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
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
|
; RUN: llc < %s -mtriple=x86_64-apple-darwin -march=x86 -mcpu=corei7-avx | FileCheck %s
define <2 x double> @floor_v2f64(<2 x double> %p)
{
; CHECK: floor_v2f64
; CHECK: vroundpd
%t = call <2 x double> @llvm.floor.v2f64(<2 x double> %p)
ret <2 x double> %t
}
declare <2 x double> @llvm.floor.v2f64(<2 x double> %p)
define <4 x float> @floor_v4f32(<4 x float> %p)
{
; CHECK: floor_v4f32
; CHECK: vroundps
%t = call <4 x float> @llvm.floor.v4f32(<4 x float> %p)
ret <4 x float> %t
}
declare <4 x float> @llvm.floor.v4f32(<4 x float> %p)
define <4 x double> @floor_v4f64(<4 x double> %p)
{
; CHECK: floor_v4f64
; CHECK: vroundpd
%t = call <4 x double> @llvm.floor.v4f64(<4 x double> %p)
ret <4 x double> %t
}
declare <4 x double> @llvm.floor.v4f64(<4 x double> %p)
define <8 x float> @floor_v8f32(<8 x float> %p)
{
; CHECK: floor_v8f32
; CHECK: vroundps
%t = call <8 x float> @llvm.floor.v8f32(<8 x float> %p)
ret <8 x float> %t
}
declare <8 x float> @llvm.floor.v8f32(<8 x float> %p)
define <2 x double> @ceil_v2f64(<2 x double> %p)
{
; CHECK: ceil_v2f64
; CHECK: vroundpd
%t = call <2 x double> @llvm.ceil.v2f64(<2 x double> %p)
ret <2 x double> %t
}
declare <2 x double> @llvm.ceil.v2f64(<2 x double> %p)
define <4 x float> @ceil_v4f32(<4 x float> %p)
{
; CHECK: ceil_v4f32
; CHECK: vroundps
%t = call <4 x float> @llvm.ceil.v4f32(<4 x float> %p)
ret <4 x float> %t
}
declare <4 x float> @llvm.ceil.v4f32(<4 x float> %p)
define <4 x double> @ceil_v4f64(<4 x double> %p)
{
; CHECK: ceil_v4f64
; CHECK: vroundpd
%t = call <4 x double> @llvm.ceil.v4f64(<4 x double> %p)
ret <4 x double> %t
}
declare <4 x double> @llvm.ceil.v4f64(<4 x double> %p)
define <8 x float> @ceil_v8f32(<8 x float> %p)
{
; CHECK: ceil_v8f32
; CHECK: vroundps
%t = call <8 x float> @llvm.ceil.v8f32(<8 x float> %p)
ret <8 x float> %t
}
declare <8 x float> @llvm.ceil.v8f32(<8 x float> %p)
define <2 x double> @trunc_v2f64(<2 x double> %p)
{
; CHECK: trunc_v2f64
; CHECK: vroundpd
%t = call <2 x double> @llvm.trunc.v2f64(<2 x double> %p)
ret <2 x double> %t
}
declare <2 x double> @llvm.trunc.v2f64(<2 x double> %p)
define <4 x float> @trunc_v4f32(<4 x float> %p)
{
; CHECK: trunc_v4f32
; CHECK: vroundps
%t = call <4 x float> @llvm.trunc.v4f32(<4 x float> %p)
ret <4 x float> %t
}
declare <4 x float> @llvm.trunc.v4f32(<4 x float> %p)
define <4 x double> @trunc_v4f64(<4 x double> %p)
{
; CHECK: trunc_v4f64
; CHECK: vroundpd
%t = call <4 x double> @llvm.trunc.v4f64(<4 x double> %p)
ret <4 x double> %t
}
declare <4 x double> @llvm.trunc.v4f64(<4 x double> %p)
define <8 x float> @trunc_v8f32(<8 x float> %p)
{
; CHECK: trunc_v8f32
; CHECK: vroundps
%t = call <8 x float> @llvm.trunc.v8f32(<8 x float> %p)
ret <8 x float> %t
}
declare <8 x float> @llvm.trunc.v8f32(<8 x float> %p)
define <2 x double> @rint_v2f64(<2 x double> %p)
{
; CHECK: rint_v2f64
; CHECK: vroundpd
%t = call <2 x double> @llvm.rint.v2f64(<2 x double> %p)
ret <2 x double> %t
}
declare <2 x double> @llvm.rint.v2f64(<2 x double> %p)
define <4 x float> @rint_v4f32(<4 x float> %p)
