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
|
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
; RUN: llc -mtriple=riscv32 -mattr=+d,+zfh,+experimental-zvfh,+v -verify-machineinstrs < %s | FileCheck %s
; RUN: llc -mtriple=riscv64 -mattr=+d,+zfh,+experimental-zvfh,+v -verify-machineinstrs < %s | FileCheck %s
declare <vscale x 2 x half> @llvm.vp.fptrunc.nxv2f16.nxv2f32(<vscale x 2 x float>, <vscale x 2 x i1>, i32)
define <vscale x 2 x half> @vfptrunc_nxv2f16_nxv2f32(<vscale x 2 x float> %a, <vscale x 2 x i1> %m, i32 zeroext %vl) {
; CHECK-LABEL: vfptrunc_nxv2f16_nxv2f32:
; CHECK: # %bb.0:
; CHECK-NEXT: vsetvli zero, a0, e16, mf2, ta, mu
; CHECK-NEXT: vfncvt.f.f.w v9, v8, v0.t
; CHECK-NEXT: vmv1r.v v8, v9
; CHECK-NEXT: ret
%v = call <vscale x 2 x half> @llvm.vp.fptrunc.nxv2f16.nxv2f32(<vscale x 2 x float> %a, <vscale x 2 x i1> %m, i32 %vl)
ret <vscale x 2 x half> %v
}
define <vscale x 2 x half> @vfptrunc_nxv2f16_nxv2f32_unmasked(<vscale x 2 x float> %a, i32 zeroext %vl) {
; CHECK-LABEL: vfptrunc_nxv2f16_nxv2f32_unmasked:
; CHECK: # %bb.0:
; CHECK-NEXT: vsetvli zero, a0, e16, mf2, ta, mu
; CHECK-NEXT: vfncvt.f.f.w v9, v8
; CHECK-NEXT: vmv1r.v v8, v9
; CHECK-NEXT: ret
%v = call <vscale x 2 x half> @llvm.vp.fptrunc.nxv2f16.nxv2f32(<vscale x 2 x float> %a, <vscale x 2 x i1> shufflevector (<vscale x 2 x i1> insertelement (<vscale x 2 x i1> undef, i1 true, i32 0), <vscale x 2 x i1> undef, <vscale x 2 x i32> zeroinitializer), i32 %vl)
ret <vscale x 2 x half> %v
}
declare <vscale x 2 x half> @llvm.vp.fptrunc.nxv2f16.nxv2f64(<vscale x 2 x double>, <vscale x 2 x i1>, i32)
define <vscale x 2 x half> @vfptrunc_nxv2f16_nxv2f64(<vscale x 2 x double> %a, <vscale x 2 x i1> %m, i32 zeroext %vl) {
; CHECK-LABEL: vfptrunc_nxv2f16_nxv2f64:
; CHECK: # %bb.0:
; CHECK-NEXT: vsetvli zero, a0, e32, m1, ta, mu
; CHECK-NEXT: vfncvt.rod.f.f.w v10, v8, v0.t
; CHECK-NEXT: vsetvli zero, zero, e16, mf2, ta, mu
; CHECK-NEXT: vfncvt.f.f.w v8, v10, v0.t
; CHECK-NEXT: ret
%v = call <vscale x 2 x half> @llvm.vp.fptrunc.nxv2f16.nxv2f64(<vscale x 2 x double> %a, <vscale x 2 x i1> %m, i32 %vl)
ret <vscale x 2 x half> %v
}
define <vscale x 2 x half> @vfptrunc_nxv2f16_nxv2f64_unmasked(<vscale x 2 x double> %a, i32 zeroext %vl) {
; CHECK-LABEL: vfptrunc_nxv2f16_nxv2f64_unmasked:
; CHECK: # %bb.0:
; CHECK-NEXT: vsetvli zero, a0, e32, m1, ta, mu
; CHECK-NEXT: vfncvt.rod.f.f.w v10, v8
