1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
|
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
; RUN: llc -mtriple=aarch64-linux-gnu -mattr=+sve2 < %s | FileCheck %s
; RUN: llc -mtriple=aarch64-linux-gnu -mattr=+sme < %s | FileCheck %s
;
; FMLALB (Vectors)
;
define <vscale x 4 x float> @fmlalb_h(<vscale x 4 x float> %a, <vscale x 8 x half> %b, <vscale x 8 x half> %c) {
; CHECK-LABEL: fmlalb_h:
; CHECK: // %bb.0:
; CHECK-NEXT: fmlalb z0.s, z1.h, z2.h
; CHECK-NEXT: ret
%out = call <vscale x 4 x float> @llvm.aarch64.sve.fmlalb.nxv4f32(<vscale x 4 x float> %a,
<vscale x 8 x half> %b,
<vscale x 8 x half> %c)
ret <vscale x 4 x float> %out
}
;
; FMLALB (Indexed)
;
define <vscale x 4 x float> @fmlalb_lane_h(<vscale x 4 x float> %a, <vscale x 8 x half> %b, <vscale x 8 x half> %c) {
; CHECK-LABEL: fmlalb_lane_h:
; CHECK: // %bb.0:
; CHECK-NEXT: fmlalb z0.s, z1.h, z2.h[0]
; CHECK-NEXT: ret
%out = call <vscale x 4 x float> @llvm.aarch64.sve.fmlalb.lane.nxv4f32(<vscale x 4 x float> %a,
<vscale x 8 x half> %b,
<vscale x 8 x half> %c,
i32 0)
ret <vscale x 4 x float> %out
}
;
; FMLALT (Vectors)
;
define <vscale x 4 x float> @fmlalt_h(<vscale x 4 x float> %a, <vscale x 8 x half> %b, <vscale x 8 x half> %c) {
; CHECK-LABEL: fmlalt_h:
; CHECK: // %bb.0:
; CHECK-NEXT: fmlalt z0.s, z1.h, z2.h
; CHECK-NEXT: ret
%out = call <vscale x 4 x float> @llvm.aarch64.sve.fmlalt.nxv4f32(<vscale x 4 x float> %a,
<vscale x 8 x half> %b,
<vscale x 8 x half> %c)
ret <vscale x 4 x float> %out
}
;
; FMLALT (Indexed)
;
define <vscale x 4 x float> @fmlalt_lane_h(<vscale x 4 x float> %a, <vscale x 8 x half> %b, <vscale x 8 x half> %c) {
; CHECK-LABEL: fmlalt_lane_h:
; CHECK: // %bb.0:
; CHECK-NEXT: fmlalt z0.s, z1.h, z2.h[1]
; CHECK-NEXT: ret
%out = call <vscale x 4 x float> @llvm.aarch64.sve.fmlalt.lane.nxv4f32(<vscale x 4 x float> %a,
<vscale x 8 x half> %b,
<vscale x 8 x half> %c,
i32 1)
ret <vscale x 4 x float> %out
}
;
; FMLSLB (Vectors)
;
define <vscale x 4 x float> @fmlslb_h(<vscale x 4 x float> %a, <vscale x 8 x half> %b, <vscale x 8 x half> %c) {
; CHECK-LABEL: fmlslb_h:
; CHECK: // %bb.0:
; CHECK-NEXT: fmlslb z0.s, z1.h, z2.h
; CHECK-NEXT: ret
%out = call <vscale x 4 x float> @llvm.aarch64.sve.fmlslb.nxv4f32(<vscale x 4 x float> %a,
<vscale x 8 x half> %b,
<vscale x 8 x half> %c)
ret <vscale x 4 x float> %out
}
;
; FMLSLB (Indexed)
;
define <vscale x 4 x float> @fmlslb_lane_h(<vscale x 4 x float> %a, <vscale x 8 x half> %b, <vscale x 8 x half> %c) {
; CHECK-LABEL: fmlslb_lane_h:
; CHECK: // %bb.0:
; CHECK-NEXT: fmlslb z0.s, z1.h, z2.h[2]
; CHECK-NEXT: ret
%out = call <vscale x 4 x float> @llvm.aarch64.sve.fmlslb.lane.nxv4f32(<vscale x 4 x float> %a,
<vscale x 8 x half> %b,
<vscale x 8 x half> %c,
i32 2)
ret <vscale x 4 x float> %out
}
;
; FMLSLT (Vectors)
;
define <vscale x 4 x float> @fmlslt_h(<vscale x 4 x float> %a, <vscale x 8 x half> %b, <vscale x 8 x half> %c) {
; CHECK-LABEL: fmlslt_h:
; CHECK: // %bb.0:
; CHECK-NEXT: fmlslt z0.s, z1.h, z2.h
; CHECK-NEXT: ret
%out = call <vscale x 4 x float> @llvm.aarch64.sve.fmlslt.nxv4f32(<vscale x 4 x float> %a,
<vscale x 8 x half> %b,
<vscale x 8 x half> %c)
ret <vscale x 4 x float> %out
}
;
; FMLSLT (Indexed)
;
define <vscale x 4 x float> @fmlslt_lane_h(<vscale x 4 x float> %a, <vscale x 8 x half> %b, <vscale x 8 x half> %c) {
; CHECK-LABEL: fmlslt_lane_h:
; CHECK: // %bb.0:
; CHECK-NEXT: fmlslt z0.s, z1.h, z2.h[3]
; CHECK-NEXT: ret
%out = call <vscale x 4 x float> @llvm.aarch64.sve.fmlslt.lane.nxv4f32(<vscale x 4 x float> %a,
<vscale x 8 x half> %b,
<vscale x 8 x half> %c,
i32 3)
ret <vscale x 4 x float> %out
}
declare <vscale x 4 x float> @llvm.aarch64.sve.fmlalb.nxv4f32(<vscale x 4 x float>, <vscale x 8 x half>, <vscale x 8 x half>)
declare <vscale x 4 x float> @llvm.aarch64.sve.fmlalb.lane.nxv4f32(<vscale x 4 x float>, <vscale x 8 x half>, <vscale x 8 x half>, i32)
declare <vscale x 4 x float> @llvm.aarch64.sve.fmlalt.nxv4f32(<vscale x 4 x float>, <vscale x 8 x half>, <vscale x 8 x half>)
declare <vscale x 4 x float> @llvm.aarch64.sve.fmlalt.lane.nxv4f32(<vscale x 4 x float>, <vscale x 8 x half>, <vscale x 8 x half>, i32)
declare <vscale x 4 x float> @llvm.aarch64.sve.fmlslb.nxv4f32(<vscale x 4 x float>, <vscale x 8 x half>, <vscale x 8 x half>)
declare <vscale x 4 x float> @llvm.aarch64.sve.fmlslb.lane.nxv4f32(<vscale x 4 x float>, <vscale x 8 x half>, <vscale x 8 x half>, i32)
declare <vscale x 4 x float> @llvm.aarch64.sve.fmlslt.nxv4f32(<vscale x 4 x float>, <vscale x 8 x half>, <vscale x 8 x half>)
declare <vscale x 4 x float> @llvm.aarch64.sve.fmlslt.lane.nxv4f32(<vscale x 4 x float>, <vscale x 8 x half>, <vscale x 8 x half>, i32)
|
