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
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
|
; RUN: llc -mtriple=aarch64-linux-gnu -mattr=+sve < %s | FileCheck %s
;
; CLS
;
define <vscale x 16 x i8> @cls_i8(<vscale x 16 x i8> %a, <vscale x 16 x i1> %pg, <vscale x 16 x i8> %b) {
; CHECK-LABEL: cls_i8:
; CHECK: cls z0.b, p0/m, z1.b
; CHECK-NEXT: ret
%out = call <vscale x 16 x i8> @llvm.aarch64.sve.cls.nxv16i8(<vscale x 16 x i8> %a,
<vscale x 16 x i1> %pg,
<vscale x 16 x i8> %b)
ret <vscale x 16 x i8> %out
}
define <vscale x 8 x i16> @cls_i16(<vscale x 8 x i16> %a, <vscale x 8 x i1> %pg, <vscale x 8 x i16> %b) {
; CHECK-LABEL: cls_i16:
; CHECK: cls z0.h, p0/m, z1.h
; CHECK-NEXT: ret
%out = call <vscale x 8 x i16> @llvm.aarch64.sve.cls.nxv8i16(<vscale x 8 x i16> %a,
<vscale x 8 x i1> %pg,
<vscale x 8 x i16> %b)
ret <vscale x 8 x i16> %out
}
define <vscale x 4 x i32> @cls_i32(<vscale x 4 x i32> %a, <vscale x 4 x i1> %pg, <vscale x 4 x i32> %b) {
; CHECK-LABEL: cls_i32:
; CHECK: cls z0.s, p0/m, z1.s
; CHECK-NEXT: ret
%out = call <vscale x 4 x i32> @llvm.aarch64.sve.cls.nxv4i32(<vscale x 4 x i32> %a,
<vscale x 4 x i1> %pg,
<vscale x 4 x i32> %b)
ret <vscale x 4 x i32> %out
}
define <vscale x 2 x i64> @cls_i64(<vscale x 2 x i64> %a, <vscale x 2 x i1> %pg, <vscale x 2 x i64> %b) {
; CHECK-LABEL: cls_i64:
; CHECK: cls z0.d, p0/m, z1.d
; CHECK-NEXT: ret
%out = call <vscale x 2 x i64> @llvm.aarch64.sve.cls.nxv2i64(<vscale x 2 x i64> %a,
<vscale x 2 x i1> %pg,
<vscale x 2 x i64> %b)
ret <vscale x 2 x i64> %out
}
;
; CLZ
;
define <vscale x 16 x i8> @clz_i8(<vscale x 16 x i8> %a, <vscale x 16 x i1> %pg, <vscale x 16 x i8> %b) {
; CHECK-LABEL: clz_i8:
; CHECK: clz z0.b, p0/m, z1.b
; CHECK-NEXT: ret
%out = call <vscale x 16 x i8> @llvm.aarch64.sve.clz.nxv16i8(<vscale x 16 x i8> %a,
<vscale x 16 x i1> %pg,
<vscale x 16 x i8> %b)
ret <vscale x 16 x i8> %out
}
define <vscale x 8 x i16> @clz_i16(<vscale x 8 x i16> %a, <vscale x 8 x i1> %pg, <vscale x 8 x i16> %b) {
; CHECK-LABEL: clz_i16:
; CHECK: clz z0.h, p0/m, z1.h
; CHECK-NEXT: ret
%out = call <vscale x 8 x i16> @llvm.aarch64.sve.clz.nxv8i16(<vscale x 8 x i16> %a,
<vscale x 8 x i1> %pg,
<vscale x 8 x i16> %b)
ret <vscale x 8 x i16> %out
}
define <vscale x 4 x i32> @clz_i32(<vscale x 4 x i32> %a, <vscale x 4 x i1> %pg, <vscale x 4 x i32> %b) {
; CHECK-LABEL: clz_i32:
; CHECK: clz z0.s, p0/m, z1.s
; CHECK-NEXT: ret
%out = call <vscale x 4 x i32> @llvm.aarch64.sve.clz.nxv4i32(<vscale x 4 x i32> %a,
<vscale x 4 x i1> %pg,
<vscale x 4 x i32> %b)
