mve-scatter-increment.ll
10.1 KB
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
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
; RUN: llc -mtriple=thumbv8.1m.main-none-none-eabi -mattr=+mve.fp -enable-arm-maskedldst %s -o - | FileCheck %s
define arm_aapcs_vfpcc void @scatter_inc_minipred_4i32(<4 x i32> %data, i32* %dst, <4 x i32> %offs) {
; CHECK-LABEL: scatter_inc_minipred_4i32:
; CHECK: @ %bb.0:
; CHECK-NEXT: movw r1, #3855
; CHECK-NEXT: vmov.i32 q2, #0x4
; CHECK-NEXT: vadd.i32 q1, q1, q2
; CHECK-NEXT: vmsr p0, r1
; CHECK-NEXT: vpst
; CHECK-NEXT: vstrwt.32 q0, [r0, q1, uxtw #2]
; CHECK-NEXT: bx lr
%1 = add <4 x i32> %offs, <i32 4, i32 4, i32 4, i32 4>
%2 = getelementptr inbounds i32, i32* %dst, <4 x i32> %1
call void @llvm.masked.scatter.v4i32.v4p0i32(<4 x i32> %data, <4 x i32*> %2, i32 4, <4 x i1> <i1 true, i1 false, i1 true, i1 false>)
ret void
}
define arm_aapcs_vfpcc void @scatter_inc_mini_8i16(<8 x i16> %data, i16* %dst, <8 x i32> %offs) {
; CHECK-LABEL: scatter_inc_mini_8i16:
; CHECK: @ %bb.0:
; CHECK-NEXT: .vsave {d8, d9}
; CHECK-NEXT: vpush {d8, d9}
; CHECK-NEXT: vshl.i32 q1, q1, #1
; CHECK-NEXT: vmov.i32 q3, #0x10
; CHECK-NEXT: vadd.i32 q1, q1, r0
; CHECK-NEXT: vmov.u16 r2, q0[0]
; CHECK-NEXT: vadd.i32 q4, q1, q3
; CHECK-NEXT: vshl.i32 q1, q2, #1
; CHECK-NEXT: vmov r1, s16
; CHECK-NEXT: vadd.i32 q1, q1, r0
; CHECK-NEXT: vmov r0, s17
; CHECK-NEXT: vadd.i32 q1, q1, q3
; CHECK-NEXT: strh r2, [r1]
; CHECK-NEXT: vmov.u16 r1, q0[1]
; CHECK-NEXT: strh r1, [r0]
; CHECK-NEXT: vmov r0, s18
; CHECK-NEXT: vmov.u16 r1, q0[2]
; CHECK-NEXT: strh r1, [r0]
; CHECK-NEXT: vmov r0, s19
; CHECK-NEXT: vmov.u16 r1, q0[3]
; CHECK-NEXT: strh r1, [r0]
; CHECK-NEXT: vmov r0, s4
; CHECK-NEXT: vmov.u16 r1, q0[4]
; CHECK-NEXT: strh r1, [r0]
; CHECK-NEXT: vmov r0, s5
; CHECK-NEXT: vmov.u16 r1, q0[5]
; CHECK-NEXT: strh r1, [r0]
; CHECK-NEXT: vmov r0, s6
; CHECK-NEXT: vmov.u16 r1, q0[6]
; CHECK-NEXT: strh r1, [r0]
; CHECK-NEXT: vmov r0, s7
; CHECK-NEXT: vmov.u16 r1, q0[7]
; CHECK-NEXT: strh r1, [r0]
; CHECK-NEXT: vpop {d8, d9}
; CHECK-NEXT: bx lr
