[llvm] r231458 - Add a new pass "Loop Interchange"
Kaylor, Andrew
andrew.kaylor at intel.com
Fri Apr 24 10:32:53 PDT 2015
I can commit the change. I have a workspace set up with just that patch in it.
-Andy
From: Karthik Bhat [mailto:blitz.opensource at gmail.com]
Sent: Friday, April 24, 2015 9:38 AM
To: Aaron Ballman
Cc: Karthik Bhat; llvm-commits; Kaylor, Andrew
Subject: Re: [llvm] r231458 - Add a new pass "Loop Interchange"
Hi Aaron,
I feel we can use removeChildLoop as we want to move sub-loops inside the inner loop into outer loop after loops are interchanged. LoopSimplify and LoopInfo currently use this API.
But I agree that part of code was badly written and was bound to fail. Unfortunately was not able to catch it during our testing.
I feel Andrew Kaylor's patch with modification should be good. I had tested the patch in morning and it seems to work fine.
Unfortunately I'm not having a workspace access as of now to commit the patch..:( Creating the same.
Sorry once again for the noise.
Thanks and Regards
Karthik Bhat
On Fri, Apr 24, 2015 at 9:47 PM, Aaron Ballman <aaron at aaronballman.com<mailto:aaron at aaronballman.com>> wrote:
Ah, I hadn't noticed that patch! Yes, it looks good to me to commit it, but we may still want to consider whether the removal api is the best approach.
-Aaron
On Apr 24, 2015 11:53 AM, "Karthik Bhat" <blitz.opensource at gmail.com<mailto:blitz.opensource at gmail.com>> wrote:
Hi Aaron,
Sorry for the delay..
Kyler had proposed a fix which looks good to me.. I have added my comments on the same.. I think it would be submitted shortly.
http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20150420/273084.html
Shall i go ahead and make those changes?
Thanks and Regards
Karthik Bhat
On 24-Apr-2015 8:25 pm, "Aaron Ballman" <aaron at aaronballman.com<mailto:aaron at aaronballman.com>> wrote:
I am still seeing these failures on Windows; can you please address
them (I suspect reversion would be a horribly painful option I would
like to avoid).
Thanks!
~Aaron
On Thu, Apr 23, 2015 at 11:07 AM, Aaron Ballman <aaron at aaronballman.com<mailto:aaron at aaronballman.com>> wrote:
> On Fri, Mar 6, 2015 at 5:11 AM, Karthik Bhat <kv.bhat at samsung.com<mailto:kv.bhat at samsung.com>> wrote:
>> Author: karthik
>> Date: Fri Mar 6 04:11:25 2015
>> New Revision: 231458
>>
>> URL: http://llvm.org/viewvc/llvm-project?rev=231458&view=rev
>> Log:
>> Add a new pass "Loop Interchange"
>> This pass interchanges loops to provide a more cache-friendly memory access.
>>
>> For e.g. given a loop like -
>> for(int i=0;i<N;i++)
>> for(int j=0;j<N;j++)
>> A[j][i] = A[j][i]+B[j][i];
>>
>> is interchanged to -
>> for(int j=0;j<N;j++)
>> for(int i=0;i<N;i++)
>> A[j][i] = A[j][i]+B[j][i];
>>
>> This pass is currently disabled by default.
>>
>> To give a brief introduction it consists of 3 stages-
>>
>> LoopInterchangeLegality : Checks the legality of loop interchange based on Dependency matrix.
>> LoopInterchangeProfitability: A very basic heuristic has been added to check for profitibility. This will evolve over time.
>> LoopInterchangeTransform : Which does the actual transform.
>>
>> LNT Performance tests shows improvement in Polybench/linear-algebra/kernels/mvt and Polybench/linear-algebra/kernels/gemver becnmarks.
>>
>> TODO:
>> 1) Add support for reductions and lcssa phi.
>> 2) Improve profitability model.
>> 3) Improve loop selection algorithm to select best loop for interchange. Currently the innermost loop is selected for interchange.
>> 4) Improve compile time regression found in llvm lnt due to this pass.
>> 5) Fix issues in Dependency Analysis module.
>>
>> A special thanks to Hal for reviewing this code.
>> Review: http://reviews.llvm.org/D7499
>>
>>
>>
>> Added:
>> llvm/trunk/lib/Transforms/Scalar/LoopInterchange.cpp
>> llvm/trunk/test/Transforms/LoopInterchange/
>> llvm/trunk/test/Transforms/LoopInterchange/currentLimitation.ll
>> llvm/trunk/test/Transforms/LoopInterchange/interchange.ll
>> llvm/trunk/test/Transforms/LoopInterchange/profitability.ll
>> Modified:
>> llvm/trunk/include/llvm/InitializePasses.h
>> llvm/trunk/include/llvm/LinkAllPasses.h
>> llvm/trunk/include/llvm/Transforms/Scalar.h
>> llvm/trunk/lib/Transforms/IPO/PassManagerBuilder.cpp
>> llvm/trunk/lib/Transforms/Scalar/CMakeLists.txt
>> llvm/trunk/lib/Transforms/Scalar/Scalar.cpp
>>
>> Modified: llvm/trunk/include/llvm/InitializePasses.h
>> URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/include/llvm/InitializePasses.h?rev=231458&r1=231457&r2=231458&view=diff
>> ==============================================================================
>> --- llvm/trunk/include/llvm/InitializePasses.h (original)
>> +++ llvm/trunk/include/llvm/InitializePasses.h Fri Mar 6 04:11:25 2015
>> @@ -166,6 +166,7 @@ void initializeLocalStackSlotPassPass(Pa
>> void initializeLoopDeletionPass(PassRegistry&);
>> void initializeLoopExtractorPass(PassRegistry&);
>> void initializeLoopInfoWrapperPassPass(PassRegistry&);
>> +void initializeLoopInterchangePass(PassRegistry &);
>> void initializeLoopInstSimplifyPass(PassRegistry&);
>> void initializeLoopRotatePass(PassRegistry&);
>> void initializeLoopSimplifyPass(PassRegistry&);
>>
>> Modified: llvm/trunk/include/llvm/LinkAllPasses.h
>> URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/include/llvm/LinkAllPasses.h?rev=231458&r1=231457&r2=231458&view=diff
>> ==============================================================================
>> --- llvm/trunk/include/llvm/LinkAllPasses.h (original)
>> +++ llvm/trunk/include/llvm/LinkAllPasses.h Fri Mar 6 04:11:25 2015
>> @@ -95,6 +95,7 @@ namespace {
>> (void) llvm::createLICMPass();
>> (void) llvm::createLazyValueInfoPass();
>> (void) llvm::createLoopExtractorPass();
>> + (void)llvm::createLoopInterchangePass();
>> (void) llvm::createLoopSimplifyPass();
>> (void) llvm::createLoopStrengthReducePass();
>> (void) llvm::createLoopRerollPass();
>>
>> Modified: llvm/trunk/include/llvm/Transforms/Scalar.h
>> URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/include/llvm/Transforms/Scalar.h?rev=231458&r1=231457&r2=231458&view=diff
>> ==============================================================================
>> --- llvm/trunk/include/llvm/Transforms/Scalar.h (original)
>> +++ llvm/trunk/include/llvm/Transforms/Scalar.h Fri Mar 6 04:11:25 2015
>> @@ -140,6 +140,13 @@ Pass *createLICMPass();
>>
>> //===----------------------------------------------------------------------===//
>> //
>> +// LoopInterchange - This pass interchanges loops to provide a more
>> +// cache-friendly memory access patterns.
>> +//
>> +Pass *createLoopInterchangePass();
>> +
>> +//===----------------------------------------------------------------------===//
>> +//
>> // LoopStrengthReduce - This pass is strength reduces GEP instructions that use
>> // a loop's canonical induction variable as one of their indices.
>> //
>>
>> Modified: llvm/trunk/lib/Transforms/IPO/PassManagerBuilder.cpp
>> URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/lib/Transforms/IPO/PassManagerBuilder.cpp?rev=231458&r1=231457&r2=231458&view=diff
>> ==============================================================================
>> --- llvm/trunk/lib/Transforms/IPO/PassManagerBuilder.cpp (original)
>> +++ llvm/trunk/lib/Transforms/IPO/PassManagerBuilder.cpp Fri Mar 6 04:11:25 2015
>> @@ -77,6 +77,10 @@ static cl::opt<bool>
>> EnableMLSM("mlsm", cl::init(true), cl::Hidden,
>> cl::desc("Enable motion of merged load and store"));
>>
>> +static cl::opt<bool> EnableLoopInterchange(
>> + "enable-loopinterchange", cl::init(false), cl::Hidden,
>> + cl::desc("Enable the new, experimental LoopInterchange Pass"));
>> +
>> PassManagerBuilder::PassManagerBuilder() {
>> OptLevel = 2;
>> SizeLevel = 0;
>> @@ -239,6 +243,8 @@ void PassManagerBuilder::populateModuleP
>> MPM.add(createIndVarSimplifyPass()); // Canonicalize indvars
>> MPM.add(createLoopIdiomPass()); // Recognize idioms like memset.
>> MPM.add(createLoopDeletionPass()); // Delete dead loops
>> + if (EnableLoopInterchange)
>> + MPM.add(createLoopInterchangePass()); // Interchange loops
>>
>> if (!DisableUnrollLoops)
>> MPM.add(createSimpleLoopUnrollPass()); // Unroll small loops
>> @@ -454,6 +460,9 @@ void PassManagerBuilder::addLTOOptimizat
>> // More loops are countable; try to optimize them.
>> PM.add(createIndVarSimplifyPass());
>> PM.add(createLoopDeletionPass());
>> + if (EnableLoopInterchange)
>> + PM.add(createLoopInterchangePass());
>> +
>> PM.add(createLoopVectorizePass(true, LoopVectorize));
>>
>> // More scalar chains could be vectorized due to more alias information
>>
>> Modified: llvm/trunk/lib/Transforms/Scalar/CMakeLists.txt
>> URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/lib/Transforms/Scalar/CMakeLists.txt?rev=231458&r1=231457&r2=231458&view=diff
>> ==============================================================================
>> --- llvm/trunk/lib/Transforms/Scalar/CMakeLists.txt (original)
>> +++ llvm/trunk/lib/Transforms/Scalar/CMakeLists.txt Fri Mar 6 04:11:25 2015
>> @@ -18,6 +18,7 @@ add_llvm_library(LLVMScalarOpts
>> LoopDeletion.cpp
>> LoopIdiomRecognize.cpp
>> LoopInstSimplify.cpp
>> + LoopInterchange.cpp
>> LoopRerollPass.cpp
>> LoopRotation.cpp
>> LoopStrengthReduce.cpp
>>
>> Added: llvm/trunk/lib/Transforms/Scalar/LoopInterchange.cpp
>> URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/lib/Transforms/Scalar/LoopInterchange.cpp?rev=231458&view=auto
>> ==============================================================================
>> --- llvm/trunk/lib/Transforms/Scalar/LoopInterchange.cpp (added)
>> +++ llvm/trunk/lib/Transforms/Scalar/LoopInterchange.cpp Fri Mar 6 04:11:25 2015
>> @@ -0,0 +1,1193 @@
>> +//===- LoopInterchange.cpp - Loop interchange pass------------------------===//
>> +//
>> +// The LLVM Compiler Infrastructure
>> +//
>> +// This file is distributed under the University of Illinois Open Source
>> +// License. See LICENSE.TXT for details.
>> +//
>> +//===----------------------------------------------------------------------===//
>> +//
>> +// This Pass handles loop interchange transform.
>> +// This pass interchanges loops to provide a more cache-friendly memory access
>> +// patterns.
