1 ; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
2 ; RUN: opt -passes=slp-vectorizer -S < %s -mtriple=x86_64-unknown-linux -slp-threshold=-127 | FileCheck %s
3 ; RUN: opt -passes=slp-vectorizer -S < %s -mtriple=x86_64-unknown-linux -slp-threshold=-115 -mattr=+avx2 | FileCheck %s --check-prefix=AVX2
5 define void @test(i64 %p0, i64 %p1, i64 %p2, i64 %p3) {
8 ; CHECK-NEXT: [[TMP0:%.*]] = insertelement <4 x i64> poison, i64 [[P0:%.*]], i32 0
9 ; CHECK-NEXT: [[TMP1:%.*]] = insertelement <4 x i64> [[TMP0]], i64 [[P1:%.*]], i32 1
10 ; CHECK-NEXT: [[TMP2:%.*]] = insertelement <4 x i64> [[TMP1]], i64 [[P2:%.*]], i32 2
11 ; CHECK-NEXT: [[TMP3:%.*]] = insertelement <4 x i64> [[TMP2]], i64 [[P3:%.*]], i32 3
12 ; CHECK-NEXT: [[TMP4:%.*]] = add <4 x i64> [[TMP3]], [[TMP3]]
13 ; CHECK-NEXT: [[TMP5:%.*]] = mul <4 x i64> [[TMP3]], [[TMP3]]
14 ; CHECK-NEXT: [[TMP6:%.*]] = sdiv <4 x i64> [[TMP3]], [[TMP3]]
15 ; CHECK-NEXT: [[TMP7:%.*]] = sub <4 x i64> [[TMP5]], [[TMP6]]
16 ; CHECK-NEXT: [[TMP8:%.*]] = shl <4 x i64> [[TMP4]], [[TMP7]]
17 ; CHECK-NEXT: [[TMP9:%.*]] = shufflevector <4 x i64> [[TMP4]], <4 x i64> [[TMP5]], <4 x i32> <i32 0, i32 4, i32 poison, i32 poison>
18 ; CHECK-NEXT: [[TMP10:%.*]] = shufflevector <4 x i64> [[TMP6]], <4 x i64> [[TMP5]], <4 x i32> <i32 poison, i32 poison, i32 0, i32 4>
19 ; CHECK-NEXT: [[TMP11:%.*]] = shufflevector <4 x i64> [[TMP9]], <4 x i64> [[TMP10]], <4 x i32> <i32 0, i32 1, i32 6, i32 7>
20 ; CHECK-NEXT: [[TMP12:%.*]] = shufflevector <4 x i64> [[TMP4]], <4 x i64> [[TMP5]], <4 x i32> <i32 1, i32 5, i32 poison, i32 poison>
21 ; CHECK-NEXT: [[TMP13:%.*]] = shufflevector <4 x i64> [[TMP6]], <4 x i64> [[TMP5]], <4 x i32> <i32 poison, i32 poison, i32 1, i32 5>
22 ; CHECK-NEXT: [[TMP14:%.*]] = shufflevector <4 x i64> [[TMP12]], <4 x i64> [[TMP13]], <4 x i32> <i32 0, i32 1, i32 6, i32 7>
23 ; CHECK-NEXT: [[TMP15:%.*]] = or <4 x i64> [[TMP11]], [[TMP14]]
24 ; CHECK-NEXT: [[TMP16:%.*]] = trunc <4 x i64> [[TMP15]] to <4 x i32>
25 ; CHECK-NEXT: br label [[BB:%.*]]
27 ; CHECK-NEXT: [[TMP17:%.*]] = phi <4 x i32> [ [[TMP18:%.*]], [[BB]] ], [ [[TMP16]], [[ENTRY:%.*]] ]
28 ; CHECK-NEXT: [[TMP18]] = trunc <4 x i64> [[TMP8]] to <4 x i32>
29 ; CHECK-NEXT: br label [[BB]]
33 ; AVX2-NEXT: [[TMP0:%.*]] = insertelement <4 x i64> poison, i64 [[P0:%.*]], i32 0
34 ; AVX2-NEXT: [[TMP1:%.*]] = insertelement <4 x i64> [[TMP0]], i64 [[P1:%.*]], i32 1
