TMP15]], align 4
+; NO-VP-NEXT: [[TMP16:%.*]] = getelementptr inbounds i32, ptr [[C:%.*]], i64 [[TMP6]]
+; NO-VP-NEXT: [[TMP17:%.*]] = getelementptr inbounds i32, ptr [[C]], i64 [[TMP10]]
+; NO-VP-NEXT: [[TMP18:%.*]] = getelementptr inbounds i32, ptr [[TMP16]], i32 0
+; NO-VP-NEXT: [[TMP19:%.*]] = call i64 @llvm.vscale.i64()
+; NO-VP-NEXT: [[TMP20:%.*]] = getelementptr inbounds i32, ptr [[TMP16]], i64 [[TMP19]]
+; NO-VP-NEXT: [[WIDE_LOAD2:%.*]] = load <vscale x 1 x i32>, ptr [[TMP18]], align 4
+; NO-VP-NEXT: [[WIDE_LOAD3:%.*]] = load <vscale x 1 x i32>, ptr [[TMP20]], align 4
+; NO-VP-NEXT: [[TMP21:%.*]] = add nsw <vscale x 1 x i32> [[WIDE_LOAD2]], [[WIDE_LOAD]]
+; NO-VP-NEXT: [[TMP22:%.*]] = add nsw <vscale x 1 x i32> [[WIDE_LOAD3]], [[WIDE_LOAD1]]
+; NO-VP-NEXT: [[TMP23:%.*]] = getelementptr inbounds i32, ptr [[A:%.*]], i64 [[TMP6]]
+; NO-VP-NEXT: [[TMP24:%.*]] = getelementptr inbounds i32, ptr [[A]], i64 [[TMP10]]
+; NO-VP-NEXT: [[TMP25:%.*]] = getelementptr inbounds i32, ptr [[TMP23]], i32 0
+; NO-VP-NEXT: [[TMP26:%.*]] = call i64 @llvm.vscale.i64()
+; NO-VP-NEXT: [[TMP27:%.*]] = getelementptr inbounds i32, ptr [[TMP23]], i64 [[TMP26]]
+; NO-VP-NEXT: store <vscale x 1 x i32> [[TMP21]], ptr [[TMP25]], align 4
+; NO-VP-NEXT: store <vscale x 1 x i32> [[TMP22]], ptr [[TMP27]], align 4
+; NO-VP-NEXT: [[INDEX_NEXT]] = add nuw i64 [[INDEX]], [[TMP5]]
+; NO-VP-NEXT: [[TMP28:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
+; NO-VP-NEXT: br i1 [[TMP28]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]]
+; NO-VP: middle.block:
+; NO-VP-NEXT: [[CMP_N:%.*]] = icmp eq i64 [[N]], [[N_VEC]]
+; NO-VP-NEXT: br i1 [[CMP_N]], label [[FOR_COND_CLEANUP:%.*]], label [[SCALAR_PH]]
+; NO-VP: scalar.ph:
+; NO-VP-NEXT: [[BC_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ]
+; NO-VP-NEXT: br label [[FOR_BODY:%.*]]
+; NO-VP: for.body:
+; NO-VP-NEXT: [[IV:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[IV_NEXT:%.*]], [[FOR_BODY]] ]
+; NO-VP-NEXT: [[ARRAYIDX:%.*]] = getelementptr inbounds i32, ptr [[B]], i64 [[IV]]
+; NO-VP-NEXT: [[TMP29:%.*]] = load i32, ptr [[ARRAYIDX]], align 4
+; NO-VP-NEXT: [[ARRAYIDX2:%.*]] = getelementptr inbounds i32, ptr [[C]], i64 [[IV]]
+; NO-VP-NEXT: [[TMP30:%.*]] = load i32, ptr [[ARRAYIDX2]], align 4
+; NO-VP-NEXT: [[ADD:%.*]] = add nsw i32 [[TMP30]], [[TMP29]]
+; NO-VP-NEXT: [[ARRAYIDX4:%.*]] = getelementptr inbounds i32, ptr [[A]], i64 [[IV]]
+; NO-VP-NEXT: store i32 [[ADD]], ptr [[ARRAYIDX4]], align 4
+; NO-VP-NEXT: [[IV_NEXT]] = add nuw nsw i64 [[IV]], 1
+; NO-VP-NEXT: [[EXITCOND_NOT:%.*]] = icmp eq i64 [[IV_NEXT]], [[N]]
+; NO-VP-NEXT: br i1 [[EXITCOND_NOT]], label [[FOR_COND_CLEANUP]], label [[FOR_BODY]], !llvm.loop [[LOOP3:![0-9]+]]
+; NO-VP: for.cond.cleanup:
+; NO-VP-NEXT: ret void
+;
+entry:
+ br label %for.body
+
+for.body:
+ %iv = phi i64 [ 0, %entry ], [ %iv.next, %for.body ]
+ %arrayidx = getelementptr inbounds i32, ptr %b, i64 %iv
+ %0 = load i32, ptr %arrayidx, align 4
+ %arrayidx2 = getelementptr inbounds i32, ptr %c, i64 %iv
+ %1 = load i32, ptr %arrayidx2, align 4
+ %add = add nsw i32 %1, %0
+ %arrayidx4 = getelementptr inbounds i32, ptr %a, i64 %iv
+ store i32 %add, ptr %arrayidx4, align 4
+ %iv.next = add nuw nsw i64 %iv, 1
+ %exitcond.not = icmp eq i64 %iv.next, %N
+ br i1 %exitcond.not, label %for.cond.cleanup, label %for.body, !llvm.loop !0
+
+for.cond.cleanup:
+ ret void
+}
+
+!0 = distinct !{!0, !1, !2}
+!1 = !{!"llvm.loop.interleave.count", i32 2}
+!2 = !{!"llvm.loop.vectorize.enable", i1 true}
diff --git a/llvm/test/Transforms/LoopVectorize/vectorize-vp-intrinsics-iv32.ll b/llvm/test/Transforms/LoopVectorize/vectorize-vp-intrinsics-iv32.ll
new file mode 100644
index 00000000000000..5df0229ded2837
--- /dev/null
+++ b/llvm/test/Transforms/LoopVectorize/vectorize-vp-intrinsics-iv32.ll
@@ -0,0 +1,85 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
+; RUN: opt -passes=loop-vectorize \
+; RUN: -force-tail-folding-style=data-with-evl \
+; RUN: -prefer-predicate-over-epilogue=predicate-dont-vectorize \
+; RUN: -force-target-supports-scalable-vectors -scalable-vectorization=on -S < %s | FileCheck --check-prefix=IF-EVL %s
+
+; RUN: opt -passes=loop-vectorize \
+; RUN: -force-tail-folding-style=none \
+; RUN: -prefer-predicate-over-epilogue=predicate-dont-vectorize \
+; RUN: -force-target-supports-scalable-vectors -scalable-vectorization=on -S < %s | FileCheck --check-prefix=NO-VP %s
+
+; The target does not support predicated vectorization.
+define void @iv32(ptr noalias %a, ptr noalias %b, i32 %N) {
+; IF-EVL-LABEL: @iv32(
+; IF-EVL-NEXT: entry:
+; IF-EVL-NEXT: br label [[FOR_BODY:%.*]]
+; IF-EVL: for.body:
+; IF-EVL-NEXT: [[IV:%.*]] = phi i32 [ 0, [[ENTRY:%.*]] ], [ [[IV_NEXT:%.*]], [[FOR_BODY]] ]
+; IF-EVL-NEXT: [[ARRAYIDX:%.*]] = getelementptr inbounds i32, ptr [[B:%.*]], i32 [[IV]]
+; IF-EVL-NEXT: [[TMP0:%.*]] = load i32, ptr [[ARRAYIDX]], align 4
+; IF-EVL-NEXT: [[ARRAYIDX4:%.*]] = getelementptr inbounds i32, ptr [[A:%.*]], i32 [[IV]]
+; IF-EVL-NEXT: store i32 [[TMP0]], ptr [[ARRAYIDX4]], align 4
+; IF-EVL-NEXT: [[IV_NEXT]] = add nuw nsw i32 [[IV]], 1
+; IF-EVL-NEXT: [[EXITCOND_NOT:%.*]] = icmp eq i32 [[IV_NEXT]], [[N:%.*]]
+; IF-EVL-NEXT: br i1 [[EXITCOND_NOT]], label [[FOR_COND_CLEANUP:%.*]], label [[FOR_BODY]]
+; IF-EVL: for.cond.cleanup:
+; IF-EVL-NEXT: ret void
+;
+; NO-VP-LABEL: @iv32(
+; NO-VP-NEXT: entry:
+; NO-VP-NEXT: [[TMP0:%.*]] = call i32 @llvm.vscale.i32()
+; NO-VP-NEXT: [[MIN_ITERS_CHECK:%.*]] = icmp ult i32 [[N:%.*]], [[TMP0]]
+; NO-VP-NEXT: br i1 [[MIN_ITERS_CHECK]], label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
+; NO-VP: vector.ph:
+; NO-VP-NEXT: [[TMP1:%.*]] = call i32 @llvm.vscale.i32()
+; NO-VP-NEXT: [[N_MOD_VF:%.*]] = urem i32 [[N]], [[TMP1]]
+; NO-VP-NEXT: [[N_VEC:%.*]] = sub i32 [[N]], [[N_MOD_VF]]
+; NO-VP-NEXT: [[TMP2:%.*]] = call i32 @llvm.vscale.i32()
+; NO-VP-NEXT: br label [[VECTOR_BODY:%.*]]
+; NO-VP: vector.body:
+; NO-VP-NEXT: [[INDEX:%.*]] = phi i32 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
+; NO-VP-NEXT: [[TMP3:%.*]] = add i32 [[INDEX]], 0
+; NO-VP-NEXT: [[TMP4:%.*]] = getelementptr inbounds i32, ptr [[B:%.*]], i32 [[TMP3]]
+; NO-VP-NEXT: [[TMP5:%.*]] = getelementptr inbounds i32, ptr [[TMP4]], i32 0
+; NO-VP-NEXT: [[WIDE_LOAD:%.*]] = load <vscale x 1 x i32>, ptr [[TMP5]], align 4
+; NO-VP-NEXT: [[TMP6:%.*]] = getelementptr inbounds i32, ptr [[A:%.*]], i32 [[TMP3]]
+; NO-VP-NEXT: [[TMP7:%.*]] = getelementptr inbounds i32, ptr [[TMP6]], i32 0
+; NO-VP-NEXT: store <vscale x 1 x i32> [[WIDE_LOAD]], ptr [[TMP7]], align 4
+; NO-VP-NEXT: [[INDEX_NEXT]] = add nuw i32 [[INDEX]], [[TMP2]]
+; NO-VP-NEXT: [[TMP8:%.*]] = icmp eq i32 [[INDEX_NEXT]], [[N_VEC]]
+; NO-VP-NEXT: br i1 [[TMP8]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]]
+; NO-VP: middle.block:
+; NO-VP-NEXT: [[CMP_N:%.*]] = icmp eq i32 [[N]], [[N_VEC]]
+; NO-VP-NEXT: br i1 [[CMP_N]], label [[FOR_COND_CLEANUP:%.*]], label [[SCALAR_PH]]
+; NO-VP: scalar.ph:
+; NO-VP-NEXT: [[BC_RESUME_VAL:%.*]] = phi i32 [ [[N_VEC]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ]
+; NO-VP-NEXT: br label [[FOR_BODY:%.*]]
+; NO-VP: for.body:
+; NO-VP-NEXT: [[IV:%.*]] = phi i32 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[IV_NEXT:%.*]], [[FOR_BODY]] ]
+; NO-VP-NEXT: [[ARRAYIDX:%.*]] = getelementptr inbounds i32, ptr [[B]], i32 [[IV]]
+; NO-VP-NEXT: [[TMP9:%.*]] = load i32, ptr [[ARRAYIDX]], align 4
+; NO-VP-NEXT: [[ARRAYIDX4:%.*]] = getelementptr inbounds i32, ptr [[A]], i32 [[IV]]
+; NO-VP-NEXT: store i32 [[TMP9]], ptr [[ARRAYIDX4]], align 4
+; NO-VP-NEXT: [[IV_NEXT]] = add nuw nsw i32 [[IV]], 1
+; NO-VP-NEXT: [[EXITCOND_NOT:%.*]] = icmp eq i32 [[IV_NEXT]], [[N]]
+; NO-VP-NEXT: br i1 [[EXITCOND_NOT]], label [[FOR_COND_CLEANUP]], label [[FOR_BODY]], !llvm.loop [[LOOP3:![0-9]+]]
+; NO-VP: for.cond.cleanup:
+; NO-VP-NEXT: ret void
+;
+entry:
+ br label %for.body
+
+for.body:
+ %iv = phi i32 [ 0, %entry ], [ %iv.next, %for.body ]
+ %arrayidx = getelementptr inbounds i32, ptr %b, i32 %iv
+ %0 = load i32, ptr %arrayidx, align 4
+ %arrayidx4 = getelementptr inbounds i32, ptr %a, i32 %iv
+ store i32 %0, ptr %arrayidx4, align 4
+ %iv.next = add nuw nsw i32 %iv, 1
+ %exitcond.not = icmp eq i32 %iv.next, %N
+ br i1 %exitcond.not, label %for.cond.cleanup, label %for.body
+
+for.cond.cleanup:
+ ret void
+}
diff --git a/llvm/test/Transforms/LoopVectorize/vectorize-vp-intrinsics-masked-loadstore.ll b/llvm/test/Transforms/LoopVectorize/vectorize-vp-intrinsics-masked-loadstore.ll
new file mode 100644
index 00000000000000..efa45a3374681b
--- /dev/null
+++ b/llvm/test/Transforms/LoopVectorize/vectorize-vp-intrinsics-masked-loadstore.ll
@@ -0,0 +1,60 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
+; RUN: opt -passes=loop-vectorize \
+; RUN: -force-tail-folding-style=data-with-evl \
+; RUN: -prefer-predicate-over-epilogue=predicate-dont-vectorize \
+; RUN: -force-target-supports-scalable-vectors -scalable-vectorization=on -S < %s | FileCheck %s
+
+; RUN: opt -passes=loop-vectorize \
+; RUN: -force-tail-folding-style=none \
+; RUN: -prefer-predicate-over-epilogue=predicate-dont-vectorize \
+; RUN: -force-target-supports-scalable-vectors -scalable-vectorization=on -S < %s | FileCheck %s
+
+; The target does not support predicated vectorization.
+define void @masked_loadstore(ptr noalias %a, ptr noalias %b, i64 %n) {
+; CHECK-LABEL: @masked_loadstore(
+; CHECK-NEXT: entry:
+; CHECK-NEXT: br label [[FOR_BODY:%.*]]
+; CHECK: for.body:
+; CHECK-NEXT: [[I_011:%.*]] = phi i64 [ [[INC:%.*]], [[FOR_INC:%.*]] ], [ 0, [[ENTRY:%.*]] ]
+; CHECK-NEXT: [[ARRAYIDX:%.*]] = getelementptr inbounds i32, ptr [[B:%.*]], i64 [[I_011]]
+; CHECK-NEXT: [[TMP0:%.*]] = load i32, ptr [[ARRAYIDX]], align 4
+; CHECK-NEXT: [[CMP1:%.*]] = icmp ne i32 [[TMP0]], 0
+; CHECK-NEXT: br i1 [[CMP1]], label [[IF_THEN:%.*]], label [[FOR_INC]]
+; CHECK: if.then:
+; CHECK-NEXT: [[ARRAYIDX3:%.*]] = getelementptr inbounds i32, ptr [[A:%.*]], i64 [[I_011]]
+; CHECK-NEXT: [[TMP1:%.*]] = load i32, ptr [[ARRAYIDX3]], align 4
+; CHECK-NEXT: [[ADD:%.*]] = add i32 [[TMP0]], [[TMP1]]
+; CHECK-NEXT: store i32 [[ADD]], ptr [[ARRAYIDX3]], align 4
+; CHECK-NEXT: br label [[FOR_INC]]
+; CHECK: for.inc:
+; CHECK-NEXT: [[INC]] = add nuw nsw i64 [[I_011]], 1
+; CHECK-NEXT: [[EXITCOND_NOT:%.*]] = icmp eq i64 [[INC]], [[N:%.*]]
+; CHECK-NEXT: br i1 [[EXITCOND_NOT]], label [[EXIT:%.*]], label [[FOR_BODY]]
+; CHECK: exit:
+; CHECK-NEXT: ret void
+;
+entry:
+ br label %for.body
+
+for.body:
+ %i.011 = phi i64 [ %inc, %for.inc ], [ 0, %entry ]
+ %arrayidx = getelementptr inbounds i32, ptr %b, i64 %i.011
+ %0 = load i32, ptr %arrayidx, align 4
+ %cmp1 = icmp ne i32 %0, 0
+ br i1 %cmp1, label %if.then, label %for.inc
+
+if.then:
+ %arrayidx3 = getelementptr inbounds i32, ptr %a, i64 %i.011
+ %1 = load i32, ptr %arrayidx3, align 4
+ %add = add i32 %0, %1
+ store i32 %add, ptr %arrayidx3, align 4
+ br label %for.inc
+
+for.inc:
+ %inc = add nuw nsw i64 %i.011, 1
+ %exitcond.not = icmp eq i64 %inc, %n
+ br i1 %exitcond.not, label %exit, label %for.body
+
+exit:
+ ret void
+}
diff --git a/llvm/test/Transforms/LoopVectorize/vectorize-vp-intrinsics-no-masking.ll b/llvm/test/Transforms/LoopVectorize/vectorize-vp-intrinsics-no-masking.ll
new file mode 100644
index 00000000000000..ed736602565e89
--- /dev/null
+++ b/llvm/test/Transforms/LoopVectorize/vectorize-vp-intrinsics-no-masking.ll
@@ -0,0 +1,36 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
+; RUN: opt -passes=loop-vectorize \
+; RUN: -force-tail-folding-style=data-with-evl \
+; RUN: -prefer-predicate-over-epilogue=predicate-dont-vectorize \
+; RUN: -force-target-supports-scalable-vectors -scalable-vectorization=on -S < %s | FileCheck %s
+
+; RUN: opt -passes=loop-vectorize \
+; RUN: -force-tail-folding-style=none \
+; RUN: -prefer-predicate-over-epilogue=predicate-dont-vectorize \
+; RUN: -force-target-supports-scalable-vectors -scalable-vectorization=on -S < %s | FileCheck %s
+
+; No need to emit predicated vector code if the vector instructions with masking are not required.
