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
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
|
; NOTE: Assertions have been autogenerated by utils/update_mir_test_checks.py
; RUN: llc -global-isel -march=amdgcn -stop-after=irtranslator -verify-machineinstrs %s -o - | FileCheck %s
define float @v_constained_fadd_f32_fpexcept_strict(float %x, float %y) #0 {
; CHECK-LABEL: name: v_constained_fadd_f32_fpexcept_strict
; CHECK: bb.1 (%ir-block.0):
; CHECK-NEXT: liveins: $vgpr0, $vgpr1, $sgpr30_sgpr31
; CHECK-NEXT: {{ $}}
; CHECK-NEXT: [[COPY:%[0-9]+]]:_(s32) = COPY $vgpr0
; CHECK-NEXT: [[COPY1:%[0-9]+]]:_(s32) = COPY $vgpr1
; CHECK-NEXT: [[COPY2:%[0-9]+]]:sgpr_64 = COPY $sgpr30_sgpr31
; CHECK-NEXT: [[STRICT_FADD:%[0-9]+]]:_(s32) = G_STRICT_FADD [[COPY]], [[COPY1]]
; CHECK-NEXT: $vgpr0 = COPY [[STRICT_FADD]](s32)
; CHECK-NEXT: [[COPY3:%[0-9]+]]:ccr_sgpr_64 = COPY [[COPY2]]
; CHECK-NEXT: S_SETPC_B64_return [[COPY3]], implicit $vgpr0
%val = call float @llvm.experimental.constrained.fadd.f32(float %x, float %y, metadata !"round.tonearest", metadata !"fpexcept.strict")
ret float %val
}
define float @v_constained_fadd_f32_fpexcept_strict_flags(float %x, float %y) #0 {
; CHECK-LABEL: name: v_constained_fadd_f32_fpexcept_strict_flags
; CHECK: bb.1 (%ir-block.0):
; CHECK-NEXT: liveins: $vgpr0, $vgpr1, $sgpr30_sgpr31
; CHECK-NEXT: {{ $}}
; CHECK-NEXT: [[COPY:%[0-9]+]]:_(s32) = COPY $vgpr0
; CHECK-NEXT: [[COPY1:%[0-9]+]]:_(s32) = COPY $vgpr1
; CHECK-NEXT: [[COPY2:%[0-9]+]]:sgpr_64 = COPY $sgpr30_sgpr31
; CHECK-NEXT: [[STRICT_FADD:%[0-9]+]]:_(s32) = nsz G_STRICT_FADD [[COPY]], [[COPY1]]
; CHECK-NEXT: $vgpr0 = COPY [[STRICT_FADD]](s32)
; CHECK-NEXT: [[COPY3:%[0-9]+]]:ccr_sgpr_64 = COPY [[COPY2]]
; CHECK-NEXT: S_SETPC_B64_return [[COPY3]], implicit $vgpr0
%val = call nsz float @llvm.experimental.constrained.fadd.f32(float %x, float %y, metadata !"round.tonearest", metadata !"fpexcept.strict")
ret float %val
}
define float @v_constained_fadd_f32_fpexcept_ignore(float %x, float %y) #0 {
; CHECK-LABEL: name: v_constained_fadd_f32_fpexcept_ignore
; CHECK: bb.1 (%ir-block.0):
; CHECK-NEXT: liveins: $vgpr0, $vgpr1, $sgpr30_sgpr31
; CHECK-NEXT: {{ $}}
; CHECK-NEXT: [[COPY:%[0-9]+]]:_(s32) = COPY $vgpr0
; CHECK-NEXT: [[COPY1:%[0-9]+]]:_(s32) = COPY $vgpr1
; CHECK-NEXT: [[COPY2:%[0-9]+]]:sgpr_64 = COPY $sgpr30_sgpr31
