[VPlan] Model wrap flags directly, remove *NUW opcodes (NFC)

Model wrap flags directly using VPRecipeWithIRFlags and clean up the
duplicated *NUW opcodes.

D157144 will build on this and also model FMFs for VPInstruction.

Reviewed By: Ayal

Differential Revision: https://reviews.llvm.org/D157194
This commit is contained in:
Florian Hahn committed 2023-08-08 12:12:30 +01:00
1 parent b560d5c7e3
commit af635a5547
15 files changed
+96 -85

No files matched your search

@@ -8727,9 +8727,8 @@ static void addCanonicalIVRecipes(VPlan &Plan, Type *IdxTy, DebugLoc DL,
// IV by VF * UF.
bool HasNUW = Style == TailFoldingStyle::None;
auto *CanonicalIVIncrement =
new VPInstruction(HasNUW ? VPInstruction::CanonicalIVIncrementNUW
: VPInstruction::CanonicalIVIncrement,
{CanonicalIVPHI}, DL, "index.next");
new VPInstruction(VPInstruction::CanonicalIVIncrement, {CanonicalIVPHI},
{HasNUW, false}, DL, "index.next");
CanonicalIVPHI->addOperand(CanonicalIVIncrement);
VPBasicBlock *EB = TopRegion->getExitingBasicBlock();
@@ -8742,9 +8741,8 @@ static void addCanonicalIVRecipes(VPlan &Plan, Type *IdxTy, DebugLoc DL,
// we have to take unrolling into account. Each part needs to start at
// Part * VF
auto *CanonicalIVIncrementParts =
new VPInstruction(HasNUW ? VPInstruction::CanonicalIVIncrementForPartNUW
: VPInstruction::CanonicalIVIncrementForPart,
{StartV}, DL, "index.part.next");
new VPInstruction(VPInstruction::CanonicalIVIncrementForPart, {StartV},
{HasNUW, false}, DL, "index.part.next");
VecPreheader->appendRecipe(CanonicalIVIncrementParts);
// Create the ActiveLaneMask instruction using the correct start values.
@@ -8781,9 +8779,8 @@ static void addCanonicalIVRecipes(VPlan &Plan, Type *IdxTy, DebugLoc DL,
// Create the active lane mask for the next iteration of the loop.
CanonicalIVIncrementParts =
new VPInstruction(HasNUW ? VPInstruction::CanonicalIVIncrementForPartNUW
: VPInstruction::CanonicalIVIncrementForPart,
{IncrementValue}, DL);
new VPInstruction(VPInstruction::CanonicalIVIncrementForPart,
{IncrementValue}, {HasNUW, false}, DL);
EB->appendRecipe(CanonicalIVIncrementParts);
auto *ALM = new VPInstruction(VPInstruction::ActiveLaneMask,
+2 -4
View File
@@ -763,11 +763,9 @@ void VPlan::prepareToExecute(Value *TripCountV, Value *VectorTripCountV,
return true;
auto *VPI = cast<VPInstruction>(U);
return VPI->getOpcode() ==
VPInstruction::CanonicalIVIncrement ||
VPI->getOpcode() ==
VPInstruction::CanonicalIVIncrementNUW;
VPInstruction::CanonicalIVIncrement;
}) &&
"the canonical IV should only be used by its increments or "
"the canonical IV should only be used by its increment or "
"ScalarIVSteps when resetting the start value");
IV->setOperand(0, VPV);
}
+35 -10
View File
@@ -820,10 +820,16 @@ class VPRecipeWithIRFlags : public VPRecipeBase {
FPMathOp,
Other
};
public:
struct WrapFlagsTy {
char HasNUW : 1;
char HasNSW : 1;
WrapFlagsTy(bool HasNUW, bool HasNSW) : HasNUW(HasNUW), HasNSW(HasNSW) {}
};
private:
struct ExactFlagsTy {
char IsExact : 1;
};
@@ -863,8 +869,7 @@ public:
: VPRecipeWithIRFlags(SC, Operands) {
if (auto *Op = dyn_cast<OverflowingBinaryOperator>(&I)) {
OpType = OperationType::OverflowingBinOp;
WrapFlags.HasNUW = Op->hasNoUnsignedWrap();
WrapFlags.HasNSW = Op->hasNoSignedWrap();
WrapFlags = {Op->hasNoUnsignedWrap(), Op->hasNoSignedWrap()};
} else if (auto *Op = dyn_cast<PossiblyExactOperator>(&I)) {
OpType = OperationType::PossiblyExactOp;
ExactFlags.IsExact = Op->isExact();
@@ -884,8 +889,15 @@ public:
}
}
template <typename IterT>
VPRecipeWithIRFlags(const unsigned char SC, IterT Operands,
WrapFlagsTy WrapFlags)
: VPRecipeBase(SC, Operands), OpType(OperationType::OverflowingBinOp),
WrapFlags(WrapFlags) {}
static inline bool classof(const VPRecipeBase *R) {
return R->getVPDefID() == VPRecipeBase::VPWidenSC ||
return R->getVPDefID() == VPRecipeBase::VPInstructionSC ||
R->getVPDefID() == VPRecipeBase::VPWidenSC ||
R->getVPDefID() == VPRecipeBase::VPWidenGEPSC ||
R->getVPDefID() == VPRecipeBase::VPReplicateSC;
}
@@ -949,6 +961,18 @@ public:
FastMathFlags getFastMathFlags() const;
bool hasNoUnsignedWrap() const {
assert(OpType == OperationType::OverflowingBinOp &&
"recipe doesn't have a NUW flag");
return WrapFlags.HasNUW;
}
bool hasNoSignedWrap() const {
assert(OpType == OperationType::OverflowingBinOp &&
"recipe doesn't have a NUW flag");
return WrapFlags.HasNSW;
}
#if !defined(NDEBUG) || defined(LLVM_ENABLE_DUMP)
void printFlags(raw_ostream &O) const;
#endif
@@ -958,7 +982,7 @@ public:
/// While as any Recipe it may generate a sequence of IR instructions when
/// executed, these instructions would always form a single-def expression as
/// the VPInstruction is also a single def-use vertex.
class VPInstruction : public VPRecipeBase, public VPValue {
class VPInstruction : public VPRecipeWithIRFlags, public VPValue {
friend class VPlanSlp;
public:
@@ -974,11 +998,9 @@ public:
ActiveLaneMask,
CalculateTripCountMinusVF,
CanonicalIVIncrement,
CanonicalIVIncrementNUW,
// The next two are similar to the above, but instead increment the
// The next op is similar to the above, but instead increment the
// canonical IV separately for each unrolled part.