+define i32 @no_masking() {
+; CHECK-LABEL: @no_masking(
+; CHECK-NEXT: entry:
+; CHECK-NEXT: br label [[BODY:%.*]]
+; CHECK: body:
+; CHECK-NEXT: [[P:%.*]] = phi i32 [ 1, [[ENTRY:%.*]] ], [ [[INC:%.*]], [[BODY]] ]
+; CHECK-NEXT: [[INC]] = add i32 [[P]], 1
+; CHECK-NEXT: [[CMP:%.*]] = icmp eq i32 [[INC]], 0
+; CHECK-NEXT: br i1 [[CMP]], label [[END:%.*]], label [[BODY]]
+; CHECK: end:
+; CHECK-NEXT: ret i32 0
+;
+entry:
+ br label %body
+
+body:
+ %p = phi i32 [ 1, %entry ], [ %inc, %body ]
+ %inc = add i32 %p, 1
+ %cmp = icmp eq i32 %inc, 0
+ br i1 %cmp, label %end, label %body
+
+end:
+ ret i32 0
+}
diff --git a/llvm/test/Transforms/LoopVectorize/vectorize-vp-intrinsics-reverse-load-store.ll b/llvm/test/Transforms/LoopVectorize/vectorize-vp-intrinsics-reverse-load-store.ll
new file mode 100644
index 00000000000000..f455cf633ff7ce
--- /dev/null
+++ b/llvm/test/Transforms/LoopVectorize/vectorize-vp-intrinsics-reverse-load-store.ll
@@ -0,0 +1,49 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
+; RUN: opt -passes=loop-vectorize \
+; RUN: -force-tail-folding-style=data-with-evl \
+; RUN: -prefer-predicate-over-epilogue=predicate-dont-vectorize \
+; RUN: -force-target-supports-scalable-vectors -scalable-vectorization=on -S < %s | FileCheck %s
+
+; RUN: opt -passes=loop-vectorize \
+; RUN: -force-tail-folding-style=none \
+; RUN: -prefer-predicate-over-epilogue=predicate-dont-vectorize \
+; RUN: -force-target-supports-scalable-vectors -scalable-vectorization=on -S < %s | FileCheck %s
+
+; FIXME: reversed loads/stores are not supported yet with predicated vectorization.
+
+define void @reverse_load_store(i64 %startval, ptr noalias %ptr, ptr noalias %ptr2) {
+; CHECK-LABEL: @reverse_load_store(
+; CHECK-NEXT: entry:
+; CHECK-NEXT: br label [[FOR_BODY:%.*]]
+; CHECK: for.body:
+; CHECK-NEXT: [[ADD_PHI:%.*]] = phi i64 [ [[STARTVAL:%.*]], [[ENTRY:%.*]] ], [ [[ADD:%.*]], [[FOR_BODY]] ]
+; CHECK-NEXT: [[I:%.*]] = phi i32 [ 0, [[ENTRY]] ], [ [[INC:%.*]], [[FOR_BODY]] ]
+; CHECK-NEXT: [[ADD]] = add i64 [[ADD_PHI]], -1
+; CHECK-NEXT: [[GEPL:%.*]] = getelementptr inbounds i32, ptr [[PTR:%.*]], i64 [[ADD]]
+; CHECK-NEXT: [[TMP:%.*]] = load i32, ptr [[GEPL]], align 4
+; CHECK-NEXT: [[GEPS:%.*]] = getelementptr inbounds i32, ptr [[PTR2:%.*]], i64 [[ADD]]
+; CHECK-NEXT: store i32 [[TMP]], ptr [[GEPS]], align 4
+; CHECK-NEXT: [[INC]] = add i32 [[I]], 1
+; CHECK-NEXT: [[EXITCOND:%.*]] = icmp ne i32 [[INC]], 1024
+; CHECK-NEXT: br i1 [[EXITCOND]], label [[FOR_BODY]], label [[LOOPEND:%.*]]
+; CHECK: loopend:
+; CHECK-NEXT: ret void
+;
+entry:
+ br label %for.body
+
+for.body:
+ %add.phi = phi i64 [ %startval, %entry ], [ %add, %for.body ]
+ %i = phi i32 [ 0, %entry ], [ %inc, %for.body ]
+ %add = add i64 %add.phi, -1
+ %gepl = getelementptr inbounds i32, ptr %ptr, i64 %add
+ %tmp = load i32, ptr %gepl, align 4
+ %geps = getelementptr inbounds i32, ptr %ptr2, i64 %add
+ store i32 %tmp, ptr %geps, align 4
+ %inc = add i32 %i, 1
+ %exitcond = icmp ne i32 %inc, 1024
+ br i1 %exitcond, label %for.body, label %loopend
+
+loopend:
+ ret void
+}
diff --git a/llvm/test/Transforms/LoopVectorize/vectorize-vp-intrinsics.ll b/llvm/test/Transforms/LoopVectorize/vectorize-vp-intrinsics.ll
new file mode 100644
index 00000000000000..a3625658fcd69c
--- /dev/null
+++ b/llvm/test/Transforms/LoopVectorize/vectorize-vp-intrinsics.ll
@@ -0,0 +1,98 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
+; RUN: opt -passes=loop-vectorize \
+; RUN: -force-tail-folding-style=data-with-evl \
+; RUN: -prefer-predicate-over-epilogue=predicate-dont-vectorize \
+; RUN: -force-target-supports-scalable-vectors -scalable-vectorization=on -S < %s | FileCheck --check-prefix=IF-EVL %s
+
+; RUN: opt -passes=loop-vectorize \
+; RUN: -force-tail-folding-style=none \
+; RUN: -prefer-predicate-over-epilogue=predicate-dont-vectorize \
+; RUN: -force-target-supports-scalable-vectors -scalable-vectorization=on -S < %s | FileCheck --check-prefix=NO-VP %s
+
+; The target does not support predicated vectorization.
+define void @foo(ptr noalias %a, ptr noalias %b, ptr noalias %c, i64 %N) {
+; IF-EVL-LABEL: @foo(
+; IF-EVL-NEXT: entry:
+; IF-EVL-NEXT: br label [[FOR_BODY:%.*]]
+; IF-EVL: for.body:
+; IF-EVL-NEXT: [[IV:%.*]] = phi i64 [ 0, [[ENTRY:%.*]] ], [ [[IV_NEXT:%.*]], [[FOR_BODY]] ]
+; IF-EVL-NEXT: [[ARRAYIDX:%.*]] = getelementptr inbounds i32, ptr [[B:%.*]], i64 [[IV]]
+; IF-EVL-NEXT: [[TMP0:%.*]] = load i32, ptr [[ARRAYIDX]], align 4
+; IF-EVL-NEXT: [[ARRAYIDX2:%.*]] = getelementptr inbounds i32, ptr [[C:%.*]], i64 [[IV]]
+; IF-EVL-NEXT: [[TMP1:%.*]] = load i32, ptr [[ARRAYIDX2]], align 4
+; IF-EVL-NEXT: [[ADD:%.*]] = add nsw i32 [[TMP1]], [[TMP0]]
+; IF-EVL-NEXT: [[ARRAYIDX4:%.*]] = getelementptr inbounds i32, ptr [[A:%.*]], i64 [[IV]]
+; IF-EVL-NEXT: store i32 [[ADD]], ptr [[ARRAYIDX4]], align 4
+; IF-EVL-NEXT: [[IV_NEXT]] = add nuw nsw i64 [[IV]], 1
+; IF-EVL-NEXT: [[EXITCOND_NOT:%.*]] = icmp eq i64 [[IV_NEXT]], [[N:%.*]]
+; IF-EVL-NEXT: br i1 [[EXITCOND_NOT]], label [[FOR_COND_CLEANUP:%.*]], label [[FOR_BODY]]
+; IF-EVL: for.cond.cleanup:
+; IF-EVL-NEXT: ret void
+;
+; NO-VP-LABEL: @foo(
+; NO-VP-NEXT: entry:
+; NO-VP-NEXT: [[TMP0:%.*]] = call i64 @llvm.vscale.i64()
+; NO-VP-NEXT: [[MIN_ITERS_CHECK:%.*]] = icmp ult i64 [[N:%.*]], [[TMP0]]
+; NO-VP-NEXT: br i1 [[MIN_ITERS_CHECK]], label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
+; NO-VP: vector.ph:
+; NO-VP-NEXT: [[TMP1:%.*]] = call i64 @llvm.vscale.i64()
+; NO-VP-NEXT: [[N_MOD_VF:%.*]] = urem i64 [[N]], [[TMP1]]
+; NO-VP-NEXT: [[N_VEC:%.*]] = sub i64 [[N]], [[N_MOD_VF]]
+; NO-VP-NEXT: [[TMP2:%.*]] = call i64 @llvm.vscale.i64()
+; NO-VP-NEXT: br label [[VECTOR_BODY:%.*]]
+; NO-VP: vector.body:
+; NO-VP-NEXT: [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
+; NO-VP-NEXT: [[TMP3:%.*]] = add i64 [[INDEX]], 0
+; NO-VP-NEXT: [[TMP4:%.*]] = getelementptr inbounds i32, ptr [[B:%.*]], i64 [[TMP3]]
+; NO-VP-NEXT: [[TMP5:%.*]] = getelementptr inbounds i32, ptr [[TMP4]], i32 0
+; NO-VP-NEXT: [[WIDE_LOAD:%.*]] = load <vscale x 1 x i32>, ptr [[TMP5]], align 4
+; NO-VP-NEXT: [[TMP6:%.*]] = getelementptr inbounds i32, ptr [[C:%.*]], i64 [[TMP3]]
+; NO-VP-NEXT: [[TMP7:%.*]] = getelementptr inbounds i32, ptr [[TMP6]], i32 0
+; NO-VP-NEXT: [[WIDE_LOAD1:%.*]] = load <vscale x 1 x i32>, ptr [[TMP7]], align 4
+; NO-VP-NEXT: [[TMP8:%.*]] = add nsw <vscale x 1 x i32> [[WIDE_LOAD1]], [[WIDE_LOAD]]
+; NO-VP-NEXT: [[TMP9:%.*]] = getelementptr inbounds i32, ptr [[A:%.*]], i64 [[TMP3]]
+; NO-VP-NEXT: [[TMP10:%.*]] = getelementptr inbounds i32, ptr [[TMP9]], i32 0
+; NO-VP-NEXT: store <vscale x 1 x i32> [[TMP8]], ptr [[TMP10]], align 4
+; NO-VP-NEXT: [[INDEX_NEXT]] = add nuw i64 [[INDEX]], [[TMP2]]
+; NO-VP-NEXT: [[TMP11:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
+; NO-VP-NEXT: br i1 [[TMP11]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]]
+; NO-VP: middle.block:
+; NO-VP-NEXT: [[CMP_N:%.*]] = icmp eq i64 [[N]], [[N_VEC]]
+; NO-VP-NEXT: br i1 [[CMP_N]], label [[FOR_COND_CLEANUP:%.*]], label [[SCALAR_PH]]
+; NO-VP: scalar.ph:
+; NO-VP-NEXT: [[BC_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ]
+; NO-VP-NEXT: br label [[FOR_BODY:%.*]]
+; NO-VP: for.body:
+; NO-VP-NEXT: [[IV:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[IV_NEXT:%.*]], [[FOR_BODY]] ]
+; NO-VP-NEXT: [[ARRAYIDX:%.*]] = getelementptr inbounds i32, ptr [[B]], i64 [[IV]]
+; NO-VP-NEXT: [[TMP12:%.*]] = load i32, ptr [[ARRAYIDX]], align 4
+; NO-VP-NEXT: [[ARRAYIDX2:%.*]] = getelementptr inbounds i32, ptr [[C]], i64 [[IV]]
+; NO-VP-NEXT: [[TMP13:%.*]] = load i32, ptr [[ARRAYIDX2]], align 4
+; NO-VP-NEXT: [[ADD:%.*]] = add nsw i32 [[TMP13]], [[TMP12]]
+; NO-VP-NEXT: [[ARRAYIDX4:%.*]] = getelementptr inbounds i32, ptr [[A]], i64 [[IV]]
+; NO-VP-NEXT: store i32 [[ADD]], ptr [[ARRAYIDX4]], align 4
+; NO-VP-NEXT: [[IV_NEXT]] = add nuw nsw i64 [[IV]], 1
+; NO-VP-NEXT: [[EXITCOND_NOT:%.*]] = icmp eq i64 [[IV_NEXT]], [[N]]
+; NO-VP-NEXT: br i1 [[EXITCOND_NOT]], label [[FOR_COND_CLEANUP]], label [[FOR_BODY]], !llvm.loop [[LOOP3:![0-9]+]]
+; NO-VP: for.cond.cleanup:
+; NO-VP-NEXT: ret void
+;
+entry:
+ br label %for.body
+
+for.body:
+ %iv = phi i64 [ 0, %entry ], [ %iv.next, %for.body ]
+ %arrayidx = getelementptr inbounds i32, ptr %b, i64 %iv
+ %0 = load i32, ptr %arrayidx, align 4
+ %arrayidx2 = getelementptr inbounds i32, ptr %c, i64 %iv
+ %1 = load i32, ptr %arrayidx2, align 4
+ %add = add nsw i32 %1, %0
+ %arrayidx4 = getelementptr inbounds i32, ptr %a, i64 %iv
+ store i32 %add, ptr %arrayidx4, align 4
+ %iv.next = add nuw nsw i64 %iv, 1
+ %exitcond.not = icmp eq i64 %iv.next, %N
+ br i1 %exitcond.not, label %for.cond.cleanup, label %for.body
+
+for.cond.cleanup:
+ ret void
+}
diff --git a/llvm/test/Transforms/LoopVectorize/vplan-vp-intrinsics.ll b/llvm/test/Transforms/LoopVectorize/vplan-vp-intrinsics.ll
new file mode 100644
index 00000000000000..30e0f055acfc30
--- /dev/null
+++ b/llvm/test/Transforms/LoopVectorize/vplan-vp-intrinsics.ll
@@ -0,0 +1,37 @@
+; REQUIRES: asserts
+
+; RUN: opt -passes=loop-vectorize -debug-only=loop-vectorize \
+; RUN: -force-tail-folding-style=data-with-evl \
+; RUN: -force-target-supports-scalable-vectors -scalable-vectorization=on \
+; RUN: -disable-output < %s 2>&1 | FileCheck --check-prefixes=NO-VP %s
+
+; RUN: opt -passes=loop-vectorize -debug-only=loop-vectorize \
+; RUN: -force-tail-folding-style=none \
+; RUN: -prefer-predicate-over-epilogue=predicate-dont-vectorize \
+; RUN: -force-target-supports-scalable-vectors -scalable-vectorization=on \
+; RUN: -disable-output < %s 2>&1 | FileCheck --check-prefixes=NO-VP %s
+
+; The target does not support predicated vectorization.
+define void @foo(ptr noalias %a, ptr noalias %b, ptr noalias %c, i64 %N) {
+; NO-VP-NOT: EXPLICIT-VECTOR-LENGTH-BASED-IV-PHI
+
+entry:
+ br label %for.body
+
+for.body:
+ %iv = phi i64 [ 0, %entry ], [ %iv.next, %for.body ]
+ %arrayidx = getelementptr inbounds i32, ptr %b, i64 %iv
+ %0 = load i32, ptr %arrayidx, align 4
+ %arrayidx2 = getelementptr inbounds i32, ptr %c, i64 %iv
+ %1 = load i32, ptr %arrayidx2, align 4
+ %add = add nsw i32 %1, %0
+ %arrayidx4 = getelementptr inbounds i32, ptr %a, i64 %iv
+ store i32 %add, ptr %arrayidx4, align 4
+ %iv.next = add nuw nsw i64 %iv, 1
+ %exitcond.not = icmp eq i64 %iv.next, %N
+ br i1 %exitcond.not, label %for.cond.cleanup, label %for.body
+
+for.cond.cleanup:
+ ret void
+}
+
>From f707d5d491405093b370493a3999812c7f5bfa11 Mon Sep 17 00:00:00 2001
From: Alexey Bataev <a.bataev at outlook.com>
Date: Mon, 5 Feb 2024 07:59:30 -0800
Subject: [PATCH 2/9] Address comments
---
.../llvm/Analysis/TargetTransformInfo.h | 2 +-
.../Target/RISCV/RISCVTargetTransformInfo.h | 2 +-
.../Transforms/Vectorize/LoopVectorize.cpp | 73 +++++------
llvm/lib/Transforms/Vectorize/VPlan.h | 1 +
.../Transforms/Vectorize/VPlanAnalysis.cpp | 2 +-
.../lib/Transforms/Vectorize/VPlanRecipes.cpp | 25 ++--
.../Transforms/Vectorize/VPlanTransforms.cpp | 2 +-
.../Transforms/Vectorize/VPlanVerifier.cpp | 23 ++--
.../PowerPC/vectorize-vp-intrinsics.ll | 51 ++++++++
.../PowerPC/vplan-vp-intrinsics.ll | 117 ++++++++++++++++++
.../vectorize-vp-intrinsics-interleave.ll | 50 ++++----
11 files changed, 259 insertions(+), 89 deletions(-)
create mode 100644 llvm/test/Transforms/LoopVectorize/PowerPC/vectorize-vp-intrinsics.ll
create mode 100644 llvm/test/Transforms/LoopVectorize/PowerPC/vplan-vp-intrinsics.ll
diff --git a/llvm/include/llvm/Analysis/TargetTransformInfo.h b/llvm/include/llvm/Analysis/TargetTransformInfo.h
index 5874e89b85b9aa..489e3bd1c1e31b 100644
--- a/llvm/include/llvm/Analysis/TargetTransformInfo.h
+++ b/llvm/include/llvm/Analysis/TargetTransformInfo.h
@@ -192,7 +192,7 @@ enum class TailFoldingStyle {
/// and such that the scalar epilogue loop can always be removed.
