diff options
| author | Matt Arsenault <Matthew.Arsenault@amd.com> | 2016-04-27 21:05:08 +0000 |
|---|---|---|
| committer | Matt Arsenault <Matthew.Arsenault@amd.com> | 2016-04-27 21:05:08 +0000 |
| commit | 0547b016b18631db0f20855a509986650a3b4487 (patch) | |
| tree | d5430e74cf7eea99d5edd0e2858e82235664dedd /llvm/test | |
| parent | bc38ea9596d08fb7d5d4d9ac524ffd45595fba3c (diff) | |
| download | bcm5719-llvm-0547b016b18631db0f20855a509986650a3b4487.tar.gz bcm5719-llvm-0547b016b18631db0f20855a509986650a3b4487.zip | |
AMDGPU: Account for globals in AMDGPUPromoteAlloca pass
Patch by Bas Nieuwenhuizen
llvm-svn: 267791
Diffstat (limited to 'llvm/test')
| -rw-r--r-- | llvm/test/CodeGen/AMDGPU/promote-alloca-globals.ll | 32 |
1 files changed, 32 insertions, 0 deletions
diff --git a/llvm/test/CodeGen/AMDGPU/promote-alloca-globals.ll b/llvm/test/CodeGen/AMDGPU/promote-alloca-globals.ll new file mode 100644 index 00000000000..8cfe2df3038 --- /dev/null +++ b/llvm/test/CodeGen/AMDGPU/promote-alloca-globals.ll @@ -0,0 +1,32 @@ +; RUN: opt -S -mtriple=amdgcn-unknown-unknown -amdgpu-promote-alloca < %s | FileCheck -check-prefix=IR %s +; RUN: llc -march=amdgcn -mcpu=tonga < %s | FileCheck -check-prefix=ASM %s + + +@global_array= internal unnamed_addr addrspace(3) global [1500 x [10 x i32]] undef, align 4 + +; IR-LABEL: define void @promote_alloca_size_256(i32 addrspace(1)* nocapture %out, i32 addrspace(1)* nocapture %in) { +; IR: alloca [10 x i32] +; ASM-LABEL: {{^}}promote_alloca_size_256: +; ASM: ; LDSByteSize: 60000 bytes/workgroup (compile time only) + +define void @promote_alloca_size_256(i32 addrspace(1)* nocapture %out, i32 addrspace(1)* nocapture %in) { +entry: + %stack = alloca [10 x i32], align 4 + %tmp = load i32, i32 addrspace(1)* %in, align 4 + %arrayidx1 = getelementptr inbounds [10 x i32], [10 x i32]* %stack, i32 0, i32 %tmp + store i32 4, i32* %arrayidx1, align 4 + %arrayidx2 = getelementptr inbounds i32, i32 addrspace(1)* %in, i32 1 + %tmp1 = load i32, i32 addrspace(1)* %arrayidx2, align 4 + %arrayidx3 = getelementptr inbounds [10 x i32], [10 x i32]* %stack, i32 0, i32 %tmp1 + store i32 5, i32* %arrayidx3, align 4 + %arrayidx10 = getelementptr inbounds [10 x i32], [10 x i32]* %stack, i32 0, i32 0 + %tmp2 = load i32, i32* %arrayidx10, align 4 + store i32 %tmp2, i32 addrspace(1)* %out, align 4 + %arrayidx12 = getelementptr inbounds [10 x i32], [10 x i32]* %stack, i32 0, i32 1 + %tmp3 = load i32, i32* %arrayidx12 + %arrayidx13 = getelementptr inbounds i32, i32 addrspace(1)* %out, i32 1 + store i32 %tmp3, i32 addrspace(1)* %arrayidx13 + %v = getelementptr inbounds [1500 x [10 x i32]], [1500 x [10 x i32]] addrspace(3)* @global_array, i32 0, i32 0, i32 0 + store i32 %tmp3, i32 addrspace(3)* %v + ret void +} |
