[LoopVectorizer] Fix for cost values of memory accesses.

This commit is a combination of two patches:

* "Fix in getScalarizationOverhead()"

   If target returns false in TTI.prefersVectorizedAddressing(), it means the
   address registers will not need to be extracted. Therefore, there should
   be no operands scalarization overhead for a load instruction.

* "Don't pass the instruction pointer from getMemInstScalarizationCost."

   Since VF is always > 1, this is a cost query for an instruction in the
   vectorized loop and it should not be evaluated within the scalar
   context of the instruction.

Review: Ulrich Weigand, Hal Finkel
https://reviews.llvm.org/D52351
https://reviews.llvm.org/D52417

llvm-svn: 345603

lib/Transforms/Vectorize/LoopVectorize.cpp

@@ -2982,6 +2982,10 @@ static unsigned getScalarizationOverhead(Instruction *I, unsigned VF,
        !TTI.supportsEfficientVectorElementLoadStore()))
     Cost += TTI.getScalarizationOverhead(RetTy, true, false);
 
+  // Some targets keep addresses scalar.
+  if (isa<LoadInst>(I) && !TTI.prefersVectorizedAddressing())
+    return Cost;
+
   if (CallInst *CI = dyn_cast<CallInst>(I)) {
     SmallVector<const Value *, 4> Operands(CI->arg_operands());
     Cost += TTI.getOperandsScalarizationOverhead(Operands, VF);
@@ -5372,6 +5376,7 @@ static bool isStrideMul(Instruction *I, LoopVectorizationLegality *Legal) {
 
 unsigned LoopVectorizationCostModel::getMemInstScalarizationCost(Instruction *I,
                                                                  unsigned VF) {
+  assert(VF > 1 && "Scalarization cost of instruction implies vectorization.");
   Type *ValTy = getMemInstValueType(I);
   auto SE = PSE.getSE();
 
@@ -5387,9 +5392,11 @@ unsigned LoopVectorizationCostModel::getMemInstScalarizationCost(Instruction *I,
   // Get the cost of the scalar memory instruction and address computation.
   unsigned Cost = VF * TTI.getAddressComputationCost(PtrTy, SE, PtrSCEV);
 
+  // Don't pass *I here, since it is scalar but will actually be part of a
+  // vectorized loop where the user of it is a vectorized instruction.
   Cost += VF *
           TTI.getMemoryOpCost(I->getOpcode(), ValTy->getScalarType(), Alignment,
-                              AS, I);
+                              AS);
 
   // Get the overhead of the extractelement and insertelement instructions
   // we might create due to scalarization.

test/Transforms/LoopVectorize/SystemZ/load-scalarization-cost-0.ll

@@ -0,0 +1,27 @@
+; RUN: opt -mtriple=s390x-unknown-linux -mcpu=z13 -loop-vectorize \
+; RUN:   -force-vector-width=2 -debug-only=loop-vectorize \
+; RUN:   -disable-output < %s 2>&1 | FileCheck %s
+; REQUIRES: asserts
+;
+; Check that a scalarized load does not get operands scalarization costs added.
+
+define void @fun(i64* %data, i64 %n, i64 %s, double* %Src) {
+entry:
+  br label %for.body
+
+for.body:
+  %iv = phi i64 [ 0, %entry ], [ %iv.next, %for.body ]
+  %mul = mul nsw i64 %iv, %s
+  %gep = getelementptr inbounds double, double* %Src, i64 %mul
+  %bct = bitcast double* %gep to i64*
+  %ld = load i64, i64* %bct
+  %iv.next = add nuw nsw i64 %iv, 1
+  %cmp110.us = icmp slt i64 %iv.next, %n
+  br i1 %cmp110.us, label %for.body, label %for.end
+
+for.end:
+  ret void
+
+; CHECK: LV: Found an estimated cost of 2 for VF 2 For instruction:   %mul = mul nsw i64 %iv, %s
+; CHECK: LV: Found an estimated cost of 2 for VF 2 For instruction:   %ld = load i64, i64* %bct
+}

test/Transforms/LoopVectorize/SystemZ/load-scalarization-cost-1.ll

@@ -0,0 +1,28 @@
+; RUN: opt -mtriple=s390x-unknown-linux -mcpu=z13 -loop-vectorize \
+; RUN:   -force-vector-width=4 -debug-only=loop-vectorize \
+; RUN:   -enable-interleaved-mem-accesses=false -disable-output < %s 2>&1 \
+; RUN:   | FileCheck %s
+; REQUIRES: asserts
+;
+; Check that a scalarized load does not get a zero cost in a vectorized
+; loop. It can only be folded into the add operand in the scalar loop.
+
+define i32 @fun(i64* %data, i64 %n, i64 %s, i32* %Src) {
+entry:
+  br label %for.body
+
+for.body:
+  %iv = phi i64 [ 0, %entry ], [ %iv.next, %for.body ]
+  %acc = phi i32 [ 0, %entry ], [ %acc_next, %for.body ]
+  %gep = getelementptr inbounds i32, i32* %Src, i64 %iv
+  %ld = load i32, i32* %gep
+  %acc_next = add i32 %acc, %ld
+  %iv.next = add nuw nsw i64 %iv, 2
+  %cmp110.us = icmp slt i64 %iv.next, %n
+  br i1 %cmp110.us, label %for.body, label %for.end
+
+for.end:
+  ret i32 %acc_next
+
+; CHECK: Found an estimated cost of 4 for VF 4 For instruction:   %ld = load i32, i32* %gep
+}