%1 = add <8 x i32> %offs, <i32 8, i32 8, i32 8, i32 8, i32 8, i32 8, i32 8, i32 8>
%2 = getelementptr inbounds i16, i16* %dst, <8 x i32> %1
call void @llvm.masked.scatter.v8i16.v8p0i16(<8 x i16> %data, <8 x i16*> %2, i32 4, <8 x i1> <i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true>)
ret void
}
define arm_aapcs_vfpcc void @scatter_inc_mini_16i8(<16 x i8> %data, i8* %dst, <16 x i32> %offs) {
; CHECK-LABEL: scatter_inc_mini_16i8:
; CHECK: @ %bb.0:
; CHECK-NEXT: .vsave {d8, d9, d10, d11}
; CHECK-NEXT: vpush {d8, d9, d10, d11}
; CHECK-NEXT: vmov.i32 q5, #0x10
; CHECK-NEXT: vadd.i32 q1, q1, r0
; CHECK-NEXT: vadd.i32 q4, q1, q5
; CHECK-NEXT: vmov.u8 r2, q0[0]
; CHECK-NEXT: vmov r1, s16
; CHECK-NEXT: vadd.i32 q3, q3, r0
; CHECK-NEXT: vadd.i32 q2, q2, r0
; CHECK-NEXT: vadd.i32 q3, q3, q5
; CHECK-NEXT: vadd.i32 q2, q2, q5
; CHECK-NEXT: strb r2, [r1]
; CHECK-NEXT: add r1, sp, #32
; CHECK-NEXT: vldrw.u32 q1, [r1]
; CHECK-NEXT: vmov.u8 r1, q0[1]
; CHECK-NEXT: vadd.i32 q1, q1, r0
; CHECK-NEXT: vmov r0, s17
; CHECK-NEXT: vadd.i32 q1, q1, q5
; CHECK-NEXT: strb r1, [r0]
; CHECK-NEXT: vmov r0, s18
; CHECK-NEXT: vmov.u8 r1, q0[2]
; CHECK-NEXT: strb r1, [r0]
; CHECK-NEXT: vmov r0, s19
; CHECK-NEXT: vmov.u8 r1, q0[3]
; CHECK-NEXT: strb r1, [r0]
; CHECK-NEXT: vmov r0, s8
; CHECK-NEXT: vmov.u8 r1, q0[4]
; CHECK-NEXT: strb r1, [r0]
; CHECK-NEXT: vmov r0, s9
; CHECK-NEXT: vmov.u8 r1, q0[5]
; CHECK-NEXT: strb r1, [r0]
; CHECK-NEXT: vmov r0, s10
; CHECK-NEXT: vmov.u8 r1, q0[6]
; CHECK-NEXT: strb r1, [r0]
; CHECK-NEXT: vmov r0, s11
; CHECK-NEXT: vmov.u8 r1, q0[7]
; CHECK-NEXT: strb r1, [r0]
; CHECK-NEXT: vmov r0, s12
; CHECK-NEXT: vmov.u8 r1, q0[8]
; CHECK-NEXT: strb r1, [r0]
; CHECK-NEXT: vmov r0, s13
; CHECK-NEXT: vmov.u8 r1, q0[9]
; CHECK-NEXT: strb r1, [r0]
; CHECK-NEXT: vmov r0, s14
; CHECK-NEXT: vmov.u8 r1, q0[10]
; CHECK-NEXT: strb r1, [r0]
; CHECK-NEXT: vmov r0, s15
; CHECK-NEXT: vmov.u8 r1, q0[11]
; CHECK-NEXT: strb r1, [r0]
; CHECK-NEXT: vmov r0, s4
; CHECK-NEXT: vmov.u8 r1, q0[12]
; CHECK-NEXT: strb r1, [r0]
; CHECK-NEXT: vmov r0, s5
; CHECK-NEXT: vmov.u8 r1, q0[13]
; CHECK-NEXT: strb r1, [r0]
