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
| ; RUN: llc -march=amdgcn -verify-machineinstrs < %s | FileCheck -check-prefix=SI %s
; RUN: opt -mtriple=amdgcn-- -S -amdgpu-unify-divergent-exit-nodes -verify %s | FileCheck -check-prefix=IR %s
; SI-LABEL: {{^}}infinite_loop:
; SI: v_mov_b32_e32 [[REG:v[0-9]+]], 0x3e7
; SI: [[LOOP:BB[0-9]+_[0-9]+]]: ; %loop
; SI: s_waitcnt lgkmcnt(0)
; SI: buffer_store_dword [[REG]]
; SI: s_branch [[LOOP]]
define amdgpu_kernel void @infinite_loop(i32 addrspace(1)* %out) {
entry:
br label %loop
loop:
store volatile i32 999, i32 addrspace(1)* %out, align 4
br label %loop
}
; IR-LABEL: @infinite_loop_ret(
; IR: br i1 %cond, label %loop, label %UnifiedReturnBlock
; IR: loop:
; IR: store volatile i32 999, i32 addrspace(1)* %out, align 4
; IR: br i1 true, label %loop, label %UnifiedReturnBlock
; IR: UnifiedReturnBlock:
; IR: ret void
; SI-LABEL: {{^}}infinite_loop_ret:
; SI: s_cbranch_execz [[RET:BB[0-9]+_[0-9]+]]
; SI: v_mov_b32_e32 [[REG:v[0-9]+]], 0x3e7
; SI: s_and_b64 vcc, exec, -1
; SI: [[LOOP:BB[0-9]+_[0-9]+]]: ; %loop
; SI: s_waitcnt lgkmcnt(0)
; SI: buffer_store_dword [[REG]]
; SI: s_cbranch_vccnz [[LOOP]]
; SI: [[RET]]: ; %UnifiedReturnBlock
; SI: s_endpgm
define amdgpu_kernel void @infinite_loop_ret(i32 addrspace(1)* %out) {
entry:
%tmp = tail call i32 @llvm.amdgcn.workitem.id.x()
%cond = icmp eq i32 %tmp, 1
br i1 %cond, label %loop, label %return
loop:
store volatile i32 999, i32 addrspace(1)* %out, align 4
br label %loop
return:
ret void
}
; IR-LABEL: @infinite_loops(
; IR: br i1 undef, label %loop1, label %loop2
; IR: loop1:
; IR: store volatile i32 999, i32 addrspace(1)* %out, align 4
; IR: br i1 true, label %loop1, label %DummyReturnBlock
; IR: loop2:
; IR: store volatile i32 888, i32 addrspace(1)* %out, align 4
; IR: br i1 true, label %loop2, label %DummyReturnBlock
; IR: DummyReturnBlock:
; IR: ret void
; SI-LABEL: {{^}}infinite_loops:
; SI: v_mov_b32_e32 [[REG1:v[0-9]+]], 0x3e7
; SI: s_and_b64 vcc, exec, -1
; SI: [[LOOP1:BB[0-9]+_[0-9]+]]: ; %loop1
; SI: s_waitcnt lgkmcnt(0)
; SI: buffer_store_dword [[REG1]]
; SI: s_cbranch_vccnz [[LOOP1]]
; SI: s_branch [[RET:BB[0-9]+_[0-9]+]]
; SI: v_mov_b32_e32 [[REG2:v[0-9]+]], 0x378
; SI: s_and_b64 vcc, exec, -1
; SI: [[LOOP2:BB[0-9]+_[0-9]+]]: ; %loop2
; SI: s_waitcnt lgkmcnt(0)
; SI: buffer_store_dword [[REG2]]
; SI: s_cbranch_vccnz [[LOOP2]]
; SI: [[RET]]: ; %DummyReturnBlock
; SI: s_endpgm
define amdgpu_kernel void @infinite_loops(i32 addrspace(1)* %out) {
entry:
br i1 undef, label %loop1, label %loop2
loop1:
store volatile i32 999, i32 addrspace(1)* %out, align 4
br label %loop1
loop2:
store volatile i32 888, i32 addrspace(1)* %out, align 4
br label %loop2
}
; IR-LABEL: @infinite_loop_nest_ret(
; IR: br i1 %cond1, label %outer_loop, label %UnifiedReturnBlock
; IR: outer_loop:
; IR: br label %inner_loop
; IR: inner_loop:
; IR: store volatile i32 999, i32 addrspace(1)* %out, align 4
; IR: %cond3 = icmp eq i32 %tmp, 3
; IR: br i1 true, label %TransitionBlock, label %UnifiedReturnBlock
; IR: TransitionBlock:
; IR: br i1 %cond3, label %inner_loop, label %outer_loop
; IR: UnifiedReturnBlock:
; IR: ret void
; SI-LABEL: {{^}}infinite_loop_nest_ret:
; SI: s_cbranch_execz [[RET:BB[0-9]+_[0-9]+]]
; SI: s_mov_b32
; SI: [[OUTER_LOOP:BB[0-9]+_[0-9]+]]: ; %outer_loop
; SI: [[INNER_LOOP:BB[0-9]+_[0-9]+]]: ; %inner_loop
; SI: s_waitcnt expcnt(0)
; SI: v_mov_b32_e32 [[REG:v[0-9]+]], 0x3e7
; SI: s_waitcnt lgkmcnt(0)
; SI: buffer_store_dword [[REG]]
; SI: s_andn2_b64 exec
; SI: s_cbranch_execnz [[INNER_LOOP]]
; SI: s_andn2_b64 exec
; SI: s_cbranch_execnz [[OUTER_LOOP]]
; SI: [[RET]]: ; %UnifiedReturnBlock
; SI: s_endpgm
define amdgpu_kernel void @infinite_loop_nest_ret(i32 addrspace(1)* %out) {
entry:
%tmp = tail call i32 @llvm.amdgcn.workitem.id.x()
%cond1 = icmp eq i32 %tmp, 1
br i1 %cond1, label %outer_loop, label %return
outer_loop:
; %cond2 = icmp eq i32 %tmp, 2
; br i1 %cond2, label %outer_loop, label %inner_loop
br label %inner_loop
inner_loop: ; preds = %LeafBlock, %LeafBlock1
store volatile i32 999, i32 addrspace(1)* %out, align 4
%cond3 = icmp eq i32 %tmp, 3
br i1 %cond3, label %inner_loop, label %outer_loop
return:
ret void
}
declare i32 @llvm.amdgcn.workitem.id.x()
|