1; RUN: opt -S -mtriple=amdgcn-- -amdgpu-lower-module-lds < %s | FileCheck %s
2; RUN: opt -S -mtriple=amdgcn-- -passes=amdgpu-lower-module-lds < %s | FileCheck %s
3
4@lds.size.1.align.1 = internal unnamed_addr addrspace(3) global [1 x i8] undef, align 1
5@lds.size.2.align.2 = internal unnamed_addr addrspace(3) global [2 x i8] undef, align 2
6@lds.size.4.align.4 = internal unnamed_addr addrspace(3) global [4 x i8] undef, align 4
7@lds.size.8.align.8 = internal unnamed_addr addrspace(3) global [8 x i8] undef, align 8
8@lds.size.16.align.16 = internal unnamed_addr addrspace(3) global [16 x i8] undef, align 16
9
10; CHECK: %llvm.amdgcn.module.lds.t = type { [8 x i8], [1 x i8] }
11; CHECK: %llvm.amdgcn.kernel.k0.lds.t = type { [16 x i8], [4 x i8], [2 x i8] }
12; CHECK: %llvm.amdgcn.kernel.k1.lds.t = type { [16 x i8], [4 x i8], [2 x i8] }
13; CHECK: %llvm.amdgcn.kernel..lds.t = type { [2 x i8] }
14; CHECK: %llvm.amdgcn.kernel..lds.t.0 = type { [4 x i8] }
15
16;.
17; CHECK: @llvm.amdgcn.module.lds = internal addrspace(3) global %llvm.amdgcn.module.lds.t undef, align 8
18; CHECK: @llvm.compiler.used = appending global [1 x i8*] [i8* addrspacecast (i8 addrspace(3)* getelementptr inbounds (%llvm.amdgcn.module.lds.t, %llvm.amdgcn.module.lds.t addrspace(3)* @llvm.amdgcn.module.lds, i32 0, i32 0, i32 0) to i8*)], section "llvm.metadata"
19; CHECK: @llvm.amdgcn.kernel.k0.lds = internal addrspace(3) global %llvm.amdgcn.kernel.k0.lds.t undef, align 16
20; CHECK: @llvm.amdgcn.kernel.k1.lds = internal addrspace(3) global %llvm.amdgcn.kernel.k1.lds.t undef, align 16
21; CHECK: @llvm.amdgcn.kernel..lds = internal addrspace(3) global %llvm.amdgcn.kernel..lds.t undef, align 2
22; CHECK: @llvm.amdgcn.kernel..lds.1 = internal addrspace(3) global %llvm.amdgcn.kernel..lds.t.0 undef, align 4
23;.
24define amdgpu_kernel void @k0() #0 {
25; CHECK-LABEL: @k0(
26; CHECK-NEXT:    %lds.size.1.align.1.bc = bitcast [1 x i8] addrspace(3)* getelementptr inbounds (%llvm.amdgcn.module.lds.t, %llvm.amdgcn.module.lds.t addrspace(3)* @llvm.amdgcn.module.lds, i32 0, i32 1) to i8 addrspace(3)*
27; CHECK-NEXT:    store i8 1, i8 addrspace(3)* %lds.size.1.align.1.bc, align 8
28; CHECK-NEXT:    %lds.size.2.align.2.bc = bitcast [2 x i8] addrspace(3)* getelementptr inbounds (%llvm.amdgcn.kernel.k0.lds.t, %llvm.amdgcn.kernel.k0.lds.t addrspace(3)* @llvm.amdgcn.kernel.k0.lds, i32 0, i32 2) to i8 addrspace(3)*
29; CHECK-NEXT:    store i8 2, i8 addrspace(3)* %lds.size.2.align.2.bc, align 4