DataAndControlFlowWithoutRuntimeCheck,
/// Use predicated EVL instructions for tail-folding.
- /// Indicates that VP intrinsics should be used if tail-folding is enabled.
+ /// Indicates that VP intrinsics should be used.
DataWithEVL,
};
diff --git a/llvm/lib/Target/RISCV/RISCVTargetTransformInfo.h b/llvm/lib/Target/RISCV/RISCVTargetTransformInfo.h
index e2ce97aef9c1db..453b139480c250 100644
--- a/llvm/lib/Target/RISCV/RISCVTargetTransformInfo.h
+++ b/llvm/lib/Target/RISCV/RISCVTargetTransformInfo.h
@@ -75,7 +75,7 @@ class RISCVTTIImpl : public BasicTTIImplBase<RISCVTTIImpl> {
const APInt &Imm, Type *Ty,
TTI::TargetCostKind CostKind);
- /// \name Vector Predication Information
+ /// \name EVL Support for predicated vectorization.
/// Whether the target supports the %evl parameter of VP intrinsic efficiently
/// in hardware, for the given opcode and type/alignment. (see LLVM Language
/// Reference - "Vector Predication Intrinsics",
diff --git a/llvm/lib/Transforms/Vectorize/LoopVectorize.cpp b/llvm/lib/Transforms/Vectorize/LoopVectorize.cpp
index 7246144d630c63..290c5d6e0a918e 100644
--- a/llvm/lib/Transforms/Vectorize/LoopVectorize.cpp
+++ b/llvm/lib/Transforms/Vectorize/LoopVectorize.cpp
@@ -253,8 +253,7 @@ static cl::opt<TailFoldingStyle> ForceTailFoldingStyle(
"data-and-control-without-rt-check",
"Similar to data-and-control, but remove the runtime check"),
clEnumValN(TailFoldingStyle::DataWithEVL, "data-with-evl",
- "Use predicated EVL instructions for tail folding if the "
- "target supports vector length predication")));
+ "Use predicated EVL instructions for tail folding")));
static cl::opt<bool> MaximizeBandwidth(
"vectorizer-maximize-bandwidth", cl::init(false), cl::Hidden,
@@ -1523,6 +1522,7 @@ class LoopVectorizationCostModel {
if (!Legal->prepareToFoldTailByMasking())
return;
+<<<<<<< HEAD
if (ForceTailFoldingStyle.getNumOccurrences()) {
ChosenTailFoldingStyle.first = ChosenTailFoldingStyle.second =
ForceTailFoldingStyle;
@@ -1558,6 +1558,16 @@ class LoopVectorizationCostModel {
}
return;
}
+||||||| parent of 73e3b4948fb4 (Address comments)
+ if (ForceTailFoldingStyle.getNumOccurrences())
+ return ForceTailFoldingStyle;
+=======
+ if (ChosenTailFoldingStyle)
+ return *ChosenTailFoldingStyle;
+
+ if (ForceTailFoldingStyle.getNumOccurrences())
+ return ForceTailFoldingStyle;
+>>>>>>> 73e3b4948fb4 (Address comments)
ChosenTailFoldingStyle.first =
TTI.getPreferredTailFoldingStyle(/*IVUpdateMayOverflow=*/true);
@@ -1582,18 +1592,7 @@ class LoopVectorizationCostModel {
/// Returns true if VP intrinsics with explicit vector length support should
/// be generated in the tail folded loop.
bool useVPIWithVPEVLVectorization() const {
- return PreferEVL && !EnableVPlanNativePath &&
- getTailFoldingStyle() == TailFoldingStyle::DataWithEVL &&
- // FIXME: implement support for max safe dependency distance.
- Legal->isSafeForAnyVectorWidth() &&
- // FIXME: remove this once reductions are supported.
- Legal->getReductionVars().empty() &&
- // FIXME: remove this once vp_reverse is supported.
- none_of(
- WideningDecisions,
- [](const std::pair<std::pair<Instruction *, ElementCount>,
- std::pair<InstWidening, InstructionCost>>
- &Data) { return Data.second.first == CM_Widen_Reverse; });
+ return getTailFoldingStyle() == TailFoldingStyle::DataWithEVL;
}
/// Returns true if the Phi is part of an inloop reduction.
@@ -1743,9 +1742,8 @@ class LoopVectorizationCostModel {
std::pair<TailFoldingStyle, TailFoldingStyle> ChosenTailFoldingStyle =
std::make_pair(TailFoldingStyle::None, TailFoldingStyle::None);
- /// Control whether to generate VP intrinsics with explicit-vector-length
- /// support in vectorized code.
- bool PreferEVL = false;
+ /// Control finally chosen tail folding style.
+ std::optional<TailFoldingStyle> ChosenTailFoldingStyle;
/// A map holding scalar costs for different vectorization factors. The
/// presence of a cost for an instruction in the mapping indicates that the
@@ -5326,8 +5324,11 @@ LoopVectorizationCostModel::selectInterleaveCount(ElementCount VF,
return 1;
// Do not interleave if EVL is preferred and no User IC is specified.
- if (useVPIWithVPEVLVectorization())
+ if (useVPIWithVPEVLVectorization()) {
+ LLVM_DEBUG(dbgs() << "LV: Preference for VP intrinsics indicated. "
+ "Unroll factor forced to be 1.\n");
return 1;
+ }
// We used the distance for the interleave count.
if (!Legal->isSafeForAnyVectorWidth())
@@ -9442,17 +9443,17 @@ void VPReplicateRecipe::execute(VPTransformState &State) {
static Instruction *
lowerStoreUsingVectorIntrinsics(IRBuilderBase &Builder, Value *Addr,
Value *StoredVal, bool IsScatter, Value *Mask,
- Value *EVLPart, const Align &Alignment) {
+ Value *EVL, const Align &Alignment) {
CallInst *Call;
if (IsScatter) {
- Call = Builder.CreateIntrinsic(Type::getVoidTy(EVLPart->getContext()),
+ Call = Builder.CreateIntrinsic(Type::getVoidTy(EVL->getContext()),
Intrinsic::vp_scatter,
- {StoredVal, Addr, Mask, EVLPart});
+ {StoredVal, Addr, Mask, EVL});
} else {
VectorBuilder VBuilder(Builder);
- VBuilder.setEVL(EVLPart).setMask(Mask);
+ VBuilder.setEVL(EVL).setMask(Mask);
Call = cast<CallInst>(VBuilder.createVectorInstruction(
- Instruction::Store, Type::getVoidTy(EVLPart->getContext()),
+ Instruction::Store, Type::getVoidTy(EVL->getContext()),
{StoredVal, Addr}));
}
Call->addParamAttr(
@@ -9465,16 +9466,16 @@ lowerStoreUsingVectorIntrinsics(IRBuilderBase &Builder, Value *Addr,
static Instruction *lowerLoadUsingVectorIntrinsics(IRBuilderBase &Builder,
VectorType *DataTy,
Value *Addr, bool IsGather,
- Value *Mask, Value *EVLPart,
+ Value *Mask, Value *EVL,
const Align &Alignment) {
CallInst *Call;
if (IsGather) {
- Call = Builder.CreateIntrinsic(DataTy, Intrinsic::vp_gather,
- {Addr, Mask, EVLPart}, nullptr,
- "wide.masked.gather");
+ Call =
+ Builder.CreateIntrinsic(DataTy, Intrinsic::vp_gather, {Addr, Mask, EVL},
+ nullptr, "wide.masked.gather");
} else {
VectorBuilder VBuilder(Builder);
- VBuilder.setEVL(EVLPart).setMask(Mask);
+ VBuilder.setEVL(EVL).setMask(Mask);
Call = cast<CallInst>(VBuilder.createVectorInstruction(
Instruction::Load, DataTy, Addr, "vp.op.load"));
}
@@ -9522,9 +9523,9 @@ void VPWidenMemoryInstructionRecipe::execute(VPTransformState &State) {
Instruction *NewSI = nullptr;
Value *StoredVal = State.get(StoredValue, Part);
if (State.EVL) {
- assert(State.UF == 1 &&
- "Expected only UF==1 for predicated vectorization.");
- Value *EVLPart = State.get(State.EVL, Part);
+ assert(State.UF == 1 && "Expected only UF == 1 when vectorizing with "
+ "explicit vector length.");
+ Value *EVL = State.get(State.EVL, VPIteration(0, 0));
// If EVL is not nullptr, then EVL must be a valid value set during plan
// creation, possibly default value = whole vector register length. EVL
// is created only if TTI prefers predicated vectorization, thus if EVL
@@ -9534,7 +9535,7 @@ void VPWidenMemoryInstructionRecipe::execute(VPTransformState &State) {
Value *MaskPart = isMaskRequired ? BlockInMaskParts[Part] : nullptr;
NewSI = lowerStoreUsingVectorIntrinsics(
Builder, State.get(getAddr(), Part), StoredVal, CreateGatherScatter,
- MaskPart, EVLPart, Alignment);
+ MaskPart, EVL, Alignment);
} else if (CreateGatherScatter) {
Value *MaskPart = isMaskRequired ? BlockInMaskParts[Part] : nullptr;
Value *VectorGep = State.get(getAddr(), Part);
@@ -9566,9 +9567,9 @@ void VPWidenMemoryInstructionRecipe::execute(VPTransformState &State) {
for (unsigned Part = 0; Part < State.UF; ++Part) {
Value *NewLI;
if (State.EVL) {
- assert(State.UF == 1 &&
- "Expected only UF==1 for predicated vectorization.");
- Value *EVLPart = State.get(State.EVL, Part);
+ assert(State.UF == 1 && "Expected only UF == 1 when vectorizing with "
+ "explicit vector length.");
+ Value *EVL = State.get(State.EVL, VPIteration(0, 0));
// If EVL is not nullptr, then EVL must be a valid value set during plan
// creation, possibly default value = whole vector register length. EVL
// is created only if TTI prefers predicated vectorization, thus if EVL
@@ -9578,7 +9579,7 @@ void VPWidenMemoryInstructionRecipe::execute(VPTransformState &State) {
Value *MaskPart = isMaskRequired ? BlockInMaskParts[Part] : nullptr;
NewLI = lowerLoadUsingVectorIntrinsics(
Builder, DataTy, State.get(getAddr(), Part), CreateGatherScatter,
- MaskPart, EVLPart, Alignment);
+ MaskPart, EVL, Alignment);
} else if (CreateGatherScatter) {
Value *MaskPart = isMaskRequired ? BlockInMaskParts[Part] : nullptr;
Value *VectorGep = State.get(getAddr(), Part);
diff --git a/llvm/lib/Transforms/Vectorize/VPlan.h b/llvm/lib/Transforms/Vectorize/VPlan.h
index c44b3118f100e5..3568778743eaff 100644
--- a/llvm/lib/Transforms/Vectorize/VPlan.h
+++ b/llvm/lib/Transforms/Vectorize/VPlan.h
@@ -2474,6 +2474,7 @@ class VPEVLBasedIVPHIRecipe : public VPHeaderPHIRecipe {
}
/// Generate phi for handling IV based on EVL over iterations correctly.
+ // TODO: investigate if it can share the code with VPCanonicalIVPHIRecipe.
void execute(VPTransformState &State) override;
/// Returns true if the recipe only uses the first lane of operand \p Op.
diff --git a/llvm/lib/Transforms/Vectorize/VPlanAnalysis.cpp b/llvm/lib/Transforms/Vectorize/VPlanAnalysis.cpp
index 9fe1da9da93deb..78e21aec26dbb7 100644
--- a/llvm/lib/Transforms/Vectorize/VPlanAnalysis.cpp
+++ b/llvm/lib/Transforms/Vectorize/VPlanAnalysis.cpp
@@ -215,7 +215,7 @@ Type *VPTypeAnalysis::inferScalarType(const VPValue *V) {
.Case<VPCanonicalIVPHIRecipe, VPFirstOrderRecurrencePHIRecipe,
VPReductionPHIRecipe, VPWidenPointerInductionRecipe,
VPEVLBasedIVPHIRecipe>([this](const auto *R) {
- // Handle header phi recipes, except VPWienIntOrFpInduction
+ // Handle header phi recipes, except VPWidenIntOrFpInduction
// which needs special handling due it being possibly truncated.
// TODO: consider inferring/caching type of siblings, e.g.,
// backedge value, here and in cases below.
diff --git a/llvm/lib/Transforms/Vectorize/VPlanRecipes.cpp b/llvm/lib/Transforms/Vectorize/VPlanRecipes.cpp
index da3cf5c3e9cb38..2a47b8de827f07 100644
--- a/llvm/lib/Transforms/Vectorize/VPlanRecipes.cpp
+++ b/llvm/lib/Transforms/Vectorize/VPlanRecipes.cpp
@@ -352,40 +352,41 @@ Value *VPInstruction::generateInstruction(VPTransformState &State,
}
case VPInstruction::ExplicitVectorLength: {
// Compute EVL
- auto GetSetVL = [=](VPTransformState &State, Value *EVL) {
- assert(EVL->getType()->isIntegerTy() &&
+ auto GetSetVL = [=](VPTransformState &State, Value *AVL) {
+ assert(AVL->getType()->isIntegerTy() &&
"Requested vector length should be an integer.");
// TODO: Add support for MaxSafeDist for correct loop emission.
assert(State.VF.isScalable() && "Expected scalable vector factor.");
Value *VFArg = State.Builder.getInt32(State.VF.getKnownMinValue());
- Value *GVL = State.Builder.CreateIntrinsic(
+ Value *EVL = State.Builder.CreateIntrinsic(
State.Builder.getInt32Ty(), Intrinsic::experimental_get_vector_length,
- {EVL, VFArg, State.Builder.getTrue()});
- return GVL;
+ {AVL, VFArg, State.Builder.getTrue()});
+ return EVL;
};
// TODO: Restructure this code with an explicit remainder loop, vsetvli can
// be outside of the main loop.
assert(Part == 0 && "No unrolling expected for predicated vectorization.");
// Compute VTC - IV as the EVL(requested vector length).
- Value *Index = State.get(getOperand(0), 0);
+ Value *Index = State.get(getOperand(0), VPIteration(0, 0));
Value *TripCount = State.get(getOperand(1), VPIteration(0, 0));
- Value *EVL = State.Builder.CreateSub(TripCount, Index);
- Value *SetVL = GetSetVL(State, EVL);
+ Value *AVL = State.Builder.CreateSub(TripCount, Index);
+ Value *EVL = GetSetVL(State, AVL);
assert(!State.EVL && "multiple EVL recipes");
State.EVL = this;
- return SetVL;
+ return EVL;
}
// TODO: remove this once a regular Add VPInstruction is supported.
case VPInstruction::ExplicitVectorLengthIVIncrement: {
assert(Part == 0 && "Expected unroll factor 1 for VP vectorization.");
- Value *Phi = State.get(getOperand(0), VPIteration(Part, 0));
- Value *EVL = State.get(getOperand(1), VPIteration(Part, 0));
+ Value *Phi = State.get(getOperand(0), VPIteration(0, 0));
+ Value *EVL = State.get(getOperand(1), VPIteration(0, 0));
assert(EVL->getType() == Phi->getType() &&
"EVL and Phi must have the same type.");
return Builder.CreateAdd(Phi, EVL, Name, hasNoUnsignedWrap(),
hasNoSignedWrap());
+ return EVL;
}
case VPInstruction::CanonicalIVIncrementForPart: {
auto *IV = State.get(getOperand(0), VPIteration(0, 0));
@@ -571,6 +572,8 @@ bool VPInstruction::onlyFirstLaneUsed(const VPValue *Op) const {
// TODO: Cover additional opcodes.
return vputils::onlyFirstLaneUsed(this);
case VPInstruction::ActiveLaneMask:
+ case VPInstruction::ExplicitVectorLength:
+ case VPInstruction::ExplicitVectorLengthIVIncrement:
case VPInstruction::CalculateTripCountMinusVF:
case VPInstruction::CanonicalIVIncrementForPart:
case VPInstruction::BranchOnCount:
diff --git a/llvm/lib/Transforms/Vectorize/VPlanTransforms.cpp b/llvm/lib/Transforms/Vectorize/VPlanTransforms.cpp
index 328888c122b15d..786ae4b3c4c828 100644
--- a/llvm/lib/Transforms/Vectorize/VPlanTransforms.cpp
+++ b/llvm/lib/Transforms/Vectorize/VPlanTransforms.cpp
@@ -1305,7 +1305,7 @@ void VPlanTransforms::addExplicitVectorLength(VPlan &Plan) {
EVLPhi->addOperand(NextEVLIV);
// Replace all uses of VPCanonicalIVPHIRecipe by
- // VPEVLBasedIVPHIRecipe except for VPInstruction::CanonicalIVIncrement.
+ // VPEVLBasedIVPHIRecipe except for the canonical IV increment..
