[VectorCombine] Fold binary op of reductions. (#121567)

Replace binary of of two reductions with one reduction of the binary op
applied to vectors. For example:

```
%v0_red = tail call i32 @llvm.vector.reduce.add.v16i32(<16 x i32> %v0)
%v1_red = tail call i32 @llvm.vector.reduce.add.v16i32(<16 x i32> %v1)
%res = add i32 %v0_red, %v1_red
```
gets transformed to:

```
%1 = add <16 x i32> %v0, %v1
%res = call i32 @llvm.vector.reduce.add.v16i32(<16 x i32> %1)
```

1 files changed, 20 insertions, 0 deletions

diff --git a/llvm/lib/Transforms/Utils/LoopUtils.cpp b/llvm/lib/Transforms/Utils/LoopUtils.cpp
index 45915c10..0506ea9 100644
--- a/llvm/lib/Transforms/Utils/LoopUtils.cpp
+++ b/llvm/lib/Transforms/Utils/LoopUtils.cpp
@@ -957,6 +957,7 @@ constexpr Intrinsic::ID llvm::getReductionIntrinsicID(RecurKind RK) {
   }
 }
 
+// This is the inverse to getReductionForBinop
 unsigned llvm::getArithmeticReductionInstruction(Intrinsic::ID RdxID) {
   switch (RdxID) {
   case Intrinsic::vector_reduce_fadd:
@@ -986,6 +987,25 @@ unsigned llvm::getArithmeticReductionInstruction(Intrinsic::ID RdxID) {
   }
 }
 
+// This is the inverse to getArithmeticReductionInstruction
+Intrinsic::ID llvm::getReductionForBinop(Instruction::BinaryOps Opc) {
+  switch (Opc) {
+  default:
+    break;
+  case Instruction::Add:
+    return Intrinsic::vector_reduce_add;
+  case Instruction::Mul:
+    return Intrinsic::vector_reduce_mul;
+  case Instruction::And:
+    return Intrinsic::vector_reduce_and;
+  case Instruction::Or:
+    return Intrinsic::vector_reduce_or;
+  case Instruction::Xor:
+    return Intrinsic::vector_reduce_xor;
+  }
+  return Intrinsic::not_intrinsic;
+}
+
 Intrinsic::ID llvm::getMinMaxReductionIntrinsicOp(Intrinsic::ID RdxID) {
   switch (RdxID) {
   default: