intel/llvm
1
0
Fork 0
mirror of https://github.com/intel/llvm.git synced 2026-02-05 13:21:04 +08:00
Code Issues Packages Projects Releases Wiki Activity

Reland "Relax constraints for reduction vectorization"

Browse Source 
Change from original commit: move test (that uses an X86 triple) into the X86
subdirectory.

Original description:
Gating vectorizing reductions on *all* fastmath flags seems unnecessary;
`reassoc` should be sufficient.

Reviewers: tvvikram, mkuper, kristof.beyls, sdesmalen, Ayal

Reviewed By: sdesmalen

Subscribers: dcaballe, huntergr, jmolloy, mcrosier, jlebar, bixia, llvm-commits

Tags: #llvm

Differential Revision: https://reviews.llvm.org/D57728

llvm-svn: 355889
This commit is contained in:
Sanjoy Das
2019-03-12 01:31:44 +00:00
parent 3751ae4a94
commit 3f5ce18658
9 changed files with 180 additions and 34 deletions
Download Patch File Download Diff File Expand all files Collapse all files

15
llvm/include/llvm/Analysis/IVDescriptors.h
View File

@@ -89,10 +89,12 @@ public:
RecurrenceDescriptor() = default;
RecurrenceDescriptor(Value *Start, Instruction *Exit, RecurrenceKind K,
MinMaxRecurrenceKind MK, Instruction *UAI, Type *RT,
bool Signed, SmallPtrSetImpl<Instruction *> &CI)
: StartValue(Start), LoopExitInstr(Exit), Kind(K), MinMaxKind(MK),
UnsafeAlgebraInst(UAI), RecurrenceType(RT), IsSigned(Signed) {
FastMathFlags FMF, MinMaxRecurrenceKind MK,
Instruction *UAI, Type *RT, bool Signed,
SmallPtrSetImpl<Instruction *> &CI)
: StartValue(Start), LoopExitInstr(Exit), Kind(K), FMF(FMF),
MinMaxKind(MK), UnsafeAlgebraInst(UAI), RecurrenceType(RT),
IsSigned(Signed) {
CastInsts.insert(CI.begin(), CI.end());
}
@@ -198,6 +200,8 @@ public:
MinMaxRecurrenceKind getMinMaxRecurrenceKind() { return MinMaxKind; }
FastMathFlags getFastMathFlags() { return FMF; }
TrackingVH<Value> getRecurrenceStartValue() { return StartValue; }
Instruction *getLoopExitInstr() { return LoopExitInstr; }
@@ -237,6 +241,9 @@ private:
Instruction *LoopExitInstr = nullptr;
// The kind of the recurrence.
RecurrenceKind Kind = RK_NoRecurrence;
// The fast-math flags on the recurrent instructions. We propagate these
// fast-math flags into the vectorized FP instructions we generate.
FastMathFlags FMF;
// If this a min/max recurrence the kind of recurrence.
MinMaxRecurrenceKind MinMaxKind = MRK_Invalid;
// First occurrence of unasfe algebra in the PHI's use-chain.

6
llvm/include/llvm/IR/Operator.h
View File

@@ -187,6 +187,12 @@ public:
FastMathFlags() = default;
static FastMathFlags getFast() {
FastMathFlags FMF;
FMF.setFast();
return FMF;
}
bool any() const { return Flags != 0; }
bool none() const { return Flags == 0; }
bool all() const { return Flags == ~0U; }

2
llvm/include/llvm/Transforms/Utils/LoopUtils.h
View File

@@ -296,6 +296,7 @@ getOrderedReduction(IRBuilder<> &Builder, Value *Acc, Value *Src, unsigned Op,
Value *getShuffleReduction(IRBuilder<> &Builder, Value *Src, unsigned Op,
RecurrenceDescriptor::MinMaxRecurrenceKind
MinMaxKind = RecurrenceDescriptor::MRK_Invalid,
FastMathFlags FMF = FastMathFlags(),
ArrayRef<Value *> RedOps = None);
/// Create a target reduction of the given vector. The reduction operation
@@ -308,6 +309,7 @@ Value *createSimpleTargetReduction(IRBuilder<> &B,
unsigned Opcode, Value *Src,
TargetTransformInfo::ReductionFlags Flags =
TargetTransformInfo::ReductionFlags(),
FastMathFlags FMF = FastMathFlags(),
ArrayRef<Value *> RedOps = None);
/// Create a generic target reduction using a recurrence descriptor \p Desc