CanonicalIVPHI->replaceAllUsesWith(EVLPhi);
CanonicalIVIncrement->setOperand(0, CanonicalIVPHI);
// TODO: support unroll factor > 1.
diff --git a/llvm/lib/Transforms/Vectorize/VPlanVerifier.cpp b/llvm/lib/Transforms/Vectorize/VPlanVerifier.cpp
index 364e9cda4fa044..37ea02b302102d 100644
--- a/llvm/lib/Transforms/Vectorize/VPlanVerifier.cpp
+++ b/llvm/lib/Transforms/Vectorize/VPlanVerifier.cpp
@@ -92,26 +92,26 @@ static bool verifyVPBasicBlock(const VPBasicBlock *VPBB,
for (const VPRecipeBase &R : *VPBB)
RecipeNumbering[&R] = Cnt++;
- // Check if EVL recipes exist only in Entry block and only once.
+ // Set of recipe types along with VPInstruction Opcodes of all EVL-related
+ // recipes that must appear at most once in Entry or Exiting blocks.
DenseSet<unsigned> EVLFound;
- const VPBlockBase *Header = nullptr;
- const VPBlockBase *Exit = nullptr;
+ bool IsHeader = false;
+ bool IsExiting = false;
const VPlan *Plan = VPBB->getPlan();
if (Plan && Plan->getEntry()->getNumSuccessors() == 1) {
- Header = Plan->getVectorLoopRegion()->getEntry();
- Exit = Plan->getVectorLoopRegion()->getExiting();
+ IsHeader = Plan->getVectorLoopRegion()->getEntry() == VPBB;
+ IsExiting = Plan->getVectorLoopRegion()->getExiting() == VPBB;
}
auto CheckEVLRecipiesInsts = [&](const VPRecipeBase *R) {
if (isa<VPEVLBasedIVPHIRecipe>(R)) {
- if (!Header || VPBB != Header) {
+ if (!IsHeader) {
errs() << "EVL PHI recipe not in entry block!\n";
return false;
}
- if (EVLFound.contains(VPDef::VPEVLBasedIVPHISC)) {
+ if (!EVLFound.insert(VPDef::VPEVLBasedIVPHISC).second) {
errs() << "EVL PHI recipe inserted more than once!\n";
return false;
}
- EVLFound.insert(VPDef::VPEVLBasedIVPHISC);
return true;
}
auto *RInst = dyn_cast<VPInstruction>(R);
@@ -119,13 +119,13 @@ static bool verifyVPBasicBlock(const VPBasicBlock *VPBB,
return true;
switch (RInst->getOpcode()) {
case VPInstruction::ExplicitVectorLength:
- if (!Header || VPBB != Header) {
+ if (!IsHeader) {
errs() << "EVL instruction not in entry block!\n";
return false;
}
break;
case VPInstruction::ExplicitVectorLengthIVIncrement:
- if (!Exit || VPBB != Exit) {
+ if (!IsExiting) {
errs() << "EVL inc instruction not in exit block!\n";
return false;
}
@@ -133,11 +133,10 @@ static bool verifyVPBasicBlock(const VPBasicBlock *VPBB,
default:
return true;
}
- if (EVLFound.contains(RInst->getOpcode() + VPDef::VPLastPHISC)) {
+ if (!EVLFound.insert(RInst->getOpcode() + VPDef::VPLastPHISC).second) {
errs() << "EVL instruction inserted more than once!\n";
return false;
}
- EVLFound.insert(RInst->getOpcode() + VPDef::VPLastPHISC);
return true;
};
diff --git a/llvm/test/Transforms/LoopVectorize/PowerPC/vectorize-vp-intrinsics.ll b/llvm/test/Transforms/LoopVectorize/PowerPC/vectorize-vp-intrinsics.ll
new file mode 100644
index 00000000000000..2ce2a45a811ab3
--- /dev/null
+++ b/llvm/test/Transforms/LoopVectorize/PowerPC/vectorize-vp-intrinsics.ll
@@ -0,0 +1,51 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
+; RUN: opt -passes=loop-vectorize \
+; RUN: -force-tail-folding-style=data-with-evl \
+; RUN: -prefer-predicate-over-epilogue=predicate-dont-vectorize \
+; RUN: -mtriple=powerpc64le-unknown-linux-gnu \
+; RUN: -mcpu=pwr10 -S < %s | FileCheck %s
+
+; RUN: opt -passes=loop-vectorize \
+; RUN: -force-tail-folding-style=none \
+; RUN: -prefer-predicate-over-epilogue=predicate-dont-vectorize \
+; RUN: -mtriple=powerpc64le-unknown-linux-gnu \
+; RUN: -mcpu=pwr10 -S < %s | FileCheck %s
+
+define void @foo(ptr noalias %a, ptr noalias %b, ptr noalias %c, i64 %N) {
+; CHECK-LABEL: @foo(
+; CHECK-NEXT: entry:
+; CHECK-NEXT: br label [[FOR_BODY:%.*]]
+; CHECK: for.body:
+; CHECK-NEXT: [[IV:%.*]] = phi i64 [ 0, [[ENTRY:%.*]] ], [ [[IV_NEXT:%.*]], [[FOR_BODY]] ]
+; CHECK-NEXT: [[ARRAYIDX:%.*]] = getelementptr inbounds i32, ptr [[B:%.*]], i64 [[IV]]
+; CHECK-NEXT: [[TMP0:%.*]] = load i32, ptr [[ARRAYIDX]], align 4
+; CHECK-NEXT: [[ARRAYIDX2:%.*]] = getelementptr inbounds i32, ptr [[C:%.*]], i64 [[IV]]
+; CHECK-NEXT: [[TMP1:%.*]] = load i32, ptr [[ARRAYIDX2]], align 4
+; CHECK-NEXT: [[ADD:%.*]] = add nsw i32 [[TMP1]], [[TMP0]]
+; CHECK-NEXT: [[ARRAYIDX4:%.*]] = getelementptr inbounds i32, ptr [[A:%.*]], i64 [[IV]]
+; CHECK-NEXT: store i32 [[ADD]], ptr [[ARRAYIDX4]], align 4
+; CHECK-NEXT: [[IV_NEXT]] = add nuw nsw i64 [[IV]], 1
+; CHECK-NEXT: [[EXITCOND_NOT:%.*]] = icmp eq i64 [[IV_NEXT]], [[N:%.*]]
+; CHECK-NEXT: br i1 [[EXITCOND_NOT]], label [[FOR_COND_CLEANUP:%.*]], label [[FOR_BODY]]
+; CHECK: for.cond.cleanup:
+; CHECK-NEXT: ret void
+;
+entry:
+ br label %for.body
+
+for.body:
+ %iv = phi i64 [ 0, %entry ], [ %iv.next, %for.body ]
+ %arrayidx = getelementptr inbounds i32, ptr %b, i64 %iv
+ %0 = load i32, ptr %arrayidx, align 4
+ %arrayidx2 = getelementptr inbounds i32, ptr %c, i64 %iv
+ %1 = load i32, ptr %arrayidx2, align 4
+ %add = add nsw i32 %1, %0
+ %arrayidx4 = getelementptr inbounds i32, ptr %a, i64 %iv
+ store i32 %add, ptr %arrayidx4, align 4
+ %iv.next = add nuw nsw i64 %iv, 1
+ %exitcond.not = icmp eq i64 %iv.next, %N
+ br i1 %exitcond.not, label %for.cond.cleanup, label %for.body
+
+for.cond.cleanup:
+ ret void
+}
diff --git a/llvm/test/Transforms/LoopVectorize/PowerPC/vplan-vp-intrinsics.ll b/llvm/test/Transforms/LoopVectorize/PowerPC/vplan-vp-intrinsics.ll
new file mode 100644
index 00000000000000..5d1a471c5d166f
--- /dev/null
+++ b/llvm/test/Transforms/LoopVectorize/PowerPC/vplan-vp-intrinsics.ll
@@ -0,0 +1,117 @@
+; REQUIRES: asserts
+
+; RUN: opt -passes=loop-vectorize -debug-only=loop-vectorize \
+; RUN: -force-tail-folding-style=data-with-evl \
+; RUN: -prefer-predicate-over-epilogue=predicate-dont-vectorize \
+; RUN: -mtriple=powerpc64le-unknown-linux-gnu \
+; RUN: -mcpu=pwr10 -disable-output < %s 2>&1 | FileCheck %s
+
+define void @foo(ptr noalias %a, ptr noalias %b, ptr noalias %c, i64 %N) {
+; CHECK-LABEL: VPlan 'Initial VPlan for VF={2,4},UF>=1' {
+; CHECK-NEXT: Live-in vp<%0> = VF * UF
+; CHECK-NEXT: Live-in vp<%1> = vector-trip-count
+; CHECK-NEXT: Live-in vp<%2> = backedge-taken count
+; CHECK-NEXT: Live-in ir<%N> = original trip-count
+; CHECK-EMPTY:
+; CHECK-NEXT: vector.ph:
+; CHECK-NEXT: Successor(s): vector loop
+; CHECK-EMPTY:
+; CHECK-NEXT: <x1> vector loop: {
+; CHECK-NEXT: vector.body:
+; CHECK-NEXT: EMIT vp<%3> = CANONICAL-INDUCTION ir<0>, vp<%16>
+; CHECK-NEXT: WIDEN-INDUCTION %iv = phi 0, %iv.next, ir<1>
+; CHECK-NEXT: EMIT vp<%5> = icmp ule ir<%iv>, vp<%2>
+; CHECK-NEXT: Successor(s): pred.store
+; CHECK-EMPTY:
+; CHECK-NEXT: <xVFxUF> pred.store: {
+; CHECK-NEXT: pred.store.entry:
+; CHECK-NEXT: BRANCH-ON-MASK vp<%5>
+; CHECK-NEXT: Successor(s): pred.store.if, pred.store.continue
+; CHECK-EMPTY:
+; CHECK-NEXT: pred.store.if:
+; CHECK-NEXT: vp<%6> = SCALAR-STEPS vp<%3>, ir<1>
+; CHECK-NEXT: REPLICATE ir<%arrayidx> = getelementptr inbounds ir<%b>, vp<%6>
+; CHECK-NEXT: REPLICATE ir<%0> = load ir<%arrayidx>
+; CHECK-NEXT: REPLICATE ir<%arrayidx2> = getelementptr inbounds ir<%c>, vp<%6>
+; CHECK-NEXT: REPLICATE ir<%1> = load ir<%arrayidx2>
+; CHECK-NEXT: REPLICATE ir<%arrayidx4> = getelementptr inbounds ir<%a>, vp<%6>
+; CHECK-NEXT: REPLICATE ir<%add> = add nsw ir<%1>, ir<%0>
+; CHECK-NEXT: REPLICATE store ir<%add>, ir<%arrayidx4>
+; CHECK-NEXT: Successor(s): pred.store.continue
+; CHECK-EMPTY:
+; CHECK-NEXT: pred.store.continue:
+; CHECK-NEXT: PHI-PREDICATED-INSTRUCTION vp<%14> = ir<%0>
+; CHECK-NEXT: PHI-PREDICATED-INSTRUCTION vp<%15> = ir<%1>
+; CHECK-NEXT: No successors
+; CHECK-NEXT: }
+; CHECK-NEXT: Successor(s): for.body.2
+; CHECK-EMPTY:
+; CHECK-NEXT: for.body.2:
+; CHECK-NEXT: EMIT vp<%16> = add vp<%3>, vp<%0>
+; CHECK-NEXT: EMIT branch-on-count vp<%16>, vp<%1>
+; CHECK-NEXT: No successors
+; CHECK-NEXT: }
+;
+entry:
+ br label %for.body
+
+for.body:
+ %iv = phi i64 [ 0, %entry ], [ %iv.next, %for.body ]
+ %arrayidx = getelementptr inbounds i32, ptr %b, i64 %iv
+ %0 = load i32, ptr %arrayidx, align 4
+ %arrayidx2 = getelementptr inbounds i32, ptr %c, i64 %iv
+ %1 = load i32, ptr %arrayidx2, align 4
+ %add = add nsw i32 %1, %0
+ %arrayidx4 = getelementptr inbounds i32, ptr %a, i64 %iv
+ store i32 %add, ptr %arrayidx4, align 4
+ %iv.next = add nuw nsw i64 %iv, 1
+ %exitcond.not = icmp eq i64 %iv.next, %N
+ br i1 %exitcond.not, label %for.cond.cleanup, label %for.body
+
+for.cond.cleanup:
+ ret void
+}
+
+define void @safe_dep(ptr %p) {
+; CHECK-LABEL: VPlan 'Initial VPlan for VF={2},UF>=1' {
+; CHECK-NEXT: Live-in vp<%0> = VF * UF
+; CHECK-NEXT: Live-in vp<%1> = vector-trip-count
+; CHECK-NEXT: Live-in ir<512> = original trip-count
+; CHECK-EMPTY:
+; CHECK-NEXT: vector.ph:
+; CHECK-NEXT: Successor(s): vector loop
+; CHECK-EMPTY:
+; CHECK-NEXT: <x1> vector loop: {
+; CHECK-NEXT: vector.body:
+; CHECK-NEXT: EMIT vp<%2> = CANONICAL-INDUCTION ir<0>, vp<%10>
+; CHECK-NEXT: vp<%3> = SCALAR-STEPS vp<%2>, ir<1>
+; CHECK-NEXT: CLONE ir<%a1> = getelementptr ir<%p>, vp<%3>
+; CHECK-NEXT: vp<%5> = vector-pointer ir<%a1>
+; CHECK-NEXT: WIDEN ir<%v> = load vp<%5>
+; CHECK-NEXT: CLONE ir<%offset> = add vp<%3>, ir<100>
+; CHECK-NEXT: CLONE ir<%a2> = getelementptr ir<%p>, ir<%offset>
+; CHECK-NEXT: vp<%9> = vector-pointer ir<%a2>
+; CHECK-NEXT: WIDEN store vp<%9>, ir<%v>
+; CHECK-NEXT: EMIT vp<%10> = add nuw vp<%2>, vp<%0>
+; CHECK-NEXT: EMIT branch-on-count vp<%10>, vp<%1>
+; CHECK-NEXT: No successors
+; CHECK-NEXT: }
+;
+entry:
+ br label %loop
+
+loop:
+ %iv = phi i64 [0, %entry], [%iv.next, %loop]
+ %a1 = getelementptr i64, ptr %p, i64 %iv
+ %v = load i64, ptr %a1, align 32
+ %offset = add i64 %iv, 100
+ %a2 = getelementptr i64, ptr %p, i64 %offset
+ store i64 %v, ptr %a2, align 32
+ %iv.next = add i64 %iv, 1
+ %cmp = icmp ne i64 %iv, 511
+ br i1 %cmp, label %loop, label %exit
+
+exit:
+ ret void
+}
+
diff --git a/llvm/test/Transforms/LoopVectorize/vectorize-vp-intrinsics-interleave.ll b/llvm/test/Transforms/LoopVectorize/vectorize-vp-intrinsics-interleave.ll
index 9d0ae16c265da4..5f528e8cd9ebfc 100644
--- a/llvm/test/Transforms/LoopVectorize/vectorize-vp-intrinsics-interleave.ll
+++ b/llvm/test/Transforms/LoopVectorize/vectorize-vp-intrinsics-interleave.ll
@@ -21,39 +21,37 @@ define void @interleave(ptr noalias %a, ptr noalias %b, ptr noalias %c, i64 %N)
; IF-EVL-NEXT: [[TRIP_COUNT_MINUS_1:%.*]] = sub i64 [[N]], 1
; IF-EVL-NEXT: br label [[VECTOR_BODY:%.*]]
; IF-EVL: vector.body:
-; IF-EVL-NEXT: [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[PRED_STORE_CONTINUE3:%.*]] ]
-; IF-EVL-NEXT: [[VEC_IV:%.*]] = add i64 [[INDEX]], 0
-; IF-EVL-NEXT: [[VEC_IV1:%.*]] = add i64 [[INDEX]], 1
-; IF-EVL-NEXT: [[TMP0:%.*]] = icmp ule i64 [[VEC_IV]], [[TRIP_COUNT_MINUS_1]]
-; IF-EVL-NEXT: [[TMP1:%.*]] = icmp ule i64 [[VEC_IV1]], [[TRIP_COUNT_MINUS_1]]
-; IF-EVL-NEXT: br i1 [[TMP0]], label [[PRED_STORE_IF:%.*]], label [[PRED_STORE_CONTINUE:%.*]]
+; IF-EVL-NEXT: [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[PRED_STORE_CONTINUE2:%.*]] ]
+; IF-EVL-NEXT: [[TMP0:%.*]] = add i64 [[INDEX]], 0
+; IF-EVL-NEXT: [[TMP1:%.*]] = add i64 [[INDEX]], 1
+; IF-EVL-NEXT: [[TMP2:%.*]] = icmp ule i64 [[TMP0]], [[TRIP_COUNT_MINUS_1]]
+; IF-EVL-NEXT: [[TMP3:%.*]] = icmp ule i64 [[TMP1]], [[TRIP_COUNT_MINUS_1]]
+; IF-EVL-NEXT: br i1 [[TMP2]], label [[PRED_STORE_IF:%.*]], label [[PRED_STORE_CONTINUE:%.*]]
; IF-EVL: pred.store.if:
-; IF-EVL-NEXT: [[TMP2:%.*]] = add i64 [[INDEX]], 0
-; IF-EVL-NEXT: [[TMP3:%.*]] = getelementptr inbounds i32, ptr [[B:%.*]], i64 [[TMP2]]
-; IF-EVL-NEXT: [[TMP4:%.*]] = load i32, ptr [[TMP3]], align 4
-; IF-EVL-NEXT: [[TMP5:%.*]] = getelementptr inbounds i32, ptr [[C:%.*]], i64 [[TMP2]]
-; IF-EVL-NEXT: [[TMP6:%.*]] = load i32, ptr [[TMP5]], align 4
-; IF-EVL-NEXT: [[TMP7:%.*]] = getelementptr inbounds i32, ptr [[A:%.*]], i64 [[TMP2]]
-; IF-EVL-NEXT: [[TMP8:%.*]] = add nsw i32 [[TMP6]], [[TMP4]]
-; IF-EVL-NEXT: store i32 [[TMP8]], ptr [[TMP7]], align 4
+; IF-EVL-NEXT: [[TMP4:%.*]] = getelementptr inbounds i32, ptr [[B:%.*]], i64 [[TMP0]]
+; IF-EVL-NEXT: [[TMP5:%.*]] = load i32, ptr [[TMP4]], align 4
