File: legalize-fmad.s64.mir

package info (click to toggle)
swiftlang 6.0.3-2
  • links: PTS, VCS
  • area: main
  • in suites: forky, sid, trixie
  • size: 2,519,992 kB
  • sloc: cpp: 9,107,863; ansic: 2,040,022; asm: 1,135,751; python: 296,500; objc: 82,456; f90: 60,502; lisp: 34,951; pascal: 19,946; sh: 18,133; perl: 7,482; ml: 4,937; javascript: 4,117; makefile: 3,840; awk: 3,535; xml: 914; fortran: 619; cs: 573; ruby: 573
file content (125 lines) | stat: -rw-r--r-- 5,592 bytes parent folder | download | duplicates (15)
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
# NOTE: Assertions have been autogenerated by utils/update_mir_test_checks.py
# RUN: llc -mtriple=amdgcn-mesa-mesa3d -mcpu=tahiti -run-pass=legalizer %s -o - | FileCheck %s
# RUN: llc -mtriple=amdgcn-mesa-mesa3d -mcpu=fiji -run-pass=legalizer %s -o - | FileCheck %s
# RUN: llc -mtriple=amdgcn-mesa-mesa3d -mcpu=gfx1010 -run-pass=legalizer %s -o - | FileCheck %s
# RUN: llc -mtriple=amdgcn-mesa-mesa3d -mcpu=gfx1100 -run-pass=legalizer %s -o - | FileCheck %s

---
name: test_fmad_s64_flush
machineFunctionInfo:
  mode:
    fp64-fp16-output-denormals: false
    fp64-fp16-input-denormals: false

body: |
  bb.0:
    liveins: $vgpr0_vgpr1, $vgpr2_vgpr3,  $vgpr4_vgpr5

    ; CHECK-LABEL: name: test_fmad_s64_flush
    ; CHECK: liveins: $vgpr0_vgpr1, $vgpr2_vgpr3, $vgpr4_vgpr5
    ; CHECK-NEXT: {{  $}}
    ; CHECK-NEXT: [[COPY:%[0-9]+]]:_(s64) = COPY $vgpr0_vgpr1
    ; CHECK-NEXT: [[COPY1:%[0-9]+]]:_(s64) = COPY $vgpr2_vgpr3
    ; CHECK-NEXT: [[COPY2:%[0-9]+]]:_(s64) = COPY $vgpr4_vgpr5
    ; CHECK-NEXT: [[FMUL:%[0-9]+]]:_(s64) = G_FMUL [[COPY]], [[COPY1]]
    ; CHECK-NEXT: [[FADD:%[0-9]+]]:_(s64) = G_FADD [[FMUL]], [[COPY2]]
    ; CHECK-NEXT: $vgpr0_vgpr1 = COPY [[FADD]](s64)
    %0:_(s64) = COPY $vgpr0_vgpr1
    %1:_(s64) = COPY $vgpr2_vgpr3
    %2:_(s64) = COPY $vgpr4_vgpr5
    %3:_(s64) = G_FMAD %0, %1, %2
    $vgpr0_vgpr1 = COPY %3
...

---
name: test_fmad_v2s64_flush
machineFunctionInfo:
  mode:
    fp64-fp16-input-denormals: false
    fp64-fp16-output-denormals: false

body: |
  bb.0:
    liveins: $vgpr0_vgpr1_vgpr2_vgpr3, $vgpr4_vgpr5_vgpr6_vgpr7, $vgpr8_vgpr9_vgpr10_vgpr11

    ; CHECK-LABEL: name: test_fmad_v2s64_flush
    ; CHECK: liveins: $vgpr0_vgpr1_vgpr2_vgpr3, $vgpr4_vgpr5_vgpr6_vgpr7, $vgpr8_vgpr9_vgpr10_vgpr11
    ; CHECK-NEXT: {{  $}}
    ; CHECK-NEXT: [[COPY:%[0-9]+]]:_(<2 x s64>) = COPY $vgpr0_vgpr1_vgpr2_vgpr3
    ; CHECK-NEXT: [[COPY1:%[0-9]+]]:_(<2 x s64>) = COPY $vgpr4_vgpr5_vgpr6_vgpr7
    ; CHECK-NEXT: [[COPY2:%[0-9]+]]:_(<2 x s64>) = COPY $vgpr8_vgpr9_vgpr10_vgpr11
    ; CHECK-NEXT: [[UV:%[0-9]+]]:_(s64), [[UV1:%[0-9]+]]:_(s64) = G_UNMERGE_VALUES [[COPY]](<2 x s64>)
    ; CHECK-NEXT: [[UV2:%[0-9]+]]:_(s64), [[UV3:%[0-9]+]]:_(s64) = G_UNMERGE_VALUES [[COPY1]](<2 x s64>)
    ; CHECK-NEXT: [[UV4:%[0-9]+]]:_(s64), [[UV5:%[0-9]+]]:_(s64) = G_UNMERGE_VALUES [[COPY2]](<2 x s64>)
    ; CHECK-NEXT: [[FMUL:%[0-9]+]]:_(s64) = G_FMUL [[UV]], [[UV2]]
    ; CHECK-NEXT: [[FADD:%[0-9]+]]:_(s64) = G_FADD [[FMUL]], [[UV4]]
    ; CHECK-NEXT: [[FMUL1:%[0-9]+]]:_(s64) = G_FMUL [[UV1]], [[UV3]]
    ; CHECK-NEXT: [[FADD1:%[0-9]+]]:_(s64) = G_FADD [[FMUL1]], [[UV5]]
    ; CHECK-NEXT: [[BUILD_VECTOR:%[0-9]+]]:_(<2 x s64>) = G_BUILD_VECTOR [[FADD]](s64), [[FADD1]](s64)
    ; CHECK-NEXT: $vgpr0_vgpr1_vgpr2_vgpr3 = COPY [[BUILD_VECTOR]](<2 x s64>)
    %0:_(<2 x s64>) = COPY $vgpr0_vgpr1_vgpr2_vgpr3
    %1:_(<2 x s64>) = COPY $vgpr4_vgpr5_vgpr6_vgpr7
    %2:_(<2 x s64>) = COPY $vgpr8_vgpr9_vgpr10_vgpr11
    %3:_(<2 x s64>) = G_FMAD %0, %1, %2
    $vgpr0_vgpr1_vgpr2_vgpr3 = COPY %3
...