CanonicalIVIncrementForPart,
CanonicalIVIncrementForPartNUW,
BranchOnCount,
BranchOnCond
};
@@ -1004,13 +1026,18 @@ protected:
public:
VPInstruction(unsigned Opcode, ArrayRef<VPValue *> Operands, DebugLoc DL,
const Twine &Name = "")
: VPRecipeBase(VPDef::VPInstructionSC, Operands), VPValue(this),
: VPRecipeWithIRFlags(VPDef::VPInstructionSC, Operands), VPValue(this),
Opcode(Opcode), DL(DL), Name(Name.str()) {}
VPInstruction(unsigned Opcode, std::initializer_list<VPValue *> Operands,
DebugLoc DL = {}, const Twine &Name = "")
: VPInstruction(Opcode, ArrayRef<VPValue *>(Operands), DL, Name) {}
VPInstruction(unsigned Opcode, std::initializer_list<VPValue *> Operands,
WrapFlagsTy WrapFlags, DebugLoc DL = {}, const Twine &Name = "")
: VPRecipeWithIRFlags(VPDef::VPInstructionSC, Operands, WrapFlags),
VPValue(this), Opcode(Opcode), DL(DL), Name(Name.str()) {}
VP_CLASSOF_IMPL(VPDef::VPInstructionSC)
VPInstruction *clone() const {
@@ -1079,9 +1106,7 @@ public:
case VPInstruction::ActiveLaneMask:
case VPInstruction::CalculateTripCountMinusVF:
case VPInstruction::CanonicalIVIncrement:
case VPInstruction::CanonicalIVIncrementNUW:
case VPInstruction::CanonicalIVIncrementForPart:
case VPInstruction::CanonicalIVIncrementForPartNUW:
case VPInstruction::BranchOnCount:
return true;
};
+8 -17
View File
@@ -299,23 +299,20 @@ Value *VPInstruction::generateInstruction(VPTransformState &State,
Value *Zero = ConstantInt::get(ScalarTC->getType(), 0);
return Builder.CreateSelect(Cmp, Sub, Zero);
}
case VPInstruction::CanonicalIVIncrement:
case VPInstruction::CanonicalIVIncrementNUW: {
case VPInstruction::CanonicalIVIncrement: {
if (Part == 0) {
bool IsNUW = getOpcode() == VPInstruction::CanonicalIVIncrementNUW;
auto *Phi = State.get(getOperand(0), 0);
// The loop step is equal to the vectorization factor (num of SIMD
// elements) times the unroll factor (num of SIMD instructions).
Value *Step =
createStepForVF(Builder, Phi->getType(), State.VF, State.UF);
return Builder.CreateAdd(Phi, Step, Name, IsNUW, false);
return Builder.CreateAdd(Phi, Step, Name, hasNoUnsignedWrap(),
hasNoSignedWrap());
}
return State.get(this, 0);
}
case VPInstruction::CanonicalIVIncrementForPart:
case VPInstruction::CanonicalIVIncrementForPartNUW: {
bool IsNUW = getOpcode() == VPInstruction::CanonicalIVIncrementForPartNUW;
case VPInstruction::CanonicalIVIncrementForPart: {
auto *IV = State.get(getOperand(0), VPIteration(0, 0));
if (Part == 0)
return IV;
@@ -323,7 +320,7 @@ Value *VPInstruction::generateInstruction(VPTransformState &State,
// The canonical IV is incremented by the vectorization factor (num of SIMD
// elements) times the unroll part.