+; IF-EVL-NEXT: [[TMP6:%.*]] = getelementptr inbounds i32, ptr [[C:%.*]], i64 [[TMP0]]
+; IF-EVL-NEXT: [[TMP7:%.*]] = load i32, ptr [[TMP6]], align 4
+; IF-EVL-NEXT: [[TMP8:%.*]] = getelementptr inbounds i32, ptr [[A:%.*]], i64 [[TMP0]]
+; IF-EVL-NEXT: [[TMP9:%.*]] = add nsw i32 [[TMP7]], [[TMP5]]
+; IF-EVL-NEXT: store i32 [[TMP9]], ptr [[TMP8]], align 4
; IF-EVL-NEXT: br label [[PRED_STORE_CONTINUE]]
; IF-EVL: pred.store.continue:
-; IF-EVL-NEXT: [[TMP9:%.*]] = phi i32 [ poison, [[VECTOR_BODY]] ], [ [[TMP4]], [[PRED_STORE_IF]] ]
-; IF-EVL-NEXT: [[TMP10:%.*]] = phi i32 [ poison, [[VECTOR_BODY]] ], [ [[TMP6]], [[PRED_STORE_IF]] ]
-; IF-EVL-NEXT: br i1 [[TMP1]], label [[PRED_STORE_IF2:%.*]], label [[PRED_STORE_CONTINUE3]]
-; IF-EVL: pred.store.if2:
-; IF-EVL-NEXT: [[TMP11:%.*]] = add i64 [[INDEX]], 1
-; IF-EVL-NEXT: [[TMP12:%.*]] = getelementptr inbounds i32, ptr [[B]], i64 [[TMP11]]
+; IF-EVL-NEXT: [[TMP10:%.*]] = phi i32 [ poison, [[VECTOR_BODY]] ], [ [[TMP5]], [[PRED_STORE_IF]] ]
+; IF-EVL-NEXT: [[TMP11:%.*]] = phi i32 [ poison, [[VECTOR_BODY]] ], [ [[TMP7]], [[PRED_STORE_IF]] ]
+; IF-EVL-NEXT: br i1 [[TMP3]], label [[PRED_STORE_IF1:%.*]], label [[PRED_STORE_CONTINUE2]]
+; IF-EVL: pred.store.if1:
+; IF-EVL-NEXT: [[TMP12:%.*]] = getelementptr inbounds i32, ptr [[B]], i64 [[TMP1]]
; IF-EVL-NEXT: [[TMP13:%.*]] = load i32, ptr [[TMP12]], align 4
-; IF-EVL-NEXT: [[TMP14:%.*]] = getelementptr inbounds i32, ptr [[C]], i64 [[TMP11]]
+; IF-EVL-NEXT: [[TMP14:%.*]] = getelementptr inbounds i32, ptr [[C]], i64 [[TMP1]]
; IF-EVL-NEXT: [[TMP15:%.*]] = load i32, ptr [[TMP14]], align 4
-; IF-EVL-NEXT: [[TMP16:%.*]] = getelementptr inbounds i32, ptr [[A]], i64 [[TMP11]]
+; IF-EVL-NEXT: [[TMP16:%.*]] = getelementptr inbounds i32, ptr [[A]], i64 [[TMP1]]
; IF-EVL-NEXT: [[TMP17:%.*]] = add nsw i32 [[TMP15]], [[TMP13]]
; IF-EVL-NEXT: store i32 [[TMP17]], ptr [[TMP16]], align 4
-; IF-EVL-NEXT: br label [[PRED_STORE_CONTINUE3]]
-; IF-EVL: pred.store.continue3:
-; IF-EVL-NEXT: [[TMP18:%.*]] = phi i32 [ poison, [[PRED_STORE_CONTINUE]] ], [ [[TMP13]], [[PRED_STORE_IF2]] ]
-; IF-EVL-NEXT: [[TMP19:%.*]] = phi i32 [ poison, [[PRED_STORE_CONTINUE]] ], [ [[TMP15]], [[PRED_STORE_IF2]] ]
+; IF-EVL-NEXT: br label [[PRED_STORE_CONTINUE2]]
+; IF-EVL: pred.store.continue2:
+; IF-EVL-NEXT: [[TMP18:%.*]] = phi i32 [ poison, [[PRED_STORE_CONTINUE]] ], [ [[TMP13]], [[PRED_STORE_IF1]] ]
+; IF-EVL-NEXT: [[TMP19:%.*]] = phi i32 [ poison, [[PRED_STORE_CONTINUE]] ], [ [[TMP15]], [[PRED_STORE_IF1]] ]
; IF-EVL-NEXT: [[INDEX_NEXT]] = add i64 [[INDEX]], 2
; IF-EVL-NEXT: [[TMP20:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
; IF-EVL-NEXT: br i1 [[TMP20]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]]
>From 1abcb77ee66351d6159986aa0a9d8a089490394c Mon Sep 17 00:00:00 2001
From: Alexey Bataev <a.bataev at outlook.com>
Date: Wed, 7 Feb 2024 09:26:24 -0800
Subject: [PATCH 3/9] Adjust comments, lambda name
---
llvm/lib/Transforms/Vectorize/VPlan.h | 2 +-
.../lib/Transforms/Vectorize/VPlanRecipes.cpp | 4 +-
.../Transforms/Vectorize/VPlanTransforms.cpp | 44 +++++++++----------
3 files changed, 25 insertions(+), 25 deletions(-)
diff --git a/llvm/lib/Transforms/Vectorize/VPlan.h b/llvm/lib/Transforms/Vectorize/VPlan.h
index 3568778743eaff..51ab2edc4e553a 100644
--- a/llvm/lib/Transforms/Vectorize/VPlan.h
+++ b/llvm/lib/Transforms/Vectorize/VPlan.h
@@ -2474,7 +2474,7 @@ class VPEVLBasedIVPHIRecipe : public VPHeaderPHIRecipe {
}
/// Generate phi for handling IV based on EVL over iterations correctly.
- // TODO: investigate if it can share the code with VPCanonicalIVPHIRecipe.
+ /// TODO: investigate if it can share the code with VPCanonicalIVPHIRecipe.
void execute(VPTransformState &State) override;
/// Returns true if the recipe only uses the first lane of operand \p Op.
diff --git a/llvm/lib/Transforms/Vectorize/VPlanRecipes.cpp b/llvm/lib/Transforms/Vectorize/VPlanRecipes.cpp
index 2a47b8de827f07..a97e844d6dc1d1 100644
--- a/llvm/lib/Transforms/Vectorize/VPlanRecipes.cpp
+++ b/llvm/lib/Transforms/Vectorize/VPlanRecipes.cpp
@@ -352,7 +352,7 @@ Value *VPInstruction::generateInstruction(VPTransformState &State,
}
case VPInstruction::ExplicitVectorLength: {
// Compute EVL
- auto GetSetVL = [=](VPTransformState &State, Value *AVL) {
+ auto GetEVL = [=](VPTransformState &State, Value *AVL) {
assert(AVL->getType()->isIntegerTy() &&
"Requested vector length should be an integer.");
@@ -372,7 +372,7 @@ Value *VPInstruction::generateInstruction(VPTransformState &State,
Value *Index = State.get(getOperand(0), VPIteration(0, 0));
Value *TripCount = State.get(getOperand(1), VPIteration(0, 0));
Value *AVL = State.Builder.CreateSub(TripCount, Index);
- Value *EVL = GetSetVL(State, AVL);
+ Value *EVL = GetEVL(State, AVL);
assert(!State.EVL && "multiple EVL recipes");
State.EVL = this;
return EVL;
diff --git a/llvm/lib/Transforms/Vectorize/VPlanTransforms.cpp b/llvm/lib/Transforms/Vectorize/VPlanTransforms.cpp
index 786ae4b3c4c828..e43eae98ae4a72 100644
--- a/llvm/lib/Transforms/Vectorize/VPlanTransforms.cpp
+++ b/llvm/lib/Transforms/Vectorize/VPlanTransforms.cpp
@@ -1243,28 +1243,28 @@ void VPlanTransforms::addActiveLaneMask(
replaceHeaderPredicateWithIdiom(Plan, *LaneMask);
}
-// Add a VPEVLBasedIVPHIRecipe and related recipes to \p Plan and
-// replaces all uses except the canonical IV increment of VPCanonicalIVPHIRecipe
-// with a VPEVLBasedIVPHIRecipe. VPCanonicalIVPHIRecipe is used only
-// for loop iterations counting after this transformation.
-//
-// The function uses the following definitions:
-// %StartV is the canonical induction start value.
-//
-// The function adds the following recipes:
-//
-// vector.ph:
-// ...
-//
-// vector.body:
-// ...
-// %P = EXPLICIT-VECTOR-LENGTH-BASED-IV-PHI [ %StartV, %vector.ph ], [ %NextEVL,
-// %vector.body ]
-// %EVL = EXPLICIT-VECTOR-LENGTH %P, original TC
-// ...
-// %NextEVL = EXPLICIT-VECTOR-LENGTH + %P, %EVL
-// ...
-//
+/// Add a VPEVLBasedIVPHIRecipe and related recipes to \p Plan and
+/// replaces all uses except the canonical IV increment of VPCanonicalIVPHIRecipe
+/// with a VPEVLBasedIVPHIRecipe. VPCanonicalIVPHIRecipe is used only
+/// for loop iterations counting after this transformation.
+///
+/// The function uses the following definitions:
+/// %StartV is the canonical induction start value.
+///
+/// The function adds the following recipes:
+///
+/// vector.ph:
+/// ...
+///
+/// vector.body:
+/// ...
+/// %P = EXPLICIT-VECTOR-LENGTH-BASED-IV-PHI [ %StartV, %vector.ph ], [ %NextEVL,
+/// %vector.body ]
+/// %EVL = EXPLICIT-VECTOR-LENGTH %P, original TC
+/// ...
+/// %NextEVL = EXPLICIT-VECTOR-LENGTH + %P, %EVL
+/// ...
+///
void VPlanTransforms::addExplicitVectorLength(VPlan &Plan) {
VPBasicBlock *Header = Plan.getVectorLoopRegion()->getEntryBasicBlock();
auto *CanonicalIVPHI = Plan.getCanonicalIV();
>From 97efcc5e1a9586b7ccd2d3994d48d5819caf2fa3 Mon Sep 17 00:00:00 2001
From: Alexey Bataev <a.bataev at outlook.com>
Date: Wed, 7 Feb 2024 09:50:22 -0800
Subject: [PATCH 4/9] Fix formatting
---
llvm/lib/Transforms/Vectorize/VPlanTransforms.cpp | 11 +++++------
1 file changed, 5 insertions(+), 6 deletions(-)
diff --git a/llvm/lib/Transforms/Vectorize/VPlanTransforms.cpp b/llvm/lib/Transforms/Vectorize/VPlanTransforms.cpp
index e43eae98ae4a72..683e2dee367378 100644
--- a/llvm/lib/Transforms/Vectorize/VPlanTransforms.cpp
+++ b/llvm/lib/Transforms/Vectorize/VPlanTransforms.cpp
@@ -1244,9 +1244,9 @@ void VPlanTransforms::addActiveLaneMask(
}
/// Add a VPEVLBasedIVPHIRecipe and related recipes to \p Plan and
-/// replaces all uses except the canonical IV increment of VPCanonicalIVPHIRecipe
-/// with a VPEVLBasedIVPHIRecipe. VPCanonicalIVPHIRecipe is used only
-/// for loop iterations counting after this transformation.
+/// replaces all uses except the canonical IV increment of
+/// VPCanonicalIVPHIRecipe with a VPEVLBasedIVPHIRecipe. VPCanonicalIVPHIRecipe
+/// is used only for loop iterations counting after this transformation.
///
/// The function uses the following definitions:
/// %StartV is the canonical induction start value.
@@ -1258,9 +1258,8 @@ void VPlanTransforms::addActiveLaneMask(
///
/// vector.body:
/// ...
-/// %P = EXPLICIT-VECTOR-LENGTH-BASED-IV-PHI [ %StartV, %vector.ph ], [ %NextEVL,
-/// %vector.body ]
-/// %EVL = EXPLICIT-VECTOR-LENGTH %P, original TC
+/// %P = EXPLICIT-VECTOR-LENGTH-BASED-IV-PHI [ %StartV, %vector.ph ], [
+/// %NextEVL, %vector.body ] %EVL = EXPLICIT-VECTOR-LENGTH %P, original TC
/// ...
/// %NextEVL = EXPLICIT-VECTOR-LENGTH + %P, %EVL
/// ...
>From f51f4621a274d776c497177e08c401bdf498d04a Mon Sep 17 00:00:00 2001
From: Alexey Bataev <a.bataev at outlook.com>
Date: Fri, 16 Feb 2024 06:10:19 -0800
Subject: [PATCH 5/9] Rebase
---
llvm/lib/Transforms/Vectorize/VPlanTransforms.cpp | 4 +---
1 file changed, 1 insertion(+), 3 deletions(-)
diff --git a/llvm/lib/Transforms/Vectorize/VPlanTransforms.cpp b/llvm/lib/Transforms/Vectorize/VPlanTransforms.cpp
index 683e2dee367378..342b352650a00d 100644
--- a/llvm/lib/Transforms/Vectorize/VPlanTransforms.cpp
+++ b/llvm/lib/Transforms/Vectorize/VPlanTransforms.cpp
@@ -1334,9 +1334,7 @@ void VPlanTransforms::dropPoisonGeneratingRecipes(
// handled.
if (isa<VPWidenMemoryInstructionRecipe>(CurRec) ||
isa<VPInterleaveRecipe>(CurRec) ||
- isa<VPScalarIVStepsRecipe>(CurRec) ||
- isa<VPCanonicalIVPHIRecipe>(CurRec) ||
- isa<VPActiveLaneMaskPHIRecipe>(CurRec))
+ isa<VPScalarIVStepsRecipe>(CurRec) || isa<VPHeaderPHIRecipe>(CurRec))
continue;
// This recipe contributes to the address computation of a widen
>From 737898126bba4d875e0b2db80929205925a3a404 Mon Sep 17 00:00:00 2001
From: Alexey Bataev <a.bataev at outlook.com>
Date: Wed, 28 Feb 2024 08:39:29 -0800
Subject: [PATCH 6/9] Rebase
---
.../Transforms/Vectorize/LoopVectorize.cpp | 22 +--
llvm/lib/Transforms/Vectorize/VPlan.cpp | 16 ++-
.../lib/Transforms/Vectorize/VPlanRecipes.cpp | 2 +-
.../LoopVectorize/RISCV/inloop-reduction.ll | 12 +-
.../RISCV/vectorize-vp-intrinsics.ll | 2 +-
.../RISCV/vplan-vp-intrinsics.ll | 2 +-
.../X86/vectorize-vp-intrinsics.ll | 128 +++++++++++++-----
.../LoopVectorize/X86/vplan-vp-intrinsics.ll | 2 +-
.../vectorize-vp-intrinsics-interleave.ll | 2 +-
.../vectorize-vp-intrinsics-iv32.ll | 2 +-
.../LoopVectorize/vectorize-vp-intrinsics.ll | 2 +-
11 files changed, 122 insertions(+), 70 deletions(-)
diff --git a/llvm/lib/Transforms/Vectorize/LoopVectorize.cpp b/llvm/lib/Transforms/Vectorize/LoopVectorize.cpp
index 290c5d6e0a918e..81992883ac7049 100644
--- a/llvm/lib/Transforms/Vectorize/LoopVectorize.cpp
+++ b/llvm/lib/Transforms/Vectorize/LoopVectorize.cpp
@@ -1522,7 +1522,6 @@ class LoopVectorizationCostModel {
if (!Legal->prepareToFoldTailByMasking())
return;
-<<<<<<< HEAD
if (ForceTailFoldingStyle.getNumOccurrences()) {
ChosenTailFoldingStyle.first = ChosenTailFoldingStyle.second =
ForceTailFoldingStyle;
@@ -1558,16 +1557,6 @@ class LoopVectorizationCostModel {
}
return;
}
-||||||| parent of 73e3b4948fb4 (Address comments)
- if (ForceTailFoldingStyle.getNumOccurrences())
- return ForceTailFoldingStyle;
-=======
- if (ChosenTailFoldingStyle)
- return *ChosenTailFoldingStyle;
-
- if (ForceTailFoldingStyle.getNumOccurrences())
- return ForceTailFoldingStyle;
->>>>>>> 73e3b4948fb4 (Address comments)
ChosenTailFoldingStyle.first =
TTI.getPreferredTailFoldingStyle(/*IVUpdateMayOverflow=*/true);
@@ -1742,9 +1731,6 @@ class LoopVectorizationCostModel {
std::pair<TailFoldingStyle, TailFoldingStyle> ChosenTailFoldingStyle =
std::make_pair(TailFoldingStyle::None, TailFoldingStyle::None);
- /// Control finally chosen tail folding style.
- std::optional<TailFoldingStyle> ChosenTailFoldingStyle;
-
/// A map holding scalar costs for different vectorization factors. The
/// presence of a cost for an instruction in the mapping indicates that the
/// instruction will be scalarized when vectorizing with the associated
@@ -9534,8 +9520,8 @@ void VPWidenMemoryInstructionRecipe::execute(VPTransformState &State) {
// FIXME: Support reverse store after vp_reverse is added.