; CHECK-NEXT: vsetvli zero, zero, e16, mf2, ta, mu
; CHECK-NEXT: vfncvt.f.f.w v8, v10
; CHECK-NEXT: ret
%v = call <vscale x 2 x half> @llvm.vp.fptrunc.nxv2f16.nxv2f64(<vscale x 2 x double> %a, <vscale x 2 x i1> shufflevector (<vscale x 2 x i1> insertelement (<vscale x 2 x i1> undef, i1 true, i32 0), <vscale x 2 x i1> undef, <vscale x 2 x i32> zeroinitializer), i32 %vl)
ret <vscale x 2 x half> %v
}
declare <vscale x 2 x float> @llvm.vp.fptrunc.nxv2f64.nxv2f32(<vscale x 2 x double>, <vscale x 2 x i1>, i32)
define <vscale x 2 x float> @vfptrunc_nxv2f32_nxv2f64(<vscale x 2 x double> %a, <vscale x 2 x i1> %m, i32 zeroext %vl) {
; CHECK-LABEL: vfptrunc_nxv2f32_nxv2f64:
; CHECK: # %bb.0:
; CHECK-NEXT: vsetvli zero, a0, e32, m1, ta, mu
; CHECK-NEXT: vfncvt.f.f.w v10, v8, v0.t
; CHECK-NEXT: vmv.v.v v8, v10
; CHECK-NEXT: ret
%v = call <vscale x 2 x float> @llvm.vp.fptrunc.nxv2f64.nxv2f32(<vscale x 2 x double> %a, <vscale x 2 x i1> %m, i32 %vl)
ret <vscale x 2 x float> %v
}
define <vscale x 2 x float> @vfptrunc_nxv2f32_nxv2f64_unmasked(<vscale x 2 x double> %a, i32 zeroext %vl) {
; CHECK-LABEL: vfptrunc_nxv2f32_nxv2f64_unmasked:
; CHECK: # %bb.0:
; CHECK-NEXT: vsetvli zero, a0, e32, m1, ta, mu
; CHECK-NEXT: vfncvt.f.f.w v10, v8
; CHECK-NEXT: vmv.v.v v8, v10
; CHECK-NEXT: ret
%v = call <vscale x 2 x float> @llvm.vp.fptrunc.nxv2f64.nxv2f32(<vscale x 2 x double> %a, <vscale x 2 x i1> shufflevector (<vscale x 2 x i1> insertelement (<vscale x 2 x i1> undef, i1 true, i32 0), <vscale x 2 x i1> undef, <vscale x 2 x i32> zeroinitializer), i32 %vl)
ret <vscale x 2 x float> %v
}
declare <vscale x 7 x float> @llvm.vp.fptrunc.nxv7f64.nxv7f32(<vscale x 7 x double>, <vscale x 7 x i1>, i32)
define <vscale x 7 x float> @vfptrunc_nxv7f32_nxv7f64(<vscale x 7 x double> %a, <vscale x 7 x i1> %m, i32 zeroext %vl) {
; CHECK-LABEL: vfptrunc_nxv7f32_nxv7f64:
; CHECK: # %bb.0:
; CHECK-NEXT: vsetvli zero, a0, e32, m4, ta, mu
; CHECK-NEXT: vfncvt.f.f.w v16, v8, v0.t
; CHECK-NEXT: vmv.v.v v8, v16
; CHECK-NEXT: ret
%v = call <vscale x 7 x float> @llvm.vp.fptrunc.nxv7f64.nxv7f32(<vscale x 7 x double> %a, <vscale x 7 x i1> %m, i32 %vl)
ret <vscale x 7 x float> %v
}
declare <vscale x 16 x float> @llvm.vp.fptrunc.nxv16f64.nxv16f32(<vscale x 16 x double>, <vscale x 16 x i1>, i32)
define <vscale x 16 x float> @vfptrunc_nxv16f32_nxv16f64(<vscale x 16 x double> %a, <vscale x 16 x i1> %m, i32 zeroext %vl) {