35 ; AVX2-NEXT: [[TMP2:%.*]] = insertelement <4 x i64> [[TMP1]], i64 [[P2:%.*]], i32 2
36 ; AVX2-NEXT: [[TMP3:%.*]] = insertelement <4 x i64> [[TMP2]], i64 [[P3:%.*]], i32 3
37 ; AVX2-NEXT: [[TMP4:%.*]] = add <4 x i64> [[TMP3]], [[TMP3]]
38 ; AVX2-NEXT: [[TMP5:%.*]] = mul <4 x i64> [[TMP3]], [[TMP3]]
39 ; AVX2-NEXT: [[TMP6:%.*]] = sdiv <4 x i64> [[TMP3]], [[TMP3]]
40 ; AVX2-NEXT: [[TMP7:%.*]] = sub <4 x i64> [[TMP5]], [[TMP6]]
41 ; AVX2-NEXT: [[TMP8:%.*]] = shl <4 x i64> [[TMP4]], [[TMP7]]
42 ; AVX2-NEXT: [[TMP9:%.*]] = shufflevector <4 x i64> [[TMP4]], <4 x i64> [[TMP5]], <4 x i32> <i32 0, i32 4, i32 poison, i32 4>
43 ; AVX2-NEXT: [[TMP10:%.*]] = shufflevector <4 x i64> [[TMP9]], <4 x i64> [[TMP6]], <4 x i32> <i32 0, i32 1, i32 4, i32 3>
44 ; AVX2-NEXT: [[TMP11:%.*]] = shufflevector <4 x i64> [[TMP4]], <4 x i64> [[TMP5]], <4 x i32> <i32 1, i32 5, i32 poison, i32 5>
45 ; AVX2-NEXT: [[TMP12:%.*]] = shufflevector <4 x i64> [[TMP11]], <4 x i64> [[TMP6]], <4 x i32> <i32 0, i32 1, i32 5, i32 3>
46 ; AVX2-NEXT: [[TMP13:%.*]] = or <4 x i64> [[TMP10]], [[TMP12]]
47 ; AVX2-NEXT: [[TMP14:%.*]] = trunc <4 x i64> [[TMP13]] to <4 x i32>
48 ; AVX2-NEXT: br label [[BB:%.*]]
50 ; AVX2-NEXT: [[TMP15:%.*]] = phi <4 x i32> [ [[TMP16:%.*]], [[BB]] ], [ [[TMP14]], [[ENTRY:%.*]] ]
51 ; AVX2-NEXT: [[TMP16]] = trunc <4 x i64> [[TMP8]] to <4 x i32>
52 ; AVX2-NEXT: br label [[BB]]
55 %a0 = add i64 %p0, %p0
56 %a1 = add i64 %p1, %p1
57 %a2 = add i64 %p2, %p2
58 %a3 = add i64 %p3, %p3
59 %m0 = mul i64 %p0, %p0
60 %m1 = mul i64 %p1, %p1
61 %m2 = mul i64 %p2, %p2
62 %m3 = mul i64 %p3, %p3
63 %d0 = sdiv i64 %p0, %p0
64 %d1 = sdiv i64 %p1, %p1
65 %d2 = sdiv i64 %p2, %p2
66 %d3 = sdiv i64 %p3, %p3
67 %s0 = sub i64 %m0, %d0
68 %s1 = sub i64 %m1, %d1
69 %s2 = sub i64 %m2, %d2
70 %s3 = sub i64 %m3, %d3
71 %shl1 = shl i64 %a0, %s0
72 %shl2 = shl i64 %a1, %s1
73 %shl3 = shl i64 %a2, %s2
74 %shl4 = shl i64 %a3, %s3
76 %tt0 = trunc i64 %o0 to i32
78 %tt1 = trunc i64 %o1 to i32
80 %tt2 = trunc i64 %o2 to i32
82 %tt3 = trunc i64 %o3 to i32
86 %phi0 = phi i32 [ %t1, %bb ], [ %tt0, %entry ]
87 %phi1 = phi i32 [ %t2, %bb ], [ %tt1, %entry ]
88 %phi2 = phi i32 [ %t3, %bb ], [ %tt2, %entry ]
89 %phi3 = phi i32 [ %t4, %bb ], [ %tt3, %entry ]
90 %t1 = trunc i64 %shl1 to i32
91 %t2 = trunc i64 %shl2 to i32
92 %t3 = trunc i64 %shl3 to i32
93 %t4 = trunc i64 %shl4 to i32