---
name: test_fmad_s64_denorm
machineFunctionInfo:
  mode:
    fp64-fp16-input-denormals: true
    fp64-fp16-output-denormals: true

body: |
  bb.0:
    liveins: $vgpr0_vgpr1, $vgpr2_vgpr3,  $vgpr4_vgpr5

    ; CHECK-LABEL: name: test_fmad_s64_denorm
    ; CHECK: liveins: $vgpr0_vgpr1, $vgpr2_vgpr3, $vgpr4_vgpr5
    ; CHECK-NEXT: {{  $}}
    ; CHECK-NEXT: [[COPY:%[0-9]+]]:_(s64) = COPY $vgpr0_vgpr1
    ; CHECK-NEXT: [[COPY1:%[0-9]+]]:_(s64) = COPY $vgpr2_vgpr3
    ; CHECK-NEXT: [[COPY2:%[0-9]+]]:_(s64) = COPY $vgpr4_vgpr5
    ; CHECK-NEXT: [[FMUL:%[0-9]+]]:_(s64) = G_FMUL [[COPY]], [[COPY1]]
    ; CHECK-NEXT: [[FADD:%[0-9]+]]:_(s64) = G_FADD [[FMUL]], [[COPY2]]
    ; CHECK-NEXT: $vgpr0_vgpr1 = COPY [[FADD]](s64)
    %0:_(s64) = COPY $vgpr0_vgpr1
    %1:_(s64) = COPY $vgpr2_vgpr3
    %2:_(s64) = COPY $vgpr4_vgpr5
    %3:_(s64) = G_FMAD %0, %1, %2
    $vgpr0_vgpr1 = COPY %3
...

---
name: test_fmad_v2s64_denorm
machineFunctionInfo:
  mode:
    fp64-fp16-input-denormals: true
    fp64-fp16-output-denormals: true

body: |
  bb.0:
    liveins: $vgpr0_vgpr1_vgpr2_vgpr3, $vgpr4_vgpr5_vgpr6_vgpr7, $vgpr8_vgpr9_vgpr10_vgpr11

    ; CHECK-LABEL: name: test_fmad_v2s64_denorm
    ; CHECK: liveins: $vgpr0_vgpr1_vgpr2_vgpr3, $vgpr4_vgpr5_vgpr6_vgpr7, $vgpr8_vgpr9_vgpr10_vgpr11
    ; CHECK-NEXT: {{  $}}
    ; CHECK-NEXT: [[COPY:%[0-9]+]]:_(<2 x s64>) = COPY $vgpr0_vgpr1_vgpr2_vgpr3
    ; CHECK-NEXT: [[COPY1:%[0-9]+]]:_(<2 x s64>) = COPY $vgpr4_vgpr5_vgpr6_vgpr7
    ; CHECK-NEXT: [[COPY2:%[0-9]+]]:_(<2 x s64>) = COPY $vgpr8_vgpr9_vgpr10_vgpr11
    ; CHECK-NEXT: [[UV:%[0-9]+]]:_(s64), [[UV1:%[0-9]+]]:_(s64) = G_UNMERGE_VALUES [[COPY]](<2 x s64>)
    ; CHECK-NEXT: [[UV2:%[0-9]+]]:_(s64), [[UV3:%[0-9]+]]:_(s64) = G_UNMERGE_VALUES [[COPY1]](<2 x s64>)
    ; CHECK-NEXT: [[UV4:%[0-9]+]]:_(s64), [[UV5:%[0-9]+]]:_(s64) = G_UNMERGE_VALUES [[COPY2]](<2 x s64>)
    ; CHECK-NEXT: [[FMUL:%[0-9]+]]:_(s64) = G_FMUL [[UV]], [[UV2]]
    ; CHECK-NEXT: [[FADD:%[0-9]+]]:_(s64) = G_FADD [[FMUL]], [[UV4]]
    ; CHECK-NEXT: [[FMUL1:%[0-9]+]]:_(s64) = G_FMUL [[UV1]], [[UV3]]
    ; CHECK-NEXT: [[FADD1:%[0-9]+]]:_(s64) = G_FADD [[FMUL1]], [[UV5]]
    ; CHECK-NEXT: [[BUILD_VECTOR:%[0-9]+]]:_(<2 x s64>) = G_BUILD_VECTOR [[FADD]](s64), [[FADD1]](s64)
    ; CHECK-NEXT: $vgpr0_vgpr1_vgpr2_vgpr3 = COPY [[BUILD_VECTOR]](<2 x s64>)
    %0:_(<2 x s64>) = COPY $vgpr0_vgpr1_vgpr2_vgpr3
    %1:_(<2 x s64>) = COPY $vgpr4_vgpr5_vgpr6_vgpr7
    %2:_(<2 x s64>) = COPY $vgpr8_vgpr9_vgpr10_vgpr11
    %3:_(<2 x s64>) = G_FMAD %0, %1, %2
    $vgpr0_vgpr1_vgpr2_vgpr3 = COPY %3
...