Value *Step = createStepForVF(Builder, IV->getType(), State.VF, Part);
return Builder.CreateAdd(IV, Step, Name, IsNUW, false);
return Builder.CreateAdd(IV, Step, Name, hasNoUnsignedWrap(), false);
}
case VPInstruction::BranchOnCond: {
if (Part != 0)
@@ -425,9 +422,6 @@ void VPInstruction::print(raw_ostream &O, const Twine &Indent,
case VPInstruction::CanonicalIVIncrement:
O << "VF * UF +";
break;
case VPInstruction::CanonicalIVIncrementNUW:
O << "VF * UF +(nuw)";
break;
case VPInstruction::BranchOnCond:
O << "branch-on-cond";
break;
@@ -437,9 +431,6 @@ void VPInstruction::print(raw_ostream &O, const Twine &Indent,
case VPInstruction::CanonicalIVIncrementForPart:
O << "VF * Part +";
break;
case VPInstruction::CanonicalIVIncrementForPartNUW:
O << "VF * Part +(nuw)";
break;
case VPInstruction::BranchOnCount:
O << "branch-on-count";
break;
@@ -448,8 +439,7 @@ void VPInstruction::print(raw_ostream &O, const Twine &Indent,
}
O << FMF;
if (getNumOperands() > 0)
O << " ";
printFlags(O);
printOperands(O, SlotTracker);
if (DL) {
@@ -608,7 +598,8 @@ void VPRecipeWithIRFlags::printFlags(raw_ostream &O) const {
case OperationType::Other:
break;
}
O << " ";
if (getNumOperands() > 0)
O << " ";
}
#endif
@@ -25,7 +25,7 @@ target triple = "aarch64-unknown-linux-gnu"
; CHECK-NEXT: WIDEN ir<%lv> = load ir<%ptr.iv.2>
; CHECK-NEXT: WIDEN ir<%add> = add ir<%lv>, ir<1>
; CHECK-NEXT: WIDEN store ir<%ptr.iv.2>, ir<%add>
; CHECK-NEXT: EMIT vp<[[CAN_IV_NEXT:%.+]]> = VF * UF +(nuw) vp<[[CAN_IV]]>
; CHECK-NEXT: EMIT vp<[[CAN_IV_NEXT:%.+]]> = VF * UF + nuw vp<[[CAN_IV]]>
; CHECK-NEXT: EMIT branch-on-count vp<[[CAN_IV_NEXT]]>, vp<[[VEC_TC]]>
; CHECK-NEXT: No successors
; CHECK-NEXT: }
@@ -25,8 +25,8 @@ target triple = "aarch64-unknown-linux-gnu"
; CHECK-NEXT: REPLICATE ir<%call> = call @foo(ir<%load>)
; CHECK-NEXT: CLONE ir<%arrayidx> = getelementptr inbounds ir<%a>, vp<[[STEPS]]>
; CHECK-NEXT: WIDEN store ir<%arrayidx>, ir<%call>
; CHECK-NEXT: EMIT vp<[[CAN_IV_NEXT:%.+]]> = VF * UF +(nuw) vp<[[CAN_IV]]>
; CHECK-NEXT: EMIT branch-on-count vp<[[CAN_IV_NEXT]]>, vp<[[VTC]]>
; CHECK-NEXT: EMIT vp<[[CAN_IV_NEXT:%.+]]> = VF * UF + nuw vp<[[CAN_IV]]>
; CHECK-NEXT: EMIT branch-on-count vp<[[CAN_IV_NEXT]]>, vp<[[VTC]]>
; CHECK-NEXT: No successors
; CHECK-NEXT: }
; CHECK-NEXT: Successor(s): middle.block
@@ -51,7 +51,7 @@ target triple = "aarch64-unknown-linux-gnu"
; CHECK-NEXT: WIDEN-CALL ir<%call> = call @foo(ir<%load>) (using library function: foo_vector_fixed4_nomask)
; CHECK-NEXT: CLONE ir<%arrayidx> = getelementptr inbounds ir<%a>, vp<[[STEPS]]>
; CHECK-NEXT: WIDEN store ir<%arrayidx>, ir<%call>
; CHECK-NEXT: EMIT vp<[[CAN_IV_NEXT:%.+]]> = VF * UF +(nuw) vp<[[CAN_IV]]>
; CHECK-NEXT: EMIT vp<[[CAN_IV_NEXT:%.+]]> = VF * UF + nuw vp<[[CAN_IV]]>
; CHECK-NEXT: EMIT branch-on-count vp<[[CAN_IV_NEXT]]>, vp<[[VTC]]>
; CHECK-NEXT: No successors
; CHECK-NEXT: }
@@ -82,7 +82,7 @@ target triple = "aarch64-unknown-linux-gnu"
; CHECK-NEXT: WIDEN-CALL ir<%call> = call @foo(ir<%load>) (using library function: foo_vector_fixed2_nomask)
; CHECK-NEXT: CLONE ir<%arrayidx> = getelementptr inbounds ir<%a>, vp<[[STEPS]]>
; CHECK-NEXT: WIDEN store ir<%arrayidx>, ir<%call>
; CHECK-NEXT: EMIT vp<[[CAN_IV_NEXST:%.+]]> = VF * UF +(nuw) vp<[[CAN_IV]]>
; CHECK-NEXT: EMIT vp<[[CAN_IV_NEXST:%.+]]> = VF * UF + nuw vp<[[CAN_IV]]>
; CHECK-NEXT: EMIT branch-on-count vp<[[CAN_IV_NEXT]]>, vp<[[VTC]]>
; CHECK-NEXT: No successors