>> +//
>> +//===----------------------------------------------------------------------===//
>> +
>> +#include "llvm/ADT/SmallVector.h"
>> +#include "llvm/Analysis/AliasAnalysis.h"
>> +#include "llvm/Analysis/AliasSetTracker.h"
>> +#include "llvm/Analysis/AssumptionCache.h"
>> +#include "llvm/Analysis/BlockFrequencyInfo.h"
>> +#include "llvm/Analysis/CodeMetrics.h"
>> +#include "llvm/Analysis/DependenceAnalysis.h"
>> +#include "llvm/Analysis/LoopInfo.h"
>> +#include "llvm/Analysis/LoopIterator.h"
>> +#include "llvm/Analysis/LoopPass.h"
>> +#include "llvm/Analysis/ScalarEvolution.h"
>> +#include "llvm/Analysis/ScalarEvolutionExpander.h"
>> +#include "llvm/Analysis/ScalarEvolutionExpressions.h"
>> +#include "llvm/Analysis/TargetTransformInfo.h"
>> +#include "llvm/Analysis/ValueTracking.h"
>> +#include "llvm/Transforms/Scalar.h"
>> +#include "llvm/IR/Function.h"
>> +#include "llvm/IR/IRBuilder.h"
>> +#include "llvm/IR/IntrinsicInst.h"
>> +#include "llvm/IR/InstIterator.h"
>> +#include "llvm/IR/Dominators.h"
>> +#include "llvm/Pass.h"
>> +#include "llvm/Support/Debug.h"
>> +#include "llvm/Transforms/Utils/SSAUpdater.h"
>> +#include "llvm/Support/raw_ostream.h"
>> +#include "llvm/Transforms/Utils/LoopUtils.h"
>> +#include "llvm/Transforms/Utils/BasicBlockUtils.h"
>> +using namespace llvm;
>> +
>> +#define DEBUG_TYPE "loop-interchange"
>> +
>> +namespace {
>> +
>> +typedef SmallVector<Loop *, 8> LoopVector;
>> +
>> +// TODO: Check if we can use a sparse matrix here.
>> +typedef std::vector<std::vector<char>> CharMatrix;
>> +
>> +// Maximum number of dependencies that can be handled in the dependency matrix.
>> +static const unsigned MaxMemInstrCount = 100;
>> +
>> +// Maximum loop depth supported.
>> +static const unsigned MaxLoopNestDepth = 10;
>> +
>> +struct LoopInterchange;
>> +
>> +#ifdef DUMP_DEP_MATRICIES
>> +void printDepMatrix(CharMatrix &DepMatrix) {
>> + for (auto I = DepMatrix.begin(), E = DepMatrix.end(); I != E; ++I) {
>> + std::vector<char> Vec = *I;
>> + for (auto II = Vec.begin(), EE = Vec.end(); II != EE; ++II)
>> + DEBUG(dbgs() << *II << " ");
>> + DEBUG(dbgs() << "\n");
>> + }
>> +}
>> +#endif
>> +
>> +bool populateDependencyMatrix(CharMatrix &DepMatrix, unsigned Level, Loop *L,
>> + DependenceAnalysis *DA) {
>> + typedef SmallVector<Value *, 16> ValueVector;
>> + ValueVector MemInstr;
>> +
>> + if (Level > MaxLoopNestDepth) {
>> + DEBUG(dbgs() << "Cannot handle loops of depth greater than "
>> + << MaxLoopNestDepth << "\n");
>> + return false;
>> + }
>> +
>> + // For each block.
>> + for (Loop::block_iterator BB = L->block_begin(), BE = L->block_end();
>> + BB != BE; ++BB) {
>> + // Scan the BB and collect legal loads and stores.
>> + for (BasicBlock::iterator I = (*BB)->begin(), E = (*BB)->end(); I != E;
>> + ++I) {
>> + Instruction *Ins = dyn_cast<Instruction>(I);
>> + if (!Ins)
>> + return false;
>> + LoadInst *Ld = dyn_cast<LoadInst>(I);
>> + StoreInst *St = dyn_cast<StoreInst>(I);
>> + if (!St && !Ld)
>> + continue;
>> + if (Ld && !Ld->isSimple())
>> + return false;
>> + if (St && !St->isSimple())
>> + return false;
>> + MemInstr.push_back(I);
>> + }
>> + }
>> +
>> + DEBUG(dbgs() << "Found " << MemInstr.size()
>> + << " Loads and Stores to analyze\n");
>> +
>> + ValueVector::iterator I, IE, J, JE;
>> +
>> + for (I = MemInstr.begin(), IE = MemInstr.end(); I != IE; ++I) {
>> + for (J = I, JE = MemInstr.end(); J != JE; ++J) {
>> + std::vector<char> Dep;
>> + Instruction *Src = dyn_cast<Instruction>(*I);
>> + Instruction *Des = dyn_cast<Instruction>(*J);
>> + if (Src == Des)
>> + continue;
>> + if (isa<LoadInst>(Src) && isa<LoadInst>(Des))
>> + continue;
>> + if (auto D = DA->depends(Src, Des, true)) {
>> + DEBUG(dbgs() << "Found Dependency between Src=" << Src << " Des=" << Des
>> + << "\n");
>> + if (D->isFlow()) {
>> + // TODO: Handle Flow dependence.Check if it is sufficient to populate
>> + // the Dependence Matrix with the direction reversed.
>> + DEBUG(dbgs() << "Flow dependence not handled");
>> + return false;
>> + }
>> + if (D->isAnti()) {
>> + DEBUG(dbgs() << "Found Anti dependence \n");
>> + unsigned Levels = D->getLevels();
>> + char Direction;
>> + for (unsigned II = 1; II <= Levels; ++II) {
>> + const SCEV *Distance = D->getDistance(II);
>> + const SCEVConstant *SCEVConst =
>> + dyn_cast_or_null<SCEVConstant>(Distance);
>> + if (SCEVConst) {
>> + const ConstantInt *CI = SCEVConst->getValue();
>> + if (CI->isNegative())
>> + Direction = '<';
>> + else if (CI->isZero())
>> + Direction = '=';
>> + else
>> + Direction = '>';
>> + Dep.push_back(Direction);
>> + } else if (D->isScalar(II)) {
>> + Direction = 'S';
>> + Dep.push_back(Direction);
>> + } else {
>> + unsigned Dir = D->getDirection(II);
>> + if (Dir == Dependence::DVEntry::LT ||
>> + Dir == Dependence::DVEntry::LE)
>> + Direction = '<';
>> + else if (Dir == Dependence::DVEntry::GT ||
>> + Dir == Dependence::DVEntry::GE)
>> + Direction = '>';
>> + else if (Dir == Dependence::DVEntry::EQ)
>> + Direction = '=';
>> + else
>> + Direction = '*';
>> + Dep.push_back(Direction);
>> + }
>> + }
>> + while (Dep.size() != Level) {
>> + Dep.push_back('I');
>> + }
>> +
>> + DepMatrix.push_back(Dep);
>> + if (DepMatrix.size() > MaxMemInstrCount) {
>> + DEBUG(dbgs() << "Cannot handle more than " << MaxMemInstrCount
>> + << " dependencies inside loop\n");
>> + return false;
>> + }
>> + }
>> + }
>> + }
>> + }
>> +
>> + // We don't have a DepMatrix to check legality return false
>> + if (DepMatrix.size() == 0)
>> + return false;
>> + return true;
>> +}
>> +
>> +// A loop is moved from index 'from' to an index 'to'. Update the Dependence
>> +// matrix by exchanging the two columns.
>> +void interChangeDepedencies(CharMatrix &DepMatrix, unsigned FromIndx,
>> + unsigned ToIndx) {
>> + unsigned numRows = DepMatrix.size();
>> + for (unsigned i = 0; i < numRows; ++i) {
>> + char TmpVal = DepMatrix[i][ToIndx];
>> + DepMatrix[i][ToIndx] = DepMatrix[i][FromIndx];
>> + DepMatrix[i][FromIndx] = TmpVal;
>> + }
>> +}
>> +
>> +// Checks if outermost non '=','S'or'I' dependence in the dependence matrix is
>> +// '>'
>> +bool isOuterMostDepPositive(CharMatrix &DepMatrix, unsigned Row,
>> + unsigned Column) {
>> + for (unsigned i = 0; i <= Column; ++i) {
>> + if (DepMatrix[Row][i] == '<')
>> + return false;
>> + if (DepMatrix[Row][i] == '>')
>> + return true;
>> + }
>> + // All dependencies were '=','S' or 'I'
>> + return false;
>> +}
>> +
>> +// Checks if no dependence exist in the dependency matrix in Row before Column.
>> +bool containsNoDependence(CharMatrix &DepMatrix, unsigned Row,
>> + unsigned Column) {
>> + for (unsigned i = 0; i < Column; ++i) {
>> + if (DepMatrix[Row][i] != '=' || DepMatrix[Row][i] != 'S' ||
>> + DepMatrix[Row][i] != 'I')
>> + return false;
>> + }
>> + return true;
>> +}
>> +
>> +bool validDepInterchange(CharMatrix &DepMatrix, unsigned Row,
>> + unsigned OuterLoopId, char InnerDep, char OuterDep) {
>> +
>> + if (isOuterMostDepPositive(DepMatrix, Row, OuterLoopId))
>> + return false;
>> +
>> + if (InnerDep == OuterDep)
>> + return true;
>> +
>> + // It is legal to interchange if and only if after interchange no row has a
>> + // '>' direction as the leftmost non-'='.
>> +
>> + if (InnerDep == '=' || InnerDep == 'S' || InnerDep == 'I')
>> + return true;
>> +
>> + if (InnerDep == '<')
>> + return true;
>> +
>> + if (InnerDep == '>') {
>> + // If OuterLoopId represents outermost loop then interchanging will make the
>> + // 1st dependency as '>'
>> + if (OuterLoopId == 0)
>> + return false;
>> +
>> + // If all dependencies before OuterloopId are '=','S'or 'I'. Then
>> + // interchanging will result in this row having an outermost non '='
>> + // dependency of '>'
>> + if (!containsNoDependence(DepMatrix, Row, OuterLoopId))
>> + return true;
>> + }
>> +
>> + return false;
>> +}
>> +
>> +// Checks if it is legal to interchange 2 loops.
>> +// [Theorm] A permutation of the loops in a perfect nest is legal if and only if
>> +// the direction matrix, after the same permutation is applied to its columns,
>> +// has no ">" direction as the leftmost non-"=" direction in any row.
>> +bool isLegalToInterChangeLoops(CharMatrix &DepMatrix, unsigned InnerLoopId,
>> + unsigned OuterLoopId) {
>> +
>> + unsigned NumRows = DepMatrix.size();
>> + // For each row check if it is valid to interchange.
>> + for (unsigned Row = 0; Row < NumRows; ++Row) {
>> + char InnerDep = DepMatrix[Row][InnerLoopId];
>> + char OuterDep = DepMatrix[Row][OuterLoopId];
>> + if (InnerDep == '*' || OuterDep == '*')
>> + return false;
>> + else if (!validDepInterchange(DepMatrix, Row, OuterLoopId, InnerDep,
>> + OuterDep))
>> + return false;
>> + }
>> + return true;
>> +}
>> +
>> +static void populateWorklist(Loop &L, SmallVector<LoopVector, 8> &V) {
>> +
>> + DEBUG(dbgs() << "Calling populateWorklist called\n");
>> + LoopVector LoopList;
>> + Loop *CurrentLoop = &L;
>> + std::vector<Loop *> vec = CurrentLoop->getSubLoopsVector();
>> + while (vec.size() != 0) {
>> + // The current loop has multiple subloops in it hence it is not tightly
>> + // nested.
>> + // Discard all loops above it added into Worklist.
