Index: include/llvm/CodeGen/BasicTTIImpl.h =================================================================== --- include/llvm/CodeGen/BasicTTIImpl.h +++ include/llvm/CodeGen/BasicTTIImpl.h @@ -1421,28 +1421,43 @@ auto *ConcreteTTI = static_cast(this); std::pair LT = ConcreteTTI->getTLI()->getTypeLegalizationCost(DL, Ty); - unsigned LongVectorCount = 0; unsigned MVTLen = LT.second.isVector() ? LT.second.getVectorNumElements() : 1; while (NumVecElts > MVTLen) { NumVecElts /= 2; - Type *SubTy = VectorType::get(ScalarTy, NumVecElts); - // Assume the pairwise shuffles add a cost. - ShuffleCost += (IsPairwise + 1) * - ConcreteTTI->getShuffleCost(TTI::SK_ExtractSubvector, Ty, - NumVecElts, SubTy); - ArithCost += ConcreteTTI->getArithmeticInstrCost(Opcode, SubTy); - Ty = SubTy; - ++LongVectorCount; + Ty = VectorType::get(ScalarTy, NumVecElts); + + if (IsPairwise && NumVecElts > 1) { + // Pairwise requires 2 two source permutes for each subvector after + // type splitting. Split here so we don't ask the target about a worst + // case two soure permute on a wide illegal type. + std::pair ShufLT = + ConcreteTTI->getTLI()->getTypeLegalizationCost(DL, Ty); + Type *ShufTy = VectorType::get(ScalarTy, + ShufLT.second.getVectorNumElements()); + ShuffleCost += 2 * ShufLT.first * + ConcreteTTI->getShuffleCost(TTI::SK_PermuteTwoSrc, + ShufTy, 0, ShufTy); + } + + ArithCost += ConcreteTTI->getArithmeticInstrCost(Opcode, Ty); + --NumReduxLevels; } // The minimal length of the vector is limited by the real length of vector // operations performed on the current platform. That's why several final // reduction operations are performed on the vectors with the same // architecture-dependent length. - ShuffleCost += (NumReduxLevels - LongVectorCount) * (IsPairwise + 1) * + ShuffleCost += NumReduxLevels * ConcreteTTI->getShuffleCost(TTI::SK_PermuteSingleSrc, Ty, 0, Ty); - ArithCost += (NumReduxLevels - LongVectorCount) * + if (IsPairwise && NumReduxLevels > 1) { + // Pairwise operations only require one shuffle on the last step since + // the single even element is already in place. + ShuffleCost += (NumReduxLevels - 1) * + ConcreteTTI->getShuffleCost(TTI::SK_PermuteSingleSrc, Ty, + 0, Ty); + } + ArithCost += NumReduxLevels * ConcreteTTI->getArithmeticInstrCost(Opcode, Ty); return ShuffleCost + ArithCost + ConcreteTTI->getVectorInstrCost(Instruction::ExtractElement, Ty, 0); @@ -1470,34 +1485,48 @@ auto *ConcreteTTI = static_cast(this); std::pair LT = ConcreteTTI->getTLI()->getTypeLegalizationCost(DL, Ty); - unsigned LongVectorCount = 0; unsigned MVTLen = LT.second.isVector() ? LT.second.getVectorNumElements() : 1; while (NumVecElts > MVTLen) { NumVecElts /= 2; - Type *SubTy = VectorType::get(ScalarTy, NumVecElts); + Ty = VectorType::get(ScalarTy, NumVecElts); CondTy = VectorType::get(ScalarCondTy, NumVecElts); - // Assume the pairwise shuffles add a cost. - ShuffleCost += (IsPairwise + 1) * - ConcreteTTI->getShuffleCost(TTI::SK_ExtractSubvector, Ty, - NumVecElts, SubTy); + if (IsPairwise && NumVecElts > 1) { + // Pairwise requires 2 two source permutes for each subvector after + // type splitting. Split here so we don't ask the target about a worst + // case two soure permute on a wide illegal type. + std::pair ShufLT = + ConcreteTTI->getTLI()->getTypeLegalizationCost(DL, Ty); + Type *ShufTy = VectorType::get(ScalarTy, + ShufLT.second.getVectorNumElements()); + ShuffleCost += 2 * ShufLT.first * + ConcreteTTI->getShuffleCost(TTI::SK_PermuteTwoSrc, + ShufTy, 0, ShufTy); + } + MinMaxCost += - ConcreteTTI->getCmpSelInstrCost(CmpOpcode, SubTy, CondTy, nullptr) + - ConcreteTTI->getCmpSelInstrCost(Instruction::Select, SubTy, CondTy, + ConcreteTTI->getCmpSelInstrCost(CmpOpcode, Ty, CondTy, nullptr) + + ConcreteTTI->getCmpSelInstrCost(Instruction::Select, Ty, CondTy, nullptr); - Ty = SubTy; - ++LongVectorCount; + --NumReduxLevels; } // The minimal length of the