; CHECK-NEXT: %3:_(s32) = nofpexcept G_STRICT_FADD [[COPY]], [[COPY1]]
; CHECK-NEXT: $vgpr0 = COPY %3(s32)
; CHECK-NEXT: [[COPY3:%[0-9]+]]:ccr_sgpr_64 = COPY [[COPY2]]
; CHECK-NEXT: S_SETPC_B64_return [[COPY3]], implicit $vgpr0
%val = call float @llvm.experimental.constrained.fadd.f32(float %x, float %y, metadata !"round.tonearest", metadata !"fpexcept.ignore")
ret float %val
}
define float @v_constained_fadd_f32_fpexcept_ignore_flags(float %x, float %y) #0 {
; CHECK-LABEL: name: v_constained_fadd_f32_fpexcept_ignore_flags
; CHECK: bb.1 (%ir-block.0):
; CHECK-NEXT: liveins: $vgpr0, $vgpr1, $sgpr30_sgpr31
; CHECK-NEXT: {{ $}}
; CHECK-NEXT: [[COPY:%[0-9]+]]:_(s32) = COPY $vgpr0
; CHECK-NEXT: [[COPY1:%[0-9]+]]:_(s32) = COPY $vgpr1
; CHECK-NEXT: [[COPY2:%[0-9]+]]:sgpr_64 = COPY $sgpr30_sgpr31
; CHECK-NEXT: %3:_(s32) = nsz nofpexcept G_STRICT_FADD [[COPY]], [[COPY1]]
; CHECK-NEXT: $vgpr0 = COPY %3(s32)
; CHECK-NEXT: [[COPY3:%[0-9]+]]:ccr_sgpr_64 = COPY [[COPY2]]
; CHECK-NEXT: S_SETPC_B64_return [[COPY3]], implicit $vgpr0
%val = call nsz float @llvm.experimental.constrained.fadd.f32(float %x, float %y, metadata !"round.tonearest", metadata !"fpexcept.ignore")
ret float %val
}
define float @v_constained_fadd_f32_fpexcept_maytrap(float %x, float %y) #0 {
; CHECK-LABEL: name: v_constained_fadd_f32_fpexcept_maytrap
; CHECK: bb.1 (%ir-block.0):
; CHECK-NEXT: liveins: $vgpr0, $vgpr1, $sgpr30_sgpr31
; CHECK-NEXT: {{ $}}
; CHECK-NEXT: [[COPY:%[0-9]+]]:_(s32) = COPY $vgpr0
; CHECK-NEXT: [[COPY1:%[0-9]+]]:_(s32) = COPY $vgpr1
; CHECK-NEXT: [[COPY2:%[0-9]+]]:sgpr_64 = COPY $sgpr30_sgpr31
; CHECK-NEXT: [[STRICT_FADD:%[0-9]+]]:_(s32) = G_STRICT_FADD [[COPY]], [[COPY1]]
; CHECK-NEXT: $vgpr0 = COPY [[STRICT_FADD]](s32)
; CHECK-NEXT: [[COPY3:%[0-9]+]]:ccr_sgpr_64 = COPY [[COPY2]]
; CHECK-NEXT: S_SETPC_B64_return [[COPY3]], implicit $vgpr0
%val = call float @llvm.experimental.constrained.fadd.f32(float %x, float %y, metadata !"round.tonearest", metadata !"fpexcept.maytrap")
ret float %val
}
define <2 x float> @v_constained_fadd_v2f32_fpexcept_strict(<2 x float> %x, <2 x float> %y) #0 {
; CHECK-LABEL: name: v_constained_fadd_v2f32_fpexcept_strict
; CHECK: bb.1 (%ir-block.0):
; CHECK-NEXT: liveins: $vgpr0, $vgpr1, $vgpr2, $vgpr3, $sgpr30_sgpr31
; CHECK-NEXT: {{ $}}
; CHECK-NEXT: [[COPY:%[0-9]+]]:_(s32) = COPY $vgpr0
; CHECK-NEXT: [[COPY1:%[0-9]+]]:_(s32) = COPY $vgpr1
