1 ; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
2 ; RUN: opt -mtriple=systemz-unknown -mcpu=z13 -slp-vectorizer -S < %s | FileCheck %s
4 @bar = external global [4 x [4 x i32]], align 4
5 @dct_luma = external global [4 x [4 x i32]], align 4
7 define void @foo() local_unnamed_addr {
10 ; CHECK-NEXT: [[ADD277:%.*]] = add nsw i32 undef, undef
11 ; CHECK-NEXT: store i32 [[ADD277]], i32* getelementptr inbounds ([4 x [4 x i32]], [4 x [4 x i32]]* @bar, i64 0, i64 3, i64 1), align 4
12 ; CHECK-NEXT: [[TMP0:%.*]] = load i32, i32* getelementptr inbounds ([4 x [4 x i32]], [4 x [4 x i32]]* @bar, i64 0, i64 3, i64 0), align 4
13 ; CHECK-NEXT: [[ARRAYIDX372:%.*]] = getelementptr inbounds [4 x [4 x i32]], [4 x [4 x i32]]* @dct_luma, i64 0, i64 3, i64 0
14 ; CHECK-NEXT: [[ARRAYIDX372_1:%.*]] = getelementptr inbounds [4 x [4 x i32]], [4 x [4 x i32]]* @dct_luma, i64 0, i64 3, i64 1
15 ; CHECK-NEXT: [[TMP1:%.*]] = load i32, i32* getelementptr inbounds ([4 x [4 x i32]], [4 x [4 x i32]]* @bar, i64 0, i64 3, i64 2), align 4
16 ; CHECK-NEXT: [[ARRAYIDX372_2:%.*]] = getelementptr inbounds [4 x [4 x i32]], [4 x [4 x i32]]* @dct_luma, i64 0, i64 3, i64 2
17 ; CHECK-NEXT: [[TMP2:%.*]] = load i32, i32* getelementptr inbounds ([4 x [4 x i32]], [4 x [4 x i32]]* @bar, i64 0, i64 3, i64 3), align 4
18 ; CHECK-NEXT: [[TMP3:%.*]] = insertelement <4 x i32> undef, i32 [[TMP0]], i32 0
19 ; CHECK-NEXT: [[TMP4:%.*]] = insertelement <4 x i32> [[TMP3]], i32 [[ADD277]], i32 1
20 ; CHECK-NEXT: [[TMP5:%.*]] = insertelement <4 x i32> [[TMP4]], i32 [[TMP1]], i32 2
21 ; CHECK-NEXT: [[TMP6:%.*]] = insertelement <4 x i32> [[TMP5]], i32 [[TMP2]], i32 3
22 ; CHECK-NEXT: [[TMP7:%.*]] = add nsw <4 x i32> undef, [[TMP6]]
23 ; CHECK-NEXT: [[TMP8:%.*]] = ashr <4 x i32> [[TMP7]], <i32 6, i32 6, i32 6, i32 6>
24 ; CHECK-NEXT: [[ARRAYIDX372_3:%.*]] = getelementptr inbounds [4 x [4 x i32]], [4 x [4 x i32]]* @dct_luma, i64 0, i64 3, i64 3
25 ; CHECK-NEXT: [[TMP9:%.*]] = bitcast i32* [[ARRAYIDX372]] to <4 x i32>*
26 ; CHECK-NEXT: store <4 x i32> [[TMP8]], <4 x i32>* [[TMP9]], align 4
27 ; CHECK-NEXT: unreachable
30 %add277 = add nsw i32 undef, undef
31 store i32 %add277, i32* getelementptr inbounds ([4 x [4 x i32]], [4 x [4 x i32]]* @bar, i64 0, i64 3, i64 1), align 4
32 %0 = load i32, i32* getelementptr inbounds ([4 x [4 x i32]], [4 x [4 x i32]]* @bar, i64 0, i64 3, i64 0), align 4
33 %sub355 = add nsw i32 undef, %0
34 %shr.i = ashr i32 %sub355, 6
35 %arrayidx372 = getelementptr inbounds [4 x [4 x i32]], [4 x [4 x i32]]* @dct_luma, i64 0, i64 3, i64 0
36 store i32 %shr.i, i32* %arrayidx372, align 4
37 %sub355.1 = add nsw i32 undef, %add277
38 %shr.i.1 = ashr i32 %sub355.1, 6
39 %arrayidx372.1 = getelementptr inbounds [4 x [4 x i32]], [4 x [4 x i32]]* @dct_luma, i64 0, i64 3, i64 1
40 store i32 %shr.i.1, i32* %arrayidx372.1, align 4
41 %1 = load i32, i32* getelementptr inbounds ([4 x [4 x i32]], [4 x [4 x i32]]* @bar, i64 0, i64 3, i64 2), align 4
42 %sub355.2 = add nsw i32 undef, %1
43 %shr.i.2 = ashr i32 %sub355.2, 6
44 %arrayidx372.2 = getelementptr inbounds [4 x [4 x i32]], [4 x [4 x i32]]* @dct_luma, i64 0, i64 3, i64 2
45 store i32 %shr.i.2, i32* %arrayidx372.2, align 4
46 %2 = load i32, i32* getelementptr inbounds ([4 x [4 x i32]], [4 x [4 x i32]]* @bar, i64 0, i64 3, i64 3), align 4
47 %sub355.3 = add nsw i32 undef, %2
48 %shr.i.3 = ashr i32 %sub355.3, 6
49 %arrayidx372.3 = getelementptr inbounds [4 x [4 x i32]], [4 x [4 x i32]]* @dct_luma, i64 0, i64 3, i64 3
50 store i32 %shr.i.3, i32* %arrayidx372.3, align 4