>> + if (vec.size() != 1) {
>> + LoopList.clear();
>> + return;
>> + }
>> + LoopList.push_back(CurrentLoop);
>> + CurrentLoop = *(vec.begin());
>> + vec = CurrentLoop->getSubLoopsVector();
>> + }
>> + LoopList.push_back(CurrentLoop);
>> + V.push_back(LoopList);
>> +}
>> +
>> +static PHINode *getInductionVariable(Loop *L, ScalarEvolution *SE) {
>> + PHINode *InnerIndexVar = L->getCanonicalInductionVariable();
>> + if (InnerIndexVar)
>> + return InnerIndexVar;
>> + if (L->getLoopLatch() == nullptr || L->getLoopPredecessor() == nullptr)
>> + return nullptr;
>> + for (BasicBlock::iterator I = L->getHeader()->begin(); isa<PHINode>(I); ++I) {
>> + PHINode *PhiVar = cast<PHINode>(I);
>> + Type *PhiTy = PhiVar->getType();
>> + if (!PhiTy->isIntegerTy() && !PhiTy->isFloatingPointTy() &&
>> + !PhiTy->isPointerTy())
>> + return nullptr;
>> + const SCEVAddRecExpr *AddRec =
>> + dyn_cast<SCEVAddRecExpr>(SE->getSCEV(PhiVar));
>> + if (!AddRec || !AddRec->isAffine())
>> + continue;
>> + const SCEV *Step = AddRec->getStepRecurrence(*SE);
>> + const SCEVConstant *C = dyn_cast<SCEVConstant>(Step);
>> + if (!C)
>> + continue;
>> + // Found the induction variable.
>> + // FIXME: Handle loops with more than one induction variable. Note that,
>> + // currently, legality makes sure we have only one induction variable.
>> + return PhiVar;
>> + }
>> + return nullptr;
>> +}
>> +
>> +/// LoopInterchangeLegality checks if it is legal to interchange the loop.
>> +class LoopInterchangeLegality {
>> +public:
>> + LoopInterchangeLegality(Loop *Outer, Loop *Inner, ScalarEvolution *SE,
>> + LoopInterchange *Pass)
>> + : OuterLoop(Outer), InnerLoop(Inner), SE(SE), CurrentPass(Pass) {}
>> +
>> + /// Check if the loops can be interchanged.
>> + bool canInterchangeLoops(unsigned InnerLoopId, unsigned OuterLoopId,
>> + CharMatrix &DepMatrix);
>> + /// Check if the loop structure is understood. We do not handle triangular
>> + /// loops for now.
>> + bool isLoopStructureUnderstood(PHINode *InnerInductionVar);
>> +
>> + bool currentLimitations();
>> +
>> +private:
>> + bool tightlyNested(Loop *Outer, Loop *Inner);
>> +
>> + Loop *OuterLoop;
>> + Loop *InnerLoop;
>> +
>> + /// Scev analysis.
>> + ScalarEvolution *SE;
>> + LoopInterchange *CurrentPass;
>> +};
>> +
>> +/// LoopInterchangeProfitability checks if it is profitable to interchange the
>> +/// loop.
>> +class LoopInterchangeProfitability {
>> +public:
>> + LoopInterchangeProfitability(Loop *Outer, Loop *Inner, ScalarEvolution *SE)
>> + : OuterLoop(Outer), InnerLoop(Inner), SE(SE) {}
>> +
>> + /// Check if the loop interchange is profitable
>> + bool isProfitable(unsigned InnerLoopId, unsigned OuterLoopId,
>> + CharMatrix &DepMatrix);
>> +
>> +private:
>> + int getInstrOrderCost();
>> +
>> + Loop *OuterLoop;
>> + Loop *InnerLoop;
>> +
>> + /// Scev analysis.
>> + ScalarEvolution *SE;
>> +};
>> +
>> +/// LoopInterchangeTransform interchanges the loop
>> +class LoopInterchangeTransform {
>> +public:
>> + LoopInterchangeTransform(Loop *Outer, Loop *Inner, ScalarEvolution *SE,
>> + LoopInfo *LI, DominatorTree *DT,
>> + LoopInterchange *Pass, BasicBlock *LoopNestExit)
>> + : OuterLoop(Outer), InnerLoop(Inner), SE(SE), LI(LI), DT(DT),
>> + LoopExit(LoopNestExit) {
>> + initialize();
>> + }
>> +
>> + /// Interchange OuterLoop and InnerLoop.
>> + bool transform();
>> + void restructureLoops(Loop *InnerLoop, Loop *OuterLoop);
>> + void removeChildLoop(Loop *OuterLoop, Loop *InnerLoop);
>> + void initialize();
>> +
>> +private:
>> + void splitInnerLoopLatch(Instruction *);
>> + void splitOuterLoopLatch();
>> + void splitInnerLoopHeader();
>> + bool adjustLoopLinks();
>> + void adjustLoopPreheaders();
>> + void adjustOuterLoopPreheader();
>> + void adjustInnerLoopPreheader();
>> + bool adjustLoopBranches();
>> +
>> + Loop *OuterLoop;
>> + Loop *InnerLoop;
>> +
>> + /// Scev analysis.
>> + ScalarEvolution *SE;
>> + LoopInfo *LI;
>> + DominatorTree *DT;
>> + BasicBlock *LoopExit;
>> +};
>> +
>> +// Main LoopInterchange Pass
>> +struct LoopInterchange : public FunctionPass {
>> + static char ID;
>> + ScalarEvolution *SE;
>> + LoopInfo *LI;
>> + DependenceAnalysis *DA;
>> + DominatorTree *DT;
>> + LoopInterchange()
>> + : FunctionPass(ID), SE(nullptr), LI(nullptr), DA(nullptr), DT(nullptr) {
>> + initializeLoopInterchangePass(*PassRegistry::getPassRegistry());
>> + }
>> +
>> + void getAnalysisUsage(AnalysisUsage &AU) const override {
>> + AU.addRequired<ScalarEvolution>();
>> + AU.addRequired<AliasAnalysis>();
>> + AU.addRequired<DominatorTreeWrapperPass>();
>> + AU.addRequired<LoopInfoWrapperPass>();
>> + AU.addRequired<DependenceAnalysis>();
>> + AU.addRequiredID(LoopSimplifyID);
>> + AU.addRequiredID(LCSSAID);
>> + }
>> +
>> + bool runOnFunction(Function &F) override {
>> + SE = &getAnalysis<ScalarEvolution>();
>> + LI = &getAnalysis<LoopInfoWrapperPass>().getLoopInfo();
>> + DA = &getAnalysis<DependenceAnalysis>();
>> + auto *DTWP = getAnalysisIfAvailable<DominatorTreeWrapperPass>();
>> + DT = DTWP ? &DTWP->getDomTree() : nullptr;
>> + // Build up a worklist of loop pairs to analyze.
>> + SmallVector<LoopVector, 8> Worklist;
>> +
>> + for (Loop *L : *LI)
>> + populateWorklist(*L, Worklist);
>> +
>> + DEBUG(dbgs() << "Worklist size = " << Worklist.size() << "\n");
>> + bool Changed = true;
>> + while (!Worklist.empty()) {
>> + LoopVector LoopList = Worklist.pop_back_val();
>> + Changed = processLoopList(LoopList);
>> + }
>> + return Changed;
>> + }
>> +
>> + bool isComputableLoopNest(LoopVector LoopList) {
>> + for (auto I = LoopList.begin(), E = LoopList.end(); I != E; ++I) {
>> + Loop *L = *I;
>> + const SCEV *ExitCountOuter = SE->getBackedgeTakenCount(L);
>> + if (ExitCountOuter == SE->getCouldNotCompute()) {
>> + DEBUG(dbgs() << "Couldn't compute Backedge count\n");
>> + return false;
>> + }
>> + if (L->getNumBackEdges() != 1) {
>> + DEBUG(dbgs() << "NumBackEdges is not equal to 1\n");
>> + return false;
>> + }
>> + if (!L->getExitingBlock()) {
>> + DEBUG(dbgs() << "Loop Doesn't have unique exit block\n");
>> + return false;
>> + }
>> + }
>> + return true;
>> + }
>> +
>> + unsigned selectLoopForInterchange(LoopVector LoopList) {
>> + // TODO: Add a better heuristic to select the loop to be interchanged based
>> + // on the dependece matrix. Currently we select the innermost loop.
>> + return LoopList.size() - 1;
>> + }
>> +
>> + bool processLoopList(LoopVector LoopList) {
>> + bool Changed = false;
>> + bool containsLCSSAPHI = false;
>> + CharMatrix DependencyMatrix;
>> + if (LoopList.size() < 2) {
>> + DEBUG(dbgs() << "Loop doesn't contain minimum nesting level.\n");
>> + return false;
>> + }
>> + if (!isComputableLoopNest(LoopList)) {
>> + DEBUG(dbgs() << "Not vaild loop candidate for interchange\n");
>> + return false;
>> + }
>> + Loop *OuterMostLoop = *(LoopList.begin());
>> +
>> + DEBUG(dbgs() << "Processing LoopList of size = " << LoopList.size()
>> + << "\n");
>> +
>> + if (!populateDependencyMatrix(DependencyMatrix, LoopList.size(),
>> + OuterMostLoop, DA)) {
>> + DEBUG(dbgs() << "Populating Dependency matrix failed\n");
>> + return false;
>> + }
>> +#ifdef DUMP_DEP_MATRICIES
>> + DEBUG(dbgs() << "Dependence before inter change \n");
>> + printDepMatrix(DependencyMatrix);
>> +#endif
>> +
>> + BasicBlock *OuterMostLoopLatch = OuterMostLoop->getLoopLatch();
>> + BranchInst *OuterMostLoopLatchBI =
>> + dyn_cast<BranchInst>(OuterMostLoopLatch->getTerminator());
>> + if (!OuterMostLoopLatchBI)
>> + return false;
>> +
>> + // Since we currently do not handle LCSSA PHI's any failure in loop
>> + // condition will now branch to LoopNestExit.
>> + // TODO: This should be removed once we handle LCSSA PHI nodes.
>> +
>> + // Get the Outermost loop exit.
>> + BasicBlock *LoopNestExit;
>> + if (OuterMostLoopLatchBI->getSuccessor(0) == OuterMostLoop->getHeader())
>> + LoopNestExit = OuterMostLoopLatchBI->getSuccessor(1);
>> + else
>> + LoopNestExit = OuterMostLoopLatchBI->getSuccessor(0);
>> +
>> + for (auto I = LoopList.begin(), E = LoopList.end(); I != E; ++I) {
>> + Loop *L = *I;
>> + BasicBlock *Latch = L->getLoopLatch();
>> + BasicBlock *Header = L->getHeader();
>> + if (Latch && Latch != Header && isa<PHINode>(Latch->begin())) {
>> + containsLCSSAPHI = true;
>> + break;
>> + }
>> + }
>> +
>> + // TODO: Handle lcssa PHI's. Currently LCSSA PHI's are not handled. Handle
>> + // the same by splitting the loop latch and adjusting loop links
>> + // accordingly.
>> + if (containsLCSSAPHI)
>> + return false;
>> +
>> + unsigned SelecLoopId = selectLoopForInterchange(LoopList);
>> + // Move the selected loop outwards to the best posible position.