; CHECK-NEXT: [[BUILD_VECTOR:%[0-9]+]]:_(<2 x s32>) = G_BUILD_VECTOR [[COPY]](s32), [[COPY1]](s32)
; CHECK-NEXT: [[COPY2:%[0-9]+]]:_(s32) = COPY $vgpr2
; CHECK-NEXT: [[COPY3:%[0-9]+]]:_(s32) = COPY $vgpr3
; CHECK-NEXT: [[BUILD_VECTOR1:%[0-9]+]]:_(<2 x s32>) = G_BUILD_VECTOR [[COPY2]](s32), [[COPY3]](s32)
; CHECK-NEXT: [[COPY4:%[0-9]+]]:sgpr_64 = COPY $sgpr30_sgpr31
; CHECK-NEXT: [[STRICT_FADD:%[0-9]+]]:_(<2 x s32>) = G_STRICT_FADD [[BUILD_VECTOR]], [[BUILD_VECTOR1]]
; CHECK-NEXT: [[UV:%[0-9]+]]:_(s32), [[UV1:%[0-9]+]]:_(s32) = G_UNMERGE_VALUES [[STRICT_FADD]](<2 x s32>)
; CHECK-NEXT: $vgpr0 = COPY [[UV]](s32)
; CHECK-NEXT: $vgpr1 = COPY [[UV1]](s32)
; CHECK-NEXT: [[COPY5:%[0-9]+]]:ccr_sgpr_64 = COPY [[COPY4]]
; CHECK-NEXT: S_SETPC_B64_return [[COPY5]], implicit $vgpr0, implicit $vgpr1
%val = call <2 x float> @llvm.experimental.constrained.fadd.v2f32(<2 x float> %x, <2 x float> %y, metadata !"round.tonearest", metadata !"fpexcept.strict")
ret <2 x float> %val
}
define <2 x float> @v_constained_fadd_v2f32_fpexcept_ignore(<2 x float> %x, <2 x float> %y) #0 {
; CHECK-LABEL: name: v_constained_fadd_v2f32_fpexcept_ignore
; CHECK: bb.1 (%ir-block.0):
; CHECK-NEXT: liveins: $vgpr0, $vgpr1, $vgpr2, $vgpr3, $sgpr30_sgpr31
; CHECK-NEXT: {{ $}}
; CHECK-NEXT: [[COPY:%[0-9]+]]:_(s32) = COPY $vgpr0
; CHECK-NEXT: [[COPY1:%[0-9]+]]:_(s32) = COPY $vgpr1
; CHECK-NEXT: [[BUILD_VECTOR:%[0-9]+]]:_(<2 x s32>) = G_BUILD_VECTOR [[COPY]](s32), [[COPY1]](s32)
; CHECK-NEXT: [[COPY2:%[0-9]+]]:_(s32) = COPY $vgpr2
; CHECK-NEXT: [[COPY3:%[0-9]+]]:_(s32) = COPY $vgpr3
; CHECK-NEXT: [[BUILD_VECTOR1:%[0-9]+]]:_(<2 x s32>) = G_BUILD_VECTOR [[COPY2]](s32), [[COPY3]](s32)
; CHECK-NEXT: [[COPY4:%[0-9]+]]:sgpr_64 = COPY $sgpr30_sgpr31
; CHECK-NEXT: %7:_(<2 x s32>) = nofpexcept G_STRICT_FADD [[BUILD_VECTOR]], [[BUILD_VECTOR1]]
; CHECK-NEXT: [[UV:%[0-9]+]]:_(s32), [[UV1:%[0-9]+]]:_(s32) = G_UNMERGE_VALUES %7(<2 x s32>)
; CHECK-NEXT: $vgpr0 = COPY [[UV]](s32)
; CHECK-NEXT: $vgpr1 = COPY [[UV1]](s32)
; CHECK-NEXT: [[COPY5:%[0-9]+]]:ccr_sgpr_64 = COPY [[COPY4]]
; CHECK-NEXT: S_SETPC_B64_return [[COPY5]], implicit $vgpr0, implicit $vgpr1
%val = call <2 x float> @llvm.experimental.constrained.fadd.v2f32(<2 x float> %x, <2 x float> %y, metadata !"round.tonearest", metadata !"fpexcept.ignore")
ret <2 x float> %val
}
define <2 x float> @v_constained_fadd_v2f32_fpexcept_maytrap(<2 x float> %x, <2 x float> %y) #0 {