Value *MaskPart = isMaskRequired ? BlockInMaskParts[Part] : nullptr;
NewSI = lowerStoreUsingVectorIntrinsics(
- Builder, State.get(getAddr(), Part), StoredVal, CreateGatherScatter,
- MaskPart, EVL, Alignment);
+ Builder, State.get(getAddr(), Part, !CreateGatherScatter),
+ StoredVal, CreateGatherScatter, MaskPart, EVL, Alignment);
} else if (CreateGatherScatter) {
Value *MaskPart = isMaskRequired ? BlockInMaskParts[Part] : nullptr;
Value *VectorGep = State.get(getAddr(), Part);
@@ -9578,8 +9564,8 @@ void VPWidenMemoryInstructionRecipe::execute(VPTransformState &State) {
// FIXME: Support reverse loading after vp_reverse is added.
Value *MaskPart = isMaskRequired ? BlockInMaskParts[Part] : nullptr;
NewLI = lowerLoadUsingVectorIntrinsics(
- Builder, DataTy, State.get(getAddr(), Part), CreateGatherScatter,
- MaskPart, EVL, Alignment);
+ Builder, DataTy, State.get(getAddr(), Part, !CreateGatherScatter),
+ CreateGatherScatter, MaskPart, EVL, Alignment);
} else if (CreateGatherScatter) {
Value *MaskPart = isMaskRequired ? BlockInMaskParts[Part] : nullptr;
Value *VectorGep = State.get(getAddr(), Part);
diff --git a/llvm/lib/Transforms/Vectorize/VPlan.cpp b/llvm/lib/Transforms/Vectorize/VPlan.cpp
index 9768e4b7aa0a8a..bceddad9800c7d 100644
--- a/llvm/lib/Transforms/Vectorize/VPlan.cpp
+++ b/llvm/lib/Transforms/Vectorize/VPlan.cpp
@@ -883,13 +883,15 @@ void VPlan::execute(VPTransformState *State) {
// only a single part is generated, which provides the last part from the
// previous iteration. For non-ordered reductions all UF parts are
// generated.
- bool SinglePartNeeded = isa<VPCanonicalIVPHIRecipe>(PhiR) ||
- isa<VPFirstOrderRecurrencePHIRecipe>(PhiR) ||
- (isa<VPReductionPHIRecipe>(PhiR) &&
- cast<VPReductionPHIRecipe>(PhiR)->isOrdered());
- bool NeedsScalar = isa<VPCanonicalIVPHIRecipe>(PhiR) ||
- (isa<VPReductionPHIRecipe>(PhiR) &&
- cast<VPReductionPHIRecipe>(PhiR)->isInLoop());
+ bool SinglePartNeeded =
+ isa<VPCanonicalIVPHIRecipe>(PhiR) ||
+ isa<VPFirstOrderRecurrencePHIRecipe, VPEVLBasedIVPHIRecipe>(PhiR) ||
+ (isa<VPReductionPHIRecipe>(PhiR) &&
+ cast<VPReductionPHIRecipe>(PhiR)->isOrdered());
+ bool NeedsScalar =
+ isa<VPCanonicalIVPHIRecipe, VPEVLBasedIVPHIRecipe>(PhiR) ||
+ (isa<VPReductionPHIRecipe>(PhiR) &&
+ cast<VPReductionPHIRecipe>(PhiR)->isInLoop());
unsigned LastPartForNewPhi = SinglePartNeeded ? 1 : State->UF;
for (unsigned Part = 0; Part < LastPartForNewPhi; ++Part) {
diff --git a/llvm/lib/Transforms/Vectorize/VPlanRecipes.cpp b/llvm/lib/Transforms/Vectorize/VPlanRecipes.cpp
index a97e844d6dc1d1..98da507341fa92 100644
--- a/llvm/lib/Transforms/Vectorize/VPlanRecipes.cpp
+++ b/llvm/lib/Transforms/Vectorize/VPlanRecipes.cpp
@@ -1968,7 +1968,7 @@ void VPEVLBasedIVPHIRecipe::execute(VPTransformState &State) {
State.Builder.CreatePHI(Start->getType(), 2, "evl.based.iv");
EntryPart->addIncoming(Start, VectorPH);
EntryPart->setDebugLoc(getDebugLoc());
- State.set(this, EntryPart, 0);
+ State.set(this, EntryPart, 0, /*IsScalar=*/true);
}
#if !defined(NDEBUG) || defined(LLVM_ENABLE_DUMP)
diff --git a/llvm/test/Transforms/LoopVectorize/RISCV/inloop-reduction.ll b/llvm/test/Transforms/LoopVectorize/RISCV/inloop-reduction.ll
index dc2298ed21dfcc..b876e9d2c1a5cd 100644
--- a/llvm/test/Transforms/LoopVectorize/RISCV/inloop-reduction.ll
+++ b/llvm/test/Transforms/LoopVectorize/RISCV/inloop-reduction.ll
@@ -133,21 +133,21 @@ define i32 @add_i16_i32(ptr nocapture readonly %x, i32 %n) {
; IF-EVL-NEXT: [[N_MOD_VF:%.*]] = urem i32 [[N_RND_UP]], [[TMP1]]
; IF-EVL-NEXT: [[N_VEC:%.*]] = sub i32 [[N_RND_UP]], [[N_MOD_VF]]
; IF-EVL-NEXT: [[TRIP_COUNT_MINUS_1:%.*]] = sub i32 [[N]], 1
-; IF-EVL-NEXT: [[BROADCAST_SPLATINSERT:%.*]] = insertelement <vscale x 4 x i32> poison, i32 [[TRIP_COUNT_MINUS_1]], i64 0
-; IF-EVL-NEXT: [[BROADCAST_SPLAT:%.*]] = shufflevector <vscale x 4 x i32> [[BROADCAST_SPLATINSERT]], <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
; IF-EVL-NEXT: [[TMP5:%.*]] = call i32 @llvm.vscale.i32()
; IF-EVL-NEXT: [[TMP6:%.*]] = mul i32 [[TMP5]], 4
+; IF-EVL-NEXT: [[BROADCAST_SPLATINSERT1:%.*]] = insertelement <vscale x 4 x i32> poison, i32 [[TRIP_COUNT_MINUS_1]], i64 0
+; IF-EVL-NEXT: [[BROADCAST_SPLAT2:%.*]] = shufflevector <vscale x 4 x i32> [[BROADCAST_SPLATINSERT1]], <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
; IF-EVL-NEXT: br label [[VECTOR_BODY:%.*]]
; IF-EVL: vector.body:
; IF-EVL-NEXT: [[INDEX:%.*]] = phi i32 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
; IF-EVL-NEXT: [[VEC_PHI:%.*]] = phi <vscale x 4 x i32> [ zeroinitializer, [[VECTOR_PH]] ], [ [[TMP14:%.*]], [[VECTOR_BODY]] ]
; IF-EVL-NEXT: [[TMP7:%.*]] = add i32 [[INDEX]], 0
-; IF-EVL-NEXT: [[BROADCAST_SPLATINSERT1:%.*]] = insertelement <vscale x 4 x i32> poison, i32 [[INDEX]], i64 0
-; IF-EVL-NEXT: [[BROADCAST_SPLAT2:%.*]] = shufflevector <vscale x 4 x i32> [[BROADCAST_SPLATINSERT1]], <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
+; IF-EVL-NEXT: [[BROADCAST_SPLATINSERT:%.*]] = insertelement <vscale x 4 x i32> poison, i32 [[INDEX]], i64 0
+; IF-EVL-NEXT: [[BROADCAST_SPLAT:%.*]] = shufflevector <vscale x 4 x i32> [[BROADCAST_SPLATINSERT]], <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
; IF-EVL-NEXT: [[TMP8:%.*]] = call <vscale x 4 x i32> @llvm.experimental.stepvector.nxv4i32()
; IF-EVL-NEXT: [[TMP9:%.*]] = add <vscale x 4 x i32> zeroinitializer, [[TMP8]]
-; IF-EVL-NEXT: [[VEC_IV:%.*]] = add <vscale x 4 x i32> [[BROADCAST_SPLAT2]], [[TMP9]]
-; IF-EVL-NEXT: [[TMP10:%.*]] = icmp ule <vscale x 4 x i32> [[VEC_IV]], [[BROADCAST_SPLAT]]
+; IF-EVL-NEXT: [[VEC_IV:%.*]] = add <vscale x 4 x i32> [[BROADCAST_SPLAT]], [[TMP9]]
+; IF-EVL-NEXT: [[TMP10:%.*]] = icmp ule <vscale x 4 x i32> [[VEC_IV]], [[BROADCAST_SPLAT2]]
; IF-EVL-NEXT: [[TMP11:%.*]] = getelementptr inbounds i16, ptr [[X:%.*]], i32 [[TMP7]]
; IF-EVL-NEXT: [[TMP12:%.*]] = getelementptr inbounds i16, ptr [[TMP11]], i32 0
; IF-EVL-NEXT: [[WIDE_MASKED_LOAD:%.*]] = call <vscale x 4 x i16> @llvm.masked.load.nxv4i16.p0(ptr [[TMP12]], i32 2, <vscale x 4 x i1> [[TMP10]], <vscale x 4 x i16> poison)
diff --git a/llvm/test/Transforms/LoopVectorize/RISCV/vectorize-vp-intrinsics.ll b/llvm/test/Transforms/LoopVectorize/RISCV/vectorize-vp-intrinsics.ll
index 60d61f25e9f0ca..c69bb17f698aaf 100644
--- a/llvm/test/Transforms/LoopVectorize/RISCV/vectorize-vp-intrinsics.ll
+++ b/llvm/test/Transforms/LoopVectorize/RISCV/vectorize-vp-intrinsics.ll
@@ -6,7 +6,7 @@
; RUN: opt -passes=loop-vectorize \
; RUN: -force-tail-folding-style=none \
-; RUN: -prefer-predicate-over-epilogue=predicate-dont-vectorize \
+; RUN: -prefer-predicate-over-epilogue=predicate-else-scalar-epilogue \
; RUN: -mtriple=riscv64 -mattr=+v -S < %s | FileCheck --check-prefix=NO-VP %s
define void @foo(ptr noalias %a, ptr noalias %b, ptr noalias %c, i64 %N) {
diff --git a/llvm/test/Transforms/LoopVectorize/RISCV/vplan-vp-intrinsics.ll b/llvm/test/Transforms/LoopVectorize/RISCV/vplan-vp-intrinsics.ll
index 9f03d945aceadb..b0417c61b1c4be 100644
--- a/llvm/test/Transforms/LoopVectorize/RISCV/vplan-vp-intrinsics.ll
+++ b/llvm/test/Transforms/LoopVectorize/RISCV/vplan-vp-intrinsics.ll
@@ -7,7 +7,7 @@
; RUN: opt -passes=loop-vectorize -debug-only=loop-vectorize \
; RUN: -force-tail-folding-style=none \
-; RUN: -prefer-predicate-over-epilogue=predicate-dont-vectorize \
+; RUN: -prefer-predicate-over-epilogue=predicate-else-scalar-epilogue \
; RUN: -mtriple=riscv64 -mattr=+v -riscv-v-vector-bits-max=128 -disable-output < %s 2>&1 | FileCheck --check-prefixes=NO-VP,CHECK %s
define void @foo(ptr noalias %a, ptr noalias %b, ptr noalias %c, i64 %N) {
diff --git a/llvm/test/Transforms/LoopVectorize/X86/vectorize-vp-intrinsics.ll b/llvm/test/Transforms/LoopVectorize/X86/vectorize-vp-intrinsics.ll
index cc9711f869d5ef..1cf71360adf723 100644
--- a/llvm/test/Transforms/LoopVectorize/X86/vectorize-vp-intrinsics.ll
+++ b/llvm/test/Transforms/LoopVectorize/X86/vectorize-vp-intrinsics.ll
@@ -6,7 +6,7 @@
; RUN: opt -passes=loop-vectorize \
; RUN: -force-tail-folding-style=none \
-; RUN: -prefer-predicate-over-epilogue=predicate-dont-vectorize \
+; RUN: -prefer-predicate-over-epilogue=predicate-else-scalar-epilogue \
; RUN: -mtriple=x86_64 -mattr=+avx512f -S < %s 2>&1 | FileCheck --check-prefix=NO-VP %s
define void @foo(ptr noalias %a, ptr noalias %b, ptr noalias %c, i64 %N) {
@@ -18,16 +18,16 @@ define void @foo(ptr noalias %a, ptr noalias %b, ptr noalias %c, i64 %N) {
; IF-EVL-NEXT: [[N_MOD_VF:%.*]] = urem i64 [[N_RND_UP]], 16
; IF-EVL-NEXT: [[N_VEC:%.*]] = sub i64 [[N_RND_UP]], [[N_MOD_VF]]
; IF-EVL-NEXT: [[TRIP_COUNT_MINUS_1:%.*]] = sub i64 [[N]], 1
-; IF-EVL-NEXT: [[BROADCAST_SPLATINSERT:%.*]] = insertelement <16 x i64> poison, i64 [[TRIP_COUNT_MINUS_1]], i64 0
-; IF-EVL-NEXT: [[BROADCAST_SPLAT:%.*]] = shufflevector <16 x i64> [[BROADCAST_SPLATINSERT]], <16 x i64> poison, <16 x i32> zeroinitializer
+; IF-EVL-NEXT: [[BROADCAST_SPLATINSERT1:%.*]] = insertelement <16 x i64> poison, i64 [[TRIP_COUNT_MINUS_1]], i64 0
+; IF-EVL-NEXT: [[BROADCAST_SPLAT2:%.*]] = shufflevector <16 x i64> [[BROADCAST_SPLATINSERT1]], <16 x i64> poison, <16 x i32> zeroinitializer
; IF-EVL-NEXT: br label [[VECTOR_BODY:%.*]]
; IF-EVL: vector.body:
; IF-EVL-NEXT: [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
; IF-EVL-NEXT: [[TMP0:%.*]] = add i64 [[INDEX]], 0
-; IF-EVL-NEXT: [[BROADCAST_SPLATINSERT1:%.*]] = insertelement <16 x i64> poison, i64 [[INDEX]], i64 0
-; IF-EVL-NEXT: [[BROADCAST_SPLAT2:%.*]] = shufflevector <16 x i64> [[BROADCAST_SPLATINSERT1]], <16 x i64> poison, <16 x i32> zeroinitializer
-; IF-EVL-NEXT: [[VEC_IV:%.*]] = add <16 x i64> [[BROADCAST_SPLAT2]], <i64 0, i64 1, i64 2, i64 3, i64 4, i64 5, i64 6, i64 7, i64 8, i64 9, i64 10, i64 11, i64 12, i64 13, i64 14, i64 15>
-; IF-EVL-NEXT: [[TMP1:%.*]] = icmp ule <16 x i64> [[VEC_IV]], [[BROADCAST_SPLAT]]
+; IF-EVL-NEXT: [[BROADCAST_SPLATINSERT:%.*]] = insertelement <16 x i64> poison, i64 [[INDEX]], i64 0
+; IF-EVL-NEXT: [[BROADCAST_SPLAT:%.*]] = shufflevector <16 x i64> [[BROADCAST_SPLATINSERT]], <16 x i64> poison, <16 x i32> zeroinitializer
+; IF-EVL-NEXT: [[VEC_IV:%.*]] = add <16 x i64> [[BROADCAST_SPLAT]], <i64 0, i64 1, i64 2, i64 3, i64 4, i64 5, i64 6, i64 7, i64 8, i64 9, i64 10, i64 11, i64 12, i64 13, i64 14, i64 15>
+; IF-EVL-NEXT: [[TMP1:%.*]] = icmp ule <16 x i64> [[VEC_IV]], [[BROADCAST_SPLAT2]]
; IF-EVL-NEXT: [[TMP2:%.*]] = getelementptr inbounds i32, ptr [[B:%.*]], i64 [[TMP0]]
; IF-EVL-NEXT: [[TMP3:%.*]] = getelementptr inbounds i32, ptr [[TMP2]], i32 0
; IF-EVL-NEXT: [[WIDE_MASKED_LOAD:%.*]] = call <16 x i32> @llvm.masked.load.v16i32.p0(ptr [[TMP3]], i32 4, <16 x i1> [[TMP1]], <16 x i32> poison)
@@ -62,47 +62,111 @@ define void @foo(ptr noalias %a, ptr noalias %b, ptr noalias %c, i64 %N) {
; IF-EVL-NEXT: ret void
;
; NO-VP-LABEL: @foo(
-; NO-VP-NEXT: entry:
-; NO-VP-NEXT: [[MIN_ITERS_CHECK:%.*]] = icmp ult i64 [[N:%.*]], 16
-; NO-VP-NEXT: br i1 [[MIN_ITERS_CHECK]], label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
+; NO-VP-NEXT: iter.check:
+; NO-VP-NEXT: [[MIN_ITERS_CHECK:%.*]] = icmp ult i64 [[N:%.*]], 8
+; NO-VP-NEXT: br i1 [[MIN_ITERS_CHECK]], label [[VEC_EPILOG_SCALAR_PH:%.*]], label [[VECTOR_MAIN_LOOP_ITER_CHECK:%.*]]
+; NO-VP: vector.main.loop.iter.check:
+; NO-VP-NEXT: [[MIN_ITERS_CHECK1:%.*]] = icmp ult i64 [[N]], 64
+; NO-VP-NEXT: br i1 [[MIN_ITERS_CHECK1]], label [[VEC_EPILOG_PH:%.*]], label [[VECTOR_PH:%.*]]
; NO-VP: vector.ph:
-; NO-VP-NEXT: [[N_MOD_VF:%.*]] = urem i64 [[N]], 16
+; NO-VP-NEXT: [[N_MOD_VF:%.*]] = urem i64 [[N]], 64
; NO-VP-NEXT: [[N_VEC:%.*]] = sub i64 [[N]], [[N_MOD_VF]]
; NO-VP-NEXT: br label [[VECTOR_BODY:%.*]]
; NO-VP: vector.body:
; NO-VP-NEXT: [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
; NO-VP-NEXT: [[TMP0:%.*]] = add i64 [[INDEX]], 0