>> + for (unsigned i = SelecLoopId; i > 0; i--) {
>> + bool Interchanged =
>> + processLoop(LoopList, i, i - 1, LoopNestExit, DependencyMatrix);
>> + if (!Interchanged)
>> + return Changed;
>> + // Loops interchanged reflect the same in LoopList
>> + Loop *OldOuterLoop = LoopList[i - 1];
>> + LoopList[i - 1] = LoopList[i];
>> + LoopList[i] = OldOuterLoop;
>> +
>> + // Update the DependencyMatrix
>> + interChangeDepedencies(DependencyMatrix, i, i - 1);
>> +
>> +#ifdef DUMP_DEP_MATRICIES
>> + DEBUG(dbgs() << "Dependence after inter change \n");
>> + printDepMatrix(DependencyMatrix);
>> +#endif
>> + Changed |= Interchanged;
>> + }
>> + return Changed;
>> + }
>> +
>> + bool processLoop(LoopVector LoopList, unsigned InnerLoopId,
>> + unsigned OuterLoopId, BasicBlock *LoopNestExit,
>> + std::vector<std::vector<char>> &DependencyMatrix) {
>> +
>> + DEBUG(dbgs() << "Processing Innder Loop Id = " << InnerLoopId
>> + << " and OuterLoopId = " << OuterLoopId << "\n");
>> + Loop *InnerLoop = LoopList[InnerLoopId];
>> + Loop *OuterLoop = LoopList[OuterLoopId];
>> +
>> + LoopInterchangeLegality LIL(OuterLoop, InnerLoop, SE, this);
>> + if (!LIL.canInterchangeLoops(InnerLoopId, OuterLoopId, DependencyMatrix)) {
>> + DEBUG(dbgs() << "Not interchanging Loops. Cannot prove legality\n");
>> + return false;
>> + }
>> + DEBUG(dbgs() << "Loops are legal to interchange\n");
>> + LoopInterchangeProfitability LIP(OuterLoop, InnerLoop, SE);
>> + if (!LIP.isProfitable(InnerLoopId, OuterLoopId, DependencyMatrix)) {
>> + DEBUG(dbgs() << "Interchanging Loops not profitable\n");
>> + return false;
>> + }
>> +
>> + LoopInterchangeTransform LIT(OuterLoop, InnerLoop, SE, LI, DT, this,
>> + LoopNestExit);
>> + LIT.transform();
>> + DEBUG(dbgs() << "Loops interchanged\n");
>> + return true;
>> + }
>> +};
>> +
>> +} // end of namespace
>> +
>> +static bool containsUnsafeInstructions(BasicBlock *BB) {
>> + for (auto I = BB->begin(), E = BB->end(); I != E; ++I) {
>> + if (I->mayHaveSideEffects() || I->mayReadFromMemory())
>> + return true;
>> + }
>> + return false;
>> +}
>> +
>> +bool LoopInterchangeLegality::tightlyNested(Loop *OuterLoop, Loop *InnerLoop) {
>> + BasicBlock *OuterLoopHeader = OuterLoop->getHeader();
>> + BasicBlock *InnerLoopPreHeader = InnerLoop->getLoopPreheader();
>> + BasicBlock *OuterLoopLatch = OuterLoop->getLoopLatch();
>> +
>> + DEBUG(dbgs() << "Checking if Loops are Tightly Nested\n");
>> +
>> + // A perfectly nested loop will not have any branch in between the outer and
>> + // inner block i.e. outer header will branch to either inner preheader and
>> + // outerloop latch.
>> + BranchInst *outerLoopHeaderBI =
>> + dyn_cast<BranchInst>(OuterLoopHeader->getTerminator());
>> + if (!outerLoopHeaderBI)
>> + return false;
>> + unsigned num = outerLoopHeaderBI->getNumSuccessors();
>> + for (unsigned i = 0; i < num; i++) {
>> + if (outerLoopHeaderBI->getSuccessor(i) != InnerLoopPreHeader &&
>> + outerLoopHeaderBI->getSuccessor(i) != OuterLoopLatch)
>> + return false;
>> + }
>> +
>> + DEBUG(dbgs() << "Checking instructions in Loop header and Loop latch \n");
>> + // We do not have any basic block in between now make sure the outer header
>> + // and outer loop latch doesnt contain any unsafe instructions.
>> + if (containsUnsafeInstructions(OuterLoopHeader) ||
>> + containsUnsafeInstructions(OuterLoopLatch))
>> + return false;
>> +
>> + DEBUG(dbgs() << "Loops are perfectly nested \n");
>> + // We have a perfect loop nest.
>> + return true;
>> +}
>> +
>> +static unsigned getPHICount(BasicBlock *BB) {
>> + unsigned PhiCount = 0;
>> + for (auto I = BB->begin(); isa<PHINode>(I); ++I)
>> + PhiCount++;
>> + return PhiCount;
>> +}
>> +
>> +bool LoopInterchangeLegality::isLoopStructureUnderstood(
>> + PHINode *InnerInduction) {
>> +
>> + unsigned Num = InnerInduction->getNumOperands();
>> + BasicBlock *InnerLoopPreheader = InnerLoop->getLoopPreheader();
>> + for (unsigned i = 0; i < Num; ++i) {
>> + Value *Val = InnerInduction->getOperand(i);
>> + if (isa<Constant>(Val))
>> + continue;
>> + Instruction *I = dyn_cast<Instruction>(Val);
>> + if (!I)
>> + return false;
>> + // TODO: Handle triangular loops.
>> + // e.g. for(int i=0;i<N;i++)
>> + // for(int j=i;j<N;j++)
>> + unsigned IncomBlockIndx = PHINode::getIncomingValueNumForOperand(i);
>> + if (InnerInduction->getIncomingBlock(IncomBlockIndx) ==
>> + InnerLoopPreheader &&
>> + !OuterLoop->isLoopInvariant(I)) {
>> + return false;
>> + }
>> + }
>> + return true;
>> +}
>> +
>> +// This function indicates the current limitations in the transform as a result
>> +// of which we do not proceed.
>> +bool LoopInterchangeLegality::currentLimitations() {
>> +
>> + BasicBlock *InnerLoopPreHeader = InnerLoop->getLoopPreheader();
>> + BasicBlock *InnerLoopHeader = InnerLoop->getHeader();
>> + BasicBlock *OuterLoopHeader = OuterLoop->getHeader();
>> + BasicBlock *InnerLoopLatch = InnerLoop->getLoopLatch();
>> + BasicBlock *OuterLoopLatch = OuterLoop->getLoopLatch();
>> +
>> + PHINode *InnerInductionVar;
>> + PHINode *OuterInductionVar;
>> +
>> + // We currently handle only 1 induction variable inside the loop. We also do
>> + // not handle reductions as of now.
>> + if (getPHICount(InnerLoopHeader) > 1)
>> + return true;
>> +
>> + if (getPHICount(OuterLoopHeader) > 1)
>> + return true;
>> +
>> + InnerInductionVar = getInductionVariable(InnerLoop, SE);
>> + OuterInductionVar = getInductionVariable(OuterLoop, SE);
>> +
>> + if (!OuterInductionVar || !InnerInductionVar) {
>> + DEBUG(dbgs() << "Induction variable not found\n");
>> + return true;
>> + }
>> +
>> + // TODO: Triangular loops are not handled for now.
>> + if (!isLoopStructureUnderstood(InnerInductionVar)) {
>> + DEBUG(dbgs() << "Loop structure not understood by pass\n");
>> + return true;
>> + }
>> +
>> + // TODO: Loops with LCSSA PHI's are currently not handled.
>> + if (isa<PHINode>(OuterLoopLatch->begin())) {
>> + DEBUG(dbgs() << "Found and LCSSA PHI in outer loop latch\n");
>> + return true;
>> + }
>> + if (InnerLoopLatch != InnerLoopHeader &&
>> + isa<PHINode>(InnerLoopLatch->begin())) {
>> + DEBUG(dbgs() << "Found and LCSSA PHI in inner loop latch\n");
>> + return true;
>> + }
>> +
>> + // TODO: Current limitation: Since we split the inner loop latch at the point
>> + // were induction variable is incremented (induction.next); We cannot have
>> + // more than 1 user of induction.next since it would result in broken code
>> + // after split.
>> + // e.g.
>> + // for(i=0;i<N;i++) {
>> + // for(j = 0;j<M;j++) {
>> + // A[j+1][i+2] = A[j][i]+k;
>> + // }
>> + // }
>> + bool FoundInduction = false;
>> + Instruction *InnerIndexVarInc = nullptr;
>> + if (InnerInductionVar->getIncomingBlock(0) == InnerLoopPreHeader)
>> + InnerIndexVarInc =
>> + dyn_cast<Instruction>(InnerInductionVar->getIncomingValue(1));
>> + else
>> + InnerIndexVarInc =
>> + dyn_cast<Instruction>(InnerInductionVar->getIncomingValue(0));
>> +
>> + if (!InnerIndexVarInc)
>> + return true;
>> +
>> + // Since we split the inner loop latch on this induction variable. Make sure
>> + // we do not have any instruction between the induction variable and branch
>> + // instruction.
>> +
>> + for (auto I = InnerLoopLatch->rbegin(), E = InnerLoopLatch->rend();
>> + I != E && !FoundInduction; ++I) {
>> + if (isa<BranchInst>(*I) || isa<CmpInst>(*I) || isa<TruncInst>(*I))
>> + continue;
>> + const Instruction &Ins = *I;
>> + // We found an instruction. If this is not induction variable then it is not
>> + // safe to split this loop latch.
>> + if (!Ins.isIdenticalTo(InnerIndexVarInc))
>> + return true;
>> + else
>> + FoundInduction = true;
>> + }
>> + // The loop latch ended and we didnt find the induction variable return as
>> + // current limitation.
>> + if (!FoundInduction)
>> + return true;
>> +
>> + return false;
>> +}
>> +
>> +bool LoopInterchangeLegality::canInterchangeLoops(unsigned InnerLoopId,
>> + unsigned OuterLoopId,
>> + CharMatrix &DepMatrix) {
>> +
>> + if (!isLegalToInterChangeLoops(DepMatrix, InnerLoopId, OuterLoopId)) {
>> + DEBUG(dbgs() << "Failed interchange InnerLoopId = " << InnerLoopId
>> + << "and OuterLoopId = " << OuterLoopId
>> + << "due to dependence\n");
>> + return false;
>> + }
>> +
>> + // Create unique Preheaders if we already do not have one.
>> + BasicBlock *OuterLoopPreHeader = OuterLoop->getLoopPreheader();
>> + BasicBlock *InnerLoopPreHeader = InnerLoop->getLoopPreheader();
>> +
>> + // Create a unique outer preheader -
>> + // 1) If OuterLoop preheader is not present.
>> + // 2) If OuterLoop Preheader is same as OuterLoop Header
>> + // 3) If OuterLoop Preheader is same as Header of the previous loop.
>> + // 4) If OuterLoop Preheader is Entry node.
>> + if (!OuterLoopPreHeader || OuterLoopPreHeader == OuterLoop->getHeader() ||
>> + isa<PHINode>(OuterLoopPreHeader->begin()) ||
>> + !OuterLoopPreHeader->getUniquePredecessor()) {
>> + OuterLoopPreHeader = InsertPreheaderForLoop(OuterLoop, CurrentPass);
>> + }
>> +
>> + if (!InnerLoopPreHeader || InnerLoopPreHeader == InnerLoop->getHeader() ||
>> + InnerLoopPreHeader == OuterLoop->getHeader()) {
>> + InnerLoopPreHeader = InsertPreheaderForLoop(InnerLoop, CurrentPass);
>> + }
>> +
>> + // Check if the loops are tightly nested.
>> + if (!tightlyNested(OuterLoop, InnerLoop)) {
>> + DEBUG(dbgs() << "Loops not tightly nested\n");
>> + return false;
>> + }
>> +
>> + // TODO: The loops could not be interchanged due to current limitations in the
>> + // transform module.
>> + if (currentLimitations()) {
>> + DEBUG(dbgs() << "Not legal because of current transform limitation\n");
>> + return false;
>> + }
>> +
>> + return true;
>> +}
>> +
>> +int LoopInterchangeProfitability::getInstrOrderCost() {
>> + unsigned GoodOrder, BadOrder;
>> + BadOrder = GoodOrder = 0;
>> + for (auto BI = InnerLoop->block_begin(), BE = InnerLoop->block_end();
>> + BI != BE; ++BI) {
>> + for (auto I = (*BI)->begin(), E = (*BI)->end(); I != E; ++I) {
>> + const Instruction &Ins = *I;
>> + if (const GetElementPtrInst *GEP = dyn_cast<GetElementPtrInst>(&Ins)) {
>> + unsigned NumOp = GEP->getNumOperands();
>> + bool FoundInnerInduction = false;
>> + bool FoundOuterInduction = false;
>> + for (unsigned i = 0; i < NumOp; ++i) {
>> + const SCEV *OperandVal = SE->getSCEV(GEP->getOperand(i));
>> + const SCEVAddRecExpr *AR = dyn_cast<SCEVAddRecExpr>(OperandVal);
>> + if (!AR)
>> + continue;
>> +
>> + // If we find the inner induction after an outer induction e.g.
