1 target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128-n8:16:32:64-S128" 2 target triple = "x86_64-unknown-linux-gnu" 3 ; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -mcpu=corei7 -bb-vectorize -bb-vectorize-req-chain-depth=3 -instcombine -gvn -S | FileCheck %s 4 ; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -mcpu=corei7 -basicaa -loop-unroll -unroll-threshold=45 -unroll-allow-partial -bb-vectorize -bb-vectorize-req-chain-depth=3 -instcombine -gvn -S | FileCheck %s -check-prefix=CHECK-UNRL 5 ; The second check covers the use of alias analysis (with loop unrolling). 6 7 define void @test1(double* noalias %out, double* noalias %in1, double* noalias %in2) nounwind uwtable { 8 entry: 9 br label %for.body 10 ; CHECK-LABEL: @test1( 11 ; CHECK-UNRL-LABEL: @test1( 12 13 for.body: ; preds = %for.body, %entry 14 %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ] 15 %arrayidx = getelementptr inbounds double* %in1, i64 %indvars.iv 16 %0 = load double* %arrayidx, align 8 17 %arrayidx2 = getelementptr inbounds double* %in2, i64 %indvars.iv 18 %1 = load double* %arrayidx2, align 8 19 %mul = fmul double %0, %0 20 %mul3 = fmul double %0, %1 21 %add = fadd double %mul, %mul3 22 %add4 = fadd double %1, %1 23 %add5 = fadd double %add4, %0 24 %mul6 = fmul double %0, %add5 25 %add7 = fadd double %add, %mul6 26 %mul8 = fmul double %1, %1 27 %add9 = fadd double %0, %0 28 %add10 = fadd double %add9, %0 29 %mul11 = fmul double %mul8, %add10 30 %add12 = fadd double %add7, %mul11 31 %arrayidx14 = getelementptr inbounds double* %out, i64 %indvars.iv 32 store double %add12, double* %arrayidx14, align 8 33 %indvars.iv.next = add i64 %indvars.iv, 1 34 %lftr.wideiv = trunc i64 %indvars.iv.next to i32 35 %exitcond = icmp eq i32 %lftr.wideiv, 10 36 br i1 %exitcond, label %for.end, label %for.body 37 ; CHECK: insertelement 38 ; CHECK-NEXT: insertelement 39 ; CHECK-NEXT: fadd <2 x double> 40 ; CHECK-NEXT: insertelement 41 ; CHECK-NEXT: insertelement 42 ; CHECK-NEXT: fadd <2 x double> 43 ; CHECK-NEXT: insertelement 44 ; CHECK-NEXT: fmul <2 x double> 45 46 ; CHECK-UNRL: %mul = fmul <2 x double> %2, %2 47 ; CHECK-UNRL: %mul3 = fmul <2 x double> %2, %3 48 ; CHECK-UNRL: %add = fadd <2 x double> %mul, %mul3 49 ; CHECK-UNRL: %add4 = fadd <2 x double> %3, %3 50 ; CHECK-UNRL: %add5 = fadd <2 x double> %add4, %2 51 ; CHECK-UNRL: %mul6 = fmul <2 x double> %2, %add5 52 ; CHECK-UNRL: %add7 = fadd <2 x double> %add, %mul6 53 ; CHECK-UNRL: %mul8 = fmul <2 x double> %3, %3 54 ; CHECK-UNRL: %add9 = fadd <2 x double> %2, %2 55 ; CHECK-UNRL: %add10 = fadd <2 x double> %add9, %2 56 ; CHECK-UNRL: %mul11 = fmul <2 x double> %mul8, %add10 57 ; CHECK-UNRL: %add12 = fadd <2 x double> %add7, %mul11 58 59 for.end: ; preds = %for.body 60 ret void 61 } 62