10
llvm/lib/Analysis/IVDescriptors.cpp
View File

@@ -251,6 +251,10 @@ bool RecurrenceDescriptor::AddReductionVar(PHINode *Phi, RecurrenceKind Kind,
Worklist.push_back(Start);
VisitedInsts.insert(Start);
// Start with all flags set because we will intersect this with the reduction
// flags from all the reduction operations.
FastMathFlags FMF = FastMathFlags::getFast();
// A value in the reduction can be used:
// - By the reduction:
// - Reduction operation:
@@ -296,6 +300,8 @@ bool RecurrenceDescriptor::AddReductionVar(PHINode *Phi, RecurrenceKind Kind,
ReduxDesc = isRecurrenceInstr(Cur, Kind, ReduxDesc, HasFunNoNaNAttr);
if (!ReduxDesc.isRecurrence())
return false;
if (isa<FPMathOperator>(ReduxDesc.getPatternInst()))
FMF &= ReduxDesc.getPatternInst()->getFastMathFlags();
}
bool IsASelect = isa<SelectInst>(Cur);
@@ -441,7 +447,7 @@ bool RecurrenceDescriptor::AddReductionVar(PHINode *Phi, RecurrenceKind Kind,
// Save the description of this reduction variable.
RecurrenceDescriptor RD(
RdxStart, ExitInstruction, Kind, ReduxDesc.getMinMaxKind(),
RdxStart, ExitInstruction, Kind, FMF, ReduxDesc.getMinMaxKind(),
ReduxDesc.getUnsafeAlgebraInst(), RecurrenceType, IsSigned, CastInsts);
RedDes = RD;
@@ -550,7 +556,7 @@ RecurrenceDescriptor::InstDesc
RecurrenceDescriptor::isRecurrenceInstr(Instruction *I, RecurrenceKind Kind,
InstDesc &Prev, bool HasFunNoNaNAttr) {
Instruction *UAI = Prev.getUnsafeAlgebraInst();
if (!UAI && isa<FPMathOperator>(I) && !I->isFast())
if (!UAI && isa<FPMathOperator>(I) && !I->hasAllowReassoc())
UAI = I; // Found an unsafe (unvectorizable) algebra instruction.
switch (I->getOpcode()) {

4
llvm/lib/CodeGen/ExpandReductions.cpp
View File

@@ -118,9 +118,11 @@ bool expandReductions(Function &F, const TargetTransformInfo *TTI) {
}
if (!TTI->shouldExpandReduction(II))
continue;
FastMathFlags FMF =
isa<FPMathOperator>(II) ? II->getFastMathFlags() : FastMathFlags{};
Value *Rdx =
IsOrdered ? getOrderedReduction(Builder, Acc, Vec, getOpcode(ID), MRK)
: getShuffleReduction(Builder, Vec, getOpcode(ID), MRK);
: getShuffleReduction(Builder, Vec, getOpcode(ID), MRK, FMF);
II->replaceAllUsesWith(Rdx);
II->eraseFromParent();
Changed = true;