ret <vscale x 4 x i32> %out
}
define <vscale x 2 x i64> @clz_i64(<vscale x 2 x i64> %a, <vscale x 2 x i1> %pg, <vscale x 2 x i64> %b) {
; CHECK-LABEL: clz_i64:
; CHECK: clz z0.d, p0/m, z1.d
; CHECK-NEXT: ret
%out = call <vscale x 2 x i64> @llvm.aarch64.sve.clz.nxv2i64(<vscale x 2 x i64> %a,
<vscale x 2 x i1> %pg,
<vscale x 2 x i64> %b)
ret <vscale x 2 x i64> %out
}
;
; CNT
;
define <vscale x 16 x i8> @cnt_i8(<vscale x 16 x i8> %a, <vscale x 16 x i1> %pg, <vscale x 16 x i8> %b) {
; CHECK-LABEL: cnt_i8:
; CHECK: cnt z0.b, p0/m, z1.b
; CHECK-NEXT: ret
%out = call <vscale x 16 x i8> @llvm.aarch64.sve.cnt.nxv16i8(<vscale x 16 x i8> %a,
<vscale x 16 x i1> %pg,
<vscale x 16 x i8> %b)
ret <vscale x 16 x i8> %out
}
define <vscale x 8 x i16> @cnt_i16(<vscale x 8 x i16> %a, <vscale x 8 x i1> %pg, <vscale x 8 x i16> %b) {
; CHECK-LABEL: cnt_i16:
; CHECK: cnt z0.h, p0/m, z1.h
; CHECK-NEXT: ret
%out = call <vscale x 8 x i16> @llvm.aarch64.sve.cnt.nxv8i16(<vscale x 8 x i16> %a,
<vscale x 8 x i1> %pg,
<vscale x 8 x i16> %b)
ret <vscale x 8 x i16> %out
}
define <vscale x 4 x i32> @cnt_i32(<vscale x 4 x i32> %a, <vscale x 4 x i1> %pg, <vscale x 4 x i32> %b) {
; CHECK-LABEL: cnt_i32:
; CHECK: cnt z0.s, p0/m, z1.s
; CHECK-NEXT: ret
%out = call <vscale x 4 x i32> @llvm.aarch64.sve.cnt.nxv4i32(<vscale x 4 x i32> %a,
<vscale x 4 x i1> %pg,
<vscale x 4 x i32> %b)
ret <vscale x 4 x i32> %out
}
define <vscale x 2 x i64> @cnt_i64(<vscale x 2 x i64> %a, <vscale x 2 x i1> %pg, <vscale x 2 x i64> %b) {
; CHECK-LABEL: cnt_i64:
; CHECK: cnt z0.d, p0/m, z1.d
; CHECK-NEXT: ret
%out = call <vscale x 2 x i64> @llvm.aarch64.sve.cnt.nxv2i64(<vscale x 2 x i64> %a,
<vscale x 2 x i1> %pg,
<vscale x 2 x i64> %b)
ret <vscale x 2 x i64> %out
}
define <vscale x 8 x i16> @cnt_f16(<vscale x 8 x i16> %a, <vscale x 8 x i1> %pg, <vscale x 8 x half> %b) {
; CHECK-LABEL: cnt_f16:
; CHECK: cnt z0.h, p0/m, z1.h
; CHECK-NEXT: ret
%out = call <vscale x 8 x i16> @llvm.aarch64.sve.cnt.nxv8f16(<vscale x 8 x i16> %a,
<vscale x 8 x i1> %pg,
<vscale x 8 x half> %b)
ret <vscale x 8 x i16> %out
}
define <vscale x 8 x i16> @cnt_bf16(<vscale x 8 x i16> %a, <vscale x 8 x i1> %pg, <vscale x 8 x bfloat> %b) #0 {
; CHECK-LABEL: cnt_bf16:
; CHECK: cnt z0.h, p0/m, z1.h
; CHECK-NEXT: ret
%out = call <vscale x 8 x i16> @llvm.aarch64.sve.cnt.nxv8bf16(<vscale x 8 x i16> %a,
<vscale x 8 x i1> %pg,
<vscale x 8 x bfloat> %b)
ret <vscale x 8 x i16> %out
}
define <vscale x 4 x i32> @cnt_f32(<vscale x 4 x i32> %a, <vscale x 4 x i1> %pg, <vscale x 4 x float> %b) {