vector is limited by the real length of vector // operations performed on the current platform. That's why several final // reduction opertions are perfomed on the vectors with the same // architecture-dependent length. - ShuffleCost += (NumReduxLevels - LongVectorCount) * (IsPairwise + 1) * + ShuffleCost += NumReduxLevels * ConcreteTTI->getShuffleCost(TTI::SK_PermuteSingleSrc, Ty, 0, Ty); + if (IsPairwise && NumReduxLevels > 1) { + // Pairwise operations only require one shuffle on the last step since + // the single even element is already in place. + ShuffleCost += (NumReduxLevels - 1) * + ConcreteTTI->getShuffleCost(TTI::SK_PermuteSingleSrc, Ty, + 0, Ty); + } MinMaxCost += - (NumReduxLevels - LongVectorCount) * + NumReduxLevels * (ConcreteTTI->getCmpSelInstrCost(CmpOpcode, Ty, CondTy, nullptr) + ConcreteTTI->getCmpSelInstrCost(Instruction::Select, Ty, CondTy, nullptr)); Index: test/Analysis/CostModel/X86/reduction.ll =================================================================== --- test/Analysis/CostModel/X86/reduction.ll +++ test/Analysis/CostModel/X86/reduction.ll @@ -107,7 +107,7 @@ ; SSE2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: %rdx.shuf.1.0 = shufflevector <4 x float> %bin.rdx.0, <4 x float> undef, <4 x i32> ; SSE2-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %rdx.shuf.1.1 = shufflevector <4 x float> %bin.rdx.0, <4 x float> undef, <4 x i32> ; SSE2-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %bin.rdx.1 = fadd <4 x float> %rdx.shuf.1.0, %rdx.shuf.1.1 -; SSE2-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %r = extractelement <4 x float> %bin.rdx.1, i32 0 +; SSE2-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %r = extractelement <4 x float> %bin.rdx.1, i32 0 ; SSE2-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %r2 = fadd float %r, %f1 ; SSE2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret float %r2 ; @@ -118,7 +118,7 @@ ; SSSE3-NEXT: Cost Model: Found an estimated cost of 0 for instruction: %rdx.shuf.1.0 = shufflevector <4 x float> %bin.rdx.0, <4 x float> undef, <4 x i32> ; SSSE3-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %rdx.shuf.1.1 = shufflevector <4 x float> %bin.rdx.0, <4 x float> undef, <4 x i32> ; SSSE3-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %bin.rdx.1 = fadd <4 x float> %rdx.shuf.1.0, %rdx.shuf.1.1 -; SSSE3-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %r = extractelement <4 x float> %bin.rdx.1, i32 0 +; SSSE3-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %r = extractelement <4 x float> %bin.rdx.1, i32 0 ; SSSE3-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %r2 = fadd float %r, %f1 ; SSSE3-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret float %r2 ; @@ -168,7 +168,7 @@ ; SSE2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: %rdx.shuf.1.0 = shufflevector <4 x float> %bin.rdx.0, <4 x float> undef, <4 x i32> ; SSE2-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %rdx.shuf.1.1 = shufflevector <4 x float> %bin.rdx.0, <4 x float> undef, <4 x i32> ; SSE2-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %bin.rdx.1 = fadd <4 x float> %rdx.shuf.1.0, %rdx.shuf.1.1 -; SSE2-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %r = extractelement <4 x float> %bin.rdx.1, i32 0 +; SSE2-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %r = extractelement <4 x float> %bin.rdx.1, i32 0 ; SSE2-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %r2 = fadd float %r, %f1 ; SSE2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret float %r2 ; @@ -179,7 +179,7 @@ ; SSSE3-NEXT: Cost Model: Found an estimated cost of 0 for instruction: %rdx.shuf.1.0 = shufflevector <4 x float> %bin.rdx.0, <4 x float> undef, <4 x i32> ; SSSE3-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %rdx.shuf.1.1 = shufflevector <4 x float> %bin.rdx.0, <4 x float> undef, <4 x i32> ; SSSE3-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %bin.rdx.1 = fadd <4 x float> %rdx.shuf.1.0, %rdx.shuf.1.1 -; SSSE3-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %r = extractelement <4 x float> %bin.rdx.1, i32 0 +; SSSE3-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %r = extractelement <4 x float> %bin.rdx.1, i32 0 ; SSSE3-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %r2 = fadd float %r, %f1 ; SSSE3-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret float %r2 ; @@ -228,7 +228,7 @@ ; SSE2-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %bin.rdx.0 = fadd <4 x float> %rdx.shuf.0.0, %rdx.shuf.0.1 ; SSE2-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %rdx.shuf.1.1 = shufflevector <4 x float> %bin.rdx.0, <4 x float> undef, <4 x i32> ; SSE2-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %bin.rdx.1 = fadd <4 x float> %bin.rdx.0, %rdx.shuf.1.1 -; SSE2-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %r = extractelement <4 x float> %bin.rdx.1, i32 0 +; SSE2-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %r = extractelement <4 x float> %bin.rdx.1, i32 0 ; SSE2-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %r2 = fadd float %r, %f1 ; SSE2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret float %r2 ; @@ -238,7 +238,7 @@ ; SSSE3-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %bin.rdx.0 = fadd <4 x float> %rdx.shuf.0.0, %rdx.shuf.0.1 ; SSSE3-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %rdx.shuf.1.1 = shufflevector <4 x float> %bin.rdx.0, <4 x float> undef, <4 x i32> ; SSSE3-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %bin.rdx.1 = fadd <4 x float> %bin.rdx.0, %rdx.shuf.1.1 -; SSSE3-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %r = extractelement <4 x float> %bin.rdx.1, i32 0 +; SSSE3-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %r = extractelement <4 x float> %bin.rdx.1, i32 0 ; SSSE3-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %r2 = fadd float %r, %f1 ; SSSE3-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret float %r2 ; @@ -669,14 +669,14 @@ ; SSE2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: %rdx.shuf.1.0 = shufflevector <2 x double> %rdx, <2 x double> undef, <2 x i32> ; SSE2-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %rdx.shuf.1.1 = shufflevector <2 x double> %rdx, <2 x double> undef, <2 x i32> ; SSE2-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %bin.rdx8 = fadd <2 x double> %rdx.shuf.1.0, %rdx.shuf.1.1 -; SSE2-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %r = extractelement <2 x double> %bin.rdx8, i32 0 +; SSE2-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %r = extractelement <2 x double> %bin.rdx8, i32 0 ; SSE2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret double %r ; ; SSSE3-LABEL: 'pairwise_reduction2double' ; SSSE3-NEXT: Cost Model: Found an estimated cost of 0 for instruction: %rdx.shuf.1.0 = shufflevector <2 x double> %rdx, <2 x double> undef, <2 x i32> ; SSSE3-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %rdx.shuf.1.1 = shufflevector <2 x double> %rdx, <2 x double> undef, <2 x i32> ; SSSE3-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %bin.rdx8 = fadd <2 x double> %rdx.shuf.1.0, %rdx.shuf.1.1 -; SSSE3-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %r = extractelement <2 x double> %bin.rdx8, i32 0 +; SSSE3-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %r = extractelement <2 x double> %bin.rdx8, i32 0 ; SSSE3-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret double %r ; ; SSE42-LABEL: 'pairwise_reduction2double' @@ -709,7 +709,7 @@ ; SSE2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: %rdx.shuf.1.0 = shufflevector <4 x float> %bin.rdx, <4 x float> undef, <4 x i32> ; SSE2-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %rdx.shuf.1.1 = shufflevector <4 x float> %bin.rdx, <4 x float> undef, <4 x i32> ; SSE2-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %bin.rdx8 = fadd <4 x float> %rdx.shuf.1.0, %rdx.shuf.1.1 -; SSE2-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %r = extractelement <4 x float> %bin.rdx8, i32 0 +; SSE2-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %r = extractelement <4 x float> %bin.rdx8, i32 0 ; SSE2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret float %r ; ; SSSE3-LABEL: 'pairwise_reduction4float' @@ -719,7 +719,7 @@ ; SSSE3-NEXT: Cost Model: Found an estimated cost of 0 for instruction: %rdx.shuf.1.0 = shufflevector <4 x float> %bin.rdx, <4 x float> undef, <4 x i32> ; SSSE3-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %rdx.shuf.1.1 = shufflevector <4 x float> %bin.rdx, <4 x float> undef, <4 x i32> ; SSSE3-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %bin.rdx8 = fadd <4 x float> %rdx.shuf.1.0, %rdx.shuf.1.1 -; SSSE3-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %r = extractelement <4 x float> %bin.rdx8, i32 0 +; SSSE3-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %r = extractelement <4 x float> %bin.rdx8, i32 0 ; SSSE3-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret float %r ; ; SSE42-LABEL: 'pairwise_reduction4float' @@ -761,7 +761,7 @@ ; SSE2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: %rdx.shuf.1.0 = shufflevector <4 x double> %bin.rdx, <4 x double> undef, <4 x i32> ; SSE2-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %rdx.shuf.1.1 = shufflevector <4 x double> %bin.rdx, <4 x double> undef, <4 x i32> ; SSE2-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %bin.rdx8 = fadd <4 x double> %rdx.shuf.1.0, %rdx.shuf.1.1 -; SSE2-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %r = extractelement <4 x double> %bin.rdx8, i32 0 +; SSE2-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %r = extractelement <4 x double> %bin.rdx8, i32 0 ; SSE2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret double %r ; ; SSSE3-LABEL: 'pairwise_reduction4double' @@ -771,7 +771,7 @@ ; SSSE3-NEXT: Cost Model: Found an estimated cost of 0 for instruction: %rdx.shuf.1.0 = shufflevector <4 x double> %bin.rdx, <4 x double> undef, <4 x i32> ; SSSE3-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %rdx.shuf.1.1 = shufflevector <4 x double> %bin.rdx, <4 x double> undef, <4 x i32> ; SSSE3-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %bin.rdx8 = fadd <4 x double> %rdx.shuf.1.0, %rdx.shuf.1.1 -; SSSE3-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %r = extractelement <4 x double> %bin.rdx8, i32 0 +; SSSE3-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %r = extractelement <4 x double> %bin.rdx8, i32 0 ; SSSE3-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret double %r ; ; SSE42-LABEL: 'pairwise_reduction4double' @@ -826,7 +826,7 @@ ; SSE2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: %rdx.shuf.2.0 = shufflevector <8 x float> %bin.rdx8, <8 x float> undef, <8 x i32> ; SSE2-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %rdx.shuf.2.1 = shufflevector <8 x float> %bin.rdx8, <8 x float> undef, <8 x i32> ; SSE2-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %bin.rdx9 = fadd <8 x float> %rdx.shuf.2.0, %rdx.shuf.2.1 -; SSE2-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %r = extractelement <8 x float> %bin.rdx9, i32 0 +; SSE2-NEXT: Cost Model: Found an estimated cost of 13 for instruction: %r = extractelement <8 x float> %bin.rdx9, i32 0 ; SSE2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret float %r ; ; SSSE3-LABEL: 'pairwise_reduction8float' @@ -839,7 +839,7 @@ ; SSSE3-NEXT: Cost Model: Found an estimated cost of 0 for instruction: %rdx.shuf.2.0 = shufflevector <8 x float> %bin.rdx8, <8 x float> undef, <8 x i32> ; SSSE3-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %rdx.shuf.2.1 = shufflevector <8 x float> %bin.rdx8, <8 x float> undef, <8 x i32> ; SSSE3-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %bin.rdx9 = fadd <8 x