; CHECK-LABEL: name: v_constained_fadd_v2f32_fpexcept_maytrap
; CHECK: bb.1 (%ir-block.0):
; CHECK-NEXT: liveins: $vgpr0, $vgpr1, $vgpr2, $vgpr3, $sgpr30_sgpr31
; CHECK-NEXT: {{ $}}
; CHECK-NEXT: [[COPY:%[0-9]+]]:_(s32) = COPY $vgpr0
; CHECK-NEXT: [[COPY1:%[0-9]+]]:_(s32) = COPY $vgpr1
; CHECK-NEXT: [[BUILD_VECTOR:%[0-9]+]]:_(<2 x s32>) = G_BUILD_VECTOR [[COPY]](s32), [[COPY1]](s32)
; CHECK-NEXT: [[COPY2:%[0-9]+]]:_(s32) = COPY $vgpr2
; CHECK-NEXT: [[COPY3:%[0-9]+]]:_(s32) = COPY $vgpr3
; CHECK-NEXT: [[BUILD_VECTOR1:%[0-9]+]]:_(<2 x s32>) = G_BUILD_VECTOR [[COPY2]](s32), [[COPY3]](s32)
; CHECK-NEXT: [[COPY4:%[0-9]+]]:sgpr_64 = COPY $sgpr30_sgpr31
; CHECK-NEXT: [[STRICT_FADD:%[0-9]+]]:_(<2 x s32>) = G_STRICT_FADD [[BUILD_VECTOR]], [[BUILD_VECTOR1]]
; CHECK-NEXT: [[UV:%[0-9]+]]:_(s32), [[UV1:%[0-9]+]]:_(s32) = G_UNMERGE_VALUES [[STRICT_FADD]](<2 x s32>)
; CHECK-NEXT: $vgpr0 = COPY [[UV]](s32)
; CHECK-NEXT: $vgpr1 = COPY [[UV1]](s32)
; CHECK-NEXT: [[COPY5:%[0-9]+]]:ccr_sgpr_64 = COPY [[COPY4]]
; CHECK-NEXT: S_SETPC_B64_return [[COPY5]], implicit $vgpr0, implicit $vgpr1
%val = call <2 x float> @llvm.experimental.constrained.fadd.v2f32(<2 x float> %x, <2 x float> %y, metadata !"round.tonearest", metadata !"fpexcept.maytrap")
ret <2 x float> %val
}
define float @v_constained_fsub_f32_fpexcept_ignore_flags(float %x, float %y) #0 {
; CHECK-LABEL: name: v_constained_fsub_f32_fpexcept_ignore_flags
; CHECK: bb.1 (%ir-block.0):
; CHECK-NEXT: liveins: $vgpr0, $vgpr1, $sgpr30_sgpr31
; CHECK-NEXT: {{ $}}
; CHECK-NEXT: [[COPY:%[0-9]+]]:_(s32) = COPY $vgpr0
; CHECK-NEXT: [[COPY1:%[0-9]+]]:_(s32) = COPY $vgpr1
; CHECK-NEXT: [[COPY2:%[0-9]+]]:sgpr_64 = COPY $sgpr30_sgpr31
; CHECK-NEXT: %3:_(s32) = nsz nofpexcept G_STRICT_FSUB [[COPY]], [[COPY1]]
; CHECK-NEXT: $vgpr0 = COPY %3(s32)
; CHECK-NEXT: [[COPY3:%[0-9]+]]:ccr_sgpr_64 = COPY [[COPY2]]
; CHECK-NEXT: S_SETPC_B64_return [[COPY3]], implicit $vgpr0
%val = call nsz float @llvm.experimental.constrained.fsub.f32(float %x, float %y, metadata !"round.tonearest", metadata !"fpexcept.ignore")
ret float %val
}
define float @v_constained_fmul_f32_fpexcept_ignore_flags(float %x, float %y) #0 {
; CHECK-LABEL: name: v_constained_fmul_f32_fpexcept_ignore_flags
; CHECK: bb.1 (%ir-block.0):
; CHECK-NEXT: liveins: $vgpr0, $vgpr1, $sgpr30_sgpr31
; CHECK-NEXT: {{ $}}
