1; RUN: opt -amdgpu-lower-enqueued-block -S < %s | FileCheck %s
2
3; CHECK: @__test_block_invoke_kernel_runtime_handle = external addrspace(1) externally_initialized constant i8 addrspace(1)*
4; CHECK: @__test_block_invoke_2_kernel_runtime_handle = external addrspace(1) externally_initialized constant i8 addrspace(1)*
5
6target datalayout = "e-p:32:32-p1:64:64-p2:64:64-p3:32:32-p4:64:64-p5:32:32-i64:64-v16:16-v24:32-v32:32-v48:64-v96:128-v192:256-v256:256-v512:512-v1024:1024-v2048:2048-n32:64"
7target triple = "amdgcn-amdhsa-amd-opencl"
8
9%struct.ndrange_t = type { i32 }
10%opencl.queue_t = type opaque
11
12; CHECK: define amdgpu_kernel void @non_caller(i8 addrspace(1)* %a, i8 %b, i64 addrspace(1)* %c, i64 %d) local_unnamed_addr !kernel_arg_addr_space
13define amdgpu_kernel void @non_caller(i8 addrspace(1)* %a, i8 %b, i64 addrspace(1)* %c, i64 %d) local_unnamed_addr
14  !kernel_arg_addr_space !3 !kernel_arg_access_qual !4 !kernel_arg_type !5 !kernel_arg_base_type !5 !kernel_arg_type_qual !6 {
15  ret void
16}
17
18; CHECK: define amdgpu_kernel void @caller_indirect(i8 addrspace(1)* %a, i8 %b, i64 addrspace(1)* %c, i64 %d) local_unnamed_addr #[[AT_CALLER:[0-9]+]]
19define amdgpu_kernel void @caller_indirect(i8 addrspace(1)* %a, i8 %b, i64 addrspace(1)* %c, i64 %d) local_unnamed_addr
20  !kernel_arg_addr_space !3 !kernel_arg_access_qual !4 !kernel_arg_type !5 !kernel_arg_base_type !5 !kernel_arg_type_qual !6 {
21  call void @caller(i8 addrspace(1)* %a, i8 %b, i64 addrspace(1)* %c, i64 %d)
22  ret void
23}
24
25; CHECK: define amdgpu_kernel void @caller(i8 addrspace(1)* %a, i8 %b, i64 addrspace(1)* %c, i64 %d) local_unnamed_addr #[[AT_CALLER]]
26define amdgpu_kernel void @caller(i8 addrspace(1)* %a, i8 %b, i64 addrspace(1)* %c, i64 %d) local_unnamed_addr
27  !kernel_arg_addr_space !3 !kernel_arg_access_qual !4 !kernel_arg_type !5 !kernel_arg_base_type !5 !kernel_arg_type_qual !6 {
28entry:
29  %block = alloca <{ i32, i32, i8 addrspace(4)*, i8 addrspace(1)*, i8 }>, align 8
30  %tmp = alloca %struct.ndrange_t, align 4
31  %block2 = alloca <{ i32, i32, i8 addrspace(4)*, i8 addrspace(1)*, i64 addrspace(1)*, i64, i8 }>, align 8
32  %tmp3 = alloca %struct.ndrange_t, align 4
33  %block.size = getelementptr inbounds <{ i32, i32, i8 addrspace(4)*, i8 addrspace(1)*, i8 }>, <{ i32, i32, i8 addrspace(4)*, i8 addrspace(1)*, i8 }>* %block, i32 0, i32 0
34  store i32 25, i32* %block.size, align 8
35  %block.align = getelementptr inbounds <{ i32, i32, i8 addrspace(4)*, i8 addrspace(1)*, i8 }>, <{ i32, i32, i8 addrspace(4)*, i8 addrspace(1)*, i8 }>* %block, i32 0, i32 1
36  store i32 8, i32* %block.align, align 4
37  %block.invoke = getelementptr inbounds <{ i32, i32, i8 addrspace(4)*, i8 addrspace(1)*, i8 }>, <{ i32, i32, i8 addrspace(4)*, i8 addrspace(1)*, i8 }>* %block, i32 0, i32 2
38  store i8 addrspace(4)* addrspacecast (i8* bitcast (void (<{ i32, i32, i8 addrspace(4)*, i8 addrspace(1)*, i8 }>)* @__test_block_invoke_kernel to i8*) to i8 addrspace(4)*), i8 addrspace(4)** %block.invoke, align 8