; CHECK-NEXT: vmov r0, s6
; CHECK-NEXT: vmov.u8 r1, q0[14]
; CHECK-NEXT: strb r1, [r0]
; CHECK-NEXT: vmov r0, s7
; CHECK-NEXT: vmov.u8 r1, q0[15]
; CHECK-NEXT: strb r1, [r0]
; CHECK-NEXT: vpop {d8, d9, d10, d11}
; CHECK-NEXT: bx lr
%1 = add <16 x i32> %offs, <i32 16, i32 16, i32 16, i32 16, i32 16, i32 16, i32 16, i32 16, i32 16, i32 16, i32 16, i32 16, i32 16, i32 16, i32 16, i32 16>
%2 = getelementptr inbounds i8, i8* %dst, <16 x i32> %1
call void @llvm.masked.scatter.v16i8.v16p0i8(<16 x i8> %data, <16 x i8*> %2, i32 2, <16 x i1> <i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true>)
ret void
}
define arm_aapcs_vfpcc void @scatter_inc_v4i32_complex(<4 x i32> %data1, <4 x i32> %data2, <4 x i32> %data3, i32* %dst, i32 %n) {
; CHECK-LABEL: scatter_inc_v4i32_complex:
; CHECK: @ %bb.0: @ %entry
; CHECK-NEXT: .save {r4, lr}
; CHECK-NEXT: push {r4, lr}
; CHECK-NEXT: .vsave {d8, d9, d10, d11, d12, d13, d14, d15}
; CHECK-NEXT: vpush {d8, d9, d10, d11, d12, d13, d14, d15}
; CHECK-NEXT: .pad #24
; CHECK-NEXT: sub sp, #24
; CHECK-NEXT: cmp r1, #1
; CHECK-NEXT: blt .LBB3_5
; CHECK-NEXT: @ %bb.1: @ %vector.ph.preheader
; CHECK-NEXT: adr r4, .LCPI3_2
; CHECK-NEXT: bic r2, r1, #3
; CHECK-NEXT: vldrw.u32 q3, [r4]
; CHECK-NEXT: sub.w r12, r2, #4
; CHECK-NEXT: adr.w lr, .LCPI3_1
; CHECK-NEXT: movs r3, #1
; CHECK-NEXT: vadd.i32 q3, q3, r0
; CHECK-NEXT: add.w r3, r3, r12, lsr #2
; CHECK-NEXT: vstrw.32 q3, [sp] @ 16-byte Spill
; CHECK-NEXT: vldrw.u32 q3, [lr]
; CHECK-NEXT: adr.w r12, .LCPI3_0
; CHECK-NEXT: vadd.i32 q4, q3, r0
; CHECK-NEXT: vldrw.u32 q3, [r12]
; CHECK-NEXT: vadd.i32 q3, q3, r0
; CHECK-NEXT: .LBB3_2: @ %vector.ph
; CHECK-NEXT: @ =>This Loop Header: Depth=1
; CHECK-NEXT: @ Child Loop BB3_3 Depth 2
; CHECK-NEXT: vmov q6, q4
; CHECK-NEXT: vldrw.u32 q7, [sp] @ 16-byte Reload
; CHECK-NEXT: vmov q5, q3
; CHECK-NEXT: dls lr, r3
; CHECK-NEXT: .LBB3_3: @ %vector.body
; CHECK-NEXT: @ Parent Loop BB3_2 Depth=1
; CHECK-NEXT: @ => This Inner Loop Header: Depth=2
; CHECK-NEXT: vstrw.32 q0, [q5, #48]!
; CHECK-NEXT: vstrw.32 q1, [q6, #48]!
; CHECK-NEXT: vstrw.32 q2, [q7, #48]!