; CHECK-NEXT: }
@@ -108,7 +108,7 @@ target triple = "aarch64-unknown-linux-gnu"
; CHECK-NEXT: WIDEN-CALL ir<%call> = call @foo(ir<%load>, ir<true>) (using library function: foo_vector_fixed4_mask)
; CHECK-NEXT: CLONE ir<%arrayidx> = getelementptr inbounds ir<%a>, vp<[[STEPS]]>
; CHECK-NEXT: WIDEN store ir<%arrayidx>, ir<%call>
; CHECK-NEXT: EMIT vp<[[CAN_IV_NEXT:%.+]]> = VF * UF +(nuw) vp<[[CAN_IV]]>
; CHECK-NEXT: EMIT vp<[[CAN_IV_NEXT:%.+]]> = VF * UF + nuw vp<[[CAN_IV]]>
; CHECK-NEXT: EMIT branch-on-count vp<[[CAN_IV_NEXT]]>, vp<[[VTC]]>
; CHECK-NEXT: No successors
; CHECK-NEXT: }
@@ -138,8 +138,8 @@ target triple = "aarch64-unknown-linux-gnu"
; CHECK-NEXT: WIDEN-CALL ir<%call> = call @foo(ir<%load>) (using library function: foo_vector_fixed2_nomask)
; CHECK-NEXT: CLONE ir<%arrayidx> = getelementptr inbounds ir<%a>, vp<[[STEPS]]>
; CHECK-NEXT: WIDEN store ir<%arrayidx>, ir<%call>
; CHECK-NEXT: EMIT vp<[[CAN_IV_NEXT:%.+]]> = VF * UF +(nuw) vp<[[CAN_IV]]>
; CHECK-NEXT: EMIT branch-on-count vp<[[CAN_IV_NEXT]]>, vp<[[VTC]]>
; CHECK-NEXT: EMIT vp<[[CAN_IV_NEXT:%.+]]> = VF * UF + nuw vp<[[CAN_IV]]>
; CHECK-NEXT: EMIT branch-on-count vp<[[CAN_IV_NEXT]]>, vp<[[VTC]]>
; CHECK-NEXT: No successors
; CHECK-NEXT: }
; CHECK-NEXT: Successor(s): middle.block
@@ -164,7 +164,7 @@ target triple = "aarch64-unknown-linux-gnu"
; CHECK-NEXT: WIDEN-CALL ir<%call> = call @foo(ir<%load>) (using library function: foo_vector_fixed4_nomask)
; CHECK-NEXT: CLONE ir<%arrayidx> = getelementptr inbounds ir<%a>, vp<[[STEPS]]>
; CHECK-NEXT: WIDEN store ir<%arrayidx>, ir<%call>
; CHECK-NEXT: EMIT vp<[[CAN_IV_NEXT:%.+]]> = VF * UF +(nuw) vp<[[CAN_IV]]>
; CHECK-NEXT: EMIT vp<[[CAN_IV_NEXT:%.+]]> = VF * UF + nuw vp<[[CAN_IV]]>
; CHECK-NEXT: EMIT branch-on-count vp<[[CAN_IV_NEXT]]>, vp<[[VTC]]>
; CHECK-NEXT: No successors
; CHECK-NEXT: }
@@ -23,7 +23,7 @@ target triple = "arm64-apple-ios"
; CHECK-NEXT: WIDEN-CALL ir<%s> = call @llvm.sin.f64(ir<%conv>) (using library function: __simd_sin_v2f64)
; CHECK-NEXT: REPLICATE ir<%gep.dst> = getelementptr inbounds ir<%dst>, vp<[[STEPS]]>
; CHECK-NEXT: REPLICATE store ir<%s>, ir<%gep.dst>
; CHECK-NEXT: EMIT vp<[[CAN_IV_NEXT:%.+]]> = VF * UF +(nuw) vp<[[CAN_IV]]>
; CHECK-NEXT: EMIT vp<[[CAN_IV_NEXT:%.+]]> = VF * UF + nuw vp<[[CAN_IV]]>
; CHECK-NEXT: EMIT branch-on-count vp<[[CAN_IV_NEXT]]>, vp<[[VTC]]>
; CHECK-NEXT: No successors
; CHECK-NEXT: }
@@ -50,7 +50,7 @@ target triple = "arm64-apple-ios"
; CHECK-NEXT: WIDEN-CALL ir<%s> = call @llvm.sin.f64(ir<%conv>) (using vector intrinsic)
; CHECK-NEXT: REPLICATE ir<%gep.dst> = getelementptr inbounds ir<%dst>, vp<[[STEPS]]>
; CHECK-NEXT: REPLICATE store ir<%s>, ir<%gep.dst>
; CHECK-NEXT: EMIT vp<[[CAN_IV_NEXT:%.+]]> = VF * UF +(nuw) vp<[[CAN_IV]]>
; CHECK-NEXT: EMIT vp<[[CAN_IV_NEXT:%.+]]> = VF * UF + nuw vp<[[CAN_IV]]>
; CHECK-NEXT: EMIT branch-on-count vp<[[CAN_IV_NEXT]]>, vp<[[VTC]]>
; CHECK-NEXT: No successors
; CHECK-NEXT: }
@@ -70,7 +70,7 @@ define void @vector_reverse_i64(ptr nocapture noundef writeonly %A, ptr nocaptur
; CHECK-NEXT: WIDEN ir<%add9> = add ir<%1>, ir<1>
; CHECK-NEXT: CLONE ir<%arrayidx3> = getelementptr inbounds ir<%A>, ir<%idxprom>
; CHECK-NEXT: WIDEN store ir<%arrayidx3>, ir<%add9>
; CHECK-NEXT: EMIT vp<%11> = VF * UF +(nuw) vp<%2>
; CHECK-NEXT: EMIT vp<%11> = VF * UF + nuw vp<%2>
; CHECK-NEXT: EMIT branch-on-count vp<%11>, vp<%0>
; CHECK-NEXT: No successors
; CHECK-NEXT: }