46
llvm/lib/Transforms/Utils/LoopUtils.cpp
View File

@@ -671,13 +671,9 @@ bool llvm::hasIterationCountInvariantInParent(Loop *InnerLoop,
return true;
}
/// Adds a 'fast' flag to floating point operations.
static Value *addFastMathFlag(Value *V) {
if (isa<FPMathOperator>(V)) {
FastMathFlags Flags;
Flags.setFast();
cast<Instruction>(V)->setFastMathFlags(Flags);
}
static Value *addFastMathFlag(Value *V, FastMathFlags FMF) {
if (isa<FPMathOperator>(V))
cast<Instruction>(V)->setFastMathFlags(FMF);
return V;
}
@@ -761,7 +757,7 @@ llvm::getOrderedReduction(IRBuilder<> &Builder, Value *Acc, Value *Src,
Value *
llvm::getShuffleReduction(IRBuilder<> &Builder, Value *Src, unsigned Op,
RecurrenceDescriptor::MinMaxRecurrenceKind MinMaxKind,
ArrayRef<Value *> RedOps) {
FastMathFlags FMF, ArrayRef<Value *> RedOps) {
unsigned VF = Src->getType()->getVectorNumElements();
// VF is a power of 2 so we can emit the reduction using log2(VF) shuffles
// and vector ops, reducing the set of values being computed by half each
@@ -786,7 +782,8 @@ llvm::getShuffleReduction(IRBuilder<> &Builder, Value *Src, unsigned Op,
if (Op != Instruction::ICmp && Op != Instruction::FCmp) {
// Floating point operations had to be 'fast' to enable the reduction.
TmpVec = addFastMathFlag(Builder.CreateBinOp((Instruction::BinaryOps)Op,
TmpVec, Shuf, "bin.rdx"));
TmpVec, Shuf, "bin.rdx"),
FMF);
} else {
assert(MinMaxKind != RecurrenceDescriptor::MRK_Invalid &&
"Invalid min/max");
@@ -803,7 +800,7 @@ llvm::getShuffleReduction(IRBuilder<> &Builder, Value *Src, unsigned Op,
/// flags (if generating min/max reductions).
Value *llvm::createSimpleTargetReduction(
IRBuilder<> &Builder, const TargetTransformInfo *TTI, unsigned Opcode,
Value *Src, TargetTransformInfo::ReductionFlags Flags,
Value *Src, TargetTransformInfo::ReductionFlags Flags, FastMathFlags FMF,
ArrayRef<Value *> RedOps) {
assert(isa<VectorType>(Src->getType()) && "Type must be a vector");
@@ -873,7 +870,7 @@ Value *llvm::createSimpleTargetReduction(
}
if (TTI->useReductionIntrinsic(Opcode, Src->getType(), Flags))
return BuildFunc();
return getShuffleReduction(Builder, Src, Opcode, MinMaxKind, RedOps);
return getShuffleReduction(Builder, Src, Opcode, MinMaxKind, FMF, RedOps);
}
/// Create a vector reduction using a given recurrence descriptor.
@@ -888,28 +885,37 @@ Value *llvm::createTargetReduction(IRBuilder<> &B,
Flags.NoNaN = NoNaN;
switch (RecKind) {
case RD::RK_FloatAdd:
return createSimpleTargetReduction(B, TTI, Instruction::FAdd, Src, Flags);
return createSimpleTargetReduction(B, TTI, Instruction::FAdd, Src, Flags,
Desc.getFastMathFlags());
case RD::RK_FloatMult:
return createSimpleTargetReduction(B, TTI, Instruction::FMul, Src, Flags);
return createSimpleTargetReduction(B, TTI, Instruction::FMul, Src, Flags,
Desc.getFastMathFlags());
case RD::RK_IntegerAdd:
return createSimpleTargetReduction(B, TTI, Instruction::Add, Src, Flags);
return createSimpleTargetReduction(B, TTI, Instruction::Add, Src, Flags,
Desc.getFastMathFlags());
case RD::RK_IntegerMult:
return createSimpleTargetReduction(B, TTI, Instruction::Mul, Src, Flags);
return createSimpleTargetReduction(B, TTI, Instruction::Mul, Src, Flags,
Desc.getFastMathFlags());
case RD::RK_IntegerAnd:
return createSimpleTargetReduction(B, TTI, Instruction::And, Src, Flags);
return createSimpleTargetReduction(B, TTI, Instruction::And, Src, Flags,
Desc.getFastMathFlags());
case RD::RK_IntegerOr:
return createSimpleTargetReduction(B, TTI, Instruction::Or, Src, Flags);
return createSimpleTargetReduction(B, TTI, Instruction::Or, Src, Flags,
Desc.getFastMathFlags());
case RD::RK_IntegerXor:
return createSimpleTargetReduction(B, TTI, Instruction::Xor, Src, Flags);
return createSimpleTargetReduction(B, TTI, Instruction::Xor, Src, Flags,
Desc.getFastMathFlags());
case RD::RK_IntegerMinMax: {
RD::MinMaxRecurrenceKind MMKind = Desc.getMinMaxRecurrenceKind();
Flags.IsMaxOp = (MMKind == RD::MRK_SIntMax || MMKind == RD::MRK_UIntMax);
Flags.IsSigned = (MMKind == RD::MRK_SIntMax || MMKind == RD::MRK_SIntMin);
return createSimpleTargetReduction(B, TTI, Instruction::ICmp, Src, Flags);
return createSimpleTargetReduction(B, TTI, Instruction::ICmp, Src, Flags,
Desc.getFastMathFlags());
}
case RD::RK_FloatMinMax: {
Flags.IsMaxOp = Desc.getMinMaxRecurrenceKind() == RD::MRK_FloatMax;
return createSimpleTargetReduction(B, TTI, Instruction::FCmp, Src, Flags);
return createSimpleTargetReduction(B, TTI, Instruction::FCmp, Src, Flags,
Desc.getFastMathFlags());
}
default:
llvm_unreachable("Unhandled RecKind");