; CHECK-NEXT: [[COPY:%[0-9]+]]:_(s32) = COPY $vgpr0
; CHECK-NEXT: [[COPY1:%[0-9]+]]:_(s32) = COPY $vgpr1
; CHECK-NEXT: [[COPY2:%[0-9]+]]:sgpr_64 = COPY $sgpr30_sgpr31
; CHECK-NEXT: %3:_(s32) = nsz nofpexcept G_STRICT_FMUL [[COPY]], [[COPY1]]
; CHECK-NEXT: $vgpr0 = COPY %3(s32)
; CHECK-NEXT: [[COPY3:%[0-9]+]]:ccr_sgpr_64 = COPY [[COPY2]]
; CHECK-NEXT: S_SETPC_B64_return [[COPY3]], implicit $vgpr0
%val = call nsz float @llvm.experimental.constrained.fmul.f32(float %x, float %y, metadata !"round.tonearest", metadata !"fpexcept.ignore")
ret float %val
}
define float @v_constained_fdiv_f32_fpexcept_ignore_flags(float %x, float %y) #0 {
; CHECK-LABEL: name: v_constained_fdiv_f32_fpexcept_ignore_flags
; CHECK: bb.1 (%ir-block.0):
; CHECK-NEXT: liveins: $vgpr0, $vgpr1, $sgpr30_sgpr31
; CHECK-NEXT: {{ $}}
; CHECK-NEXT: [[COPY:%[0-9]+]]:_(s32) = COPY $vgpr0
; CHECK-NEXT: [[COPY1:%[0-9]+]]:_(s32) = COPY $vgpr1
; CHECK-NEXT: [[COPY2:%[0-9]+]]:sgpr_64 = COPY $sgpr30_sgpr31
; CHECK-NEXT: %3:_(s32) = nsz nofpexcept G_STRICT_FDIV [[COPY]], [[COPY1]]
; CHECK-NEXT: $vgpr0 = COPY %3(s32)
; CHECK-NEXT: [[COPY3:%[0-9]+]]:ccr_sgpr_64 = COPY [[COPY2]]
; CHECK-NEXT: S_SETPC_B64_return [[COPY3]], implicit $vgpr0
%val = call nsz float @llvm.experimental.constrained.fdiv.f32(float %x, float %y, metadata !"round.tonearest", metadata !"fpexcept.ignore")
ret float %val
}
define float @v_constained_frem_f32_fpexcept_ignore_flags(float %x, float %y) #0 {
; CHECK-LABEL: name: v_constained_frem_f32_fpexcept_ignore_flags
; CHECK: bb.1 (%ir-block.0):
; CHECK-NEXT: liveins: $vgpr0, $vgpr1, $sgpr30_sgpr31
; CHECK-NEXT: {{ $}}
; CHECK-NEXT: [[COPY:%[0-9]+]]:_(s32) = COPY $vgpr0
; CHECK-NEXT: [[COPY1:%[0-9]+]]:_(s32) = COPY $vgpr1
; CHECK-NEXT: [[COPY2:%[0-9]+]]:sgpr_64 = COPY $sgpr30_sgpr31
; CHECK-NEXT: %3:_(s32) = nsz nofpexcept G_STRICT_FREM [[COPY]], [[COPY1]]
; CHECK-NEXT: $vgpr0 = COPY %3(s32)
; CHECK-NEXT: [[COPY3:%[0-9]+]]:ccr_sgpr_64 = COPY [[COPY2]]
; CHECK-NEXT: S_SETPC_B64_return [[COPY3]], implicit $vgpr0
%val = call nsz float @llvm.experimental.constrained.frem.f32(float %x, float %y, metadata !"round.tonearest", metadata !"fpexcept.ignore")
ret float %val
}
define float @v_constained_fma_f32_fpexcept_ignore_flags(float %x, float %y, float %z) #0 {
; CHECK-LABEL: name: v_constained_fma_f32_fpexcept_ignore_flags
; CHECK: bb.1 (%ir-block.0):
