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() { 25; CHECK-LABEL: @k0( 26; CHECK-NEXT: call void @llvm.donothing() [ "ExplicitUse"(%llvm.amdgcn.module.lds.t addrspace(3)* @llvm.amdgcn.module.lds) ] 27; 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)* 28; CHECK-NEXT: store i8 1, i8 addrspace(3)* %lds.size.1.align.1.bc, align 8 29; 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)* 30; CHECK-NEXT: store i8 2, i8 addrspace(3)* %lds.size.2.align.2.bc, align 4 31; 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)* 32; CHECK-NEXT: store i8 4, i8 addrspace(3)* %lds.size.4.align.4.bc, align 16 33; 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)* 34; CHECK-NEXT: store i8 16, i8 addrspace(3)* %lds.size.16.align.16.bc, align 16 35; CHECK-NEXT: ret void 36; 37 %lds.size.1.align.1.bc = bitcast [1 x i8] addrspace(3)* @lds.size.1.align.1 to i8 addrspace(3)* 38 store i8 1, i8 addrspace(3)* %lds.size.1.align.1.bc, align 1 39 40 %lds.size.2.align.2.bc = bitcast [2 x i8] addrspace(3)* @lds.size.2.align.2 to i8 addrspace(3)* 41 store i8 2, i8 addrspace(3)* %lds.size.2.align.2.bc, align 2 42 43 %lds.size.4.align.4.bc = bitcast [4 x i8] addrspace(3)* @lds.size.4.align.4 to i8 addrspace(3)* 44 store i8 4, i8 addrspace(3)* %lds.size.4.align.4.bc, align 4 45 46 %lds.size.16.align.16.bc = bitcast [16 x i8] addrspace(3)* @lds.size.16.align.16 to i8 addrspace(3)* 47 store i8 16, i8 addrspace(3)* %lds.size.16.align.16.bc, align 16 48 49 ret void 50} 51 52define amdgpu_kernel void @k1() { 53; CHECK-LABEL: @k1( 54; CHECK-NEXT: call void @llvm.donothing() [ "ExplicitUse"(%llvm.amdgcn.module.lds.t addrspace(3)* @llvm.amdgcn.module.lds) ] 55; 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)* 56; CHECK-NEXT: store i8 2, i8 addrspace(3)* %lds.size.2.align.2.bc, align 4 57; 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)* 58; CHECK-NEXT: store i8 4, i8 addrspace(3)* %lds.size.4.align.4.bc, align 16 59; 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)* 60; CHECK-NEXT: store i8 16, i8 addrspace(3)* %lds.size.16.align.16.bc, align 16 61; CHECK-NEXT: ret void 62; 63 %lds.size.2.align.2.bc = bitcast [2 x i8] addrspace(3)* @lds.size.2.align.2 to i8 addrspace(3)* 64 store i8 2, i8 addrspace(3)* %lds.size.2.align.2.bc, align 2 65 66 %lds.size.4.align.4.bc = bitcast [4 x i8] addrspace(3)* @lds.size.4.align.4 to i8 addrspace(3)* 67 store i8 4, i8 addrspace(3)* %lds.size.4.align.4.bc, align 4 68 69 %lds.size.16.align.16.bc = bitcast [16 x i8] addrspace(3)* @lds.size.16.align.16 to i8 addrspace(3)* 70 store i8 16, i8 addrspace(3)* %lds.size.16.align.16.bc, align 16 71 72 ret void 73} 74 75define amdgpu_kernel void @0() { 76; CHECK-LABEL: @0( 77; CHECK-NEXT: call void @llvm.donothing() [ "ExplicitUse"(%llvm.amdgcn.module.lds.t addrspace(3)* @llvm.amdgcn.module.lds) ] 78; 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)* 79; CHECK-NEXT: store i8 2, i8 addrspace(3)* %lds.size.2.align.2.bc, align 2 80; CHECK-NEXT: ret void 81; 82 %lds.size.2.align.2.bc = bitcast [2 x i8] addrspace(3)* @lds.size.2.align.2 to i8 addrspace(3)* 83 store i8 2, i8 addrspace(3)* %lds.size.2.align.2.bc, align 2 84 85 ret void 86} 87 88define amdgpu_kernel void @1() { 89; CHECK-LABEL: @1( 90; CHECK-NEXT: call void @llvm.donothing() [ "ExplicitUse"(%llvm.amdgcn.module.lds.t addrspace(3)* @llvm.amdgcn.module.lds) ] 91; 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)* 92; CHECK-NEXT: store i8 4, i8 addrspace(3)* %lds.size.4.align.4.bc, align 4 93; CHECK-NEXT: ret void 94; 95 %lds.size.4.align.4.bc = bitcast [4 x i8] addrspace(3)* @lds.size.4.align.4 to i8 addrspace(3)* 96 store i8 4, i8 addrspace(3)* %lds.size.4.align.4.bc, align 4 97 98 ret void 99} 100 101define void @f0() { 102; CHECK-LABEL: @f0( 103; 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)* 104; CHECK-NEXT: store i8 1, i8 addrspace(3)* %lds.size.1.align.1.bc, align 8 105; 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)* 106; CHECK-NEXT: store i8 8, i8 addrspace(3)* %lds.size.8.align.8.bc, align 8 107; CHECK-NEXT: ret void 108; 109 %lds.size.1.align.1.bc = bitcast [1 x i8] addrspace(3)* @lds.size.1.align.1 to i8 addrspace(3)* 110 store i8 1, i8 addrspace(3)* %lds.size.1.align.1.bc, align 1 111 112 %lds.size.8.align.8.bc = bitcast [8 x i8] addrspace(3)* @lds.size.8.align.8 to i8 addrspace(3)* 113 store i8 8, i8 addrspace(3)* %lds.size.8.align.8.bc, align 4 114 115 ret void 116} 117;. 118; CHECK: attributes #0 = { nofree nosync nounwind readnone willreturn } 119;. 120