39  %block.captured = getelementptr inbounds <{ i32, i32, i8 addrspace(4)*, i8 addrspace(1)*, i8 }>, <{ i32, i32, i8 addrspace(4)*, i8 addrspace(1)*, i8 }>* %block, i32 0, i32 3
40  store i8 addrspace(1)* %a, i8 addrspace(1)** %block.captured, align 8
41  %block.captured1 = getelementptr inbounds <{ i32, i32, i8 addrspace(4)*, i8 addrspace(1)*, i8 }>, <{ i32, i32, i8 addrspace(4)*, i8 addrspace(1)*, i8 }>* %block, i32 0, i32 4
42  store i8 %b, i8* %block.captured1, align 8
43  %tmp1 = bitcast <{ i32, i32, i8 addrspace(4)*, i8 addrspace(1)*, i8 }>* %block to void ()*
44  %tmp2 = bitcast void ()* %tmp1 to i8*
45  %tmp4 = addrspacecast i8* %tmp2 to i8 addrspace(4)*
46  %tmp5 = call i32 @__enqueue_kernel_basic(%opencl.queue_t addrspace(1)* undef, i32 0, %struct.ndrange_t* byval nonnull %tmp, i8 addrspace(4)* nonnull %tmp4) #2
47  %block.size4 = getelementptr inbounds <{ i32, i32, i8 addrspace(4)*, i8 addrspace(1)*, i64 addrspace(1)*, i64, i8 }>, <{ i32, i32, i8 addrspace(4)*, i8 addrspace(1)*, i64 addrspace(1)*, i64, i8 }>* %block2, i32 0, i32 0
48  store i32 41, i32* %block.size4, align 8
49  %block.align5 = getelementptr inbounds <{ i32, i32, i8 addrspace(4)*, i8 addrspace(1)*, i64 addrspace(1)*, i64, i8 }>, <{ i32, i32, i8 addrspace(4)*, i8 addrspace(1)*, i64 addrspace(1)*, i64, i8 }>* %block2, i32 0, i32 1
50  store i32 8, i32* %block.align5, align 4
51  %block.invoke6 = getelementptr inbounds <{ i32, i32, i8 addrspace(4)*, i8 addrspace(1)*, i64 addrspace(1)*, i64, i8 }>, <{ i32, i32, i8 addrspace(4)*, i8 addrspace(1)*, i64 addrspace(1)*, i64, i8 }>* %block2, i32 0, i32 2
52  store i8 addrspace(4)* addrspacecast (i8* bitcast (void (<{ i32, i32, i8 addrspace(4)*, i8 addrspace(1)*, i64 addrspace(1)*, i64, i8 }>)* @__test_block_invoke_2_kernel to i8*) to i8 addrspace(4)*), i8 addrspace(4)** %block.invoke6, align 8
53  %block.captured7 = getelementptr inbounds <{ i32, i32, i8 addrspace(4)*, i8 addrspace(1)*, i64 addrspace(1)*, i64, i8 }>, <{ i32, i32, i8 addrspace(4)*, i8 addrspace(1)*, i64 addrspace(1)*, i64, i8 }>* %block2, i32 0, i32 3
54  store i8 addrspace(1)* %a, i8 addrspace(1)** %block.captured7, align 8
55  %block.captured8 = getelementptr inbounds <{ i32, i32, i8 addrspace(4)*, i8 addrspace(1)*, i64 addrspace(1)*, i64, i8 }>, <{ i32, i32, i8 addrspace(4)*, i8 addrspace(1)*, i64 addrspace(1)*, i64, i8 }>* %block2, i32 0, i32 6
56  store i8 %b, i8* %block.captured8, align 8
57  %block.captured9 = getelementptr inbounds <{ i32, i32, i8 addrspace(4)*, i8 addrspace(1)*, i64 addrspace(1)*, i64, i8 }>, <{ i32, i32, i8 addrspace(4)*, i8 addrspace(1)*, i64 addrspace(1)*, i64, i8 }>* %block2, i32 0, i32 4
58  store i64 addrspace(1)* %c, i64 addrspace(1)** %block.captured9, align 8
59  %block.captured10 = getelementptr inbounds <{ i32, i32, i8 addrspace(4)*, i8 addrspace(1)*, i64 addrspace(1)*, i64, i8 }>, <{ i32, i32, i8 addrspace(4)*, i8 addrspace(1)*, i64 addrspace(1)*, i64, i8 }>* %block2, i32 0, i32 5