; CHECK-NEXT: liveins: $vgpr0, $vgpr1, $vgpr2, $sgpr30_sgpr31
; CHECK-NEXT: {{ $}}
; CHECK-NEXT: [[COPY:%[0-9]+]]:_(s32) = COPY $vgpr0
; CHECK-NEXT: [[COPY1:%[0-9]+]]:_(s32) = COPY $vgpr1
; CHECK-NEXT: [[COPY2:%[0-9]+]]:_(s32) = COPY $vgpr2
; CHECK-NEXT: [[COPY3:%[0-9]+]]:sgpr_64 = COPY $sgpr30_sgpr31
; CHECK-NEXT: %4:_(s32) = nsz nofpexcept G_STRICT_FMA [[COPY]], [[COPY1]], [[COPY2]]
; CHECK-NEXT: $vgpr0 = COPY %4(s32)
; CHECK-NEXT: [[COPY4:%[0-9]+]]:ccr_sgpr_64 = COPY [[COPY3]]
; CHECK-NEXT: S_SETPC_B64_return [[COPY4]], implicit $vgpr0
%val = call nsz float @llvm.experimental.constrained.fma.f32(float %x, float %y, float %z, metadata !"round.tonearest", metadata !"fpexcept.ignore")
ret float %val
}
define float @v_constained_sqrt_f32_fpexcept_strict(float %x) #0 {
; CHECK-LABEL: name: v_constained_sqrt_f32_fpexcept_strict
; CHECK: bb.1 (%ir-block.0):
; CHECK-NEXT: liveins: $vgpr0, $sgpr30_sgpr31
; CHECK-NEXT: {{ $}}
; CHECK-NEXT: [[COPY:%[0-9]+]]:_(s32) = COPY $vgpr0
; CHECK-NEXT: [[COPY1:%[0-9]+]]:sgpr_64 = COPY $sgpr30_sgpr31
; CHECK-NEXT: [[STRICT_FSQRT:%[0-9]+]]:_(s32) = G_STRICT_FSQRT [[COPY]]
; CHECK-NEXT: $vgpr0 = COPY [[STRICT_FSQRT]](s32)
; CHECK-NEXT: [[COPY2:%[0-9]+]]:ccr_sgpr_64 = COPY [[COPY1]]
; CHECK-NEXT: S_SETPC_B64_return [[COPY2]], implicit $vgpr0
%val = call float @llvm.experimental.constrained.sqrt.f32(float %x, metadata !"round.tonearest", metadata !"fpexcept.strict")
ret float %val
}
declare float @llvm.experimental.constrained.fadd.f32(float, float, metadata, metadata) #1
declare <2 x float> @llvm.experimental.constrained.fadd.v2f32(<2 x float>, <2 x float>, metadata, metadata) #1
declare <3 x float> @llvm.experimental.constrained.fadd.v3f32(<3 x float>, <3 x float>, metadata, metadata) #1
declare float @llvm.experimental.constrained.fsub.f32(float, float, metadata, metadata) #1
declare float @llvm.experimental.constrained.fmul.f32(float, float, metadata, metadata) #1
declare float @llvm.experimental.constrained.fdiv.f32(float, float, metadata, metadata) #1
declare float @llvm.experimental.constrained.frem.f32(float, float, metadata, metadata) #1
declare float @llvm.experimental.constrained.fma.f32(float, float, float, metadata, metadata) #1
declare float @llvm.experimental.constrained.sqrt.f32(float, metadata, metadata) #1
attributes #0 = { strictfp }
attributes #1 = { inaccessiblememonly nounwind willreturn }
|