16
llvm/lib/Transforms/Vectorize/LoopVectorize.cpp
View File

@@ -319,11 +319,14 @@ static unsigned getReciprocalPredBlockProb() { return 2; }
/// A helper function that adds a 'fast' flag to floating-point operations.
static Value *addFastMathFlag(Value *V) {
if (isa<FPMathOperator>(V)) {
FastMathFlags Flags;
Flags.setFast();
cast<Instruction>(V)->setFastMathFlags(Flags);
}
if (isa<FPMathOperator>(V))
cast<Instruction>(V)->setFastMathFlags(FastMathFlags::getFast());
return V;
}
static Value *addFastMathFlag(Value *V, FastMathFlags FMF) {
if (isa<FPMathOperator>(V))
cast<Instruction>(V)->setFastMathFlags(FMF);
return V;
}
@@ -3612,7 +3615,8 @@ void InnerLoopVectorizer::fixReduction(PHINode *Phi) {
// Floating point operations had to be 'fast' to enable the reduction.
ReducedPartRdx = addFastMathFlag(
Builder.CreateBinOp((Instruction::BinaryOps)Op, RdxPart,
ReducedPartRdx, "bin.rdx"));
ReducedPartRdx, "bin.rdx"),
RdxDesc.getFastMathFlags());
else
ReducedPartRdx = createMinMaxOp(Builder, MinMaxKind, ReducedPartRdx,
RdxPart);

3
llvm/lib/Transforms/Vectorize/SLPVectorizer.cpp
View File

@@ -5929,7 +5929,8 @@ private:
if (!IsPairwiseReduction)
return createSimpleTargetReduction(
Builder, TTI, ReductionData.getOpcode(), VectorizedValue,
ReductionData.getFlags(), ReductionOps.back());
ReductionData.getFlags(), FastMathFlags::getFast(),
ReductionOps.back());
Value *TmpVec = VectorizedValue;
for (unsigned i = ReduxWidth / 2; i != 0; i >>= 1) {

112
llvm/test/Transforms/LoopVectorize/X86/reduction-fastmath.ll Normal file
View File