-; NO-VP-NEXT: [[TMP1:%.*]] = getelementptr inbounds i32, ptr [[B:%.*]], i64 [[TMP0]]
-; NO-VP-NEXT: [[TMP2:%.*]] = getelementptr inbounds i32, ptr [[TMP1]], i32 0
-; NO-VP-NEXT: [[WIDE_LOAD:%.*]] = load <16 x i32>, ptr [[TMP2]], align 4
-; NO-VP-NEXT: [[TMP3:%.*]] = getelementptr inbounds i32, ptr [[C:%.*]], i64 [[TMP0]]
-; NO-VP-NEXT: [[TMP4:%.*]] = getelementptr inbounds i32, ptr [[TMP3]], i32 0
-; NO-VP-NEXT: [[WIDE_LOAD1:%.*]] = load <16 x i32>, ptr [[TMP4]], align 4
-; NO-VP-NEXT: [[TMP5:%.*]] = add nsw <16 x i32> [[WIDE_LOAD1]], [[WIDE_LOAD]]
-; NO-VP-NEXT: [[TMP6:%.*]] = getelementptr inbounds i32, ptr [[A:%.*]], i64 [[TMP0]]
-; NO-VP-NEXT: [[TMP7:%.*]] = getelementptr inbounds i32, ptr [[TMP6]], i32 0
-; NO-VP-NEXT: store <16 x i32> [[TMP5]], ptr [[TMP7]], align 4
-; NO-VP-NEXT: [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 16
-; NO-VP-NEXT: [[TMP8:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
-; NO-VP-NEXT: br i1 [[TMP8]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]]
+; NO-VP-NEXT: [[TMP1:%.*]] = add i64 [[INDEX]], 16
+; NO-VP-NEXT: [[TMP2:%.*]] = add i64 [[INDEX]], 32
+; NO-VP-NEXT: [[TMP3:%.*]] = add i64 [[INDEX]], 48
+; NO-VP-NEXT: [[TMP4:%.*]] = getelementptr inbounds i32, ptr [[B:%.*]], i64 [[TMP0]]
+; NO-VP-NEXT: [[TMP5:%.*]] = getelementptr inbounds i32, ptr [[B]], i64 [[TMP1]]
+; NO-VP-NEXT: [[TMP6:%.*]] = getelementptr inbounds i32, ptr [[B]], i64 [[TMP2]]
+; NO-VP-NEXT: [[TMP7:%.*]] = getelementptr inbounds i32, ptr [[B]], i64 [[TMP3]]
+; NO-VP-NEXT: [[TMP8:%.*]] = getelementptr inbounds i32, ptr [[TMP4]], i32 0
+; NO-VP-NEXT: [[TMP9:%.*]] = getelementptr inbounds i32, ptr [[TMP4]], i32 16
+; NO-VP-NEXT: [[TMP10:%.*]] = getelementptr inbounds i32, ptr [[TMP4]], i32 32
+; NO-VP-NEXT: [[TMP11:%.*]] = getelementptr inbounds i32, ptr [[TMP4]], i32 48
+; NO-VP-NEXT: [[WIDE_LOAD:%.*]] = load <16 x i32>, ptr [[TMP8]], align 4
+; NO-VP-NEXT: [[WIDE_LOAD2:%.*]] = load <16 x i32>, ptr [[TMP9]], align 4
+; NO-VP-NEXT: [[WIDE_LOAD3:%.*]] = load <16 x i32>, ptr [[TMP10]], align 4
+; NO-VP-NEXT: [[WIDE_LOAD4:%.*]] = load <16 x i32>, ptr [[TMP11]], align 4
+; NO-VP-NEXT: [[TMP12:%.*]] = getelementptr inbounds i32, ptr [[C:%.*]], i64 [[TMP0]]
+; NO-VP-NEXT: [[TMP13:%.*]] = getelementptr inbounds i32, ptr [[C]], i64 [[TMP1]]
+; NO-VP-NEXT: [[TMP14:%.*]] = getelementptr inbounds i32, ptr [[C]], i64 [[TMP2]]
+; NO-VP-NEXT: [[TMP15:%.*]] = getelementptr inbounds i32, ptr [[C]], i64 [[TMP3]]
+; NO-VP-NEXT: [[TMP16:%.*]] = getelementptr inbounds i32, ptr [[TMP12]], i32 0
+; NO-VP-NEXT: [[TMP17:%.*]] = getelementptr inbounds i32, ptr [[TMP12]], i32 16
+; NO-VP-NEXT: [[TMP18:%.*]] = getelementptr inbounds i32, ptr [[TMP12]], i32 32
+; NO-VP-NEXT: [[TMP19:%.*]] = getelementptr inbounds i32, ptr [[TMP12]], i32 48
+; NO-VP-NEXT: [[WIDE_LOAD5:%.*]] = load <16 x i32>, ptr [[TMP16]], align 4
+; NO-VP-NEXT: [[WIDE_LOAD6:%.*]] = load <16 x i32>, ptr [[TMP17]], align 4
+; NO-VP-NEXT: [[WIDE_LOAD7:%.*]] = load <16 x i32>, ptr [[TMP18]], align 4
+; NO-VP-NEXT: [[WIDE_LOAD8:%.*]] = load <16 x i32>, ptr [[TMP19]], align 4
+; NO-VP-NEXT: [[TMP20:%.*]] = add nsw <16 x i32> [[WIDE_LOAD5]], [[WIDE_LOAD]]
+; NO-VP-NEXT: [[TMP21:%.*]] = add nsw <16 x i32> [[WIDE_LOAD6]], [[WIDE_LOAD2]]
+; NO-VP-NEXT: [[TMP22:%.*]] = add nsw <16 x i32> [[WIDE_LOAD7]], [[WIDE_LOAD3]]
+; NO-VP-NEXT: [[TMP23:%.*]] = add nsw <16 x i32> [[WIDE_LOAD8]], [[WIDE_LOAD4]]
+; NO-VP-NEXT: [[TMP24:%.*]] = getelementptr inbounds i32, ptr [[A:%.*]], i64 [[TMP0]]
+; NO-VP-NEXT: [[TMP25:%.*]] = getelementptr inbounds i32, ptr [[A]], i64 [[TMP1]]
+; NO-VP-NEXT: [[TMP26:%.*]] = getelementptr inbounds i32, ptr [[A]], i64 [[TMP2]]
+; NO-VP-NEXT: [[TMP27:%.*]] = getelementptr inbounds i32, ptr [[A]], i64 [[TMP3]]
+; NO-VP-NEXT: [[TMP28:%.*]] = getelementptr inbounds i32, ptr [[TMP24]], i32 0
+; NO-VP-NEXT: [[TMP29:%.*]] = getelementptr inbounds i32, ptr [[TMP24]], i32 16
+; NO-VP-NEXT: [[TMP30:%.*]] = getelementptr inbounds i32, ptr [[TMP24]], i32 32
+; NO-VP-NEXT: [[TMP31:%.*]] = getelementptr inbounds i32, ptr [[TMP24]], i32 48
+; NO-VP-NEXT: store <16 x i32> [[TMP20]], ptr [[TMP28]], align 4
+; NO-VP-NEXT: store <16 x i32> [[TMP21]], ptr [[TMP29]], align 4
+; NO-VP-NEXT: store <16 x i32> [[TMP22]], ptr [[TMP30]], align 4
+; NO-VP-NEXT: store <16 x i32> [[TMP23]], ptr [[TMP31]], align 4
+; NO-VP-NEXT: [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 64
+; NO-VP-NEXT: [[TMP32:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
+; NO-VP-NEXT: br i1 [[TMP32]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]]
; NO-VP: middle.block:
; NO-VP-NEXT: [[CMP_N:%.*]] = icmp eq i64 [[N]], [[N_VEC]]
-; NO-VP-NEXT: br i1 [[CMP_N]], label [[FOR_COND_CLEANUP:%.*]], label [[SCALAR_PH]]
-; NO-VP: scalar.ph:
-; NO-VP-NEXT: [[BC_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ]
+; NO-VP-NEXT: br i1 [[CMP_N]], label [[FOR_COND_CLEANUP:%.*]], label [[VEC_EPILOG_ITER_CHECK:%.*]]
+; NO-VP: vec.epilog.iter.check:
+; NO-VP-NEXT: [[N_VEC_REMAINING:%.*]] = sub i64 [[N]], [[N_VEC]]
+; NO-VP-NEXT: [[MIN_EPILOG_ITERS_CHECK:%.*]] = icmp ult i64 [[N_VEC_REMAINING]], 8
+; NO-VP-NEXT: br i1 [[MIN_EPILOG_ITERS_CHECK]], label [[VEC_EPILOG_SCALAR_PH]], label [[VEC_EPILOG_PH]]
+; NO-VP: vec.epilog.ph:
+; NO-VP-NEXT: [[VEC_EPILOG_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC]], [[VEC_EPILOG_ITER_CHECK]] ], [ 0, [[VECTOR_MAIN_LOOP_ITER_CHECK]] ]
+; NO-VP-NEXT: [[N_MOD_VF9:%.*]] = urem i64 [[N]], 8
+; NO-VP-NEXT: [[N_VEC10:%.*]] = sub i64 [[N]], [[N_MOD_VF9]]
+; NO-VP-NEXT: br label [[VEC_EPILOG_VECTOR_BODY:%.*]]
+; NO-VP: vec.epilog.vector.body:
+; NO-VP-NEXT: [[INDEX12:%.*]] = phi i64 [ [[VEC_EPILOG_RESUME_VAL]], [[VEC_EPILOG_PH]] ], [ [[INDEX_NEXT15:%.*]], [[VEC_EPILOG_VECTOR_BODY]] ]
+; NO-VP-NEXT: [[TMP33:%.*]] = add i64 [[INDEX12]], 0
+; NO-VP-NEXT: [[TMP34:%.*]] = getelementptr inbounds i32, ptr [[B]], i64 [[TMP33]]
+; NO-VP-NEXT: [[TMP35:%.*]] = getelementptr inbounds i32, ptr [[TMP34]], i32 0
+; NO-VP-NEXT: [[WIDE_LOAD13:%.*]] = load <8 x i32>, ptr [[TMP35]], align 4
+; NO-VP-NEXT: [[TMP36:%.*]] = getelementptr inbounds i32, ptr [[C]], i64 [[TMP33]]
+; NO-VP-NEXT: [[TMP37:%.*]] = getelementptr inbounds i32, ptr [[TMP36]], i32 0
+; NO-VP-NEXT: [[WIDE_LOAD14:%.*]] = load <8 x i32>, ptr [[TMP37]], align 4
+; NO-VP-NEXT: [[TMP38:%.*]] = add nsw <8 x i32> [[WIDE_LOAD14]], [[WIDE_LOAD13]]
+; NO-VP-NEXT: [[TMP39:%.*]] = getelementptr inbounds i32, ptr [[A]], i64 [[TMP33]]
+; NO-VP-NEXT: [[TMP40:%.*]] = getelementptr inbounds i32, ptr [[TMP39]], i32 0
+; NO-VP-NEXT: store <8 x i32> [[TMP38]], ptr [[TMP40]], align 4
+; NO-VP-NEXT: [[INDEX_NEXT15]] = add nuw i64 [[INDEX12]], 8
+; NO-VP-NEXT: [[TMP41:%.*]] = icmp eq i64 [[INDEX_NEXT15]], [[N_VEC10]]
+; NO-VP-NEXT: br i1 [[TMP41]], label [[VEC_EPILOG_MIDDLE_BLOCK:%.*]], label [[VEC_EPILOG_VECTOR_BODY]], !llvm.loop [[LOOP3:![0-9]+]]
+; NO-VP: vec.epilog.middle.block:
+; NO-VP-NEXT: [[CMP_N11:%.*]] = icmp eq i64 [[N]], [[N_VEC10]]
+; NO-VP-NEXT: br i1 [[CMP_N11]], label [[FOR_COND_CLEANUP]], label [[VEC_EPILOG_SCALAR_PH]]
+; NO-VP: vec.epilog.scalar.ph:
+; NO-VP-NEXT: [[BC_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC10]], [[VEC_EPILOG_MIDDLE_BLOCK]] ], [ [[N_VEC]], [[VEC_EPILOG_ITER_CHECK]] ], [ 0, [[ITER_CHECK:%.*]] ]
; NO-VP-NEXT: br label [[FOR_BODY:%.*]]
; NO-VP: for.body:
-; NO-VP-NEXT: [[IV:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[IV_NEXT:%.*]], [[FOR_BODY]] ]
+; NO-VP-NEXT: [[IV:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[VEC_EPILOG_SCALAR_PH]] ], [ [[IV_NEXT:%.*]], [[FOR_BODY]] ]
; NO-VP-NEXT: [[ARRAYIDX:%.*]] = getelementptr inbounds i32, ptr [[B]], i64 [[IV]]
-; NO-VP-NEXT: [[TMP9:%.*]] = load i32, ptr [[ARRAYIDX]], align 4
+; NO-VP-NEXT: [[TMP42:%.*]] = load i32, ptr [[ARRAYIDX]], align 4
; NO-VP-NEXT: [[ARRAYIDX2:%.*]] = getelementptr inbounds i32, ptr [[C]], i64 [[IV]]
-; NO-VP-NEXT: [[TMP10:%.*]] = load i32, ptr [[ARRAYIDX2]], align 4
-; NO-VP-NEXT: [[ADD:%.*]] = add nsw i32 [[TMP10]], [[TMP9]]
+; NO-VP-NEXT: [[TMP43:%.*]] = load i32, ptr [[ARRAYIDX2]], align 4
+; NO-VP-NEXT: [[ADD:%.*]] = add nsw i32 [[TMP43]], [[TMP42]]
; NO-VP-NEXT: [[ARRAYIDX4:%.*]] = getelementptr inbounds i32, ptr [[A]], i64 [[IV]]
; NO-VP-NEXT: store i32 [[ADD]], ptr [[ARRAYIDX4]], align 4
; NO-VP-NEXT: [[IV_NEXT]] = add nuw nsw i64 [[IV]], 1
; NO-VP-NEXT: [[EXITCOND_NOT:%.*]] = icmp eq i64 [[IV_NEXT]], [[N]]
-; NO-VP-NEXT: br i1 [[EXITCOND_NOT]], label [[FOR_COND_CLEANUP]], label [[FOR_BODY]], !llvm.loop [[LOOP3:![0-9]+]]
+; NO-VP-NEXT: br i1 [[EXITCOND_NOT]], label [[FOR_COND_CLEANUP]], label [[FOR_BODY]], !llvm.loop [[LOOP4:![0-9]+]]
; NO-VP: for.cond.cleanup:
; NO-VP-NEXT: ret void
;
diff --git a/llvm/test/Transforms/LoopVectorize/X86/vplan-vp-intrinsics.ll b/llvm/test/Transforms/LoopVectorize/X86/vplan-vp-intrinsics.ll
index 395339faaa60d6..9b49d44141db39 100644
--- a/llvm/test/Transforms/LoopVectorize/X86/vplan-vp-intrinsics.ll
+++ b/llvm/test/Transforms/LoopVectorize/X86/vplan-vp-intrinsics.ll
@@ -7,7 +7,7 @@
; RUN: opt -passes=loop-vectorize -debug-only=loop-vectorize -force-vector-width=4 \
; RUN: -force-tail-folding-style=none \
-; RUN: -prefer-predicate-over-epilogue=predicate-dont-vectorize \
+; RUN: -prefer-predicate-over-epilogue=predicate-else-scalar-epilogue \
; RUN: -mtriple=x86_64 -mattr=+avx512f -disable-output < %s 2>&1 | FileCheck --check-prefix=NO-VP %s
define void @foo(ptr noalias %a, ptr noalias %b, ptr noalias %c, i64 %N) {
diff --git a/llvm/test/Transforms/LoopVectorize/vectorize-vp-intrinsics-interleave.ll b/llvm/test/Transforms/LoopVectorize/vectorize-vp-intrinsics-interleave.ll
index 5f528e8cd9ebfc..fa7c0c472f70b9 100644
--- a/llvm/test/Transforms/LoopVectorize/vectorize-vp-intrinsics-interleave.ll
+++ b/llvm/test/Transforms/LoopVectorize/vectorize-vp-intrinsics-interleave.ll
@@ -6,7 +6,7 @@
; RUN: opt -passes=loop-vectorize \
; RUN: -force-tail-folding-style=none \
-; RUN: -prefer-predicate-over-epilogue=predicate-dont-vectorize \
+; RUN: -prefer-predicate-over-epilogue=predicate-else-scalar-epilogue \
; RUN: -force-target-supports-scalable-vectors -scalable-vectorization=on -S < %s | FileCheck --check-prefix=NO-VP %s
; FIXME: interleaved accesses are not supported yet with predicated vectorization.
diff --git a/llvm/test/Transforms/LoopVectorize/vectorize-vp-intrinsics-iv32.ll b/llvm/test/Transforms/LoopVectorize/vectorize-vp-intrinsics-iv32.ll
index 5df0229ded2837..511969ed972fb8 100644
--- a/llvm/test/Transforms/LoopVectorize/vectorize-vp-intrinsics-iv32.ll
+++ b/llvm/test/Transforms/LoopVectorize/vectorize-vp-intrinsics-iv32.ll
@@ -6,7 +6,7 @@
; RUN: opt -passes=loop-vectorize \
; RUN: -force-tail-folding-style=none \
-; RUN: -prefer-predicate-over-epilogue=predicate-dont-vectorize \
+; RUN: -prefer-predicate-over-epilogue=predicate-else-scalar-epilogue \
; RUN: -force-target-supports-scalable-vectors -scalable-vectorization=on -S < %s | FileCheck --check-prefix=NO-VP %s
; The target does not support predicated vectorization.
diff --git a/llvm/test/Transforms/LoopVectorize/vectorize-vp-intrinsics.ll b/llvm/test/Transforms/LoopVectorize/vectorize-vp-intrinsics.ll
index a3625658fcd69c..929889045dbbd7 100644
--- a/llvm/test/Transforms/LoopVectorize/vectorize-vp-intrinsics.ll
+++ b/llvm/test/Transforms/LoopVectorize/vectorize-vp-intrinsics.ll
@@ -6,7 +6,7 @@
; RUN: opt -passes=loop-vectorize \
; RUN: -force-tail-folding-style=none \
-; RUN: -prefer-predicate-over-epilogue=predicate-dont-vectorize \
+; RUN: -prefer-predicate-over-epilogue=predicate-else-scalar-epilogue \
; RUN: -force-target-supports-scalable-vectors -scalable-vectorization=on -S < %s | FileCheck --check-prefix=NO-VP %s
; The target does not support predicated vectorization.
