[llvm] r303763 - [LV] Update type in cost model for scalarization

[Yung, Douglas via llvm-commits]

Hi Matthew,

Your commit seems to have caused a crash in one of our internal code bases. I have filed the details in PR33193. Can you take a look?

Douglas Yung

> -----Original Message-----
> From: llvm-commits [mailto:llvm-commits-bounces at lists.llvm.org] On Behalf Of
> Matthew Simpson via llvm-commits
> Sent: Wednesday, May 24, 2017 8:26
> To: llvm-commits at lists.llvm.org
> Subject: [llvm] r303763 - [LV] Update type in cost model for scalarization
> 
> Author: mssimpso
> Date: Wed May 24 10:26:15 2017
> New Revision: 303763
> 
> URL: http://llvm.org/viewvc/llvm-project?rev=303763&view=rev
> Log:
> [LV] Update type in cost model for scalarization
> 
> For non-uniform instructions marked for scalarization, we should update
> `VectorTy` when computing instruction costs to reflect the scalar type. In
> addition to determining instruction costs, this type is also used to signal
> that all instructions in the loop will be scalarized. This currently affects
> memory instructions and non-pointer induction variables and their updates. (We
> also mark GEPs scalar after vectorization, but their cost is computed together
> with memory instructions.) For scalarized induction updates, this patch also
> scales the scalar cost by the vectorization factor, corresponding to each
> induction step.
> 
> Added:
>     llvm/trunk/test/Transforms/LoopVectorize/AArch64/no_vector_instructions.ll
> Modified:
>     llvm/trunk/lib/Transforms/Vectorize/LoopVectorize.cpp
> 
> Modified: llvm/trunk/lib/Transforms/Vectorize/LoopVectorize.cpp
> URL: http://llvm.org/viewvc/llvm-
> project/llvm/trunk/lib/Transforms/Vectorize/LoopVectorize.cpp?rev=303763&r1=30
> 3762&r2=303763&view=diff
> ==============================================================================
> --- llvm/trunk/lib/Transforms/Vectorize/LoopVectorize.cpp (original)
> +++ llvm/trunk/lib/Transforms/Vectorize/LoopVectorize.cpp Wed May 24
> +++ 10:26:15 2017
> @@ -7173,7 +7173,7 @@ LoopVectorizationCostModel::getInstructi
>    // Note: Even if all instructions are scalarized, return true if any memory
>    // accesses appear in the loop to get benefits from address folding etc.
>    bool TypeNotScalarized =
> -      VF > 1 && !VectorTy->isVoidTy() && TTI.getNumberOfParts(VectorTy) < VF;
> +      VF > 1 && VectorTy->isVectorTy() &&
> + TTI.getNumberOfParts(VectorTy) < VF;
>    return VectorizationCostTy(C, TypeNotScalarized);  }
> 
> @@ -7312,7 +7312,7 @@ unsigned LoopVectorizationCostModel::get
>    Type *RetTy = I->getType();
>    if (canTruncateToMinimalBitwidth(I, VF))
>      RetTy = IntegerType::get(RetTy->getContext(), MinBWs[I]);
> -  VectorTy = ToVectorTy(RetTy, VF);
> +  VectorTy = isScalarAfterVectorization(I, VF) ? RetTy :
> + ToVectorTy(RetTy, VF);
>    auto SE = PSE.getSE();
> 
>    // TODO: We need to estimate the cost of intrinsic calls.
> @@ -7445,9 +7445,10 @@ unsigned LoopVectorizationCostModel::get
>      } else if (Legal->isUniform(Op2)) {
>        Op2VK = TargetTransformInfo::OK_UniformValue;
>      }
> -    SmallVector<const Value *, 4> Operands(I->operand_values());
> -    return TTI.getArithmeticInstrCost(I->getOpcode(), VectorTy, Op1VK,
> -                                      Op2VK, Op1VP, Op2VP, Operands);
> +    SmallVector<const Value *, 4> Operands(I->operand_values());
> +    unsigned N = isScalarAfterVectorization(I, VF) ? VF : 1;
> +    return N * TTI.getArithmeticInstrCost(I->getOpcode(), VectorTy, Op1VK,
> +                                          Op2VK, Op1VP, Op2VP,
> + Operands);
>    }
>    case Instruction::Select: {
>      SelectInst *SI = cast<SelectInst>(I); @@ -7470,7 +7471,15 @@ unsigned
> LoopVectorizationCostModel::get
>    }
>    case Instruction::Store:
>    case Instruction::Load: {
> -    VectorTy = ToVectorTy(getMemInstValueType(I), VF);
> +    unsigned Width = VF;
> +    if (Width > 1) {
> +      InstWidening Decision = getWideningDecision(I, Width);
> +      assert(Decision != CM_Unknown &&
> +             "CM decision should be taken at this point");
> +      if (Decision == CM_Scalarize)
> +        Width = 1;
> +    }
> +    VectorTy = ToVectorTy(getMemInstValueType(I), Width);
>      return getMemoryInstructionCost(I, VF);
>    }
>    case Instruction::ZExt:
> 
> Added:
> llvm/trunk/test/Transforms/LoopVectorize/AArch64/no_vector_instructions.ll
> URL: http://llvm.org/viewvc/llvm-
> project/llvm/trunk/test/Transforms/LoopVectorize/AArch64/no_vector_instruction
> s.ll?rev=303763&view=auto
> ==============================================================================
> --- llvm/trunk/test/Transforms/LoopVectorize/AArch64/no_vector_instructions.ll
> (added)
> +++ llvm/trunk/test/Transforms/LoopVectorize/AArch64/no_vector_instructi
> +++ ons.ll Wed May 24 10:26:15 2017
> @@ -0,0 +1,26 @@
> +; REQUIRES: asserts
> +; RUN: opt < %s -loop-vectorize -force-vector-interleave=1 -S
> +-debug-only=loop-vectorize 2>&1 | FileCheck %s
> +
> +target datalayout = "e-m:e-i8:8:32-i16:16:32-i64:64-i128:128-n32:64-S128"
> +target triple = "aarch64--linux-gnu"
> +
> +; CHECK-LABEL: all_scalar
> +; CHECK:       LV: Found scalar instruction: %i.next = add nuw nsw i64 %i, 2
> +; CHECK:       LV: Found an estimated cost of 2 for VF 2 For instruction:
> %i.next = add nuw nsw i64 %i, 2
> +; CHECK:       LV: Not considering vector loop of width 2 because it will not
> generate any vector instructions
> +;
> +define void @all_scalar(i64* %a, i64 %n) {
> +entry:
> +  br label %for.body
> +
> +for.body:
> +  %i = phi i64 [ 0, %entry ], [ %i.next, %for.body ]
> +  %tmp0 = getelementptr i64, i64* %a, i64 %i
> +  store i64 0, i64* %tmp0, align 1
> +  %i.next = add nuw nsw i64 %i, 2
> +  %cond = icmp eq i64 %i.next, %n
> +  br i1 %cond, label %for.end, label %for.body
> +
> +for.end:
> +  ret void
> +}
> 
> 
> _______________________________________________
> llvm-commits mailing list
> llvm-commits at lists.llvm.org
> http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-commits