float> %rdx.shuf.2.0, %rdx.shuf.2.1 -; SSSE3-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %r = extractelement <8 x float> %bin.rdx9, i32 0 +; SSSE3-NEXT: Cost Model: Found an estimated cost of 13 for instruction: %r = extractelement <8 x float> %bin.rdx9, i32 0 ; SSSE3-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret float %r ; ; SSE42-LABEL: 'pairwise_reduction8float' @@ -900,14 +900,14 @@ ; SSE2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: %rdx.shuf.1.0 = shufflevector <2 x i64> %rdx, <2 x i64> undef, <2 x i32> ; SSE2-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %rdx.shuf.1.1 = shufflevector <2 x i64> %rdx, <2 x i64> undef, <2 x i32> ; SSE2-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %bin.rdx8 = add <2 x i64> %rdx.shuf.1.0, %rdx.shuf.1.1 -; SSE2-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %r = extractelement <2 x i64> %bin.rdx8, i32 0 +; SSE2-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %r = extractelement <2 x i64> %bin.rdx8, i32 0 ; SSE2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i64 %r ; ; SSSE3-LABEL: 'pairwise_reduction2i64' ; SSSE3-NEXT: Cost Model: Found an estimated cost of 0 for instruction: %rdx.shuf.1.0 = shufflevector <2 x i64> %rdx, <2 x i64> undef, <2 x i32> ; SSSE3-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %rdx.shuf.1.1 = shufflevector <2 x i64> %rdx, <2 x i64> undef, <2 x i32> ; SSSE3-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %bin.rdx8 = add <2 x i64> %rdx.shuf.1.0, %rdx.shuf.1.1 -; SSSE3-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %r = extractelement <2 x i64> %bin.rdx8, i32 0 +; SSSE3-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %r = extractelement <2 x i64> %bin.rdx8, i32 0 ; SSSE3-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i64 %r ; ; SSE42-LABEL: 'pairwise_reduction2i64' @@ -940,7 +940,7 @@ ; SSE2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: %rdx.shuf.1.0 = shufflevector <4 x i32> %bin.rdx, <4 x i32> undef, <4 x i32> ; SSE2-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %rdx.shuf.1.1 = shufflevector <4 x i32> %bin.rdx, <4 x i32> undef, <4 x i32> ; SSE2-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %bin.rdx8 = add <4 x i32> %rdx.shuf.1.0, %rdx.shuf.1.1 -; SSE2-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %r = extractelement <4 x i32> %bin.rdx8, i32 0 +; SSE2-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %r = extractelement <4 x i32> %bin.rdx8, i32 0 ; SSE2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 %r ; ; SSSE3-LABEL: 'pairwise_reduction4i32' @@ -950,7 +950,7 @@ ; SSSE3-NEXT: Cost Model: Found an estimated cost of 0 for instruction: %rdx.shuf.1.0 = shufflevector <4 x i32> %bin.rdx, <4 x i32> undef, <4 x i32> ; SSSE3-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %rdx.shuf.1.1 = shufflevector <4 x i32> %bin.rdx, <4 x i32> undef, <4 x i32> ; SSSE3-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %bin.rdx8 = add <4 x i32> %rdx.shuf.1.0, %rdx.shuf.1.1 -; SSSE3-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %r = extractelement <4 x i32> %bin.rdx8, i32 0 +; SSSE3-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %r = extractelement <4 x i32> %bin.rdx8, i32 0 ; SSSE3-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 %r ; ; SSE42-LABEL: 'pairwise_reduction4i32' @@ -992,7 +992,7 @@ ; SSE2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: %rdx.shuf.1.0 = shufflevector <4 x i64> %bin.rdx, <4 x i64> undef, <4 x i32> ; SSE2-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %rdx.shuf.1.1 = shufflevector <4 x i64> %bin.rdx, <4 x i64> undef, <4 x i32> ; SSE2-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %bin.rdx8 = add <4 x i64> %rdx.shuf.1.0, %rdx.shuf.1.1 -; SSE2-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %r = extractelement <4 x i64> %bin.rdx8, i32 0 +; SSE2-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %r = extractelement <4 x i64> %bin.rdx8, i32 0 ; SSE2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i64 %r ; ; SSSE3-LABEL: 'pairwise_reduction4i64' @@ -1002,7 +1002,7 @@ ; SSSE3-NEXT: Cost Model: Found an estimated cost of 0 for instruction: %rdx.shuf.1.0 = shufflevector <4 x i64> %bin.rdx, <4 x i64> undef, <4 x i32> ; SSSE3-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %rdx.shuf.1.1 = shufflevector <4 x i64> %bin.rdx, <4 x i64> undef, <4 x i32> ; SSSE3-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %bin.rdx8 = add <4 x i64> %rdx.shuf.1.0, %rdx.shuf.1.1 -; SSSE3-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %r = extractelement <4 x i64> %bin.rdx8, i32 0 +; SSSE3-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %r = extractelement <4 x i64> %bin.rdx8, i32 0 ; SSSE3-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i64 %r ; ; SSE42-LABEL: 'pairwise_reduction4i64' @@ -1057,7 +1057,7 @@ ; SSE2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: %rdx.shuf.2.0 = shufflevector <8 x i16> %bin.rdx8, <8 x i16> undef, <8 x i32> ; SSE2-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %rdx.shuf.2.1 = shufflevector <8 x i16> %bin.rdx8, <8 x i16> undef, <8 x i32> ; SSE2-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %bin.rdx9 = add <8 x i16> %rdx.shuf.2.0, %rdx.shuf.2.1 -; SSE2-NEXT: Cost Model: Found an estimated cost of 34 for instruction: %r = extractelement <8 x i16> %bin.rdx9, i32 0 +; SSE2-NEXT: Cost Model: Found an estimated cost of 29 for instruction: %r = extractelement <8 x i16> %bin.rdx9, i32 0 ; SSE2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i16 %r ; ; SSSE3-LABEL: 'pairwise_reduction8i16' @@ -1070,7 +1070,7 @@ ; SSSE3-NEXT: Cost Model: Found an estimated cost of 0 for instruction: %rdx.shuf.2.0 = shufflevector <8 x i16> %bin.rdx8, <8 x i16> undef, <8 x i32> ; SSSE3-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %rdx.shuf.2.1 = shufflevector <8 x i16> %bin.rdx8, <8 x i16> undef, <8 x i32> ; SSSE3-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %bin.rdx9 = add <8 x i16> %rdx.shuf.2.0, %rdx.shuf.2.1 -; SSSE3-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %r = extractelement <8 x i16> %bin.rdx9, i32 0 +; SSSE3-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %r = extractelement <8 x i16> %bin.rdx9, i32 0 ; SSSE3-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i16 %r ; ; SSE42-LABEL: 'pairwise_reduction8i16' @@ -1124,7 +1124,7 @@ ; SSE2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: %rdx.shuf.2.0 = shufflevector <8 x i32> %bin.rdx8, <8 x i32> undef, <8 x i32> ; SSE2-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %rdx.shuf.2.1 = shufflevector <8 x i32> %bin.rdx8, <8 x i32> undef, <8 x i32> ; SSE2-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %bin.rdx9 = add <8 x i32> %rdx.shuf.2.0, %rdx.shuf.2.1 -; SSE2-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %r = extractelement <8 x i32> %bin.rdx9, i32 0 +; SSE2-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %r = extractelement <8 x i32> %bin.rdx9, i32 0 ; SSE2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 %r ; ; SSSE3-LABEL: 'pairwise_reduction8i32' @@ -1137,7 +1137,7 @@ ; SSSE3-NEXT: Cost Model: Found an estimated cost of 0 for instruction: %rdx.shuf.2.0 = shufflevector <8 x i32> %bin.rdx8, <8 x i32> undef, <8 x i32> ; SSSE3-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %rdx.shuf.2.1 = shufflevector <8 x i32> %bin.rdx8, <8 x i32> undef, <8 x i32> ; SSSE3-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %bin.rdx9 = add <8 x i32> %rdx.shuf.2.0, %rdx.shuf.2.1 -; SSSE3-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %r = extractelement <8 x i32> %bin.rdx9, i32 0 +; SSSE3-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %r = extractelement <8 x i32> %bin.rdx9, i32 0 ; SSSE3-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 %r ; ; SSE42-LABEL: 'pairwise_reduction8i32'