1; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
2; RUN: opt -lower-matrix-intrinsics -S < %s | FileCheck  %s
3; RUN: opt -passes='lower-matrix-intrinsics' -S < %s | FileCheck %s
4
5
6define <8 x i32> @transpose(<8 x i32> %a) {
7; CHECK-LABEL: @transpose(
8; CHECK-NEXT:  entry:
9; CHECK-NEXT:    [[SPLIT:%.*]] = shufflevector <8 x i32> [[A:%.*]], <8 x i32> poison, <2 x i32> <i32 0, i32 1>
10; CHECK-NEXT:    [[SPLIT1:%.*]] = shufflevector <8 x i32> [[A]], <8 x i32> poison, <2 x i32> <i32 2, i32 3>
11; CHECK-NEXT:    [[SPLIT2:%.*]] = shufflevector <8 x i32> [[A]], <8 x i32> poison, <2 x i32> <i32 4, i32 5>
12; CHECK-NEXT:    [[SPLIT3:%.*]] = shufflevector <8 x i32> [[A]], <8 x i32> poison, <2 x i32> <i32 6, i32 7>
13; CHECK-NEXT:    [[TMP0:%.*]] = extractelement <2 x i32> [[SPLIT]], i64 0
14; CHECK-NEXT:    [[TMP1:%.*]] = insertelement <4 x i32> undef, i32 [[TMP0]], i64 0
15; CHECK-NEXT:    [[TMP2:%.*]] = extractelement <2 x i32> [[SPLIT1]], i64 0
16; CHECK-NEXT:    [[TMP3:%.*]] = insertelement <4 x i32> [[TMP1]], i32 [[TMP2]], i64 1
17; CHECK-NEXT:    [[TMP4:%.*]] = extractelement <2 x i32> [[SPLIT2]], i64 0
18; CHECK-NEXT:    [[TMP5:%.*]] = insertelement <4 x i32> [[TMP3]], i32 [[TMP4]], i64 2
19; CHECK-NEXT:    [[TMP6:%.*]] = extractelement <2 x i32> [[SPLIT3]], i64 0
20; CHECK-NEXT:    [[TMP7:%.*]] = insertelement <4 x i32> [[TMP5]], i32 [[TMP6]], i64 3
21; CHECK-NEXT:    [[TMP8:%.*]] = extractelement <2 x i32> [[SPLIT]], i64 1
22; CHECK-NEXT:    [[TMP9:%.*]] = insertelement <4 x i32> undef, i32 [[TMP8]], i64 0
23; CHECK-NEXT:    [[TMP10:%.*]] = extractelement <2 x i32> [[SPLIT1]], i64 1
24; CHECK-NEXT:    [[TMP11:%.*]] = insertelement <4 x i32> [[TMP9]], i32 [[TMP10]], i64 1
25; CHECK-NEXT:    [[TMP12:%.*]] = extractelement <2 x i32> [[SPLIT2]], i64 1
26; CHECK-NEXT:    [[TMP13:%.*]] = insertelement <4 x i32> [[TMP11]], i32 [[TMP12]], i64 2
27; CHECK-NEXT:    [[TMP14:%.*]] = extractelement <2 x i32> [[SPLIT3]], i64 1
28; CHECK-NEXT:    [[TMP15:%.*]] = insertelement <4 x i32> [[TMP13]], i32 [[TMP14]], i64 3
29; CHECK-NEXT:    [[TMP16:%.*]] = shufflevector <4 x i32> [[TMP7]], <4 x i32> [[TMP15]], <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
30; CHECK-NEXT:    ret <8 x i32> [[TMP16]]
31;
32entry:
33  %c  = call <8 x i32> @llvm.matrix.transpose(<8 x i32> %a, i32 2, i32 4)
34  ret <8 x i32> %c
35}
36
37declare <8 x i32> @llvm.matrix.transpose(<8 x i32>, i32, i32)
38
39define <8 x i32> @transpose_single_column(<8 x i32> %a) {
40; CHECK-LABEL: @transpose_single_column(
41; CHECK-NEXT:  entry:
42; CHECK-NEXT:    [[SPLIT:%.*]] = shufflevector <8 x i32> [[A:%.*]], <8 x i32> poison, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
43; CHECK-NEXT:    [[TMP0:%.*]] = extractelement <8 x i32> [[SPLIT]], i64 0