>From 20f08e3dd56e554b04ad731b2cfe1acfc20e4dfd Mon Sep 17 00:00:00 2001
From: Alexey Bataev <a.bataev at outlook.com>
Date: Thu, 29 Feb 2024 06:56:29 -0800
Subject: [PATCH 7/9] Removed ExplicitVectorLengthIVIncrement, replaced by
Instruction::Add
---
llvm/lib/Transforms/Vectorize/VPlan.h | 1 -
llvm/lib/Transforms/Vectorize/VPlanRecipes.cpp | 15 ---------------
llvm/lib/Transforms/Vectorize/VPlanTransforms.cpp | 10 +++++-----
llvm/lib/Transforms/Vectorize/VPlanVerifier.cpp | 14 ++------------
.../LoopVectorize/RISCV/vplan-vp-intrinsics.ll | 2 +-
5 files changed, 8 insertions(+), 34 deletions(-)
diff --git a/llvm/lib/Transforms/Vectorize/VPlan.h b/llvm/lib/Transforms/Vectorize/VPlan.h
index 51ab2edc4e553a..4f1c942907dfaf 100644
--- a/llvm/lib/Transforms/Vectorize/VPlan.h
+++ b/llvm/lib/Transforms/Vectorize/VPlan.h
@@ -1156,7 +1156,6 @@ class VPInstruction : public VPRecipeWithIRFlags {
SLPStore,
ActiveLaneMask,
ExplicitVectorLength,
- ExplicitVectorLengthIVIncrement,
CalculateTripCountMinusVF,
// Increment the canonical IV separately for each unrolled part.
CanonicalIVIncrementForPart,
diff --git a/llvm/lib/Transforms/Vectorize/VPlanRecipes.cpp b/llvm/lib/Transforms/Vectorize/VPlanRecipes.cpp
index 98da507341fa92..83816b7d5c745b 100644
--- a/llvm/lib/Transforms/Vectorize/VPlanRecipes.cpp
+++ b/llvm/lib/Transforms/Vectorize/VPlanRecipes.cpp
@@ -377,17 +377,6 @@ Value *VPInstruction::generateInstruction(VPTransformState &State,
State.EVL = this;
return EVL;
}
- // TODO: remove this once a regular Add VPInstruction is supported.
- case VPInstruction::ExplicitVectorLengthIVIncrement: {
- assert(Part == 0 && "Expected unroll factor 1 for VP vectorization.");
- Value *Phi = State.get(getOperand(0), VPIteration(0, 0));
- Value *EVL = State.get(getOperand(1), VPIteration(0, 0));
- assert(EVL->getType() == Phi->getType() &&
- "EVL and Phi must have the same type.");
- return Builder.CreateAdd(Phi, EVL, Name, hasNoUnsignedWrap(),
- hasNoSignedWrap());
- return EVL;
- }
case VPInstruction::CanonicalIVIncrementForPart: {
auto *IV = State.get(getOperand(0), VPIteration(0, 0));
if (Part == 0)
@@ -573,7 +562,6 @@ bool VPInstruction::onlyFirstLaneUsed(const VPValue *Op) const {
return vputils::onlyFirstLaneUsed(this);
case VPInstruction::ActiveLaneMask:
case VPInstruction::ExplicitVectorLength:
- case VPInstruction::ExplicitVectorLengthIVIncrement:
case VPInstruction::CalculateTripCountMinusVF:
case VPInstruction::CanonicalIVIncrementForPart:
case VPInstruction::BranchOnCount:
@@ -613,9 +601,6 @@ void VPInstruction::print(raw_ostream &O, const Twine &Indent,
case VPInstruction::ExplicitVectorLength:
O << "EXPLICIT-VECTOR-LENGTH";
break;
- case VPInstruction::ExplicitVectorLengthIVIncrement:
- O << "EXPLICIT-VECTOR-LENGTH +";
- break;
case VPInstruction::FirstOrderRecurrenceSplice:
O << "first-order splice";
break;
diff --git a/llvm/lib/Transforms/Vectorize/VPlanTransforms.cpp b/llvm/lib/Transforms/Vectorize/VPlanTransforms.cpp
index 342b352650a00d..db9b7a3c47d0c1 100644
--- a/llvm/lib/Transforms/Vectorize/VPlanTransforms.cpp
+++ b/llvm/lib/Transforms/Vectorize/VPlanTransforms.cpp
@@ -1295,11 +1295,11 @@ void VPlanTransforms::addExplicitVectorLength(VPlan &Plan) {
CanonicalIVPHI->getScalarType());
OpVPEVL->insertBefore(CanonicalIVIncrement);
}
- auto *NextEVLIV = new VPInstruction(
- VPInstruction::ExplicitVectorLengthIVIncrement, {OpVPEVL, EVLPhi},
- {CanonicalIVIncrement->hasNoUnsignedWrap(),
- CanonicalIVIncrement->hasNoSignedWrap()},
- CanonicalIVIncrement->getDebugLoc(), "index.evl.next");
+ auto *NextEVLIV =
+ new VPInstruction(Instruction::Add, {OpVPEVL, EVLPhi},
+ {CanonicalIVIncrement->hasNoUnsignedWrap(),
+ CanonicalIVIncrement->hasNoSignedWrap()},
+ CanonicalIVIncrement->getDebugLoc(), "index.evl.next");
NextEVLIV->insertBefore(CanonicalIVIncrement);
EVLPhi->addOperand(NextEVLIV);
diff --git a/llvm/lib/Transforms/Vectorize/VPlanVerifier.cpp b/llvm/lib/Transforms/Vectorize/VPlanVerifier.cpp
index 37ea02b302102d..fc604c51df7995 100644
--- a/llvm/lib/Transforms/Vectorize/VPlanVerifier.cpp
+++ b/llvm/lib/Transforms/Vectorize/VPlanVerifier.cpp
@@ -95,13 +95,9 @@ static bool verifyVPBasicBlock(const VPBasicBlock *VPBB,
// Set of recipe types along with VPInstruction Opcodes of all EVL-related
// recipes that must appear at most once in Entry or Exiting blocks.
DenseSet<unsigned> EVLFound;
- bool IsHeader = false;
- bool IsExiting = false;
const VPlan *Plan = VPBB->getPlan();
- if (Plan && Plan->getEntry()->getNumSuccessors() == 1) {
- IsHeader = Plan->getVectorLoopRegion()->getEntry() == VPBB;
- IsExiting = Plan->getVectorLoopRegion()->getExiting() == VPBB;
- }
+ bool IsHeader = Plan && Plan->getEntry()->getNumSuccessors() == 1 &&
+ Plan->getVectorLoopRegion()->getEntry() == VPBB;
auto CheckEVLRecipiesInsts = [&](const VPRecipeBase *R) {
if (isa<VPEVLBasedIVPHIRecipe>(R)) {
if (!IsHeader) {
@@ -124,12 +120,6 @@ static bool verifyVPBasicBlock(const VPBasicBlock *VPBB,
return false;
}
break;
- case VPInstruction::ExplicitVectorLengthIVIncrement:
- if (!IsExiting) {
- errs() << "EVL inc instruction not in exit block!\n";
- return false;
- }
- break;
default:
return true;
}
diff --git a/llvm/test/Transforms/LoopVectorize/RISCV/vplan-vp-intrinsics.ll b/llvm/test/Transforms/LoopVectorize/RISCV/vplan-vp-intrinsics.ll
index b0417c61b1c4be..72b881bd44c768 100644
--- a/llvm/test/Transforms/LoopVectorize/RISCV/vplan-vp-intrinsics.ll
+++ b/llvm/test/Transforms/LoopVectorize/RISCV/vplan-vp-intrinsics.ll
@@ -36,7 +36,7 @@ define void @foo(ptr noalias %a, ptr noalias %b, ptr noalias %c, i64 %N) {
; IF-EVL-NEXT: vp<[[PTR3:%[0-9]+]]> = vector-pointer ir<[[GEP3]]>
; IF-EVL-NEXT: WIDEN store vp<[[PTR3]]>, ir<[[ADD]]>, ir<true>
; IF-EVL-NEXT: SCALAR-CAST vp<[[CAST:%[0-9]+]]> = zext vp<[[EVL]]> to i64
-; IF-EVL-NEXT: EMIT vp<[[IV_NEXT]]> = EXPLICIT-VECTOR-LENGTH + vp<[[CAST]]>, vp<[[EVL_PHI]]>
+; IF-EVL-NEXT: EMIT vp<[[IV_NEXT]]> = add vp<[[CAST]]>, vp<[[EVL_PHI]]>
; IF-EVL-NEXT: EMIT vp<[[IV_NEXT_EXIT:%[0-9]+]]> = add vp<[[IV]]>, vp<[[VFUF]]>
; IF-EVL-NEXT: EMIT branch-on-count vp<[[IV_NEXT_EXIT]]>, vp<[[VTC]]>
; IF-EVL-NEXT: No successors
>From 5bac02d64ebb500e5f7eb21cb49f9288fb0b7810 Mon Sep 17 00:00:00 2001
From: Alexey Bataev <a.bataev at outlook.com>
Date: Fri, 1 Mar 2024 11:36:11 -0800
Subject: [PATCH 8/9] Address comments
---
.../Transforms/Vectorize/LoopVectorize.cpp | 6 ++--
.../Transforms/Vectorize/VPlanVerifier.cpp | 29 ++++++++++---------
2 files changed, 19 insertions(+), 16 deletions(-)
diff --git a/llvm/lib/Transforms/Vectorize/LoopVectorize.cpp b/llvm/lib/Transforms/Vectorize/LoopVectorize.cpp
index 81992883ac7049..9fe63078786571 100644
--- a/llvm/lib/Transforms/Vectorize/LoopVectorize.cpp
+++ b/llvm/lib/Transforms/Vectorize/LoopVectorize.cpp
@@ -1580,7 +1580,7 @@ class LoopVectorizationCostModel {
/// Returns true if VP intrinsics with explicit vector length support should
/// be generated in the tail folded loop.
- bool useVPIWithVPEVLVectorization() const {
+ bool foldTailWithEVL() const {
return getTailFoldingStyle() == TailFoldingStyle::DataWithEVL;
}
@@ -5310,7 +5310,7 @@ LoopVectorizationCostModel::selectInterleaveCount(ElementCount VF,
return 1;
// Do not interleave if EVL is preferred and no User IC is specified.
- if (useVPIWithVPEVLVectorization()) {
+ if (foldTailWithEVL()) {
LLVM_DEBUG(dbgs() << "LV: Preference for VP intrinsics indicated. "
"Unroll factor forced to be 1.\n");
return 1;
@@ -8564,7 +8564,7 @@ void LoopVectorizationPlanner::buildVPlansWithVPRecipes(ElementCount MinVF,
VPlanTransforms::truncateToMinimalBitwidths(
*Plan, CM.getMinimalBitwidths(), PSE.getSE()->getContext());
VPlanTransforms::optimize(*Plan, *PSE.getSE());
- if (CM.useVPIWithVPEVLVectorization())
+ if (CM.foldTailWithEVL())
VPlanTransforms::addExplicitVectorLength(*Plan);
assert(verifyVPlanIsValid(*Plan) && "VPlan is invalid");
VPlans.push_back(std::move(Plan));
diff --git a/llvm/lib/Transforms/Vectorize/VPlanVerifier.cpp b/llvm/lib/Transforms/Vectorize/VPlanVerifier.cpp
index fc604c51df7995..365584096dae46 100644
--- a/llvm/lib/Transforms/Vectorize/VPlanVerifier.cpp
+++ b/llvm/lib/Transforms/Vectorize/VPlanVerifier.cpp
@@ -93,8 +93,9 @@ static bool verifyVPBasicBlock(const VPBasicBlock *VPBB,
RecipeNumbering[&R] = Cnt++;
// Set of recipe types along with VPInstruction Opcodes of all EVL-related
- // recipes that must appear at most once in Entry or Exiting blocks.
+ // recipes that must appear at most once in the header block.
DenseSet<unsigned> EVLFound;
+ const VPRecipeBase *VPWidenMemRecipe = nullptr;
const VPlan *Plan = VPBB->getPlan();
bool IsHeader = Plan && Plan->getEntry()->getNumSuccessors() == 1 &&
Plan->getVectorLoopRegion()->getEntry() == VPBB;
@@ -110,23 +111,25 @@ static bool verifyVPBasicBlock(const VPBasicBlock *VPBB,
}
return true;
}
- auto *RInst = dyn_cast<VPInstruction>(R);
- if (!RInst)
- return true;
- switch (RInst->getOpcode()) {
- case VPInstruction::ExplicitVectorLength:
+ if (const auto *RInst = dyn_cast<VPInstruction>(R);
+ RInst && RInst->getOpcode() == VPInstruction::ExplicitVectorLength) {
if (!IsHeader) {
- errs() << "EVL instruction not in entry block!\n";
+ errs() << "EVL instruction not in the header block!\n";
+ return false;
+ }
+ if (!EVLFound.insert(RInst->getOpcode() + VPDef::VPLastPHISC).second) {
+ errs() << "EVL instruction inserted more than once!\n";
+ return false;
+ }
+ if (VPWidenMemRecipe) {
+ errs() << "Use of EVL instruction by widen memory recipe before "
+ "definition!\n";
return false;
}
- break;
- default:
return true;
}
- if (!EVLFound.insert(RInst->getOpcode() + VPDef::VPLastPHISC).second) {
- errs() << "EVL instruction inserted more than once!\n";
- return false;
- }
+ if (isa<VPWidenMemoryInstructionRecipe>(R))
+ VPWidenMemRecipe = R;
return true;
};
>From 73a5f9101591d9c85292662b9d2ddada1b299123 Mon Sep 17 00:00:00 2001
From: Alexey Bataev <a.bataev at outlook.com>
Date: Mon, 4 Mar 2024 10:07:45 -0800
Subject: [PATCH 9/9] Address comments, rebase
---
llvm/lib/Transforms/Vectorize/LoopVectorize.cpp | 6 ++++++
llvm/lib/Transforms/Vectorize/VPlanVerifier.cpp | 2 +-
2 files changed, 7 insertions(+), 1 deletion(-)
diff --git a/llvm/lib/Transforms/Vectorize/LoopVectorize.cpp b/llvm/lib/Transforms/Vectorize/LoopVectorize.cpp
index 9fe63078786571..078b8622e773b3 100644
--- a/llvm/lib/Transforms/Vectorize/LoopVectorize.cpp
+++ b/llvm/lib/Transforms/Vectorize/LoopVectorize.cpp
@@ -9511,6 +9511,9 @@ void VPWidenMemoryInstructionRecipe::execute(VPTransformState &State) {
if (State.EVL) {
assert(State.UF == 1 && "Expected only UF == 1 when vectorizing with "
"explicit vector length.");
+ assert(cast<VPInstruction>(State.EVL)->getOpcode() ==
+ VPInstruction::ExplicitVectorLength &&
+ "EVL must be VPInstruction::ExplicitVectorLength.");
Value *EVL = State.get(State.EVL, VPIteration(0, 0));
// If EVL is not nullptr, then EVL must be a valid value set during plan
// creation, possibly default value = whole vector register length. EVL
@@ -9555,6 +9558,9 @@ void VPWidenMemoryInstructionRecipe::execute(VPTransformState &State) {
if (State.EVL) {
assert(State.UF == 1 && "Expected only UF == 1 when vectorizing with "
"explicit vector length.");
+ assert(cast<VPInstruction>(State.EVL)->getOpcode() ==
+ VPInstruction::ExplicitVectorLength &&
+ "EVL must be VPInstruction::ExplicitVectorLength.");
Value *EVL = State.get(State.EVL, VPIteration(0, 0));
// If EVL is not nullptr, then EVL must be a valid value set during plan
// creation, possibly default value = whole vector register length. EVL
diff --git a/llvm/lib/Transforms/Vectorize/VPlanVerifier.cpp b/llvm/lib/Transforms/Vectorize/VPlanVerifier.cpp
index 365584096dae46..12d37fa711db9f 100644
--- a/llvm/lib/Transforms/Vectorize/VPlanVerifier.cpp
+++ b/llvm/lib/Transforms/Vectorize/VPlanVerifier.cpp
@@ -97,7 +97,7 @@ static bool verifyVPBasicBlock(const VPBasicBlock *VPBB,
DenseSet<unsigned> EVLFound;
const VPRecipeBase *VPWidenMemRecipe = nullptr;
const VPlan *Plan = VPBB->getPlan();
- bool IsHeader = Plan && Plan->getEntry()->getNumSuccessors() == 1 &&
+ bool IsHeader = Plan->getEntry()->getNumSuccessors() == 1 &&
Plan->getVectorLoopRegion()->getEntry() == VPBB;
auto CheckEVLRecipiesInsts = [&](const VPRecipeBase *R) {
if (isa<VPEVLBasedIVPHIRecipe>(R)) {
More information about the llvm-commits
mailing list