@@ -0,0 +1,112 @@
; RUN: opt -S -loop-vectorize < %s | FileCheck %s
target datalayout = "e-m:e-i64:64-f80:128-n8:16:32:64-S128"
target triple = "x86_64-unknown-linux-gnu"
define float @reduction_sum_float_ieee(i32 %n, float* %array) {
; CHECK-LABEL: define float @reduction_sum_float_ieee(
entry:
%entry.cond = icmp ne i32 0, 4096
br i1 %entry.cond, label %loop, label %loop.exit
loop:
%idx = phi i32 [ 0, %entry ], [ %idx.inc, %loop ]
%sum = phi float [ 0.000000e+00, %entry ], [ %sum.inc, %loop ]
%address = getelementptr float, float* %array, i32 %idx
%value = load float, float* %address
%sum.inc = fadd float %sum, %value
%idx.inc = add i32 %idx, 1
%be.cond = icmp ne i32 %idx.inc, 4096
br i1 %be.cond, label %loop, label %loop.exit
loop.exit:
%sum.lcssa = phi float [ %sum.inc, %loop ], [ 0.000000e+00, %entry ]
; CHECK-NOT: %wide.load = load <4 x float>, <4 x float>*
; CHECK: ret float %sum.lcssa
ret float %sum.lcssa
}
define float @reduction_sum_float_fastmath(i32 %n, float* %array) {
; CHECK-LABEL: define float @reduction_sum_float_fastmath(
; CHECK: fadd fast <4 x float>
; CHECK: fadd fast <4 x float>
; CHECK: fadd fast <4 x float>
; CHECK: fadd fast <4 x float>
; CHECK: fadd fast <4 x float>
entry:
%entry.cond = icmp ne i32 0, 4096
br i1 %entry.cond, label %loop, label %loop.exit
loop:
%idx = phi i32 [ 0, %entry ], [ %idx.inc, %loop ]
%sum = phi float [ 0.000000e+00, %entry ], [ %sum.inc, %loop ]
%address = getelementptr float, float* %array, i32 %idx
%value = load float, float* %address
%sum.inc = fadd fast float %sum, %value
%idx.inc = add i32 %idx, 1
%be.cond = icmp ne i32 %idx.inc, 4096
br i1 %be.cond, label %loop, label %loop.exit
loop.exit:
%sum.lcssa = phi float [ %sum.inc, %loop ], [ 0.000000e+00, %entry ]
; CHECK: ret float %sum.lcssa
ret float %sum.lcssa
}
define float @reduction_sum_float_only_reassoc(i32 %n, float* %array) {
; CHECK-LABEL: define float @reduction_sum_float_only_reassoc(
; CHECK-NOT: fadd fast
; CHECK: fadd reassoc <4 x float>
; CHECK: fadd reassoc <4 x float>
; CHECK: fadd reassoc <4 x float>
; CHECK: fadd reassoc <4 x float>
; CHECK: fadd reassoc <4 x float>
entry:
%entry.cond = icmp ne i32 0, 4096
br i1 %entry.cond, label %loop, label %loop.exit
loop:
%idx = phi i32 [ 0, %entry ], [ %idx.inc, %loop ]
%sum = phi float [ 0.000000e+00, %entry ], [ %sum.inc, %loop ]
%address = getelementptr float, float* %array, i32 %idx
%value = load float, float* %address
%sum.inc = fadd reassoc float %sum, %value
%idx.inc = add i32 %idx, 1
%be.cond = icmp ne i32 %idx.inc, 4096
br i1 %be.cond, label %loop, label %loop.exit
loop.exit:
%sum.lcssa = phi float [ %sum.inc, %loop ], [ 0.000000e+00, %entry ]
; CHECK: ret float %sum.lcssa
ret float %sum.lcssa
}
define float @reduction_sum_float_only_reassoc_and_contract(i32 %n, float* %array) {
; CHECK-LABEL: define float @reduction_sum_float_only_reassoc_and_contract(
; CHECK-NOT: fadd fast
; CHECK: fadd reassoc contract <4 x float>
; CHECK: fadd reassoc contract <4 x float>
; CHECK: fadd reassoc contract <4 x float>
; CHECK: fadd reassoc contract <4 x float>
; CHECK: fadd reassoc contract <4 x float>
entry:
%entry.cond = icmp ne i32 0, 4096
br i1 %entry.cond, label %loop, label %loop.exit
loop:
%idx = phi i32 [ 0, %entry ], [ %idx.inc, %loop ]
%sum = phi float [ 0.000000e+00, %entry ], [ %sum.inc, %loop ]
%address = getelementptr float, float* %array, i32 %idx
%value = load float, float* %address
%sum.inc = fadd reassoc contract float %sum, %value
%idx.inc = add i32 %idx, 1
%be.cond = icmp ne i32 %idx.inc, 4096
br i1 %be.cond, label %loop, label %loop.exit
loop.exit:
%sum.lcssa = phi float [ %sum.inc, %loop ], [ 0.000000e+00, %entry ]
; CHECK: ret float %sum.lcssa
ret float %sum.lcssa
}