@@ -206,7 +206,7 @@ define void @vector_reverse_f32(ptr nocapture noundef writeonly %A, ptr nocaptur
; CHECK-NEXT: WIDEN ir<%conv1> = fadd ir<%1>, ir<1.000000e+00>
; CHECK-NEXT: CLONE ir<%arrayidx3> = getelementptr inbounds ir<%A>, ir<%idxprom>
; CHECK-NEXT: WIDEN store ir<%arrayidx3>, ir<%conv1>
; CHECK-NEXT: EMIT vp<%11> = VF * UF +(nuw) vp<%2>
; CHECK-NEXT: EMIT vp<%11> = VF * UF + nuw vp<%2>
; CHECK-NEXT: EMIT branch-on-count vp<%11>, vp<%0>
; CHECK-NEXT: No successors
; CHECK-NEXT: }
@@ -23,8 +23,8 @@ define void @test_chained_first_order_recurrences_1(ptr %ptr) {
; CHECK-NEXT: EMIT vp<[[FOR2_SPLICE:%.+]]> = first-order splice ir<%for.2>, vp<[[FOR1_SPLICE]]>
; CHECK-NEXT: WIDEN ir<%add> = add vp<[[FOR1_SPLICE]]>, vp<[[FOR2_SPLICE]]>
; CHECK-NEXT: WIDEN store ir<%gep.ptr>, ir<%add>
; CHECK-NEXT: EMIT vp<[[CAN_IV_NEXT:%.+]]> = VF * UF +(nuw) vp<[[CAN_IV]]>
; CHECK-NEXT: EMIT branch-on-count vp<[[CAN_IV_NEXT]]>, vp<[[VTC]]>
; CHECK-NEXT: EMIT vp<[[CAN_IV_NEXT:%.+]]> = VF * UF + nuw vp<[[CAN_IV]]>
; CHECK-NEXT: EMIT branch-on-count vp<[[CAN_IV_NEXT]]>, vp<[[VTC]]>
; CHECK-NEXT: No successors
; CHECK-NEXT: }
; CHECK-NEXT: Successor(s): middle.block
@@ -76,8 +76,8 @@ define void @test_chained_first_order_recurrences_3(ptr %ptr) {
; CHECK-NEXT: WIDEN ir<%add.1> = add vp<[[FOR1_SPLICE]]>, vp<[[FOR2_SPLICE]]>
; CHECK-NEXT: WIDEN ir<%add.2> = add ir<%add.1>, vp<[[FOR3_SPLICE]]>
; CHECK-NEXT: WIDEN store ir<%gep.ptr>, ir<%add.2>
; CHECK-NEXT: EMIT vp<[[CAN_IV_NEXT:%.+]]> = VF * UF +(nuw) vp<[[CAN_IV]]>
; CHECK-NEXT: EMIT branch-on-count vp<[[CAN_IV_NEXT]]>, vp<[[VTC]]>
; CHECK-NEXT: EMIT vp<[[CAN_IV_NEXT:%.+]]> = VF * UF + nuw vp<[[CAN_IV]]>
; CHECK-NEXT: EMIT branch-on-count vp<[[CAN_IV_NEXT]]>, vp<[[VTC]]>
; CHECK-NEXT: No successors
; CHECK-NEXT: }
; CHECK-NEXT: Successor(s): middle.block
@@ -23,8 +23,8 @@
; DBG-NEXT: CLONE ir<%min> = call @llvm.smin.i32(vp<[[IV_STEPS]]>, ir<65535>)
; DBG-NEXT: CLONE ir<%arrayidx> = getelementptr inbounds ir<%dst>, vp<[[IV_STEPS]]>
; DBG-NEXT: CLONE store ir<%min>, ir<%arrayidx>
; DBG-NEXT: EMIT vp<[[INC:%.+]]> = VF * UF +(nuw) vp<[[CAN_IV]]>
; DBG-NEXT: EMIT branch-on-count vp<[[INC]]>, vp<[[VEC_TC]]>
; DBG-NEXT: EMIT vp<[[INC:%.+]]> = VF * UF + nuw vp<[[CAN_IV]]>
; DBG-NEXT: EMIT branch-on-count vp<[[INC]]>, vp<[[VEC_TC]]>
; DBG-NEXT: No successors
; DBG-NEXT: }
;
@@ -100,8 +100,8 @@ declare i32 @llvm.smin.i32(i32, i32)
; DBG-NEXT: Successor(s): cond.false.1
; DBG-EMPTY:
; DBG-NEXT: cond.false.1:
; DBG-NEXT: EMIT vp<[[CAN_IV_INC:%.+]]> = VF * UF +(nuw) vp<[[CAN_IV]]>
; DBG-NEXT: EMIT branch-on-count vp<[[CAN_IV_INC]]>, vp<[[VEC_TC]]>
; DBG-NEXT: EMIT vp<[[CAN_IV_INC:%.+]]> = VF * UF + nuw vp<[[CAN_IV]]>
; DBG-NEXT: EMIT branch-on-count vp<[[CAN_IV_INC]]>, vp<[[VEC_TC]]>
; DBG-NEXT: No successors
; DBG-NEXT: }
; DBG-NEXT: Successor(s): middle.block
@@ -191,8 +191,8 @@ exit:
; DBG-NEXT: vp<[[SCALAR_STEPS]]> = SCALAR-STEPS vp<[[DERIVED_IV]]>, ir<1>
; DBG-NEXT: EMIT vp<[[SPLICE:%.+]]> = first-order splice ir<%for>, vp<[[SCALAR_STEPS]]>
; DBG-NEXT: CLONE store vp<[[SPLICE]]>, ir<%dst>
; DBG-NEXT: EMIT vp<[[IV_INC:%.+]]> = VF * UF +(nuw) vp<[[CAN_IV]]>
; DBG-NEXT: EMIT branch-on-count vp<[[IV_INC]]>, vp<[[VTC]]>
; DBG-NEXT: EMIT vp<[[IV_INC:%.+]]> = VF * UF + nuw vp<[[CAN_IV]]>
; DBG-NEXT: EMIT branch-on-count vp<[[IV_INC]]>, vp<[[VTC]]>
; DBG-NEXT: No successors
; DBG-NEXT: }
; DBG-NEXT: Successor(s): middle.block
