From ca2396e6732dfab06cf96a475520b9266852277a Mon Sep 17 00:00:00 2001
From: Alexey Bataev <a.bataev@hotmail.com>
Date: Mon, 12 Feb 2018 14:54:48 +0000
Subject: [SLP] Take user instructions cost into consideration in insertelement
 vectorization.

Summary:
For better vectorization result we should take into consideration the
cost of the user insertelement instructions when we try to
vectorize sequences that build the whole vector. I.e. if we have the
following scalar code:
```
<Scalar code>
insertelement <ScalarCode>, ...
```
we should consider the cost of the last `insertelement ` instructions as
the cost of the scalar code.

Reviewers: RKSimon, spatel, hfinkel, mkuper

Subscribers: javed.absar, llvm-commits

Differential Revision: https://reviews.llvm.org/D42657

llvm-svn: 324893
---
 llvm/lib/Transforms/Vectorize/SLPVectorizer.cpp | 28 ++++++++++++++++++-------
 1 file changed, 20 insertions(+), 8 deletions(-)

(limited to 'llvm/lib/Transforms/Vectorize/SLPVectorizer.cpp')

diff --git a/llvm/lib/Transforms/Vectorize/SLPVectorizer.cpp b/llvm/lib/Transforms/Vectorize/SLPVectorizer.cpp
index 8b63fe1bc3c..0f0b82ba6c6 100644
--- a/llvm/lib/Transforms/Vectorize/SLPVectorizer.cpp
+++ b/llvm/lib/Transforms/Vectorize/SLPVectorizer.cpp
@@ -4702,11 +4702,11 @@ bool SLPVectorizerPass::tryToVectorizePair(Value *A, Value *B, BoUpSLP &R) {
   if (!A || !B)
     return false;
   Value *VL[] = { A, B };
-  return tryToVectorizeList(VL, R, true);
+  return tryToVectorizeList(VL, R, /*UserCost=*/0, true);
 }
 
 bool SLPVectorizerPass::tryToVectorizeList(ArrayRef<Value *> VL, BoUpSLP &R,
-                                           bool AllowReorder) {
+                                           int UserCost, bool AllowReorder) {
   if (VL.size() < 2)
     return false;
 
@@ -4815,7 +4815,7 @@ bool SLPVectorizerPass::tryToVectorizeList(ArrayRef<Value *> VL, BoUpSLP &R,
         continue;
 
       R.computeMinimumValueSizes();
-      int Cost = R.getTreeCost();
+      int Cost = R.getTreeCost() - UserCost;
       CandidateFound = true;
       MinCost = std::min(MinCost, Cost);
 
@@ -5748,9 +5748,17 @@ private:
 ///
 /// Returns true if it matches
 static bool findBuildVector(InsertElementInst *LastInsertElem,
-                            SmallVectorImpl<Value *> &BuildVectorOpds) {
+                            TargetTransformInfo *TTI,
+                            SmallVectorImpl<Value *> &BuildVectorOpds,
+                            int &UserCost) {
+  UserCost = 0;
   Value *V = nullptr;
   do {
+    if (auto *CI = dyn_cast<ConstantInt>(LastInsertElem->getOperand(2))) {
+      UserCost += TTI->getVectorInstrCost(Instruction::InsertElement,
+                                          LastInsertElem->getType(),
+                                          CI->getZExtValue());
+    }
     BuildVectorOpds.push_back(LastInsertElem->getOperand(1));
     V = LastInsertElem->getOperand(0);
     if (isa<UndefValue>(V))
@@ -5965,13 +5973,17 @@ bool SLPVectorizerPass::vectorizeInsertValueInst(InsertValueInst *IVI,
 
 bool SLPVectorizerPass::vectorizeInsertElementInst(InsertElementInst *IEI,
                                                    BasicBlock *BB, BoUpSLP &R) {
+  int UserCost;
   SmallVector<Value *, 16> BuildVectorOpds;
-  if (!findBuildVector(IEI, BuildVectorOpds))
+  if (!findBuildVector(IEI, TTI, BuildVectorOpds, UserCost) ||
+      (llvm::all_of(BuildVectorOpds,
+                    [](Value *V) { return isa<ExtractElementInst>(V); }) &&
+       isShuffle(BuildVectorOpds)))
     return false;
 
   // Vectorize starting with the build vector operands ignoring the BuildVector
   // instructions for the purpose of scheduling and user extraction.
-  return tryToVectorizeList(BuildVectorOpds, R);
+  return tryToVectorizeList(BuildVectorOpds, R, UserCost);
 }
 
 bool SLPVectorizerPass::vectorizeCmpInst(CmpInst *CI, BasicBlock *BB,
@@ -6049,8 +6061,8 @@ bool SLPVectorizerPass::vectorizeChainsInBlock(BasicBlock *BB, BoUpSLP &R) {
       // is done when there are exactly two elements since tryToVectorizeList
       // asserts that there are only two values when AllowReorder is true.
       bool AllowReorder = NumElts == 2;
-      if (NumElts > 1 &&
-          tryToVectorizeList(makeArrayRef(IncIt, NumElts), R, AllowReorder)) {
+      if (NumElts > 1 && tryToVectorizeList(makeArrayRef(IncIt, NumElts), R,
+                                            /*UserCost=*/0, AllowReorder)) {
         // Success start over because instructions might have been changed.
         HaveVectorizedPhiNodes = true;
         Changed = true;
-- 
cgit v1.2.3