44; CHECK-NEXT:    [[TMP1:%.*]] = insertelement <1 x i32> undef, i32 [[TMP0]], i64 0
45; CHECK-NEXT:    [[TMP2:%.*]] = extractelement <8 x i32> [[SPLIT]], i64 1
46; CHECK-NEXT:    [[TMP3:%.*]] = insertelement <1 x i32> undef, i32 [[TMP2]], i64 0
47; CHECK-NEXT:    [[TMP4:%.*]] = extractelement <8 x i32> [[SPLIT]], i64 2
48; CHECK-NEXT:    [[TMP5:%.*]] = insertelement <1 x i32> undef, i32 [[TMP4]], i64 0
49; CHECK-NEXT:    [[TMP6:%.*]] = extractelement <8 x i32> [[SPLIT]], i64 3
50; CHECK-NEXT:    [[TMP7:%.*]] = insertelement <1 x i32> undef, i32 [[TMP6]], i64 0
51; CHECK-NEXT:    [[TMP8:%.*]] = extractelement <8 x i32> [[SPLIT]], i64 4
52; CHECK-NEXT:    [[TMP9:%.*]] = insertelement <1 x i32> undef, i32 [[TMP8]], i64 0
53; CHECK-NEXT:    [[TMP10:%.*]] = extractelement <8 x i32> [[SPLIT]], i64 5
54; CHECK-NEXT:    [[TMP11:%.*]] = insertelement <1 x i32> undef, i32 [[TMP10]], i64 0
55; CHECK-NEXT:    [[TMP12:%.*]] = extractelement <8 x i32> [[SPLIT]], i64 6
56; CHECK-NEXT:    [[TMP13:%.*]] = insertelement <1 x i32> undef, i32 [[TMP12]], i64 0
57; CHECK-NEXT:    [[TMP14:%.*]] = extractelement <8 x i32> [[SPLIT]], i64 7
58; CHECK-NEXT:    [[TMP15:%.*]] = insertelement <1 x i32> undef, i32 [[TMP14]], i64 0
59; CHECK-NEXT:    [[TMP16:%.*]] = shufflevector <1 x i32> [[TMP1]], <1 x i32> [[TMP3]], <2 x i32> <i32 0, i32 1>
60; CHECK-NEXT:    [[TMP17:%.*]] = shufflevector <1 x i32> [[TMP5]], <1 x i32> [[TMP7]], <2 x i32> <i32 0, i32 1>
61; CHECK-NEXT:    [[TMP18:%.*]] = shufflevector <1 x i32> [[TMP9]], <1 x i32> [[TMP11]], <2 x i32> <i32 0, i32 1>
62; CHECK-NEXT:    [[TMP19:%.*]] = shufflevector <1 x i32> [[TMP13]], <1 x i32> [[TMP15]], <2 x i32> <i32 0, i32 1>
63; CHECK-NEXT:    [[TMP20:%.*]] = shufflevector <2 x i32> [[TMP16]], <2 x i32> [[TMP17]], <4 x i32> <i32 0, i32 1, i32 2, i32 3>
64; CHECK-NEXT:    [[TMP21:%.*]] = shufflevector <2 x i32> [[TMP18]], <2 x i32> [[TMP19]], <4 x i32> <i32 0, i32 1, i32 2, i32 3>
65; CHECK-NEXT:    [[TMP22:%.*]] = shufflevector <4 x i32> [[TMP20]], <4 x i32> [[TMP21]], <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
66; CHECK-NEXT:    ret <8 x i32> [[TMP22]]
67;
68entry:
69  %c  = call <8 x i32> @llvm.matrix.transpose(<8 x i32> %a, i32 8, i32 1)
70  ret <8 x i32> %c
71}
72
73declare <12 x i32> @llvm.matrix.transpose.v12i32(<12 x i32>, i32, i32)
74
75define <12 x i32> @transpose_i32_3x4(<12 x i32> %a) {
76; CHECK-LABEL: @transpose_i32_3x4(
77; CHECK-NEXT:  entry:
78; CHECK-NEXT:    [[SPLIT:%.*]] = shufflevector <12 x i32> [[A:%.*]], <12 x i32> poison, <3 x i32> <i32 0, i32 1, i32 2>
79; CHECK-NEXT:    [[SPLIT1:%.*]] = shufflevector <12 x i32> [[A]], <12 x i32> poison, <3 x i32> <i32 3, i32 4, i32 5>
80; CHECK-NEXT:    [[SPLIT2:%.*]] = shufflevector <12 x i32> [[A]], <12 x i32> poison, <3 x i32> <i32 6, i32 7, i32 8>