>> + // for(int i=0;i<N;i++)
>> + // for(int j=0;j<N;j++)
>> + // A[i][j] = A[i-1][j-1]+k;
>> + // then it is a good order.
>> + if (AR->getLoop() == InnerLoop) {
>> + // We found an InnerLoop induction after OuterLoop induction. It is
>> + // a good order.
>> + FoundInnerInduction = true;
>> + if (FoundOuterInduction) {
>> + GoodOrder++;
>> + break;
>> + }
>> + }
>> + // If we find the outer induction after an inner induction e.g.
>> + // for(int i=0;i<N;i++)
>> + // for(int j=0;j<N;j++)
>> + // A[j][i] = A[j-1][i-1]+k;
>> + // then it is a bad order.
>> + if (AR->getLoop() == OuterLoop) {
>> + // We found an OuterLoop induction after InnerLoop induction. It is
>> + // a bad order.
>> + FoundOuterInduction = true;
>> + if (FoundInnerInduction) {
>> + BadOrder++;
>> + break;
>> + }
>> + }
>> + }
>> + }
>> + }
>> + }
>> + return GoodOrder - BadOrder;
>> +}
>> +
>> +bool isProfitabileForVectorization(unsigned InnerLoopId, unsigned OuterLoopId,
>> + CharMatrix &DepMatrix) {
>> + // TODO: Improve this heuristic to catch more cases.
>> + // If the inner loop is loop independent or doesn't carry any dependency it is
>> + // profitable to move this to outer position.
>> + unsigned Row = DepMatrix.size();
>> + for (unsigned i = 0; i < Row; ++i) {
>> + if (DepMatrix[i][InnerLoopId] != 'S' && DepMatrix[i][InnerLoopId] != 'I')
>> + return false;
>> + // TODO: We need to improve this heuristic.
>> + if (DepMatrix[i][OuterLoopId] != '=')
>> + return false;
>> + }
>> + // If outer loop has dependence and inner loop is loop independent then it is
>> + // profitable to interchange to enable parallelism.
>> + return true;
>> +}
>> +
>> +bool LoopInterchangeProfitability::isProfitable(unsigned InnerLoopId,
>> + unsigned OuterLoopId,
>> + CharMatrix &DepMatrix) {
>> +
>> + // TODO: Add Better Profitibility checks.
>> + // e.g
>> + // 1) Construct dependency matrix and move the one with no loop carried dep
>> + // inside to enable vectorization.
>> +
>> + // This is rough cost estimation algorithm. It counts the good and bad order
>> + // of induction variables in the instruction and allows reordering if number
>> + // of bad orders is more than good.
>> + int Cost = 0;
>> + Cost += getInstrOrderCost();
>> + DEBUG(dbgs() << "Cost = " << Cost << "\n");
>> + if (Cost < 0)
>> + return true;
>> +
>> + // It is not profitable as per current cache profitibility model. But check if
>> + // we can move this loop outside to improve parallelism.
>> + bool ImprovesPar =
>> + isProfitabileForVectorization(InnerLoopId, OuterLoopId, DepMatrix);
>> + return ImprovesPar;
>> +}
>> +
>> +void LoopInterchangeTransform::removeChildLoop(Loop *OuterLoop,
>> + Loop *InnerLoop) {
>> + for (Loop::iterator I = OuterLoop->begin(), E = OuterLoop->end();; ++I) {
>> + assert(I != E && "Couldn't find loop");
>> + if (*I == InnerLoop) {
>> + OuterLoop->removeChildLoop(I);
>> + return;
>> + }
>> + }
>> +}
>> +void LoopInterchangeTransform::restructureLoops(Loop *InnerLoop,
>> + Loop *OuterLoop) {
>> + Loop *OuterLoopParent = OuterLoop->getParentLoop();
>> + if (OuterLoopParent) {
>> + // Remove the loop from its parent loop.
>> + removeChildLoop(OuterLoopParent, OuterLoop);
>> + removeChildLoop(OuterLoop, InnerLoop);
>> + OuterLoopParent->addChildLoop(InnerLoop);
>> + } else {
>> + removeChildLoop(OuterLoop, InnerLoop);
>> + LI->changeTopLevelLoop(OuterLoop, InnerLoop);
>> + }
>> +
>> + for (Loop::iterator I = InnerLoop->begin(), E = InnerLoop->end(); I != E; ++I)
>> + OuterLoop->addChildLoop(InnerLoop->removeChildLoop(I));
>
> This for loop is causing failed assertions in debug builds with MSVC;
> the iterator is invalidated when the child loop is removed on
> InnerLoop, so when ++I is executed, the following assertion is
> triggered. I don't think removeChildLoop() is a particularly safe API
> design given how trivial it is for the underlying container to
> invalidate all iterators.
>
> 63> FAIL: LLVM :: Transforms/LoopInterchange/reductions.ll (19260 of 22394)
> 63> ******************** TEST 'LLVM ::
> Transforms/LoopInterchange/reductions.ll' FAILED ********************
> 63> Script:
> 63> --
> 63> E:/llvm/2013/Debug/bin\opt.EXE <
> E:\llvm\llvm\test\Transforms\LoopInterchange\reductions.ll -basicaa
> -loop-interchange -S | E:/llvm/2013/Debug/bin\FileCheck.EXE
> E:\llvm\llvm\test\Transforms\LoopInterchange\reductions.ll
> 63> --
> 63> Exit Code: 2
> 63>
> 63> Command Output (stdout):
> 63> --
> 63> Command 0: "E:/llvm/2013/Debug/bin\opt.EXE" "-basicaa"
> "-loop-interchange" "-S"
> 63> Command 0 Result: -2147483645
> 63> Command 0 Output:
> 63>
> 63>
> 63> Command 0 Stderr:
> 63> 0x0F4FCEE6 (0x02AB16A0 0x02AACBD0 0x00000065 0x00322278),
> ?_Debug_message at std@@YAXPB_W0I at Z() + 0x26 bytes(s)
> 63>
> 63> 0x0108C5AB (0x0421EE7C 0xCCCCCCCC 0xCCCCCCCC 0x00000000),
> std::_Vector_const_iterator<std::_Vector_val<std::_Simple_types<llvm::Loop
> *> > >::operator++() + 0x4B bytes(s), d:\program files (x86)\microsoft
> visual studio 12.0\vc\include\vector, line 101 + 0x14 byte(s)
> 63>
> 63> 0x01D974CF (0x00331A38 0x00337B40 0x00000000 0xCCCCCCCC),
> `anonymous namespace'::LoopInterchangeTransform::restructureLoops() +
> 0x9F bytes(s), e:\llvm\llvm\lib\transforms\scalar\loopinterchange.cpp,
> line 1015 + 0xA byte(s)
> 63>
> 63> 0x01D9741E (0x0421EF20 0xCCCCCCCC 0xCCCCCCCC 0x00337B40),
> `anonymous namespace'::LoopInterchangeTransform::transform() + 0x22E
> bytes(s), e:\llvm\llvm\lib\transforms\scalar\loopinterchange.cpp, line
> 1060
> 63>
> 63> 0x01D9C9BC (0x0421EE90 0x0421EE9C 0x0421EEB0 0x00337B40),
> `anonymous namespace'::LoopInterchange::processLoop() + 0x20C
> bytes(s), e:\llvm\llvm\lib\transforms\scalar\loopinterchange.cpp, line
> 592
> 63>
> 63> 0x01D9CD64 (0x0421EF34 0x0421EF40 0x0421EF54 0x00337B40),
> `anonymous namespace'::LoopInterchange::processLoopList() + 0x304
> bytes(s), e:\llvm\llvm\lib\transforms\scalar\loopinterchange.cpp, line
> 550 + 0x29 byte(s)
> 63>
> 63> 0x01D9D4F0 (0x003213C0 0x0421F2B4 0x0421F1F8 0x00000001),
> `anonymous namespace'::LoopInterchange::runOnFunction() + 0x1D0
> bytes(s), e:\llvm\llvm\lib\transforms\scalar\loopinterchange.cpp, line
> 465 + 0x1D byte(s)
> 63>
> 63> 0x0191E1D5 (0x003213C0 0x00000000 0xCCCCCCCC 0x002F53DC),
> llvm::FPPassManager::runOnFunction() + 0x105 bytes(s),
> e:\llvm\llvm\lib\ir\legacypassmanager.cpp, line 1538 + 0x17 byte(s)
> 63>
> 63> 0x0191E365 (0x002F5408 0x0421F77C 0x0421F2C0 0x00000001),
> llvm::FPPassManager::runOnModule() + 0x75 bytes(s),
> e:\llvm\llvm\lib\ir\legacypassmanager.cpp, line 1558 + 0x15 byte(s)
> 63>
> 63> 0x0191F2D9 (0x002F5408 0x0421F304 0x7EFDE000 0xCCCCCCCC),
> `anonymous namespace'::MPPassManager::runOnModule() + 0x1C9 bytes(s),
> e:\llvm\llvm\lib\ir\legacypassmanager.cpp, line 1616 + 0x17 byte(s)
> 63>
> 63> 0x0191F971 (0x002F5408 0x0421F570 0x0421F77C 0x00B90A06),
> llvm::legacy::PassManagerImpl::run() + 0x101 bytes(s),
> e:\llvm\llvm\lib\ir\legacypassmanager.cpp, line 1723 + 0x1B byte(s)
> 63>
> 63> 0x0191A3ED (0x002F5408 0x00000000 0x00000000 0xCCCCCCCC),
> llvm::legacy::PassManager::run() + 0x1D bytes(s),
> e:\llvm\llvm\lib\ir\legacypassmanager.cpp, line 1757
> 63>
> 63> 0x00B90A06 (0x00000004 0x002EA2A0 0x002EFF88 0x75B36014), main()
> + 0x1696 bytes(s), e:\llvm\llvm\tools\opt\opt.cpp, line 614
> 63>
> 63> 0x024017A9 (0x0421F7E0 0x778D336A 0x7EFDE000 0x0421F820),
> __tmainCRTStartup() + 0x199 bytes(s),
> f:\dd\vctools\crt\crtw32\dllstuff\crtexe.c, line 626 + 0x19 byte(s)
> 63>
> 63> 0x024018ED (0x7EFDE000 0x0421F820 0x77E992B2 0x7EFDE000),
> mainCRTStartup() + 0xD bytes(s),
> f:\dd\vctools\crt\crtw32\dllstuff\crtexe.c, line 466
> 63>
> 63> 0x778D336A (0x7EFDE000 0x7AB5E46A 0x00000000 0x00000000),
> BaseThreadInitThunk() + 0x12 bytes(s)
> 63>
> 63> 0x77E992B2 (0x024018E0 0x7EFDE000 0x00000000 0x00000000),
> RtlInitializeExceptionChain() + 0x63 bytes(s)
> 63>
> 63> 0x77E99285 (0x024018E0 0x7EFDE000 0x00000000 0x00000000),
> RtlInitializeExceptionChain() + 0x36 bytes(s)
> 63>
> 63>
> 63>
> 63> Command 1: "E:/llvm/2013/Debug/bin\FileCheck.EXE"
> "E:\llvm\llvm\test\Transforms\LoopInterchange\reductions.ll"
> 63> Command 1 Result: 2
> 63> Command 1 Output:
> 63>
> 63>
> 63> Command 1 Stderr:
> 63>CUSTOMBUILD : FileCheck error : '-' is empty.