30; CHECK-NEXT:    %lds.size.4.align.4.bc = bitcast [4 x i8] addrspace(3)* getelementptr inbounds (%llvm.amdgcn.kernel.k0.lds.t, %llvm.amdgcn.kernel.k0.lds.t addrspace(3)* @llvm.amdgcn.kernel.k0.lds, i32 0, i32 1) to i8 addrspace(3)*
31; CHECK-NEXT:    store i8 4, i8 addrspace(3)* %lds.size.4.align.4.bc, align 16
32; CHECK-NEXT:    %lds.size.16.align.16.bc = bitcast [16 x i8] addrspace(3)* getelementptr inbounds (%llvm.amdgcn.kernel.k0.lds.t, %llvm.amdgcn.kernel.k0.lds.t addrspace(3)* @llvm.amdgcn.kernel.k0.lds, i32 0, i32 0) to i8 addrspace(3)*
33; CHECK-NEXT:    store i8 16, i8 addrspace(3)* %lds.size.16.align.16.bc, align 16
34; CHECK-NEXT:    ret void
35;
36  %lds.size.1.align.1.bc = bitcast [1 x i8] addrspace(3)* @lds.size.1.align.1 to i8 addrspace(3)*
37  store i8 1, i8 addrspace(3)* %lds.size.1.align.1.bc, align 1
38
39  %lds.size.2.align.2.bc = bitcast [2 x i8] addrspace(3)* @lds.size.2.align.2 to i8 addrspace(3)*
40  store i8 2, i8 addrspace(3)* %lds.size.2.align.2.bc, align 2
41
42  %lds.size.4.align.4.bc = bitcast [4 x i8] addrspace(3)* @lds.size.4.align.4 to i8 addrspace(3)*
43  store i8 4, i8 addrspace(3)* %lds.size.4.align.4.bc, align 4
44
45  %lds.size.16.align.16.bc = bitcast [16 x i8] addrspace(3)* @lds.size.16.align.16 to i8 addrspace(3)*
46  store i8 16, i8 addrspace(3)* %lds.size.16.align.16.bc, align 16
47
48  ret void
49}
50
51define amdgpu_kernel void @k1() #0 {
52; CHECK-LABEL: @k1(
53; CHECK-NEXT:    %lds.size.2.align.2.bc = bitcast [2 x i8] addrspace(3)* getelementptr inbounds (%llvm.amdgcn.kernel.k1.lds.t, %llvm.amdgcn.kernel.k1.lds.t addrspace(3)* @llvm.amdgcn.kernel.k1.lds, i32 0, i32 2) to i8 addrspace(3)*
54; CHECK-NEXT:    store i8 2, i8 addrspace(3)* %lds.size.2.align.2.bc, align 4
55; CHECK-NEXT:    %lds.size.4.align.4.bc = bitcast [4 x i8] addrspace(3)* getelementptr inbounds (%llvm.amdgcn.kernel.k1.lds.t, %llvm.amdgcn.kernel.k1.lds.t addrspace(3)* @llvm.amdgcn.kernel.k1.lds, i32 0, i32 1) to i8 addrspace(3)*
56; CHECK-NEXT:    store i8 4, i8 addrspace(3)* %lds.size.4.align.4.bc, align 16
57; CHECK-NEXT:    %lds.size.16.align.16.bc = bitcast [16 x i8] addrspace(3)* getelementptr inbounds (%llvm.amdgcn.kernel.k1.lds.t, %llvm.amdgcn.kernel.k1.lds.t addrspace(3)* @llvm.amdgcn.kernel.k1.lds, i32 0, i32 0) to i8 addrspace(3)*
58; CHECK-NEXT:    store i8 16, i8 addrspace(3)* %lds.size.16.align.16.bc, align 16
59; CHECK-NEXT:    ret void
60;
61  %lds.size.2.align.2.bc = bitcast [2 x i8] addrspace(3)* @lds.size.2.align.2 to i8 addrspace(3)*
62  store i8 2, i8 addrspace(3)* %lds.size.2.align.2.bc, align 2
63
64  %lds.size.4.align.4.bc = bitcast [4 x i8] addrspace(3)* @lds.size.4.align.4 to i8 addrspace(3)*
