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
|
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 2
; RUN: llc -mcpu=pwr9 -mtriple=powerpc64le-unknown-unknown \
; RUN: -ppc-vsr-nums-as-vr -ppc-asm-full-reg-names < %s | FileCheck %s
define float @ldexp_f32(i8 zeroext %x) {
; CHECK-LABEL: ldexp_f32:
; CHECK: # %bb.0:
; CHECK-NEXT: mflr r0
; CHECK-NEXT: stdu r1, -32(r1)
; CHECK-NEXT: std r0, 48(r1)
; CHECK-NEXT: .cfi_def_cfa_offset 32
; CHECK-NEXT: .cfi_offset lr, 16
; CHECK-NEXT: vspltisw v2, 1
; CHECK-NEXT: mr r4, r3
; CHECK-NEXT: xvcvsxwdp vs1, v2
; CHECK-NEXT: bl ldexpf
; CHECK-NEXT: nop
; CHECK-NEXT: addi r1, r1, 32
; CHECK-NEXT: ld r0, 16(r1)
; CHECK-NEXT: mtlr r0
; CHECK-NEXT: blr
%zext = zext i8 %x to i32
%ldexp = call float @llvm.ldexp.f32.i32(float 1.000000e+00, i32 %zext)
ret float %ldexp
}
define double @ldexp_f64(i8 zeroext %x) {
; CHECK-LABEL: ldexp_f64:
; CHECK: # %bb.0:
; CHECK-NEXT: mflr r0
; CHECK-NEXT: stdu r1, -32(r1)
; CHECK-NEXT: std r0, 48(r1)
; CHECK-NEXT: .cfi_def_cfa_offset 32
; CHECK-NEXT: .cfi_offset lr, 16
; CHECK-NEXT: vspltisw v2, 1
; CHECK-NEXT: mr r4, r3
; CHECK-NEXT: xvcvsxwdp vs1, v2
; CHECK-NEXT: bl ldexp
; CHECK-NEXT: nop
; CHECK-NEXT: addi r1, r1, 32
; CHECK-NEXT: ld r0, 16(r1)
; CHECK-NEXT: mtlr r0
; CHECK-NEXT: blr
%zext = zext i8 %x to i32
%ldexp = call double @llvm.ldexp.f64.i32(double 1.000000e+00, i32 %zext)
ret double %ldexp
}
define <2 x float> @ldexp_v2f32(<2 x float> %val, <2 x i32> %exp) {
; CHECK-LABEL: ldexp_v2f32:
; CHECK: # %bb.0:
; CHECK-NEXT: mflr r0
; CHECK-NEXT: stdu r1, -80(r1)
; CHECK-NEXT: std r0, 96(r1)
; CHECK-NEXT: .cfi_def_cfa_offset 80
; CHECK-NEXT: .cfi_offset lr, 16
; CHECK-NEXT: .cfi_offset v29, -48
; CHECK-NEXT: .cfi_offset v30, -32
; CHECK-NEXT: .cfi_offset v31, -16
; CHECK-NEXT: li r3, 0
; CHECK-NEXT: xxsldwi vs0, v2, v2, 3
; CHECK-NEXT: stxv v29, 32(r1) # 16-byte Folded Spill
; CHECK-NEXT: xscvspdpn f1, vs0
; CHECK-NEXT: vextuwrx r3, r3, v3
; CHECK-NEXT: stxv v30, 48(r1) # 16-byte Folded Spill
; CHECK-NEXT: stxv v31, 64(r1) # 16-byte Folded Spill
; CHECK-NEXT: extsw r4, r3
; CHECK-NEXT: vmr v31, v3
; CHECK-NEXT: vmr v30, v2
; CHECK-NEXT: bl ldexpf
; CHECK-NEXT: nop
; CHECK-NEXT: li r3, 4
; CHECK-NEXT: xxswapd vs0, v30
; CHECK-NEXT: xscvdpspn v29, f1
; CHECK-NEXT: xscvspdpn f1, vs0
; CHECK-NEXT: vextuwrx r3, r3, v31
; CHECK-NEXT: extsw r4, r3
; CHECK-NEXT: bl ldexpf
; CHECK-NEXT: nop
; CHECK-NEXT: xscvdpspn vs0, f1
; CHECK-NEXT: lxv v31, 64(r1) # 16-byte Folded Reload
; CHECK-NEXT: lxv v30, 48(r1) # 16-byte Folded Reload
; CHECK-NEXT: xxmrghw v2, vs0, v29
; CHECK-NEXT: lxv v29, 32(r1) # 16-byte Folded Reload
; CHECK-NEXT: addi r1, r1, 80
; CHECK-NEXT: ld r0, 16(r1)
; CHECK-NEXT: mtlr r0
; CHECK-NEXT: blr
%1 = call <2 x float> @llvm.ldexp.v2f32.v2i32(<2 x float> %val, <2 x i32> %exp)
ret <2 x float> %1
}
define <4 x float> @ldexp_v4f32(<4 x float> %val, <4 x i32> %exp) {
; CHECK-LABEL: ldexp_v4f32:
; CHECK: # %bb.0:
; CHECK-NEXT: mflr r0
; CHECK-NEXT: stdu r1, -96(r1)
; CHECK-NEXT: std r0, 112(r1)
; CHECK-NEXT: .cfi_def_cfa_offset 96
; CHECK-NEXT: .cfi_offset lr, 16
; CHECK-NEXT: .cfi_offset v28, -64
; CHECK-NEXT: .cfi_offset v29, -48
; CHECK-NEXT: .cfi_offset v30, -32
; CHECK-NEXT: .cfi_offset v31, -16