> 63>
> 63>
> 63>
> 63>
> 63> --
> 63>
>
> ~Aaron
>
>> +
>> + InnerLoop->addChildLoop(OuterLoop);
>> +}
>> +
>> +bool LoopInterchangeTransform::transform() {
>> +
>> + DEBUG(dbgs() << "transform\n");
>> + bool Transformed = false;
>> + Instruction *InnerIndexVar;
>> +
>> + if (InnerLoop->getSubLoops().size() == 0) {
>> + BasicBlock *InnerLoopPreHeader = InnerLoop->getLoopPreheader();
>> + DEBUG(dbgs() << "Calling Split Inner Loop\n");
>> + PHINode *InductionPHI = getInductionVariable(InnerLoop, SE);
>> + if (!InductionPHI) {
>> + DEBUG(dbgs() << "Failed to find the point to split loop latch \n");
>> + return false;
>> + }
>> +
>> + if (InductionPHI->getIncomingBlock(0) == InnerLoopPreHeader)
>> + InnerIndexVar = dyn_cast<Instruction>(InductionPHI->getIncomingValue(1));
>> + else
>> + InnerIndexVar = dyn_cast<Instruction>(InductionPHI->getIncomingValue(0));
>> +
>> + //
>> + // Split at the place were the induction variable is
>> + // incremented/decremented.
>> + // TODO: This splitting logic may not work always. Fix this.
>> + splitInnerLoopLatch(InnerIndexVar);
>> + DEBUG(dbgs() << "splitInnerLoopLatch Done\n");
>> +
>> + // Splits the inner loops phi nodes out into a seperate basic block.
>> + splitInnerLoopHeader();
>> + DEBUG(dbgs() << "splitInnerLoopHeader Done\n");
>> + }
>> +
>> + Transformed |= adjustLoopLinks();
>> + if (!Transformed) {
>> + DEBUG(dbgs() << "adjustLoopLinks Failed\n");
>> + return false;
>> + }
>> +
>> + restructureLoops(InnerLoop, OuterLoop);
>> + return true;
>> +}
>> +
>> +void LoopInterchangeTransform::initialize() {}
>> +
>> +void LoopInterchangeTransform::splitInnerLoopLatch(Instruction *inc) {
>> +
>> + BasicBlock *InnerLoopLatch = InnerLoop->getLoopLatch();
>> + BasicBlock::iterator I = InnerLoopLatch->begin();
>> + BasicBlock::iterator E = InnerLoopLatch->end();
>> + for (; I != E; ++I) {
>> + if (inc == I)
>> + break;
>> + }
>> +
>> + BasicBlock *InnerLoopLatchPred = InnerLoopLatch;
>> + InnerLoopLatch = SplitBlock(InnerLoopLatchPred, I, DT, LI);
>> +}
>> +
>> +void LoopInterchangeTransform::splitOuterLoopLatch() {
>> + BasicBlock *OuterLoopLatch = OuterLoop->getLoopLatch();
>> + BasicBlock *OuterLatchLcssaPhiBlock = OuterLoopLatch;
>> + OuterLoopLatch = SplitBlock(OuterLatchLcssaPhiBlock,
>> + OuterLoopLatch->getFirstNonPHI(), DT, LI);
>> +}
>> +
>> +void LoopInterchangeTransform::splitInnerLoopHeader() {
>> +
>> + // Split the inner loop header out.
>> + BasicBlock *InnerLoopHeader = InnerLoop->getHeader();
>> + SplitBlock(InnerLoopHeader, InnerLoopHeader->getFirstNonPHI(), DT, LI);
>> +
>> + DEBUG(dbgs() << "Output of splitInnerLoopHeader InnerLoopHeaderSucc & "
>> + "InnerLoopHeader \n");
>> +}
>> +
>> +void LoopInterchangeTransform::adjustOuterLoopPreheader() {
>> + BasicBlock *OuterLoopPreHeader = OuterLoop->getLoopPreheader();
>> + SmallVector<Instruction *, 8> Inst;
>> + for (auto I = OuterLoopPreHeader->begin(), E = OuterLoopPreHeader->end();
>> + I != E; ++I) {
>> + if (isa<BranchInst>(*I))
>> + break;
>> + Inst.push_back(I);
>> + }
>> +
>> + BasicBlock *InnerPreHeader = InnerLoop->getLoopPreheader();
>> + for (auto I = Inst.begin(), E = Inst.end(); I != E; ++I) {
>> + Instruction *Ins = cast<Instruction>(*I);
>> + Ins->moveBefore(InnerPreHeader->getTerminator());
>> + }
>> +}
>> +
>> +void LoopInterchangeTransform::adjustInnerLoopPreheader() {
>> +
>> + BasicBlock *InnerLoopPreHeader = InnerLoop->getLoopPreheader();
>> + SmallVector<Instruction *, 8> Inst;
>> + for (auto I = InnerLoopPreHeader->begin(), E = InnerLoopPreHeader->end();
>> + I != E; ++I) {
>> + if (isa<BranchInst>(*I))
>> + break;
>> + Inst.push_back(I);
>> + }
>> + BasicBlock *OuterHeader = OuterLoop->getHeader();
>> + for (auto I = Inst.begin(), E = Inst.end(); I != E; ++I) {
>> + Instruction *Ins = cast<Instruction>(*I);
>> + Ins->moveBefore(OuterHeader->getTerminator());
>> + }
>> +}
>> +
>> +bool LoopInterchangeTransform::adjustLoopBranches() {
>> +
>> + DEBUG(dbgs() << "adjustLoopBranches called\n");
>> + // Adjust the loop preheader
>> + BasicBlock *InnerLoopHeader = InnerLoop->getHeader();
>> + BasicBlock *OuterLoopHeader = OuterLoop->getHeader();
>> + BasicBlock *InnerLoopLatch = InnerLoop->getLoopLatch();
>> + BasicBlock *OuterLoopLatch = OuterLoop->getLoopLatch();
>> + BasicBlock *OuterLoopPreHeader = OuterLoop->getLoopPreheader();
>> + BasicBlock *InnerLoopPreHeader = InnerLoop->getLoopPreheader();
>> + BasicBlock *OuterLoopPredecessor = OuterLoopPreHeader->getUniquePredecessor();
>> + BasicBlock *InnerLoopLatchPredecessor =
>> + InnerLoopLatch->getUniquePredecessor();
>> + BasicBlock *InnerLoopLatchSuccessor;
>> + BasicBlock *OuterLoopLatchSuccessor;
>> +
>> + BranchInst *OuterLoopLatchBI =
>> + dyn_cast<BranchInst>(OuterLoopLatch->getTerminator());
>> + BranchInst *InnerLoopLatchBI =
>> + dyn_cast<BranchInst>(InnerLoopLatch->getTerminator());
>> + BranchInst *OuterLoopHeaderBI =
>> + dyn_cast<BranchInst>(OuterLoopHeader->getTerminator());
>> + BranchInst *InnerLoopHeaderBI =
>> + dyn_cast<BranchInst>(InnerLoopHeader->getTerminator());
>> +
>> + if (!OuterLoopPredecessor || !InnerLoopLatchPredecessor ||
>> + !OuterLoopLatchBI || !InnerLoopLatchBI || !OuterLoopHeaderBI ||
>> + !InnerLoopHeaderBI)
>> + return false;
>> +
>> + BranchInst *InnerLoopLatchPredecessorBI =
>> + dyn_cast<BranchInst>(InnerLoopLatchPredecessor->getTerminator());
>> + BranchInst *OuterLoopPredecessorBI =
>> + dyn_cast<BranchInst>(OuterLoopPredecessor->getTerminator());
>> +
>> + if (!OuterLoopPredecessorBI || !InnerLoopLatchPredecessorBI)
>> + return false;
>> + BasicBlock *InnerLoopHeaderSucessor = InnerLoopHeader->getUniqueSuccessor();
>> + if (!InnerLoopHeaderSucessor)
>> + return false;
>> +
>> + // Adjust Loop Preheader and headers
>> +
>> + unsigned NumSucc = OuterLoopPredecessorBI->getNumSuccessors();
>> + for (unsigned i = 0; i < NumSucc; ++i) {
>> + if (OuterLoopPredecessorBI->getSuccessor(i) == OuterLoopPreHeader)
>> + OuterLoopPredecessorBI->setSuccessor(i, InnerLoopPreHeader);
>> + }
>> +
>> + NumSucc = OuterLoopHeaderBI->getNumSuccessors();
>> + for (unsigned i = 0; i < NumSucc; ++i) {
>> + if (OuterLoopHeaderBI->getSuccessor(i) == OuterLoopLatch)
>> + OuterLoopHeaderBI->setSuccessor(i, LoopExit);
>> + else if (OuterLoopHeaderBI->getSuccessor(i) == InnerLoopPreHeader)
>> + OuterLoopHeaderBI->setSuccessor(i, InnerLoopHeaderSucessor);
>> + }
>> +
>> + BranchInst::Create(OuterLoopPreHeader, InnerLoopHeaderBI);
>> + InnerLoopHeaderBI->eraseFromParent();
>> +
>> + // -------------Adjust loop latches-----------
>> + if (InnerLoopLatchBI->getSuccessor(0) == InnerLoopHeader)
>> + InnerLoopLatchSuccessor = InnerLoopLatchBI->getSuccessor(1);
>> + else
>> + InnerLoopLatchSuccessor = InnerLoopLatchBI->getSuccessor(0);
>> +
>> + NumSucc = InnerLoopLatchPredecessorBI->getNumSuccessors();
>> + for (unsigned i = 0; i < NumSucc; ++i) {
>> + if (InnerLoopLatchPredecessorBI->getSuccessor(i) == InnerLoopLatch)
>> + InnerLoopLatchPredecessorBI->setSuccessor(i, InnerLoopLatchSuccessor);
>> + }
>> +
>> + if (OuterLoopLatchBI->getSuccessor(0) == OuterLoopHeader)
>> + OuterLoopLatchSuccessor = OuterLoopLatchBI->getSuccessor(1);
>> + else
>> + OuterLoopLatchSuccessor = OuterLoopLatchBI->getSuccessor(0);
>> +
>> + if (InnerLoopLatchBI->getSuccessor(1) == InnerLoopLatchSuccessor)
>> + InnerLoopLatchBI->setSuccessor(1, OuterLoopLatchSuccessor);
>> + else
>> + InnerLoopLatchBI->setSuccessor(0, OuterLoopLatchSuccessor);
>> +
>> + if (OuterLoopLatchBI->getSuccessor(0) == OuterLoopLatchSuccessor) {
>> + OuterLoopLatchBI->setSuccessor(0, InnerLoopLatch);
>> + } else {
>> + OuterLoopLatchBI->setSuccessor(1, InnerLoopLatch);
>> + }
>> +
>> + return true;
>> +}
>> +void LoopInterchangeTransform::adjustLoopPreheaders() {
>> +
>> + // We have interchanged the preheaders so we need to interchange the data in
>> + // the preheader as well.
>> + // This is because the content of inner preheader was previously executed
>> + // inside the outer loop.
>> + BasicBlock *OuterLoopPreHeader = OuterLoop->getLoopPreheader();
>> + BasicBlock *InnerLoopPreHeader = InnerLoop->getLoopPreheader();
>> + BasicBlock *OuterLoopHeader = OuterLoop->getHeader();
>> + BranchInst *InnerTermBI =
>> + cast<BranchInst>(InnerLoopPreHeader->getTerminator());
>> +
>> + SmallVector<Value *, 16> OuterPreheaderInstr;
>> + SmallVector<Value *, 16> InnerPreheaderInstr;
>> +
>> + for (auto I = OuterLoopPreHeader->begin(); !isa<BranchInst>(I); ++I)
>> + OuterPreheaderInstr.push_back(I);
>> +
>> + for (auto I = InnerLoopPreHeader->begin(); !isa<BranchInst>(I); ++I)
>> + InnerPreheaderInstr.push_back(I);
>> +
>> + BasicBlock *HeaderSplit =
>> + SplitBlock(OuterLoopHeader, OuterLoopHeader->getTerminator(), DT, LI);
>> + Instruction *InsPoint = HeaderSplit->getFirstNonPHI();
>> + // These instructions should now be executed inside the loop.
>> + // Move instruction into a new block after outer header.
>> + for (auto I = InnerPreheaderInstr.begin(), E = InnerPreheaderInstr.end();
>> + I != E; ++I) {
>> + Instruction *Ins = cast<Instruction>(*I);
>> + Ins->moveBefore(InsPoint);
>> + }
>> + // These instructions were not executed previously in the loop so move them to
>> + // the older inner loop preheader.