@@ -33,7 +33,7 @@ define void @print_call_and_memory(i64 %n, ptr noalias %y, ptr noalias %x) nounw
; CHECK-NEXT: " WIDEN-CALL ir\<%call\> = call @llvm.sqrt.f32(ir\<%lv\>) (using vector intrinsic)\l" +
; CHECK-NEXT: " CLONE ir\<%arrayidx2\> = getelementptr inbounds ir\<%x\>, vp\<[[STEPS]]\>\l" +
; CHECK-NEXT: " WIDEN store ir\<%arrayidx2\>, ir\<%call\>\l" +
; CHECK-NEXT: " EMIT vp\<[[CAN_IV_NEXT:%.+]]\> = VF * UF +(nuw) vp\<[[CAN_IV]]\>\l" +
; CHECK-NEXT: " EMIT vp\<[[CAN_IV_NEXT:%.+]]\> = VF * UF + nuw vp\<[[CAN_IV]]\>\l" +
; CHECK-NEXT: " EMIT branch-on-count vp\<[[CAN_IV_NEXT]]\>, vp\<{{.+}}\>\l" +
; CHECK-NEXT: "No successors\l"
; CHECK-NEXT: ]
@@ -17,8 +17,8 @@ define void @iv_no_binary_op_in_descriptor(i1 %c, ptr %dst) {
; CHECK-NEXT: vp<[[STEPS:%.+]]> = SCALAR-STEPS vp<[[CAN_IV]]>, ir<1>
; CHECK-NEXT: CLONE ir<%gep> = getelementptr inbounds ir<%dst>, vp<[[STEPS:%.+]]>
; CHECK-NEXT: WIDEN store ir<%gep>, ir<%iv>
; CHECK-NEXT: EMIT vp<[[CAN_INC:%.+]]> = VF * UF +(nuw) vp<[[CAN_IV]]>
; CHECK-NEXT: EMIT branch-on-count vp<[[CAN_INC]]>, vp<[[VEC_TC]]>
; CHECK-NEXT: EMIT vp<[[CAN_INC:%.+]]> = VF * UF + nuw vp<[[CAN_IV]]>
; CHECK-NEXT: EMIT branch-on-count vp<[[CAN_INC]]>, vp<[[VEC_TC]]>
; CHECK-NEXT: No successors
; CHECK-NEXT: }
; CHECK-NEXT: Successor(s): middle.block
@@ -24,7 +24,7 @@ define void @print_call_and_memory(i64 %n, ptr noalias %y, ptr noalias %x) nounw
; CHECK-NEXT: WIDEN-CALL ir<%call> = call @llvm.sqrt.f32(ir<%lv>)
; CHECK-NEXT: CLONE ir<%arrayidx2> = getelementptr inbounds ir<%x>, vp<[[STEPS]]>
; CHECK-NEXT: WIDEN store ir<%arrayidx2>, ir<%call>
; CHECK-NEXT: EMIT vp<[[CAN_IV_NEXT:%.+]]> = VF * UF +(nuw) vp<[[CAN_IV]]>
; CHECK-NEXT: EMIT vp<[[CAN_IV_NEXT:%.+]]> = VF * UF + nuw vp<[[CAN_IV]]>
; CHECK-NEXT: EMIT branch-on-count vp<[[CAN_IV_NEXT]]>, vp<[[VEC_TC]]>
; CHECK-NEXT: No successors
; CHECK-NEXT: }
@@ -74,7 +74,7 @@ define void @print_widen_gep_and_select(i64 %n, ptr noalias %y, ptr noalias %x,
; CHECK-NEXT: WIDEN ir<%add> = fadd ir<%lv>, ir<%sel>
; CHECK-NEXT: CLONE ir<%arrayidx2> = getelementptr inbounds ir<%x>, vp<[[STEPS]]>
; CHECK-NEXT: WIDEN store ir<%arrayidx2>, ir<%add>
; CHECK-NEXT: EMIT vp<[[CAN_IV_NEXT:%.+]]> = VF * UF +(nuw) vp<[[CAN_IV]]>
; CHECK-NEXT: EMIT vp<[[CAN_IV_NEXT:%.+]]> = VF * UF + nuw vp<[[CAN_IV]]>
; CHECK-NEXT: EMIT branch-on-count vp<[[CAN_IV_NEXT]]>, vp<[[VEC_TC]]>
; CHECK-NEXT: No successors
; CHECK-NEXT: }
@@ -122,7 +122,7 @@ define float @print_reduction(i64 %n, ptr noalias %y) {
; CHECK-NEXT: CLONE ir<%arrayidx> = getelementptr inbounds ir<%y>, vp<[[STEPS]]>
; CHECK-NEXT: WIDEN ir<%lv> = load ir<%arrayidx>
; CHECK-NEXT: REDUCE ir<%red.next> = ir<%red> + fast reduce.fadd (ir<%lv>)
; CHECK-NEXT: EMIT vp<[[CAN_IV_NEXT:%.+]]> = VF * UF +(nuw) vp<[[CAN_IV]]>
; CHECK-NEXT: EMIT vp<[[CAN_IV_NEXT:%.+]]> = VF * UF + nuw vp<[[CAN_IV]]>
; CHECK-NEXT: EMIT branch-on-count vp<[[CAN_IV_NEXT]]>, vp<[[VEC_TC]]>
; CHECK-NEXT: No successors
; CHECK-NEXT: }
@@ -168,7 +168,7 @@ define void @print_reduction_with_invariant_store(i64 %n, ptr noalias %y, ptr no
; CHECK-NEXT: CLONE ir<%arrayidx> = getelementptr inbounds ir<%y>, vp<[[IV]]>
; CHECK-NEXT: WIDEN ir<%lv> = load ir<%arrayidx>
; CHECK-NEXT: REDUCE ir<%red.next> = ir<%red> + fast reduce.fadd (ir<%lv>) (with final reduction value stored in invariant address sank outside of loop)
; CHECK-NEXT: EMIT vp<[[CAN_IV_NEXT:%.+]]> = VF * UF +(nuw) vp<[[CAN_IV]]>