; CHECK-NEXT: le lr, .LBB3_3
; CHECK-NEXT: @ %bb.4: @ %middle.block
; CHECK-NEXT: @ in Loop: Header=BB3_2 Depth=1
; CHECK-NEXT: cmp r2, r1
; CHECK-NEXT: bne .LBB3_2
; CHECK-NEXT: .LBB3_5: @ %for.cond.cleanup
; CHECK-NEXT: add sp, #24
; CHECK-NEXT: vpop {d8, d9, d10, d11, d12, d13, d14, d15}
; CHECK-NEXT: pop {r4, pc}
; CHECK-NEXT: .p2align 4
; CHECK-NEXT: @ %bb.6:
; CHECK-NEXT: .LCPI3_0:
; CHECK-NEXT: .long 4294967248 @ 0xffffffd0
; CHECK-NEXT: .long 4294967260 @ 0xffffffdc
; CHECK-NEXT: .long 4294967272 @ 0xffffffe8
; CHECK-NEXT: .long 4294967284 @ 0xfffffff4
; CHECK-NEXT: .LCPI3_1:
; CHECK-NEXT: .long 4294967252 @ 0xffffffd4
; CHECK-NEXT: .long 4294967264 @ 0xffffffe0
; CHECK-NEXT: .long 4294967276 @ 0xffffffec
; CHECK-NEXT: .long 4294967288 @ 0xfffffff8
; CHECK-NEXT: .LCPI3_2:
; CHECK-NEXT: .long 4294967256 @ 0xffffffd8
; CHECK-NEXT: .long 4294967268 @ 0xffffffe4
; CHECK-NEXT: .long 4294967280 @ 0xfffffff0
; CHECK-NEXT: .long 4294967292 @ 0xfffffffc
entry:
%cmp22 = icmp sgt i32 %n, 0
br i1 %cmp22, label %vector.ph, label %for.cond.cleanup
vector.ph: ; preds = %for.body.preheader
%n.vec = and i32 %n, -4
br label %vector.body
vector.body: ; preds = %vector.body, %vector.ph
%index = phi i32 [ 0, %vector.ph ], [ %index.next, %vector.body ]
%vec.ind = phi <4 x i32> [ <i32 0, i32 1, i32 2, i32 3>, %vector.ph ], [ %vec.ind.next, %vector.body ]
%0 = mul nuw nsw <4 x i32> %vec.ind, <i32 3, i32 3, i32 3, i32 3>
%1 = getelementptr inbounds i32, i32* %dst, <4 x i32> %0
call void @llvm.masked.scatter.v4i32.v4p0i32(<4 x i32> %data1, <4 x i32*> %1, i32 4, <4 x i1> <i1 true, i1 true, i1 true, i1 true>)
%2 = add nuw nsw <4 x i32> %0, <i32 1, i32 1, i32 1, i32 1>
%3 = getelementptr inbounds i32, i32* %dst, <4 x i32> %2
call void @llvm.masked.scatter.v4i32.v4p0i32(<4 x i32> %data2, <4 x i32*> %3, i32 4, <4 x i1> <i1 true, i1 true, i1 true, i1 true>)
%4 = add nuw nsw <4 x i32> %0, <i32 2, i32 2, i32 2, i32 2>
%5 = getelementptr inbounds i32, i32* %dst, <4 x i32> %4
call void @llvm.masked.scatter.v4i32.v4p0i32(<4 x i32> %data3, <4 x i32*> %5, i32 4, <4 x i1> <i1 true, i1 true, i1 true, i1 true>)
%index.next = add i32 %index, 4
%vec.ind.next = add <4 x i32> %vec.ind, <i32 4, i32 4, i32 4, i32 4>
%6 = icmp eq i32 %index.next, %n.vec
br i1 %6, label %middle.block, label %vector.body
middle.block: ; preds = %vector.body
%cmp.n = icmp eq i32 %n.vec, %n
br i1 %cmp.n, label %for.cond.cleanup, label %vector.ph
for.cond.cleanup: ; preds = %for.body, %middle.block, %entry
ret void
}
declare void @llvm.masked.scatter.v8i8.v8p0i8(<8 x i8>, <8 x i8*>, i32, <8 x i1>)
declare void @llvm.masked.scatter.v8i16.v8p0i16(<8 x i16>, <8 x i16*>, i32, <8 x i1>)
declare void @llvm.masked.scatter.v8f16.v8p0f16(<8 x half>, <8 x half*>, i32, <8 x i1>)
declare void @llvm.masked.scatter.v16i8.v16p0i8(<16 x i8>, <16 x i8*>, i32, <16 x i1>)
declare void @llvm.masked.scatter.v4i8.v4p0i8(<4 x i8>, <4 x i8*>, i32, <4 x i1>)
declare void @llvm.masked.scatter.v4i16.v4p0i16(<4 x i16>, <4 x i16*>, i32, <4 x i1>)
declare void @llvm.masked.scatter.v4f16.v4p0f16(<4 x half>, <4 x half*>, i32, <4 x i1>)
declare void @llvm.masked.scatter.v4i32.v4p0i32(<4 x i32>, <4 x i32*>, i32, <4 x i1>)
declare void @llvm.masked.scatter.v4f32.v4p0f32(<4 x float>, <4 x float*>, i32, <4 x i1>)