>> + for (auto I = OuterPreheaderInstr.begin(), E = OuterPreheaderInstr.end();
>> + I != E; ++I) {
>> + Instruction *Ins = cast<Instruction>(*I);
>> + Ins->moveBefore(InnerTermBI);
>> + }
>> +}
>> +
>> +bool LoopInterchangeTransform::adjustLoopLinks() {
>> +
>> + // Adjust all branches in the inner and outer loop.
>> + bool Changed = adjustLoopBranches();
>> + if (Changed)
>> + adjustLoopPreheaders();
>> + return Changed;
>> +}
>> +
>> +char LoopInterchange::ID = 0;
>> +INITIALIZE_PASS_BEGIN(LoopInterchange, "loop-interchange",
>> + "Interchanges loops for cache reuse", false, false)
>> +INITIALIZE_AG_DEPENDENCY(AliasAnalysis)
>> +INITIALIZE_PASS_DEPENDENCY(DependenceAnalysis)
>> +INITIALIZE_PASS_DEPENDENCY(DominatorTreeWrapperPass)
>> +INITIALIZE_PASS_DEPENDENCY(ScalarEvolution)
>> +INITIALIZE_PASS_DEPENDENCY(LoopSimplify)
>> +INITIALIZE_PASS_DEPENDENCY(LCSSA)
>> +INITIALIZE_PASS_DEPENDENCY(LoopInfoWrapperPass)
>> +
>> +INITIALIZE_PASS_END(LoopInterchange, "loop-interchange",
>> + "Interchanges loops for cache reuse", false, false)
>> +
>> +Pass *llvm::createLoopInterchangePass() { return new LoopInterchange(); }
>>
>> Modified: llvm/trunk/lib/Transforms/Scalar/Scalar.cpp
>> URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/lib/Transforms/Scalar/Scalar.cpp?rev=231458&r1=231457&r2=231458&view=diff
>> ==============================================================================
>> --- llvm/trunk/lib/Transforms/Scalar/Scalar.cpp (original)
>> +++ llvm/trunk/lib/Transforms/Scalar/Scalar.cpp Fri Mar 6 04:11:25 2015
>> @@ -48,6 +48,7 @@ void llvm::initializeScalarOpts(PassRegi
>> initializeLoopDeletionPass(Registry);
>> initializeLoopAccessAnalysisPass(Registry);
>> initializeLoopInstSimplifyPass(Registry);
>> + initializeLoopInterchangePass(Registry);
>> initializeLoopRotatePass(Registry);
>> initializeLoopStrengthReducePass(Registry);
>> initializeLoopRerollPass(Registry);
>>
>> Added: llvm/trunk/test/Transforms/LoopInterchange/currentLimitation.ll
>> URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/test/Transforms/LoopInterchange/currentLimitation.ll?rev=231458&view=auto
>> ==============================================================================
>> --- llvm/trunk/test/Transforms/LoopInterchange/currentLimitation.ll (added)
>> +++ llvm/trunk/test/Transforms/LoopInterchange/currentLimitation.ll Fri Mar 6 04:11:25 2015
>> @@ -0,0 +1,58 @@
>> +; RUN: opt < %s -basicaa -loop-interchange -S | FileCheck %s
>> +;; These are test that fail to interchange due to current limitation. This will go off once we extend the loop interchange pass.
>> +
>> +target datalayout = "e-m:e-i64:64-f80:128-n8:16:32:64-S128"
>> +target triple = "x86_64-unknown-linux-gnu"
>> +
>> + at A = common global [100 x [100 x i32]] zeroinitializer
>> + at B = common global [100 x [100 x [100 x i32]]] zeroinitializer
>> +
>> +;;--------------------------------------Test case 01------------------------------------
>> +;; [FIXME] This loop though valid is currently not interchanged due to the limitation that we cannot split the inner loop latch due to multiple use of inner induction
>> +;; variable.(used to increment the loop counter and to access A[j+1][i+1]
>> +;; for(int i=0;i<N-1;i++)
>> +;; for(int j=1;j<N-1;j++)
>> +;; A[j+1][i+1] = A[j+1][i+1] + k;
>> +
>> +define void @interchange_01(i32 %k, i32 %N) {
>> + entry:
>> + %sub = add nsw i32 %N, -1
>> + %cmp26 = icmp sgt i32 %N, 1
>> + br i1 %cmp26, label %for.cond1.preheader.lr.ph<http://for.cond1.preheader.lr.ph>, label %for.end17
>> +
>> + for.cond1.preheader.lr.ph<http://for.cond1.preheader.lr.ph>:
>> + %cmp324 = icmp sgt i32 %sub, 1
>> + %0 = add i32 %N, -2
>> + %1 = sext i32 %sub to i64
>> + br label %for.cond1.preheader
>> +
>> + for.cond.loopexit:
>> + %cmp = icmp slt i64 %indvars.iv.next29, %1
>> + br i1 %cmp, label %for.cond1.preheader, label %for.end17
>> +
>> + for.cond1.preheader:
>> + %indvars.iv28 = phi i64 [ 0, %for.cond1.preheader.lr.ph<http://for.cond1.preheader.lr.ph> ], [ %indvars.iv.next29, %for.cond.loopexit ]
>> + %indvars.iv.next29 = add nuw nsw i64 %indvars.iv28, 1
>> + br i1 %cmp324, label %for.body4, label %for.cond.loopexit
>> +
>> + for.body4:
>> + %indvars.iv = phi i64 [ %indvars.iv.next, %for.body4 ], [ 1, %for.cond1.preheader ]
>> + %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
>> + %arrayidx7 = getelementptr inbounds [100 x [100 x i32]], [100 x [100 x i32]]* @A, i64 0, i64 %indvars.iv.next, i64 %indvars.iv.next29
>> + %2 = load i32, i32* %arrayidx7
>> + %add8 = add nsw i32 %2, %k
>> + store i32 %add8, i32* %arrayidx7
>> + %lftr.wideiv = trunc i64 %indvars.iv to i32
>> + %exitcond = icmp eq i32 %lftr.wideiv, %0
>> + br i1 %exitcond, label %for.cond.loopexit, label %for.body4
>> +
>> + for.end17:
>> + ret void
>> +}
>> +;; Inner loop not split so it is not interchanged.
>> +; CHECK-LABEL: @interchange_01
>> +; CHECK: for.body4:
>> +; CHECK-NEXT: %indvars.iv = phi i64 [ %indvars.iv.next, %for.body4 ], [ 1, %for.body4.preheader ]
>> +; CHECK-NEXT: %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
>> +; CHECK-NEXT: %arrayidx7 = getelementptr inbounds [100 x [100 x i32]], [100 x [100 x i32]]* @A, i64 0, i64 %indvars.iv.next, i64 %indvars.iv.next29
>> +
>>
>> Added: llvm/trunk/test/Transforms/LoopInterchange/interchange.ll
>> URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/test/Transforms/LoopInterchange/interchange.ll?rev=231458&view=auto
>> ==============================================================================
>> --- llvm/trunk/test/Transforms/LoopInterchange/interchange.ll (added)
>> +++ llvm/trunk/test/Transforms/LoopInterchange/interchange.ll Fri Mar 6 04:11:25 2015
>> @@ -0,0 +1,557 @@
>> +; RUN: opt < %s -basicaa -loop-interchange -S | FileCheck %s
>> +;; We test the complete .ll for adjustment in outer loop header/latch and inner loop header/latch.
>> +
>> +target datalayout = "e-m:e-i64:64-f80:128-n8:16:32:64-S128"
>> +target triple = "x86_64-unknown-linux-gnu"
>> +
>> + at A = common global [100 x [100 x i32]] zeroinitializer
>> + at B = common global [100 x i32] zeroinitializer
>> + at C = common global [100 x [100 x i32]] zeroinitializer
>> + at D = common global [100 x [100 x [100 x i32]]] zeroinitializer
>> +
>> +declare void @foo(...)
>> +
>> +;;--------------------------------------Test case 01------------------------------------
>> +;; for(int i=0;i<N;i++)
>> +;; for(int j=1;j<N;j++)
>> +;; A[j][i] = A[j][i]+k;
>> +
>> +define void @interchange_01(i32 %k, i32 %N) {
>> +entry:
>> + %cmp21 = icmp sgt i32 %N, 0
>> + br i1 %cmp21, label %for.cond1.preheader.lr.ph<http://for.cond1.preheader.lr.ph>, label %for.end12
>> +
>> +for.cond1.preheader.lr.ph<http://for.cond1.preheader.lr.ph>:
>> + %cmp219 = icmp sgt i32 %N, 1
>> + %0 = add i32 %N, -1
>> + br label %for.cond1.preheader
>> +
>> +for.cond1.preheader:
>> + %indvars.iv23 = phi i64 [ 0, %for.cond1.preheader.lr.ph<http://for.cond1.preheader.lr.ph> ], [ %indvars.iv.next24, %for.inc10 ]
>> + br i1 %cmp219, label %for.body3, label %for.inc10
>> +
>> +for.body3:
>> + %indvars.iv = phi i64 [ %indvars.iv.next, %for.body3 ], [ 1, %for.cond1.preheader ]
>> + %arrayidx5 = getelementptr inbounds [100 x [100 x i32]], [100 x [100 x i32]]* @A, i64 0, i64 %indvars.iv, i64 %indvars.iv23
>> + %1 = load i32, i32* %arrayidx5
>> + %add = add nsw i32 %1, %k
>> + store i32 %add, i32* %arrayidx5
>> + %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
>> + %lftr.wideiv = trunc i64 %indvars.iv to i32
>> + %exitcond = icmp eq i32 %lftr.wideiv, %0
>> + br i1 %exitcond, label %for.inc10, label %for.body3
>> +
>> +for.inc10:
>> + %indvars.iv.next24 = add nuw nsw i64 %indvars.iv23, 1
>> + %lftr.wideiv25 = trunc i64 %indvars.iv23 to i32
>> + %exitcond26 = icmp eq i32 %lftr.wideiv25, %0
>> + br i1 %exitcond26, label %for.end12, label %for.cond1.preheader
>> +
>> +for.end12:
>> + ret void
>> +}
>> +
>> +; CHECK-LABEL: @interchange_01
>> +; CHECK: entry:
>> +; CHECK: %cmp21 = icmp sgt i32 %N, 0
>> +; CHECK: br i1 %cmp21, label %for.body3.preheader, label %for.end12
>> +; CHECK: for.cond1.preheader.lr.ph<http://for.cond1.preheader.lr.ph>:
>> +; CHECK: br label %for.cond1.preheader
>> +; CHECK: for.cond1.preheader:
>> +; CHECK: %indvars.iv23 = phi i64 [ 0, %for.cond1.preheader.lr.ph<http://for.cond1.preheader.lr.ph> ], [ %indvars.iv.next24, %for.inc10 ]
>> +; CHECK: br i1 %cmp219, label %for.body3.split1, label %for.end12.loopexit
>> +; CHECK: for.body3.preheader:
>> +; CHECK: %cmp219 = icmp sgt i32 %N, 1
>> +; CHECK: %0 = add i32 %N, -1
>> +; CHECK: br label %for.body3
>> +; CHECK: for.body3:
>> +; CHECK: %indvars.iv = phi i64 [ %indvars.iv.next, %for.body3.split ], [ 1, %for.body3.preheader ]
>> +; CHECK: br label %for.cond1.preheader.lr.ph<http://for.cond1.preheader.lr.ph>
>> +; CHECK: for.body3.split1:
>> +; CHECK: %arrayidx5 = getelementptr inbounds [100 x [100 x i32]], [100 x [100 x i32]]* @A, i64 0, i64 %indvars.iv, i64 %indvars.iv23