; CHECK-NEXT: EMIT vp<[[CAN_IV_NEXT:%.+]]> = VF * UF + nuw vp<[[CAN_IV]]>
; CHECK-NEXT: EMIT branch-on-count vp<[[CAN_IV_NEXT]]>, vp<[[VEC_TC]]>
; CHECK-NEXT: No successors
; CHECK-NEXT: }
@@ -237,7 +237,7 @@ define void @print_replicate_predicated_phi(i64 %n, ptr %x) {
; CHECK-NEXT: BLEND %d = ir<0>/vp<[[NOT]]> vp<[[PRED]]>/ir<%cmp>
; CHECK-NEXT: CLONE ir<%idx> = getelementptr ir<%x>, vp<[[STEPS]]>
; CHECK-NEXT: WIDEN store ir<%idx>, ir<%d>
; CHECK-NEXT: EMIT vp<[[CAN_IV_NEXT:%.+]]> = VF * UF +(nuw) vp<[[CAN_IV]]>
; CHECK-NEXT: EMIT vp<[[CAN_IV_NEXT:%.+]]> = VF * UF + nuw vp<[[CAN_IV]]>
; CHECK-NEXT: EMIT branch-on-count vp<[[CAN_IV_NEXT]]>, vp<[[VEC_TC]]>
; CHECK-NEXT: No successors
; CHECK-NEXT: }
@@ -301,7 +301,7 @@ define void @print_interleave_groups(i32 %C, i32 %D) {
; CHECK-NEXT: store ir<1> to index 1
; CHECK-NEXT: store ir<2> to index 2
; CHECK-NEXT: store ir<%AB.3> to index 3
; CHECK-NEXT: EMIT vp<[[CAN_IV_NEXT:%.+]]> = VF * UF +(nuw) vp<[[CAN_IV]]>
; CHECK-NEXT: EMIT vp<[[CAN_IV_NEXT:%.+]]> = VF * UF + nuw vp<[[CAN_IV]]>
; CHECK-NEXT: EMIT branch-on-count vp<[[CAN_IV_NEXT]]>, vp<[[VEC_TC]]>
; CHECK-NEXT: No successors
; CHECK-NEXT: }
@@ -362,7 +362,7 @@ define float @print_fmuladd_strict(ptr %a, ptr %b, i64 %n) {
; CHECK-NEXT: WIDEN ir<%l.b> = load ir<%arrayidx2>
; CHECK-NEXT: EMIT vp<[[FMUL:%.+]]> = fmul nnan ninf nsz ir<%l.a>, ir<%l.b>
; CHECK-NEXT: REDUCE ir<[[MULADD:%.+]]> = ir<%sum.07> + nnan ninf nsz reduce.fadd (vp<[[FMUL]]>)
; CHECK-NEXT: EMIT vp<[[CAN_IV_NEXT:%.+]]> = VF * UF +(nuw) vp<[[CAN_IV]]>
; CHECK-NEXT: EMIT vp<[[CAN_IV_NEXT:%.+]]> = VF * UF + nuw vp<[[CAN_IV]]>
; CHECK-NEXT: EMIT branch-on-count vp<[[CAN_IV_NEXT]]>, vp<[[VEC_TC]]>
; CHECK-NEXT: No successors
; CHECK-NEXT: }
@@ -436,7 +436,7 @@ define void @debug_loc_vpinstruction(ptr nocapture %asd, ptr nocapture %bsd) !db
; CHECK-NEXT: EMIT vp<[[SEL2:%.+]]> = select vp<[[NOT1]]>, vp<[[NOT2]]>, ir<false>
; CHECK-NEXT: BLEND %ysd.0 = vp<[[PHI]]>/vp<[[OR1]]> ir<%psd>/vp<[[SEL2]]>
; CHECK-NEXT: WIDEN store ir<%isd>, ir<%ysd.0>
; CHECK-NEXT: EMIT vp<[[CAN_IV_NEXT:%.+]]> = VF * UF +(nuw) vp<[[CAN_IV]]>
; CHECK-NEXT: EMIT vp<[[CAN_IV_NEXT:%.+]]> = VF * UF + nuw vp<[[CAN_IV]]>
; CHECK-NEXT: EMIT branch-on-count vp<[[CAN_IV_NEXT]]>, vp<[[VEC_TC]]>
; CHECK-NEXT: No successors
; CHECK-NEXT:}
@@ -504,7 +504,7 @@ define void @print_expand_scev(i64 %y, ptr %ptr) {
; CHECK-NEXT: WIDEN ir<%v3> = add nuw ir<%v2>, ir<1>
; CHECK-NEXT: REPLICATE ir<%gep> = getelementptr inbounds ir<%ptr>, vp<[[STEPS]]>
; CHECK-NEXT: REPLICATE store ir<%v3>, ir<%gep>
; CHECK-NEXT: EMIT vp<[[CAN_INC:%.+]]> = VF * UF +(nuw) vp<[[CAN_IV]]>
; CHECK-NEXT: EMIT vp<[[CAN_INC:%.+]]> = VF * UF + nuw vp<[[CAN_IV]]>
; CHECK-NEXT: EMIT branch-on-count vp<[[CAN_INC]]>, vp<[[VTC]]>
; CHECK-NEXT: No successors
; CHECK-NEXT: }
@@ -551,8 +551,8 @@ define i32 @print_exit_value(ptr %ptr, i32 %off) {
; CHECK-NEXT: CLONE ir<%gep> = getelementptr inbounds ir<%ptr>, vp<[[STEPS]]>
; CHECK-NEXT: WIDEN ir<%add> = add ir<%iv>, ir<%off>
; CHECK-NEXT: WIDEN store ir<%gep>, ir<0>
; CHECK-NEXT: EMIT vp<[[CAN_IV_NEXT:%.+]]> = VF * UF +(nuw) vp<[[CAN_IV]]>
; CHECK-NEXT: EMIT branch-on-count vp<[[CAN_IV_NEXT]]>, vp<[[VEC_TC]]>
; CHECK-NEXT: EMIT vp<[[CAN_IV_NEXT:%.+]]> = VF * UF + nuw vp<[[CAN_IV]]>
; CHECK-NEXT: EMIT branch-on-count vp<[[CAN_IV_NEXT]]>, vp<[[VEC_TC]]>
; CHECK-NEXT: No successors
; CHECK-NEXT: }
; CHECK-NEXT: Successor(s): middle.block
