1 ; RUN: opt < %s -basicaa -slp-vectorizer -dce -S -mtriple=x86_64-apple-macosx10.8.0 -mcpu=corei7-avx | FileCheck %s 2 3 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" 4 target triple = "x86_64-apple-macosx10.8.0" 5 6 ;CHECK-LABEL: @foo( 7 ;CHECK: load <4 x i32> 8 ;CHECK: add <4 x i32> 9 ;CHECK: store <4 x i32> 10 ;CHECK: load <4 x i32> 11 ;CHECK: add <4 x i32> 12 ;CHECK: store <4 x i32> 13 ;CHECK: ret 14 define i32 @foo(i32* nocapture %A, i32 %n) #0 { 15 entry: 16 %cmp62 = icmp sgt i32 %n, 0 17 br i1 %cmp62, label %for.body, label %for.end 18 19 for.body: ; preds = %entry, %for.body 20 %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ] 21 %arrayidx = getelementptr inbounds i32* %A, i64 %indvars.iv 22 %0 = load i32* %arrayidx, align 4 23 %add1 = add nsw i32 %0, %n 24 store i32 %add1, i32* %arrayidx, align 4 25 %1 = or i64 %indvars.iv, 1 26 %arrayidx4 = getelementptr inbounds i32* %A, i64 %1 27 %2 = load i32* %arrayidx4, align 4 28 %add5 = add nsw i32 %2, %n 29 store i32 %add5, i32* %arrayidx4, align 4 30 %3 = or i64 %indvars.iv, 2 31 %arrayidx8 = getelementptr inbounds i32* %A, i64 %3 32 %4 = load i32* %arrayidx8, align 4 33 %add9 = add nsw i32 %4, %n 34 store i32 %add9, i32* %arrayidx8, align 4 35 %5 = or i64 %indvars.iv, 3 36 %arrayidx12 = getelementptr inbounds i32* %A, i64 %5 37 %6 = load i32* %arrayidx12, align 4 38 %add13 = add nsw i32 %6, %n 39 store i32 %add13, i32* %arrayidx12, align 4 40 %7 = or i64 %indvars.iv, 4 41 %arrayidx16 = getelementptr inbounds i32* %A, i64 %7 42 %8 = load i32* %arrayidx16, align 4 43 %add17 = add nsw i32 %8, %n 44 store i32 %add17, i32* %arrayidx16, align 4 45 %9 = or i64 %indvars.iv, 5 46 %arrayidx20 = getelementptr inbounds i32* %A, i64 %9 47 %10 = load i32* %arrayidx20, align 4 48 %add21 = add nsw i32 %10, %n 49 store i32 %add21, i32* %arrayidx20, align 4 50 %11 = or i64 %indvars.iv, 6 51 %arrayidx24 = getelementptr inbounds i32* %A, i64 %11 52 %12 = load i32* %arrayidx24, align 4 53 %add25 = add nsw i32 %12, %n 54 store i32 %add25, i32* %arrayidx24, align 4 55 %13 = or i64 %indvars.iv, 7 56 %arrayidx28 = getelementptr inbounds i32* %A, i64 %13 57 %14 = load i32* %arrayidx28, align 4 58 %add29 = add nsw i32 %14, %n 59 store i32 %add29, i32* %arrayidx28, align 4 60 %indvars.iv.next = add i64 %indvars.iv, 8 61 %15 = trunc i64 %indvars.iv.next to i32 62 %cmp = icmp slt i32 %15, %n 63 br i1 %cmp, label %for.body, label %for.end 64 65 for.end: ; preds = %for.body, %entry 66 ret i32 undef 67 } 68 69 attributes #0 = { nounwind ssp uwtable "less-precise-fpmad"="false" "no-frame-pointer-elim"="true" "no-frame-pointer-elim-non-leaf"="true" "no-infs-fp-math"="false" "no-nans-fp-math"="false" "unsafe-fp-math"="false" "use-soft-float"="false" } 70