; CHECK-NEXT: li r3, 4
; CHECK-NEXT: xxswapd vs0, v2
; CHECK-NEXT: stxv v28, 32(r1) # 16-byte Folded Spill
; CHECK-NEXT: xscvspdpn f1, vs0
; CHECK-NEXT: vextuwrx r3, r3, v3
; CHECK-NEXT: stxv v29, 48(r1) # 16-byte Folded Spill
; CHECK-NEXT: stxv v30, 64(r1) # 16-byte Folded Spill
; CHECK-NEXT: stxv v31, 80(r1) # 16-byte Folded Spill
; CHECK-NEXT: vmr v31, v3
; CHECK-NEXT: extsw r4, r3
; CHECK-NEXT: vmr v30, v2
; CHECK-NEXT: bl ldexpf
; CHECK-NEXT: nop
; CHECK-NEXT: li r3, 12
; CHECK-NEXT: xscpsgndp v29, f1, f1
; CHECK-NEXT: xscvspdpn f1, v30
; CHECK-NEXT: vextuwrx r3, r3, v31
; CHECK-NEXT: extsw r4, r3
; CHECK-NEXT: bl ldexpf
; CHECK-NEXT: nop
; CHECK-NEXT: xxmrghd vs0, vs1, v29
; CHECK-NEXT: li r3, 0
; CHECK-NEXT: vextuwrx r3, r3, v31
; CHECK-NEXT: xvcvdpsp v28, vs0
; CHECK-NEXT: xxsldwi vs0, v30, v30, 3
; CHECK-NEXT: extsw r4, r3
; CHECK-NEXT: xscvspdpn f1, vs0
; CHECK-NEXT: bl ldexpf
; CHECK-NEXT: nop
; CHECK-NEXT: xxsldwi vs0, v30, v30, 1
; CHECK-NEXT: mfvsrwz r3, v31
; CHECK-NEXT: xscpsgndp v29, f1, f1
; CHECK-NEXT: extsw r4, r3
; CHECK-NEXT: xscvspdpn f1, vs0
; CHECK-NEXT: bl ldexpf
; CHECK-NEXT: nop
; CHECK-NEXT: xxmrghd vs0, vs1, v29
; CHECK-NEXT: lxv v31, 80(r1) # 16-byte Folded Reload
; CHECK-NEXT: lxv v30, 64(r1) # 16-byte Folded Reload
; CHECK-NEXT: lxv v29, 48(r1) # 16-byte Folded Reload
; CHECK-NEXT: xvcvdpsp v2, vs0
; CHECK-NEXT: vmrgew v2, v28, v2
; CHECK-NEXT: lxv v28, 32(r1) # 16-byte Folded Reload
; CHECK-NEXT: addi r1, r1, 96
; CHECK-NEXT: ld r0, 16(r1)
; CHECK-NEXT: mtlr r0
; CHECK-NEXT: blr
%1 = call <4 x float> @llvm.ldexp.v4f32.v4i32(<4 x float> %val, <4 x i32> %exp)
ret <4 x float> %1
}
define half @ldexp_f16(half %arg0, i32 %arg1) {
; CHECK-LABEL: ldexp_f16:
; CHECK: # %bb.0:
; CHECK-NEXT: mflr r0
; CHECK-NEXT: stdu r1, -32(r1)
; CHECK-NEXT: std r0, 48(r1)
; CHECK-NEXT: .cfi_def_cfa_offset 32
; CHECK-NEXT: .cfi_offset lr, 16
; CHECK-NEXT: xscvdphp f0, f1
; CHECK-NEXT: extsw r4, r4
; CHECK-NEXT: mffprwz r3, f0
; CHECK-NEXT: clrlwi r3, r3, 16
; CHECK-NEXT: mtfprwz f0, r3
; CHECK-NEXT: xscvhpdp f1, f0
; CHECK-NEXT: bl ldexpf
; CHECK-NEXT: nop
; CHECK-NEXT: addi r1, r1, 32
; CHECK-NEXT: ld r0, 16(r1)
; CHECK-NEXT: mtlr r0
; CHECK-NEXT: blr
%ldexp = call half @llvm.ldexp.f16.i32(half %arg0, i32 %arg1)
ret half %ldexp
}
define ppc_fp128 @ldexp_fp128(ppc_fp128 %arg0, i32 %arg1) {
; CHECK-LABEL: ldexp_fp128:
; CHECK: # %bb.0:
; CHECK-NEXT: mflr r0
; CHECK-NEXT: stdu r1, -32(r1)
; CHECK-NEXT: std r0, 48(r1)
; CHECK-NEXT: .cfi_def_cfa_offset 32
; CHECK-NEXT: .cfi_offset lr, 16
; CHECK-NEXT: clrldi r5, r5, 32
; CHECK-NEXT: bl ldexpl
; CHECK-NEXT: nop
; CHECK-NEXT: addi r1, r1, 32
; CHECK-NEXT: ld r0, 16(r1)
; CHECK-NEXT: mtlr r0
; CHECK-NEXT: blr
%ldexp = call ppc_fp128 @llvm.ldexp.ppcf128.i32(ppc_fp128 %arg0, i32 %arg1)
ret ppc_fp128 %ldexp
}
declare double @llvm.ldexp.f64.i32(double, i32) #0
declare float @llvm.ldexp.f32.i32(float, i32) #0
declare <2 x float> @llvm.ldexp.v2f32.v2i32(<2 x float>, <2 x i32>) #0
declare <4 x float> @llvm.ldexp.v4f32.v4i32(<4 x float>, <4 x i32>) #0
declare half @llvm.ldexp.f16.i32(half, i32) #0
declare ppc_fp128 @llvm.ldexp.ppcf128.i32(ppc_fp128, i32) #0
attributes #0 = { nocallback nofree nosync nounwind speculatable willreturn memory(none) }
|
