[VectorCombine] Add foldShuffleToIdentity

This patch adds a basic version of a combine that attempts to fold away
shuffles that when combines simplify away to an identity shuffle.
For example:
  %ab = shufflevector <8 x half> %a, <8 x half> poison, <4 x i32> <i32 3, i32 2, i32 1, i32 0>
  %at = shufflevector <8 x half> %a, <8 x half> poison, <4 x i32> <i32 7, i32 6, i32 5, i32 4>
  %abt = fneg <4 x half> %at
  %abb = fneg <4 x half> %ab
  %r = shufflevector <4 x half> %abt, <4 x half> %abb, <8 x i32> <i32 7, i32 6, i32 5, i32 4, i32 3, i32 2, i32 1, i32 0>
By looking through the shuffles, it can be simplified to:
  %r = fneg <8 x half> %a

The code tracks each lane starting from the original shuffle, keeping a track
of a vector of {src, idx}. As we propagate up through the instructions we will
either look through intermediate instructions (binops and unops) or see a
collections of lanes that all have the same src and incrementing idx (an
identity). We can also see a single value with identical lanes, which we can
treat like a splat.

Only the basic version is added here, handling identites, splats, binops and
unops. In follow-up patches other instructions can be added such as constants,
intrinsics, cmp/sel and zext/sext/trunc.

Author: davemgreen

llvm/lib/Transforms/Vectorize/VectorCombine.cpp

@@ -113,6 +113,7 @@ class VectorCombine {
   bool scalarizeLoadExtract(Instruction &I);
   bool foldShuffleOfBinops(Instruction &I);
   bool foldShuffleOfCastops(Instruction &I);
+  bool foldShuffleToIdentity(Instruction &I);
   bool foldShuffleFromReductions(Instruction &I);
   bool foldTruncFromReductions(Instruction &I);
   bool foldSelectShuffle(Instruction &I, bool FromReduction = false);
@@ -1552,6 +1553,148 @@ bool VectorCombine::foldShuffleOfCastops(Instruction &I) {
   return true;
 }
 
