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: store <4 x i32> 8 ;CHECK: ret 9 define i32 @foo(i32* nocapture %A, i32 %n, i32 %m) { 10 entry: 11 %mul = mul nsw i32 %n, 5 12 %add = add nsw i32 %mul, 9 13 store i32 %add, i32* %A, align 4 14 %mul1 = mul nsw i32 %n, 9 15 %add2 = add nsw i32 %mul1, 9 16 %arrayidx3 = getelementptr inbounds i32, i32* %A, i64 1 17 store i32 %add2, i32* %arrayidx3, align 4 18 %mul4 = shl i32 %n, 3 19 %add5 = add nsw i32 %mul4, 9 20 %arrayidx6 = getelementptr inbounds i32, i32* %A, i64 2 21 store i32 %add5, i32* %arrayidx6, align 4 22 %mul7 = mul nsw i32 %n, 10 23 %add8 = add nsw i32 %mul7, 9 24 %arrayidx9 = getelementptr inbounds i32, i32* %A, i64 3 25 store i32 %add8, i32* %arrayidx9, align 4 26 %externaluse1 = add nsw i32 %add, %m 27 %externaluse2 = mul nsw i32 %add, %m ; we should add the extract cost only once and the store will be vectorized 28 %add10 = add nsw i32 %externaluse1, %externaluse2 29 ret i32 %add10 30 } 31