; CHECK-LABEL: cnt_f32:
; CHECK: cnt z0.s, p0/m, z1.s
; CHECK-NEXT: ret
%out = call <vscale x 4 x i32> @llvm.aarch64.sve.cnt.nxv4f32(<vscale x 4 x i32> %a,
<vscale x 4 x i1> %pg,
<vscale x 4 x float> %b)
ret <vscale x 4 x i32> %out
}
define <vscale x 2 x i64> @cnt_f64(<vscale x 2 x i64> %a, <vscale x 2 x i1> %pg, <vscale x 2 x double> %b) {
; CHECK-LABEL: cnt_f64:
; CHECK: cnt z0.d, p0/m, z1.d
; CHECK-NEXT: ret
%out = call <vscale x 2 x i64> @llvm.aarch64.sve.cnt.nxv2f64(<vscale x 2 x i64> %a,
<vscale x 2 x i1> %pg,
<vscale x 2 x double> %b)
ret <vscale x 2 x i64> %out
}
declare <vscale x 16 x i8> @llvm.aarch64.sve.cls.nxv16i8(<vscale x 16 x i8>, <vscale x 16 x i1>, <vscale x 16 x i8>)
declare <vscale x 8 x i16> @llvm.aarch64.sve.cls.nxv8i16(<vscale x 8 x i16>, <vscale x 8 x i1>, <vscale x 8 x i16>)
declare <vscale x 4 x i32> @llvm.aarch64.sve.cls.nxv4i32(<vscale x 4 x i32>, <vscale x 4 x i1>, <vscale x 4 x i32>)
declare <vscale x 2 x i64> @llvm.aarch64.sve.cls.nxv2i64(<vscale x 2 x i64>, <vscale x 2 x i1>, <vscale x 2 x i64>)
declare <vscale x 16 x i8> @llvm.aarch64.sve.clz.nxv16i8(<vscale x 16 x i8>, <vscale x 16 x i1>, <vscale x 16 x i8>)
declare <vscale x 8 x i16> @llvm.aarch64.sve.clz.nxv8i16(<vscale x 8 x i16>, <vscale x 8 x i1>, <vscale x 8 x i16>)
declare <vscale x 4 x i32> @llvm.aarch64.sve.clz.nxv4i32(<vscale x 4 x i32>, <vscale x 4 x i1>, <vscale x 4 x i32>)
declare <vscale x 2 x i64> @llvm.aarch64.sve.clz.nxv2i64(<vscale x 2 x i64>, <vscale x 2 x i1>, <vscale x 2 x i64>)
declare <vscale x 16 x i8> @llvm.aarch64.sve.cnt.nxv16i8(<vscale x 16 x i8>, <vscale x 16 x i1>, <vscale x 16 x i8>)
declare <vscale x 8 x i16> @llvm.aarch64.sve.cnt.nxv8i16(<vscale x 8 x i16>, <vscale x 8 x i1>, <vscale x 8 x i16>)
declare <vscale x 4 x i32> @llvm.aarch64.sve.cnt.nxv4i32(<vscale x 4 x i32>, <vscale x 4 x i1>, <vscale x 4 x i32>)
declare <vscale x 2 x i64> @llvm.aarch64.sve.cnt.nxv2i64(<vscale x 2 x i64>, <vscale x 2 x i1>, <vscale x 2 x i64>)
declare <vscale x 8 x i16> @llvm.aarch64.sve.cnt.nxv8f16(<vscale x 8 x i16>, <vscale x 8 x i1>, <vscale x 8 x half>)
declare <vscale x 8 x i16> @llvm.aarch64.sve.cnt.nxv8bf16(<vscale x 8 x i16>, <vscale x 8 x i1>, <vscale x 8 x bfloat>)
declare <vscale x 4 x i32> @llvm.aarch64.sve.cnt.nxv4f32(<vscale x 4 x i32>, <vscale x 4 x i1>, <vscale x 4 x float>)
declare <vscale x 2 x i64> @llvm.aarch64.sve.cnt.nxv2f64(<vscale x 2 x i64>, <vscale x 2 x i1>, <vscale x 2 x double>)
; +bf16 is required for the bfloat version.
attributes #0 = { "target-features"="+sve,+bf16" }
|