{
; CHECK: rint_v4f32
; CHECK: vroundps
%t = call <4 x float> @llvm.rint.v4f32(<4 x float> %p)
ret <4 x float> %t
}
declare <4 x float> @llvm.rint.v4f32(<4 x float> %p)
define <4 x double> @rint_v4f64(<4 x double> %p)
{
; CHECK: rint_v4f64
; CHECK: vroundpd
%t = call <4 x double> @llvm.rint.v4f64(<4 x double> %p)
ret <4 x double> %t
}
declare <4 x double> @llvm.rint.v4f64(<4 x double> %p)
define <8 x float> @rint_v8f32(<8 x float> %p)
{
; CHECK: rint_v8f32
; CHECK: vroundps
%t = call <8 x float> @llvm.rint.v8f32(<8 x float> %p)
ret <8 x float> %t
}
declare <8 x float> @llvm.rint.v8f32(<8 x float> %p)
define <2 x double> @nearbyint_v2f64(<2 x double> %p)
{
; CHECK: nearbyint_v2f64
; CHECK: vroundpd
%t = call <2 x double> @llvm.nearbyint.v2f64(<2 x double> %p)
ret <2 x double> %t
}
declare <2 x double> @llvm.nearbyint.v2f64(<2 x double> %p)
define <4 x float> @nearbyint_v4f32(<4 x float> %p)
{
; CHECK: nearbyint_v4f32
; CHECK: vroundps
%t = call <4 x float> @llvm.nearbyint.v4f32(<4 x float> %p)
ret <4 x float> %t
}
declare <4 x float> @llvm.nearbyint.v4f32(<4 x float> %p)
define <4 x double> @nearbyint_v4f64(<4 x double> %p)
{
; CHECK: nearbyint_v4f64
; CHECK: vroundpd
%t = call <4 x double> @llvm.nearbyint.v4f64(<4 x double> %p)
ret <4 x double> %t
}
declare <4 x double> @llvm.nearbyint.v4f64(<4 x double> %p)
define <8 x float> @nearbyint_v8f32(<8 x float> %p)
{
; CHECK: nearbyint_v8f32
; CHECK: vroundps
%t = call <8 x float> @llvm.nearbyint.v8f32(<8 x float> %p)
ret <8 x float> %t
}
declare <8 x float> @llvm.nearbyint.v8f32(<8 x float> %p)
;
; Constant Folding
;
define <2 x double> @const_floor_v2f64() {
; CHECK: const_floor_v2f64
; CHECK: movaps {{.*#+}} xmm0 = [-2.000000e+00,2.000000e+00]
%t = call <2 x double> @llvm.floor.v2f64(<2 x double> <double -1.5, double 2.5>)
ret <2 x double> %t
}
define <4 x float> @const_floor_v4f32() {
; CHECK: const_floor_v4f32
; CHECK: movaps {{.*#+}} xmm0 = [-4.000000e+00,6.000000e+00,-9.000000e+00,2.000000e+00]
%t = call <4 x float> @llvm.floor.v4f32(<4 x float> <float -3.5, float 6.0, float -9.0, float 2.5>)
ret <4 x float> %t
}
define <2 x double> @const_ceil_v2f64() {
; CHECK: const_ceil_v2f64
; CHECK: movaps {{.*#+}} xmm0 = [-1.000000e+00,3.000000e+00]
%t = call <2 x double> @llvm.ceil.v2f64(<2 x double> <double -1.5, double 2.5>)
ret <2 x double> %t
}
define <4 x float> @const_ceil_v4f32() {
; CHECK: const_ceil_v4f32
; CHECK: movaps {{.*#+}} xmm0 = [-3.000000e+00,6.000000e+00,-9.000000e+00,3.000000e+00]
%t = call <4 x float> @llvm.ceil.v4f32(<4 x float> <float -3.5, float 6.0, float -9.0, float 2.5>)
ret <4 x float> %t
}
define <2 x double> @const_trunc_v2f64() {
; CHECK: const_trunc_v2f64
; CHECK: movaps {{.*#+}} xmm0 = [-1.000000e+00,2.000000e+00]
%t = call <2 x double> @llvm.trunc.v2f64(<2 x double> <double -1.5, double 2.5>)
ret <2 x double> %t
}
define <4 x float> @const_trunc_v4f32() {
; CHECK: const_trunc_v4f32
; CHECK: movaps {{.*#+}} xmm0 = [-3.000000e+00,6.000000e+00,-9.000000e+00,2.000000e+00]
%t = call <4 x float> @llvm.trunc.v4f32(<4 x float> <float -3.5, float 6.0, float -9.0, float 2.5>)
ret <4 x float> %t
}
|