; CHECK-LABEL: vfptrunc_nxv16f32_nxv16f64:
; CHECK: # %bb.0:
; CHECK-NEXT: addi sp, sp, -16
; CHECK-NEXT: .cfi_def_cfa_offset 16
; CHECK-NEXT: csrr a1, vlenb
; CHECK-NEXT: slli a1, a1, 3
; CHECK-NEXT: sub sp, sp, a1
; CHECK-NEXT: vmv1r.v v24, v0
; CHECK-NEXT: addi a1, sp, 16
; CHECK-NEXT: vs8r.v v8, (a1) # Unknown-size Folded Spill
; CHECK-NEXT: li a2, 0
; CHECK-NEXT: csrr a1, vlenb
; CHECK-NEXT: srli a4, a1, 3
; CHECK-NEXT: vsetvli a3, zero, e8, mf4, ta, mu
; CHECK-NEXT: sub a3, a0, a1
; CHECK-NEXT: vslidedown.vx v0, v0, a4
; CHECK-NEXT: bltu a0, a3, .LBB7_2
; CHECK-NEXT: # %bb.1:
; CHECK-NEXT: mv a2, a3
; CHECK-NEXT: .LBB7_2:
; CHECK-NEXT: vsetvli zero, a2, e32, m4, ta, mu
; CHECK-NEXT: vfncvt.f.f.w v12, v16, v0.t
; CHECK-NEXT: bltu a0, a1, .LBB7_4
; CHECK-NEXT: # %bb.3:
; CHECK-NEXT: mv a0, a1
; CHECK-NEXT: .LBB7_4:
; CHECK-NEXT: vsetvli zero, a0, e32, m4, ta, mu
; CHECK-NEXT: vmv1r.v v0, v24
; CHECK-NEXT: addi a0, sp, 16
; CHECK-NEXT: vl8re8.v v16, (a0) # Unknown-size Folded Reload
; CHECK-NEXT: vfncvt.f.f.w v8, v16, v0.t
; CHECK-NEXT: csrr a0, vlenb
; CHECK-NEXT: slli a0, a0, 3
; CHECK-NEXT: add sp, sp, a0
; CHECK-NEXT: addi sp, sp, 16
; CHECK-NEXT: ret
%v = call <vscale x 16 x float> @llvm.vp.fptrunc.nxv16f64.nxv16f32(<vscale x 16 x double> %a, <vscale x 16 x i1> %m, i32 %vl)
ret <vscale x 16 x float> %v
}
declare <vscale x 32 x float> @llvm.vp.fptrunc.nxv32f64.nxv32f32(<vscale x 32 x double>, <vscale x 32 x i1>, i32)
define <vscale x 32 x float> @vfptrunc_nxv32f32_nxv32f64(<vscale x 32 x double> %a, <vscale x 32 x i1> %m, i32 zeroext %vl) {
; CHECK-LABEL: vfptrunc_nxv32f32_nxv32f64:
; CHECK: # %bb.0:
; CHECK-NEXT: addi sp, sp, -16
; CHECK-NEXT: .cfi_def_cfa_offset 16
; CHECK-NEXT: csrr a1, vlenb
; CHECK-NEXT: slli a1, a1, 4
; CHECK-NEXT: sub sp, sp, a1
; CHECK-NEXT: vmv1r.v v24, v0
; CHECK-NEXT: csrr a1, vlenb
; CHECK-NEXT: slli a1, a1, 3
; CHECK-NEXT: add a1, sp, a1
; CHECK-NEXT: addi a1, a1, 16
; CHECK-NEXT: vs8r.v v8, (a1) # Unknown-size Folded Spill
; CHECK-NEXT: csrr a1, vlenb
; CHECK-NEXT: slli a4, a1, 1
; CHECK-NEXT: srli a3, a1, 3
; CHECK-NEXT: mv a5, a2
; CHECK-NEXT: bltu a2, a4, .LBB8_2
; CHECK-NEXT: # %bb.1:
; CHECK-NEXT: mv a5, a4
; CHECK-NEXT: .LBB8_2:
; CHECK-NEXT: li a6, 0
; CHECK-NEXT: vsetvli a7, zero, e8, mf4, ta, mu
; CHECK-NEXT: sub a7, a5, a1
; CHECK-NEXT: vslidedown.vx v0, v24, a3