65  store i8 4, i8 addrspace(3)* %lds.size.4.align.4.bc, align 4
66
67  %lds.size.16.align.16.bc = bitcast [16 x i8] addrspace(3)* @lds.size.16.align.16 to i8 addrspace(3)*
68  store i8 16, i8 addrspace(3)* %lds.size.16.align.16.bc, align 16
69
70  ret void
71}
72
73define amdgpu_kernel void @0() #0 {
74; CHECK-LABEL: @0(
75; CHECK-NEXT:    %lds.size.2.align.2.bc = bitcast [2 x i8] addrspace(3)* getelementptr inbounds (%llvm.amdgcn.kernel..lds.t, %llvm.amdgcn.kernel..lds.t addrspace(3)* @llvm.amdgcn.kernel..lds, i32 0, i32 0) to i8 addrspace(3)*
76; CHECK-NEXT:    store i8 2, i8 addrspace(3)* %lds.size.2.align.2.bc, align 2
77; CHECK-NEXT:    ret void
78;
79  %lds.size.2.align.2.bc = bitcast [2 x i8] addrspace(3)* @lds.size.2.align.2 to i8 addrspace(3)*
80  store i8 2, i8 addrspace(3)* %lds.size.2.align.2.bc, align 2
81
82  ret void
83}
84
85define amdgpu_kernel void @1() #0 {
86; CHECK-LABEL: @1(
87; CHECK-NEXT:    %lds.size.4.align.4.bc = bitcast [4 x i8] addrspace(3)* getelementptr inbounds (%llvm.amdgcn.kernel..lds.t.0, %llvm.amdgcn.kernel..lds.t.0 addrspace(3)* @llvm.amdgcn.kernel..lds.1, i32 0, i32 0) to i8 addrspace(3)*
88; CHECK-NEXT:    store i8 4, i8 addrspace(3)* %lds.size.4.align.4.bc, align 4
89; CHECK-NEXT:    ret void
90;
91  %lds.size.4.align.4.bc = bitcast [4 x i8] addrspace(3)* @lds.size.4.align.4 to i8 addrspace(3)*
92  store i8 4, i8 addrspace(3)* %lds.size.4.align.4.bc, align 4
93
94  ret void
95}
96
97define void @f0() {
98; CHECK-LABEL: @f0(
99; CHECK-NEXT:    %lds.size.1.align.1.bc = bitcast [1 x i8] addrspace(3)* getelementptr inbounds (%llvm.amdgcn.module.lds.t, %llvm.amdgcn.module.lds.t addrspace(3)* @llvm.amdgcn.module.lds, i32 0, i32 1) to i8 addrspace(3)*
100; CHECK-NEXT:    store i8 1, i8 addrspace(3)* %lds.size.1.align.1.bc, align 8
101; CHECK-NEXT:    %lds.size.8.align.8.bc = bitcast [8 x i8] addrspace(3)* getelementptr inbounds (%llvm.amdgcn.module.lds.t, %llvm.amdgcn.module.lds.t addrspace(3)* @llvm.amdgcn.module.lds, i32 0, i32 0) to i8 addrspace(3)*
102; CHECK-NEXT:    store i8 8, i8 addrspace(3)* %lds.size.8.align.8.bc, align 8
103; CHECK-NEXT:    ret void
104;
105  %lds.size.1.align.1.bc = bitcast [1 x i8] addrspace(3)* @lds.size.1.align.1 to i8 addrspace(3)*
106  store i8 1, i8 addrspace(3)* %lds.size.1.align.1.bc, align 1
107
108  %lds.size.8.align.8.bc = bitcast [8 x i8] addrspace(3)* @lds.size.8.align.8 to i8 addrspace(3)*
109  store i8 8, i8 addrspace(3)* %lds.size.8.align.8.bc, align 4
110
111  ret void
112}
113
114attributes #0 = { "amdgpu-elide-module-lds" }
115; CHECK: attributes #0 = { "amdgpu-elide-module-lds" }
116