@@ -600,7 +600,7 @@ define void @print_fast_math_flags(i64 %n, ptr noalias %y, ptr noalias %x, ptr %
; CHECK-NEXT: WIDEN ir<%div> = fdiv reassoc nsz contract ir<%mul>, ir<2.000000e+00>
; CHECK-NEXT: CLONE ir<%gep.x> = getelementptr inbounds ir<%x>, vp<[[STEPS]]>
; CHECK-NEXT: WIDEN store ir<%gep.x>, ir<%div>
; CHECK-NEXT: EMIT vp<[[CAN_IV_NEXT:%.+]]> = VF * UF +(nuw) vp<[[CAN_IV]]>
; CHECK-NEXT: EMIT vp<[[CAN_IV_NEXT:%.+]]> = VF * UF + nuw vp<[[CAN_IV]]>
; CHECK-NEXT: EMIT branch-on-count vp<[[CAN_IV_NEXT]]>, vp<[[VEC_TC]]>
; CHECK-NEXT: No successors
; CHECK-NEXT: }
@@ -649,7 +649,7 @@ define void @print_exact_flags(i64 %n, ptr noalias %x) {
; CHECK-NEXT: WIDEN ir<%div.2> = udiv ir<%lv>, ir<60>
; CHECK-NEXT: WIDEN ir<%add> = add nuw nsw ir<%div.1>, ir<%div.2>
; CHECK-NEXT: WIDEN store ir<%gep.x>, ir<%add>
; CHECK-NEXT: EMIT vp<[[CAN_IV_NEXT:%.+]]> = VF * UF +(nuw) vp<[[CAN_IV]]>
; CHECK-NEXT: EMIT vp<[[CAN_IV_NEXT:%.+]]> = VF * UF + nuw vp<[[CAN_IV]]>
; CHECK-NEXT: EMIT branch-on-count vp<[[CAN_IV_NEXT]]>, vp<[[VEC_TC]]>
; CHECK-NEXT: No successors
; CHECK-NEXT: }
@@ -719,7 +719,7 @@ define void @print_call_flags(ptr readonly %src, ptr noalias %dest, i64 %n) {
; CHECK-NEXT: BLEND %st.value = ir<%ld.value>/vp<%11> ir<%fadd>/ir<%ifcond>
; CHECK-NEXT: CLONE ir<%st.addr> = getelementptr inbounds ir<%dest>, vp<%2>
; CHECK-NEXT: WIDEN store ir<%st.addr>, ir<%st.value>
; CHECK-NEXT: EMIT vp<[[CAN_IV_NEXT:%.+]]> = VF * UF +(nuw) vp<[[CAN_IV]]>
; CHECK-NEXT: EMIT vp<[[CAN_IV_NEXT:%.+]]> = VF * UF + nuw vp<[[CAN_IV]]>
; CHECK-NEXT: EMIT branch-on-count vp<[[CAN_IV_NEXT]]>, vp<[[VEC_TC]]>
; CHECK-NEXT: No successors
; CHECK-NEXT: }
@@ -39,7 +39,7 @@ define void @sink_with_sideeffects(i1 %c, ptr %ptr) {
; CHECK-NEXT: }
; CHECK: if.then.0:
; CHECK-NEXT: EMIT vp<[[CAN_IV_NEXT:%.+]]> = VF * UF +(nuw) vp<[[CAN_IV]]>
; CHECK-NEXT: EMIT vp<[[CAN_IV_NEXT:%.+]]> = VF * UF + nuw vp<[[CAN_IV]]>
; CHECK-NEXT: EMIT branch-on-count vp<[[CAN_IV_NEXT]]>, vp<[[VEC_TC]]>
; CHECK-NEXT: No successors
; CHECK-NEXT: }
@@ -918,7 +918,7 @@ define void @update_multiple_users(ptr noalias %src, ptr noalias %dst, i1 %c) {
; CHECK-NEXT: Successor(s): loop.then.1
; CHECK-EMPTY:
; CHECK-NEXT: loop.then.1:
; CHECK-NEXT: EMIT vp<[[CAN_IV_NEXT:%.+]]> = VF * UF +(nuw) vp<[[CAN_IV]]>
; CHECK-NEXT: EMIT vp<[[CAN_IV_NEXT:%.+]]> = VF * UF + nuw vp<[[CAN_IV]]>
; CHECK-NEXT: EMIT branch-on-count vp<[[CAN_IV_NEXT]]>, vp<[[VEC_TC]]>
; CHECK-NEXT: No successors
; CHECK-NEXT: }
@@ -983,7 +983,7 @@ define void @sinking_requires_duplication(ptr %addr) {
; CHECK-NEXT: Successor(s): then.0
; CHECK-EMPTY:
; CHECK-NEXT: then.0:
; CHECK-NEXT: EMIT vp<[[CAN_IV_NEXT:%.+]]> = VF * UF +(nuw) vp<[[CAN_IV]]>
; CHECK-NEXT: EMIT vp<[[CAN_IV_NEXT:%.+]]> = VF * UF + nuw vp<[[CAN_IV]]>
; CHECK-NEXT: EMIT branch-on-count vp<[[CAN_IV_NEXT]]>, vp<[[VEC_TC]]>
; CHECK-NEXT: No successors
; CHECK-NEXT: }
@@ -1121,8 +1121,8 @@ define void @ptr_induction_remove_dead_recipe(ptr %start, ptr %end) {
; CHECK-NEXT: Successor(s): if.then.0
; CHECK-EMPTY:
; CHECK-NEXT: if.then.0:
; CHECK-NEXT: EMIT vp<[[CAN_IV_NEXT:%.+]]> = VF * UF +(nuw) vp<[[CAN_IV]]>
; CHECK-NEXT: EMIT branch-on-count vp<[[CAN_IV_NEXT]]>, vp<[[VEC_TC]]>
; CHECK-NEXT: EMIT vp<[[CAN_IV_NEXT:%.+]]> = VF * UF + nuw vp<[[CAN_IV]]>
; CHECK-NEXT: EMIT branch-on-count vp<[[CAN_IV_NEXT]]>, vp<[[VEC_TC]]>
; CHECK-NEXT: No successors
; CHECK-NEXT: }
; CHECK-NEXT: Successor(s): middle.block