mul_vec.mir
4.65 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
# NOTE: Assertions have been autogenerated by utils/update_mir_test_checks.py
# RUN: llc -mtriple=mipsel-linux-gnu -mcpu=mips32r5 -mattr=+msa,+fp64,+nan2008 -run-pass=instruction-select -verify-machineinstrs %s -o - | FileCheck %s -check-prefixes=P5600
--- |
define void @mul_v16i8(<16 x i8>* %a, <16 x i8>* %b, <16 x i8>* %c) { entry: ret void }
define void @mul_v8i16(<8 x i16>* %a, <8 x i16>* %b, <8 x i16>* %c) { entry: ret void }
define void @mul_v4i32(<4 x i32>* %a, <4 x i32>* %b, <4 x i32>* %c) { entry: ret void }
define void @mul_v2i64(<2 x i64>* %a, <2 x i64>* %b, <2 x i64>* %c) { entry: ret void }
...
---
name: mul_v16i8
alignment: 4
legalized: true
regBankSelected: true
tracksRegLiveness: true
body: |
bb.1.entry:
liveins: $a0, $a1, $a2
; P5600-LABEL: name: mul_v16i8
; P5600: liveins: $a0, $a1, $a2
; P5600: [[COPY:%[0-9]+]]:gpr32 = COPY $a0
; P5600: [[COPY1:%[0-9]+]]:gpr32 = COPY $a1
; P5600: [[COPY2:%[0-9]+]]:gpr32 = COPY $a2
; P5600: [[LD_B:%[0-9]+]]:msa128b = LD_B [[COPY]], 0 :: (load 16 from %ir.a)
; P5600: [[LD_B1:%[0-9]+]]:msa128b = LD_B [[COPY1]], 0 :: (load 16 from %ir.b)
; P5600: [[MULV_B:%[0-9]+]]:msa128b = MULV_B [[LD_B1]], [[LD_B]]
; P5600: ST_B [[MULV_B]], [[COPY2]], 0 :: (store 16 into %ir.c)
; P5600: RetRA
%0:gprb(p0) = COPY $a0
%1:gprb(p0) = COPY $a1
%2:gprb(p0) = COPY $a2
%3:fprb(<16 x s8>) = G_LOAD %0(p0) :: (load 16 from %ir.a)
%4:fprb(<16 x s8>) = G_LOAD %1(p0) :: (load 16 from %ir.b)
%5:fprb(<16 x s8>) = G_MUL %4, %3
G_STORE %5(<16 x s8>), %2(p0) :: (store 16 into %ir.c)
RetRA
...
---
name: mul_v8i16
alignment: 4
legalized: true
regBankSelected: true
tracksRegLiveness: true
body: |
bb.1.entry:
liveins: $a0, $a1, $a2
; P5600-LABEL: name: mul_v8i16
; P5600: liveins: $a0, $a1, $a2
; P5600: [[COPY:%[0-9]+]]:gpr32 = COPY $a0
; P5600: [[COPY1:%[0-9]+]]:gpr32 = COPY $a1
; P5600: [[COPY2:%[0-9]+]]:gpr32 = COPY $a2
; P5600: [[LD_H:%[0-9]+]]:msa128h = LD_H [[COPY]], 0 :: (load 16 from %ir.a)
; P5600: [[LD_H1:%[0-9]+]]:msa128h = LD_H [[COPY1]], 0 :: (load 16 from %ir.b)
; P5600: [[MULV_H:%[0-9]+]]:msa128h = MULV_H [[LD_H1]], [[LD_H]]
; P5600: ST_H [[MULV_H]], [[COPY2]], 0 :: (store 16 into %ir.c)
; P5600: RetRA
%0:gprb(p0) = COPY $a0
%1:gprb(p0) = COPY $a1
%2:gprb(p0) = COPY $a2
%3:fprb(<8 x s16>) = G_LOAD %0(p0) :: (load 16 from %ir.a)
%4:fprb(<8 x s16>) = G_LOAD %1(p0) :: (load 16 from %ir.b)
%5:fprb(<8 x s16>) = G_MUL %4, %3
G_STORE %5(<8 x s16>), %2(p0) :: (store 16 into %ir.c)
RetRA
...
---
name: mul_v4i32
alignment: 4
legalized: true
regBankSelected: true
tracksRegLiveness: true
body: |
bb.1.entry:
liveins: $a0, $a1, $a2
; P5600-LABEL: name: mul_v4i32
; P5600: liveins: $a0, $a1, $a2
; P5600: [[COPY:%[0-9]+]]:gpr32 = COPY $a0
; P5600: [[COPY1:%[0-9]+]]:gpr32 = COPY $a1
; P5600: [[COPY2:%[0-9]+]]:gpr32 = COPY $a2
; P5600: [[LD_W:%[0-9]+]]:msa128w = LD_W [[COPY]], 0 :: (load 16 from %ir.a)
; P5600: [[LD_W1:%[0-9]+]]:msa128w = LD_W [[COPY1]], 0 :: (load 16 from %ir.b)
; P5600: [[MULV_W:%[0-9]+]]:msa128w = MULV_W [[LD_W1]], [[LD_W]]
; P5600: ST_W [[MULV_W]], [[COPY2]], 0 :: (store 16 into %ir.c)
; P5600: RetRA
%0:gprb(p0) = COPY $a0
%1:gprb(p0) = COPY $a1
%2:gprb(p0) = COPY $a2
%3:fprb(<4 x s32>) = G_LOAD %0(p0) :: (load 16 from %ir.a)
%4:fprb(<4 x s32>) = G_LOAD %1(p0) :: (load 16 from %ir.b)
%5:fprb(<4 x s32>) = G_MUL %4, %3
G_STORE %5(<4 x s32>), %2(p0) :: (store 16 into %ir.c)
RetRA
...
---
name: mul_v2i64
alignment: 4
legalized: true
regBankSelected: true
tracksRegLiveness: true
body: |
bb.1.entry:
liveins: $a0, $a1, $a2
; P5600-LABEL: name: mul_v2i64
; P5600: liveins: $a0, $a1, $a2
; P5600: [[COPY:%[0-9]+]]:gpr32 = COPY $a0
; P5600: [[COPY1:%[0-9]+]]:gpr32 = COPY $a1
; P5600: [[COPY2:%[0-9]+]]:gpr32 = COPY $a2
; P5600: [[LD_D:%[0-9]+]]:msa128d = LD_D [[COPY]], 0 :: (load 16 from %ir.a)
; P5600: [[LD_D1:%[0-9]+]]:msa128d = LD_D [[COPY1]], 0 :: (load 16 from %ir.b)
; P5600: [[MULV_D:%[0-9]+]]:msa128d = MULV_D [[LD_D1]], [[LD_D]]
; P5600: ST_D [[MULV_D]], [[COPY2]], 0 :: (store 16 into %ir.c)
; P5600: RetRA
%0:gprb(p0) = COPY $a0
%1:gprb(p0) = COPY $a1
%2:gprb(p0) = COPY $a2
%3:fprb(<2 x s64>) = G_LOAD %0(p0) :: (load 16 from %ir.a)
%4:fprb(<2 x s64>) = G_LOAD %1(p0) :: (load 16 from %ir.b)
%5:fprb(<2 x s64>) = G_MUL %4, %3
G_STORE %5(<2 x s64>), %2(p0) :: (store 16 into %ir.c)
RetRA
...