>> +; CHECK: %1 = load i32, i32* %arrayidx5
>> +; CHECK: %add = add nsw i32 %1, %k
>> +; CHECK: store i32 %add, i32* %arrayidx5
>> +; CHECK: br label %for.inc10.loopexit
>> +; CHECK: for.body3.split:
>> +; CHECK: %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
>> +; CHECK: %lftr.wideiv = trunc i64 %indvars.iv to i32
>> +; CHECK: %exitcond = icmp eq i32 %lftr.wideiv, %0
>> +; CHECK: br i1 %exitcond, label %for.end12.loopexit, label %for.body3
>> +; CHECK: for.inc10.loopexit:
>> +; CHECK: br label %for.inc10
>> +; CHECK: for.inc10:
>> +; CHECK: %indvars.iv.next24 = add nuw nsw i64 %indvars.iv23, 1
>> +; CHECK: %lftr.wideiv25 = trunc i64 %indvars.iv23 to i32
>> +; CHECK: %exitcond26 = icmp eq i32 %lftr.wideiv25, %0
>> +; CHECK: br i1 %exitcond26, label %for.body3.split, label %for.cond1.preheader
>> +; CHECK: for.end12.loopexit:
>> +; CHECK: br label %for.end12
>> +; CHECK: for.end12:
>> +; CHECK: ret void
>> +
>> +;;--------------------------------------Test case 02-------------------------------------
>> +
>> +;; for(int i=0;i<100;i++)
>> +;; for(int j=100;j>=0;j--)
>> +;; A[j][i] = A[j][i]+k;
>> +
>> +define void @interchange_02(i32 %k) {
>> +entry:
>> + br label %for.cond1.preheader
>> +
>> +for.cond1.preheader:
>> + %indvars.iv19 = phi i64 [ 0, %entry ], [ %indvars.iv.next20, %for.inc10 ]
>> + br label %for.body3
>> +
>> +for.body3:
>> + %indvars.iv = phi i64 [ 100, %for.cond1.preheader ], [ %indvars.iv.next, %for.body3 ]
>> + %arrayidx5 = getelementptr inbounds [100 x [100 x i32]], [100 x [100 x i32]]* @A, i64 0, i64 %indvars.iv, i64 %indvars.iv19
>> + %0 = load i32, i32* %arrayidx5
>> + %add = add nsw i32 %0, %k
>> + store i32 %add, i32* %arrayidx5
>> + %indvars.iv.next = add nsw i64 %indvars.iv, -1
>> + %cmp2 = icmp sgt i64 %indvars.iv, 0
>> + br i1 %cmp2, label %for.body3, label %for.inc10
>> +
>> +for.inc10:
>> + %indvars.iv.next20 = add nuw nsw i64 %indvars.iv19, 1
>> + %exitcond = icmp eq i64 %indvars.iv.next20, 100
>> + br i1 %exitcond, label %for.end11, label %for.cond1.preheader
>> +
>> +for.end11:
>> + ret void
>> +}
>> +
>> +; CHECK-LABEL: @interchange_02
>> +; CHECK: entry:
>> +; CHECK: br label %for.body3.preheader
>> +; CHECK: for.cond1.preheader.preheader:
>> +; CHECK: br label %for.cond1.preheader
>> +; CHECK: for.cond1.preheader:
>> +; CHECK: %indvars.iv19 = phi i64 [ %indvars.iv.next20, %for.inc10 ], [ 0, %for.cond1.preheader.preheader ]
>> +; CHECK: br label %for.body3.split1
>> +; CHECK: for.body3.preheader:
>> +; CHECK: br label %for.body3
>> +; CHECK: for.body3:
>> +; CHECK: %indvars.iv = phi i64 [ %indvars.iv.next, %for.body3.split ], [ 100, %for.body3.preheader ]
>> +; CHECK: br label %for.cond1.preheader.preheader
>> +; CHECK: for.body3.split1: ; preds = %for.cond1.preheader
>> +; CHECK: %arrayidx5 = getelementptr inbounds [100 x [100 x i32]], [100 x [100 x i32]]* @A, i64 0, i64 %indvars.iv, i64 %indvars.iv19
>> +; CHECK: %0 = load i32, i32* %arrayidx5
>> +; CHECK: %add = add nsw i32 %0, %k
>> +; CHECK: store i32 %add, i32* %arrayidx5
>> +; CHECK: br label %for.inc10
>> +; CHECK: for.body3.split:
>> +; CHECK: %indvars.iv.next = add nsw i64 %indvars.iv, -1
>> +; CHECK: %cmp2 = icmp sgt i64 %indvars.iv, 0
>> +; CHECK: br i1 %cmp2, label %for.body3, label %for.end11
>> +; CHECK: for.inc10:
>> +; CHECK: %indvars.iv.next20 = add nuw nsw i64 %indvars.iv19, 1
>> +; CHECK: %exitcond = icmp eq i64 %indvars.iv.next20, 100
>> +; CHECK: br i1 %exitcond, label %for.body3.split, label %for.cond1.preheader
>> +; CHECK: for.end11:
>> +; CHECK: ret void
>> +
>> +;;--------------------------------------Test case 03-------------------------------------
>> +;; Loops should not be interchanged in this case as it is not profitable.
>> +;; for(int i=0;i<100;i++)
>> +;; for(int j=0;j<100;j++)
>> +;; A[i][j] = A[i][j]+k;
>> +
>> +define void @interchange_03(i32 %k) {
>> +entry:
>> + br label %for.cond1.preheader
>> +
>> +for.cond1.preheader:
>> + %indvars.iv21 = phi i64 [ 0, %entry ], [ %indvars.iv.next22, %for.inc10 ]
>> + br label %for.body3
>> +
>> +for.body3:
>> + %indvars.iv = phi i64 [ 0, %for.cond1.preheader ], [ %indvars.iv.next, %for.body3 ]
>> + %arrayidx5 = getelementptr inbounds [100 x [100 x i32]], [100 x [100 x i32]]* @A, i64 0, i64 %indvars.iv21, i64 %indvars.iv
>> + %0 = load i32, i32* %arrayidx5
>> + %add = add nsw i32 %0, %k
>> + store i32 %add, i32* %arrayidx5
>> + %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
>> + %exitcond = icmp eq i64 %indvars.iv.next, 100
>> + br i1 %exitcond, label %for.inc10, label %for.body3
>> +
>> +for.inc10:
>> + %indvars.iv.next22 = add nuw nsw i64 %indvars.iv21, 1
>> + %exitcond23 = icmp eq i64 %indvars.iv.next22, 100
>> + br i1 %exitcond23, label %for.end12, label %for.cond1.preheader
>> +
>> +for.end12:
>> + ret void
>> +}
>> +
>> +; CHECK-LABEL: @interchange_03
>> +; CHECK: entry:
>> +; CHECK: br label %for.cond1.preheader.preheader
>> +; CHECK: for.cond1.preheader.preheader: ; preds = %entry
>> +; CHECK: br label %for.cond1.preheader
>> +; CHECK: for.cond1.preheader: ; preds = %for.cond1.preheader.preheader, %for.inc10
>> +; CHECK: %indvars.iv21 = phi i64 [ %indvars.iv.next22, %for.inc10 ], [ 0, %for.cond1.preheader.preheader ]
>> +; CHECK: br label %for.body3.preheader
>> +; CHECK: for.body3.preheader: ; preds = %for.cond1.preheader
>> +; CHECK: br label %for.body3
>> +; CHECK: for.body3: ; preds = %for.body3.preheader, %for.body3
>> +; CHECK: %indvars.iv = phi i64 [ %indvars.iv.next, %for.body3 ], [ 0, %for.body3.preheader ]
>> +; CHECK: %arrayidx5 = getelementptr inbounds [100 x [100 x i32]], [100 x [100 x i32]]* @A, i64 0, i64 %indvars.iv21, i64 %indvars.iv
>> +; CHECK: %0 = load i32, i32* %arrayidx5
>> +; CHECK: %add = add nsw i32 %0, %k
>> +; CHECK: store i32 %add, i32* %arrayidx5
>> +; CHECK: %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
>> +; CHECK: %exitcond = icmp eq i64 %indvars.iv.next, 100
>> +; CHECK: br i1 %exitcond, label %for.inc10, label %for.body3
>> +; CHECK: for.inc10: ; preds = %for.body3
>> +; CHECK: %indvars.iv.next22 = add nuw nsw i64 %indvars.iv21, 1
>> +; CHECK: %exitcond23 = icmp eq i64 %indvars.iv.next22, 100
>> +; CHECK: br i1 %exitcond23, label %for.end12, label %for.cond1.preheader
>> +; CHECK: for.end12: ; preds = %for.inc10
>> +; CHECK: ret void
>> +
>> +
>> +;;--------------------------------------Test case 04-------------------------------------
>> +;; Loops should not be interchanged in this case as it is not legal due to dependency.
>> +;; for(int j=0;j<99;j++)
>> +;; for(int i=0;i<99;i++)
>> +;; A[j][i+1] = A[j+1][i]+k;
>> +
>> +define void @interchange_04(i32 %k){
>> +entry:
>> + br label %for.cond1.preheader
>> +
>> +for.cond1.preheader:
>> + %indvars.iv23 = phi i64 [ 0, %entry ], [ %indvars.iv.next24, %for.inc12 ]
>> + %indvars.iv.next24 = add nuw nsw i64 %indvars.iv23, 1
>> + br label %for.body3
>> +
>> +for.body3:
>> + %indvars.iv = phi i64 [ 0, %for.cond1.preheader ], [ %indvars.iv.next, %for.body3 ]
>> + %arrayidx5 = getelementptr inbounds [100 x [100 x i32]], [100 x [100 x i32]]* @A, i64 0, i64 %indvars.iv.next24, i64 %indvars.iv
>> + %0 = load i32, i32* %arrayidx5
>> + %add6 = add nsw i32 %0, %k
>> + %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
>> + %arrayidx11 = getelementptr inbounds [100 x [100 x i32]], [100 x [100 x i32]]* @A, i64 0, i64 %indvars.iv23, i64 %indvars.iv.next
>> + store i32 %add6, i32* %arrayidx11
>> + %exitcond = icmp eq i64 %indvars.iv.next, 99
>> + br i1 %exitcond, label %for.inc12, label %for.body3
>> +
>> +for.inc12:
>> + %exitcond25 = icmp eq i64 %indvars.iv.next24, 99
>> + br i1 %exitcond25, label %for.end14, label %for.cond1.preheader
>> +
>> +for.end14:
>> + ret void
>> +}
>> +
>> +; CHECK-LABEL: @interchange_04
>> +; CHECK: entry:
>> +; CHECK: br label %for.cond1.preheader
>> +; CHECK: for.cond1.preheader: ; preds = %for.inc12, %entry
>> +; CHECK: %indvars.iv23 = phi i64 [ 0, %entry ], [ %indvars.iv.next24, %for.inc12 ]
>> +; CHECK: %indvars.iv.next24 = add nuw nsw i64 %indvars.iv23, 1
>> +; CHECK: br label %for.body3
>> +; CHECK: for.body3: ; preds = %for.body3, %for.cond1.preheader
>> +; CHECK: %indvars.iv = phi i64 [ 0, %for.cond1.preheader ], [ %indvars.iv.next, %for.body3 ]
>> +; CHECK: %arrayidx5 = getelementptr inbounds [100 x [100 x i32]], [100 x [100 x i32]]* @A, i64 0, i64 %indvars.iv.next24, i64 %indvars.iv
>> +; CHECK: %0 = load i32, i32* %arrayidx5
>> +; CHECK: %add6 = add nsw i32 %0, %k
>> +; CHECK: %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
>> +; CHECK: %arrayidx11 = getelementptr inbounds [100 x [100 x i32]], [100 x [100 x i32]]* @A, i64 0, i64 %indvars.iv23, i64 %indvars.iv.next
>> +; CHECK: store i32 %add6, i32* %arrayidx11
>> +; CHECK: %exitcond = icmp eq i64 %indvars.iv.next, 99
>>
...
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20150424/78069518/attachment.html>
More information about the llvm-commits
mailing list