60  store i64 %d, i64* %block.captured10, align 8
61  %tmp6 = bitcast <{ i32, i32, i8 addrspace(4)*, i8 addrspace(1)*, i64 addrspace(1)*, i64, i8 }>* %block2 to void ()*
62  %tmp7 = bitcast void ()* %tmp6 to i8*
63  %tmp8 = addrspacecast i8* %tmp7 to i8 addrspace(4)*
64  %tmp9 = call i32 @__enqueue_kernel_basic(%opencl.queue_t addrspace(1)* undef, i32 0, %struct.ndrange_t* byval nonnull %tmp3, i8 addrspace(4)* nonnull %tmp8) #2
65  ret void
66}
67
68; CHECK: define dso_local amdgpu_kernel void @__test_block_invoke_kernel({{.*}}) #[[AT1:[0-9]+]]
69define internal amdgpu_kernel void @__test_block_invoke_kernel(<{ i32, i32, i8 addrspace(4)*, i8 addrspace(1)*, i8 }> %arg) #0
70  !kernel_arg_addr_space !14 !kernel_arg_access_qual !15 !kernel_arg_type !16 !kernel_arg_base_type !16 !kernel_arg_type_qual !17 {
71entry:
72  %.fca.3.extract = extractvalue <{ i32, i32, i8 addrspace(4)*, i8 addrspace(1)*, i8 }> %arg, 3
73  %.fca.4.extract = extractvalue <{ i32, i32, i8 addrspace(4)*, i8 addrspace(1)*, i8 }> %arg, 4
74  store i8 %.fca.4.extract, i8 addrspace(1)* %.fca.3.extract, align 1
75  ret void
76}
77
78declare i32 @__enqueue_kernel_basic(%opencl.queue_t addrspace(1)*, i32, %struct.ndrange_t*, i8 addrspace(4)*) local_unnamed_addr
79
80; CHECK: define dso_local amdgpu_kernel void @__test_block_invoke_2_kernel({{.*}}) #[[AT2:[0-9]+]]
81define internal amdgpu_kernel void @__test_block_invoke_2_kernel(<{ i32, i32, i8 addrspace(4)*, i8 addrspace(1)*,
82  i64 addrspace(1)*, i64, i8 }> %arg) #0 !kernel_arg_addr_space !14 !kernel_arg_access_qual !15
83  !kernel_arg_type !16 !kernel_arg_base_type !16 !kernel_arg_type_qual !17 {
84entry:
85  %.fca.3.extract = extractvalue <{ i32, i32, i8 addrspace(4)*, i8 addrspace(1)*, i64 addrspace(1)*, i64, i8 }> %arg, 3
86  %.fca.4.extract = extractvalue <{ i32, i32, i8 addrspace(4)*, i8 addrspace(1)*, i64 addrspace(1)*, i64, i8 }> %arg, 4
87  %.fca.5.extract = extractvalue <{ i32, i32, i8 addrspace(4)*, i8 addrspace(1)*, i64 addrspace(1)*, i64, i8 }> %arg, 5
88  %.fca.6.extract = extractvalue <{ i32, i32, i8 addrspace(4)*, i8 addrspace(1)*, i64 addrspace(1)*, i64, i8 }> %arg, 6
89  store i8 %.fca.6.extract, i8 addrspace(1)* %.fca.3.extract, align 1
90  store i64 %.fca.5.extract, i64 addrspace(1)* %.fca.4.extract, align 8
91  ret void
92}
93
94; CHECK: attributes #[[AT_CALLER]] = { "calls-enqueue-kernel" }
95; CHECK: attributes #[[AT1]] = {{.*}}"runtime-handle"="__test_block_invoke_kernel_runtime_handle"
96; CHECK: attributes #[[AT2]] = {{.*}}"runtime-handle"="__test_block_invoke_2_kernel_runtime_handle"
97
98attributes #0 = { "enqueued-block" }
99
100!3 = !{i32 1, i32 0, i32 1, i32 0}
101!4 = !{!"none", !"none", !"none", !"none"}
102!5 = !{!"char*", !"char", !"long*", !"long"}
103!6 = !{!"", !"", !"", !""}
104!14 = !{i32 0}
105!15 = !{!"none"}
106!16 = !{!"__block_literal"}
107!17 = !{!""}
108