+// Starting from a shuffle, look up through operands tracking the shuffled index
+// of each lane. If we can simplify away the shuffles to identities then
+// do so.
+bool VectorCombine::foldShuffleToIdentity(Instruction &I) {
+  FixedVectorType *Ty = dyn_cast<FixedVectorType>(I.getType());
+  if (!Ty || !isa<Instruction>(I.getOperand(0)) ||
+      !isa<Instruction>(I.getOperand(1)))
+    return false;
+
+  using InstLane = std::pair<Value *, int>;
+
+  auto LookThroughShuffles = [](Value *V, int Lane) -> InstLane {
+    while (auto *SV = dyn_cast<ShuffleVectorInst>(V)) {
+      unsigned NumElts =
+          cast<FixedVectorType>(SV->getOperand(0)->getType())->getNumElements();
+      int M = SV->getMaskValue(Lane);
+      if (M < 0)
+        return {nullptr, -1};
+      else if (M < (int)NumElts) {
+        V = SV->getOperand(0);
+        Lane = M;
+      } else {
+        V = SV->getOperand(1);
+        Lane = M - NumElts;
+      }
+    }
+    return InstLane{V, Lane};
+  };
+
+  auto GenerateInstLaneVectorFromOperand =
+      [&LookThroughShuffles](const SmallVector<InstLane> &Item, int Op) {
+        SmallVector<InstLane> NItem;
+        for (InstLane V : Item) {
+          NItem.emplace_back(
+              !V.first
+                  ? InstLane{nullptr, -1}
+                  : LookThroughShuffles(
+                        cast<Instruction>(V.first)->getOperand(Op), V.second));
+        }
+        return NItem;
+      };
+
+  SmallVector<InstLane> Start;
+  for (unsigned M = 0; M < Ty->getNumElements(); ++M)
+    Start.push_back(LookThroughShuffles(&I, M));
+
+  SmallVector<SmallVector<InstLane>> Worklist;
+  Worklist.push_back(Start);
+  SmallPtrSet<Value *, 4> IdentityLeafs, SplatLeafs;
+  unsigned NumVisited = 0;
+
+  while (!Worklist.empty()) {
+    SmallVector<InstLane> Item = Worklist.pop_back_val();
+    if (++NumVisited > MaxInstrsToScan)
+      return false;
+
+    // If we found an undef first lane then bail out to keep things simple.
+    if (!Item[0].first)
+      return false;
+
+    // Look for an identity value.
+    if (Item[0].second == 0 && Item[0].first->getType() == Ty &&
+        all_of(drop_begin(enumerate(Item)), [&](const auto &E) {
+          return !E.value().first || (E.value().first == Item[0].first &&
+                                      E.value().second == (int)E.index());
+        })) {
+      IdentityLeafs.insert(Item[0].first);
+      continue;
+    }
+    // Look for a splat value.
+    if (all_of(drop_begin(Item), [&](InstLane &IL) {
+          return !IL.first ||
+                 (IL.first == Item[0].first && IL.second == Item[0].second);
+        })) {
+      SplatLeafs.insert(Item[0].first);
+      continue;
+    }
+
+    // We need each element to be the same type of value, and check that each
+    // element has a single use.
+    if (!all_of(drop_begin(Item), [&](InstLane IL) {
+          if (!IL.first)
+            return true;
+          if (isa<Instruction>(IL.first) &&
+              !cast<Instruction>(IL.first)->hasOneUse())
+            return false;
+          return IL.first->getValueID() == Item[0].first->getValueID() &&
+                 (!isa<IntrinsicInst>(IL.first) ||
+                  cast<IntrinsicInst>(IL.first)->getIntrinsicID() ==
+                      cast<IntrinsicInst>(Item[0].first)->getIntrinsicID());
+        }))
+      return false;
+
+    // Check the operator is one that we support.
+    if (isa<BinaryOperator>(Item[0].first)) {
+      Worklist.push_back(GenerateInstLaneVectorFromOperand(Item, 0));
+      Worklist.push_back(GenerateInstLaneVectorFromOperand(Item, 1));
+    } else if (isa<UnaryOperator>(Item[0].first)) {
+      Worklist.push_back(GenerateInstLaneVectorFromOperand(Item, 0));
+    } else {
+      return false;
+    }
+  }
+
+  // If we got this far, we know the shuffles are superfluous and can be
+  // removed. Scan through again and generate the new tree of instructions.
+  std::function<Value *(const SmallVector<InstLane> &)> generate =
+      [&](const SmallVector<InstLane> &Item) -> Value * {
+    if (IdentityLeafs.contains(Item[0].first) &&
+        all_of(drop_begin(enumerate(Item)), [&](const auto &E) {
+          return !E.value().first || (E.value().first == Item[0].first &&
+                                      E.value().second == (int)E.index());
+        })) {
+      return Item[0].first;
+    } else if (SplatLeafs.contains(Item[0].first)) {
+      if (auto ILI = dyn_cast<Instruction>(Item[0].first))
+        Builder.SetInsertPoint(*ILI->getInsertionPointAfterDef());
+      else if (isa<Argument>(Item[0].first))
+        Builder.SetInsertPointPastAllocas(I.getParent()->getParent());
+      SmallVector<int, 16> Mask(Ty->getNumElements(), Item[0].second);
+      return Builder.CreateShuffleVector(Item[0].first, Mask);
+    }
+
+    auto *I = cast<Instruction>(Item[0].first);
+    SmallVector<Value *> Ops;
+    unsigned E = I->getNumOperands();
+    for (unsigned Idx = 0; Idx < E; Idx++)
+      Ops.push_back(generate(GenerateInstLaneVectorFromOperand(Item, Idx)));
+    Builder.SetInsertPoint(I);
+    if (auto BI = dyn_cast<BinaryOperator>(I))
+      return Builder.CreateBinOp((Instruction::BinaryOps)BI->getOpcode(),
+                                 Ops[0], Ops[1]);
+    if (auto UI = dyn_cast<UnaryOperator>(I))
+      return Builder.CreateUnOp((Instruction::UnaryOps)UI->getOpcode(), Ops[0]);
+    llvm_unreachable("Unhandled instruction in generate");
+  };
+
+  Value *V = generate(Start);
+  replaceValue(I, *V);
+  return true;
+}
+
 /// Given a commutative reduction, the order of the input lanes does not alter
 /// the results. We can use this to remove certain shuffles feeding the
 /// reduction, removing the need to shuffle at all.
@@ -2108,6 +2251,7 @@ bool VectorCombine::run() {
         MadeChange |= foldShuffleOfBinops(I);
         MadeChange |= foldShuffleOfCastops(I);
         MadeChange |= foldSelectShuffle(I);
+        MadeChange |= foldShuffleToIdentity(I);
         break;
       case Instruction::BitCast:
         MadeChange |= foldBitcastShuffle(I);