1 ; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
2 ; RUN: opt < %s -passes=slp-vectorizer,dce -S -mtriple=x86_64-apple-macosx10.8.0 -mcpu=corei7-avx | FileCheck %s
4 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"
5 target triple = "x86_64-apple-macosx10.8.0"
7 define i32 @foo(ptr nocapture %A, i32 %n, i32 %m) {
10 ; CHECK-NEXT: [[TMP0:%.*]] = insertelement <4 x i32> poison, i32 [[N:%.*]], i32 0
11 ; CHECK-NEXT: [[SHUFFLE:%.*]] = shufflevector <4 x i32> [[TMP0]], <4 x i32> poison, <4 x i32> zeroinitializer
12 ; CHECK-NEXT: [[TMP1:%.*]] = mul nsw <4 x i32> [[SHUFFLE]], <i32 5, i32 9, i32 3, i32 10>
13 ; CHECK-NEXT: [[TMP2:%.*]] = shl <4 x i32> [[SHUFFLE]], <i32 5, i32 9, i32 3, i32 10>
14 ; CHECK-NEXT: [[TMP3:%.*]] = shufflevector <4 x i32> [[TMP1]], <4 x i32> [[TMP2]], <4 x i32> <i32 0, i32 1, i32 6, i32 3>
15 ; CHECK-NEXT: [[TMP4:%.*]] = add nsw <4 x i32> [[TMP3]], <i32 9, i32 9, i32 9, i32 9>
16 ; CHECK-NEXT: store <4 x i32> [[TMP4]], ptr [[A:%.*]], align 4
17 ; CHECK-NEXT: [[TMP6:%.*]] = extractelement <4 x i32> [[TMP4]], i32 0
18 ; CHECK-NEXT: [[EXTERNALUSE1:%.*]] = add nsw i32 [[TMP6]], [[M:%.*]]
19 ; CHECK-NEXT: [[EXTERNALUSE2:%.*]] = mul nsw i32 [[TMP6]], [[M]]
20 ; CHECK-NEXT: [[ADD10:%.*]] = add nsw i32 [[EXTERNALUSE1]], [[EXTERNALUSE2]]
21 ; CHECK-NEXT: ret i32 [[ADD10]]
24 %mul = mul nsw i32 %n, 5
25 %add = add nsw i32 %mul, 9
26 store i32 %add, ptr %A, align 4
27 %mul1 = mul nsw i32 %n, 9
28 %add2 = add nsw i32 %mul1, 9
29 %arrayidx3 = getelementptr inbounds i32, ptr %A, i64 1
30 store i32 %add2, ptr %arrayidx3, align 4
32 %add5 = add nsw i32 %mul4, 9
33 %arrayidx6 = getelementptr inbounds i32, ptr %A, i64 2
34 store i32 %add5, ptr %arrayidx6, align 4
35 %mul7 = mul nsw i32 %n, 10
36 %add8 = add nsw i32 %mul7, 9
37 %arrayidx9 = getelementptr inbounds i32, ptr %A, i64 3
38 store i32 %add8, ptr %arrayidx9, align 4
39 %externaluse1 = add nsw i32 %add, %m
40 %externaluse2 = mul nsw i32 %add, %m ; we should add the extract cost only once and the store will be vectorized
41 %add10 = add nsw i32 %externaluse1, %externaluse2