81; CHECK-NEXT:    [[SPLIT3:%.*]] = shufflevector <12 x i32> [[A]], <12 x i32> poison, <3 x i32> <i32 9, i32 10, i32 11>
82; CHECK-NEXT:    [[TMP0:%.*]] = extractelement <3 x i32> [[SPLIT]], i64 0
83; CHECK-NEXT:    [[TMP1:%.*]] = insertelement <4 x i32> undef, i32 [[TMP0]], i64 0
84; CHECK-NEXT:    [[TMP2:%.*]] = extractelement <3 x i32> [[SPLIT1]], i64 0
85; CHECK-NEXT:    [[TMP3:%.*]] = insertelement <4 x i32> [[TMP1]], i32 [[TMP2]], i64 1
86; CHECK-NEXT:    [[TMP4:%.*]] = extractelement <3 x i32> [[SPLIT2]], i64 0
87; CHECK-NEXT:    [[TMP5:%.*]] = insertelement <4 x i32> [[TMP3]], i32 [[TMP4]], i64 2
88; CHECK-NEXT:    [[TMP6:%.*]] = extractelement <3 x i32> [[SPLIT3]], i64 0
89; CHECK-NEXT:    [[TMP7:%.*]] = insertelement <4 x i32> [[TMP5]], i32 [[TMP6]], i64 3
90; CHECK-NEXT:    [[TMP8:%.*]] = extractelement <3 x i32> [[SPLIT]], i64 1
91; CHECK-NEXT:    [[TMP9:%.*]] = insertelement <4 x i32> undef, i32 [[TMP8]], i64 0
92; CHECK-NEXT:    [[TMP10:%.*]] = extractelement <3 x i32> [[SPLIT1]], i64 1
93; CHECK-NEXT:    [[TMP11:%.*]] = insertelement <4 x i32> [[TMP9]], i32 [[TMP10]], i64 1
94; CHECK-NEXT:    [[TMP12:%.*]] = extractelement <3 x i32> [[SPLIT2]], i64 1
95; CHECK-NEXT:    [[TMP13:%.*]] = insertelement <4 x i32> [[TMP11]], i32 [[TMP12]], i64 2
96; CHECK-NEXT:    [[TMP14:%.*]] = extractelement <3 x i32> [[SPLIT3]], i64 1
97; CHECK-NEXT:    [[TMP15:%.*]] = insertelement <4 x i32> [[TMP13]], i32 [[TMP14]], i64 3
98; CHECK-NEXT:    [[TMP16:%.*]] = extractelement <3 x i32> [[SPLIT]], i64 2
99; CHECK-NEXT:    [[TMP17:%.*]] = insertelement <4 x i32> undef, i32 [[TMP16]], i64 0
100; CHECK-NEXT:    [[TMP18:%.*]] = extractelement <3 x i32> [[SPLIT1]], i64 2
101; CHECK-NEXT:    [[TMP19:%.*]] = insertelement <4 x i32> [[TMP17]], i32 [[TMP18]], i64 1
102; CHECK-NEXT:    [[TMP20:%.*]] = extractelement <3 x i32> [[SPLIT2]], i64 2
103; CHECK-NEXT:    [[TMP21:%.*]] = insertelement <4 x i32> [[TMP19]], i32 [[TMP20]], i64 2
104; CHECK-NEXT:    [[TMP22:%.*]] = extractelement <3 x i32> [[SPLIT3]], i64 2
105; CHECK-NEXT:    [[TMP23:%.*]] = insertelement <4 x i32> [[TMP21]], i32 [[TMP22]], i64 3
106; CHECK-NEXT:    [[TMP24:%.*]] = shufflevector <4 x i32> [[TMP7]], <4 x i32> [[TMP15]], <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
107; CHECK-NEXT:    [[TMP25:%.*]] = shufflevector <4 x i32> [[TMP23]], <4 x i32> poison, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 undef, i32 undef, i32 undef, i32 undef>
108; CHECK-NEXT:    [[TMP26:%.*]] = shufflevector <8 x i32> [[TMP24]], <8 x i32> [[TMP25]], <12 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 8, i32 9, i32 10, i32 11>
109; CHECK-NEXT:    ret <12 x i32> [[TMP26]]
110;
111entry:
112  %c  = call <12 x i32> @llvm.matrix.transpose.v12i32(<12 x i32> %a, i32 3, i32 4)
113  ret <12 x i32> %c
114}
115