; CHECK-NEXT: bltu a5, a7, .LBB8_4
; CHECK-NEXT: # %bb.3:
; CHECK-NEXT: mv a6, a7
; CHECK-NEXT: .LBB8_4:
; CHECK-NEXT: srli a7, a1, 2
; CHECK-NEXT: slli t0, a1, 3
; CHECK-NEXT: vsetvli zero, a6, e32, m4, ta, mu
; CHECK-NEXT: vfncvt.f.f.w v12, v16, v0.t
; CHECK-NEXT: bltu a5, a1, .LBB8_6
; CHECK-NEXT: # %bb.5:
; CHECK-NEXT: mv a5, a1
; CHECK-NEXT: .LBB8_6:
; CHECK-NEXT: li a6, 0
; CHECK-NEXT: vsetvli t1, zero, e8, mf2, ta, mu
; CHECK-NEXT: vslidedown.vx v1, v24, a7
; CHECK-NEXT: add a7, a0, t0
; CHECK-NEXT: vsetvli zero, a5, e32, m4, ta, mu
; CHECK-NEXT: sub a4, a2, a4
; CHECK-NEXT: vmv1r.v v0, v24
; CHECK-NEXT: csrr a5, vlenb
; CHECK-NEXT: slli a5, a5, 3
; CHECK-NEXT: add a5, sp, a5
; CHECK-NEXT: addi a5, a5, 16
; CHECK-NEXT: vl8re8.v v16, (a5) # Unknown-size Folded Reload
; CHECK-NEXT: vfncvt.f.f.w v8, v16, v0.t
; CHECK-NEXT: bltu a2, a4, .LBB8_8
; CHECK-NEXT: # %bb.7:
; CHECK-NEXT: mv a6, a4
; CHECK-NEXT: .LBB8_8:
; CHECK-NEXT: vsetvli a2, zero, e8, mf4, ta, mu
; CHECK-NEXT: vl8re64.v v16, (a7)
; CHECK-NEXT: csrr a2, vlenb
; CHECK-NEXT: slli a2, a2, 3
; CHECK-NEXT: add a2, sp, a2
; CHECK-NEXT: addi a2, a2, 16
; CHECK-NEXT: vs8r.v v16, (a2) # Unknown-size Folded Spill
; CHECK-NEXT: li a2, 0
; CHECK-NEXT: sub a4, a6, a1
; CHECK-NEXT: vslidedown.vx v0, v1, a3
; CHECK-NEXT: bltu a6, a4, .LBB8_10
; CHECK-NEXT: # %bb.9:
; CHECK-NEXT: mv a2, a4
; CHECK-NEXT: .LBB8_10:
; CHECK-NEXT: vl8re64.v v16, (a0)
; CHECK-NEXT: addi a0, sp, 16
; CHECK-NEXT: vs8r.v v16, (a0) # Unknown-size Folded Spill
; CHECK-NEXT: vsetvli zero, a2, e32, m4, ta, mu
; CHECK-NEXT: csrr a0, vlenb
; CHECK-NEXT: slli a0, a0, 3
; CHECK-NEXT: add a0, sp, a0
; CHECK-NEXT: addi a0, a0, 16
; CHECK-NEXT: vl8re8.v v24, (a0) # Unknown-size Folded Reload
; CHECK-NEXT: vfncvt.f.f.w v20, v24, v0.t
; CHECK-NEXT: bltu a6, a1, .LBB8_12
; CHECK-NEXT: # %bb.11:
; CHECK-NEXT: mv a6, a1
; CHECK-NEXT: .LBB8_12:
; CHECK-NEXT: vsetvli zero, a6, e32, m4, ta, mu
; CHECK-NEXT: vmv1r.v v0, v1
; CHECK-NEXT: addi a0, sp, 16
; CHECK-NEXT: vl8re8.v v24, (a0) # Unknown-size Folded Reload
; CHECK-NEXT: vfncvt.f.f.w v16, v24, v0.t
; CHECK-NEXT: csrr a0, vlenb
; CHECK-NEXT: slli a0, a0, 4
; CHECK-NEXT: add sp, sp, a0
; CHECK-NEXT: addi sp, sp, 16
; CHECK-NEXT: ret
%v = call <vscale x 32 x float> @llvm.vp.fptrunc.nxv32f64.nxv32f32(<vscale x 32 x double> %a, <vscale x 32 x i1> %m, i32 %vl)
ret <vscale x 32 x float> %v
}
|
