diff options
Diffstat (limited to 'test/CodeGen/AMDGPU/llvm.amdgcn.mqsad.u32.u8.ll')
| -rw-r--r-- | test/CodeGen/AMDGPU/llvm.amdgcn.mqsad.u32.u8.ll | 57 |
1 files changed, 34 insertions, 23 deletions
diff --git a/test/CodeGen/AMDGPU/llvm.amdgcn.mqsad.u32.u8.ll b/test/CodeGen/AMDGPU/llvm.amdgcn.mqsad.u32.u8.ll index a8d03bf6bbac..685b5e0f29c4 100644 --- a/test/CodeGen/AMDGPU/llvm.amdgcn.mqsad.u32.u8.ll +++ b/test/CodeGen/AMDGPU/llvm.amdgcn.mqsad.u32.u8.ll @@ -3,45 +3,56 @@ declare <4 x i32> @llvm.amdgcn.mqsad.u32.u8(i64, i32, <4 x i32>) #0 -; GCN-LABEL: {{^}}v_mqsad_u32_u8_use_non_inline_constant: -; GCN: v_mqsad_u32_u8 v[{{[0-9]+:[0-9]+}}], s[{{[0-9]+:[0-9]+}}], v{{[0-9]+}}, v[{{[0-9]+:[0-9]+}}] -define amdgpu_kernel void @v_mqsad_u32_u8_use_non_inline_constant(<4 x i32> addrspace(1)* %out, i64 %src) { - %result = call <4 x i32> @llvm.amdgcn.mqsad.u32.u8(i64 %src, i32 100, <4 x i32> <i32 100, i32 100, i32 100, i32 100>) #0 - store <4 x i32> %result, <4 x i32> addrspace(1)* %out, align 4 +; GCN-LABEL: {{^}}v_mqsad_u32_u8_inline_integer_immediate: +; GCN-DAG: v_mov_b32_e32 v0, v2 +; GCN-DAG: v_mov_b32_e32 v1, v3 +; GCN: v_mqsad_u32_u8 v[2:5], v[0:1], v6, v[{{[0-9]+:[0-9]+}}] +define amdgpu_kernel void @v_mqsad_u32_u8_inline_integer_immediate(<4 x i32> addrspace(1)* %out, i64 %src, i32 %a) { + %tmp = call i64 asm "v_lsrlrev_b64 $0, $1, 1", "={v[2:3]},v"(i64 %src) #0 + %tmp1 = call i32 asm "v_mov_b32 $0, $1", "={v4},v"(i32 %a) #0 + %tmp2 = call <4 x i32> @llvm.amdgcn.mqsad.u32.u8(i64 %tmp, i32 %tmp1, <4 x i32> <i32 10, i32 20, i32 30, i32 40>) #0 + %tmp3 = call <4 x i32> asm ";; force constraint", "=v,{v[2:5]}"(<4 x i32> %tmp2) #0 + store <4 x i32> %tmp3, <4 x i32> addrspace(1)* %out, align 4 ret void } ; GCN-LABEL: {{^}}v_mqsad_u32_u8_non_immediate: -; GCN: v_mqsad_u32_u8 v[{{[0-9]+:[0-9]+}}], s[{{[0-9]+:[0-9]+}}], v{{[0-9]+}}, v[{{[0-9]+:[0-9]+}}] +; GCN-DAG: v_mov_b32_e32 v0, v2 +; GCN-DAG: v_mov_b32_e32 v1, v3 +; GCN: v_mqsad_u32_u8 v[2:5], v[0:1], v6, v[{{[0-9]+:[0-9]+}}] define amdgpu_kernel void @v_mqsad_u32_u8_non_immediate(<4 x i32> addrspace(1)* %out, i64 %src, i32 %a, <4 x i32> %b) { - %result = call <4 x i32> @llvm.amdgcn.mqsad.u32.u8(i64 %src, i32 %a, <4 x i32> %b) #0 - store <4 x i32> %result, <4 x i32> addrspace(1)* %out, align 4 - ret void -} - -; GCN-LABEL: {{^}}v_mqsad_u32_u8_inline_integer_immediate: -; GCN: v_mqsad_u32_u8 v[{{[0-9]+:[0-9]+}}], s[{{[0-9]+:[0-9]+}}], v{{[0-9]+}}, v[{{[0-9]+:[0-9]+}}] -define amdgpu_kernel void @v_mqsad_u32_u8_inline_integer_immediate(<4 x i32> addrspace(1)* %out, i64 %src, i32 %a) { - %result = call <4 x i32> @llvm.amdgcn.mqsad.u32.u8(i64 %src, i32 %a, <4 x i32> <i32 10, i32 20, i32 30, i32 40>) #0 - store <4 x i32> %result, <4 x i32> addrspace(1)* %out, align 4 + %tmp = call i64 asm "v_lsrlrev_b64 $0, $1, 1", "={v[2:3]},v"(i64 %src) #0 + %tmp1 = call i32 asm "v_mov_b32 $0, $1", "={v4},v"(i32 %a) #0 + %tmp2 = call <4 x i32> @llvm.amdgcn.mqsad.u32.u8(i64 %tmp, i32 %tmp1, <4 x i32> %b) #0 + %tmp3 = call <4 x i32> asm ";; force constraint", "=v,{v[2:5]}"(<4 x i32> %tmp2) #0 + store <4 x i32> %tmp3, <4 x i32> addrspace(1)* %out, align 4 ret void } ; GCN-LABEL: {{^}}v_mqsad_u32_u8_inline_fp_immediate: -; GCN: v_mqsad_u32_u8 v[{{[0-9]+:[0-9]+}}], s[{{[0-9]+:[0-9]+}}], v{{[0-9]+}}, v[{{[0-9]+:[0-9]+}}] +; GCN-DAG: v_mov_b32_e32 v0, v2 +; GCN-DAG: v_mov_b32_e32 v1, v3 +; GCN: v_mqsad_u32_u8 v[2:5], v[0:1], v6, v[{{[0-9]+:[0-9]+}}] define amdgpu_kernel void @v_mqsad_u32_u8_inline_fp_immediate(<4 x i32> addrspace(1)* %out, i64 %src, i32 %a) { - %result = call <4 x i32> @llvm.amdgcn.mqsad.u32.u8(i64 %src, i32 %a, <4 x i32> <i32 1065353216, i32 0, i32 0, i32 0>) #0 - store <4 x i32> %result, <4 x i32> addrspace(1)* %out, align 4 + %tmp = call i64 asm "v_lsrlrev_b64 $0, $1, 1", "={v[2:3]},v"(i64 %src) #0 + %tmp1 = call i32 asm "v_mov_b32 $0, $1", "={v4},v"(i32 %a) #0 + %tmp2 = call <4 x i32> @llvm.amdgcn.mqsad.u32.u8(i64 %tmp, i32 %tmp1, <4 x i32> <i32 1065353216, i32 0, i32 0, i32 0>) #0 + %tmp3 = call <4 x i32> asm ";; force constraint", "=v,{v[2:5]}"(<4 x i32> %tmp2) #0 + store <4 x i32> %tmp3, <4 x i32> addrspace(1)* %out, align 4 ret void } ; GCN-LABEL: {{^}}v_mqsad_u32_u8_use_sgpr_vgpr: -; GCN: v_mqsad_u32_u8 v[{{[0-9]+:[0-9]+}}], s[{{[0-9]+:[0-9]+}}], v{{[0-9]+}}, v[{{[0-9]+:[0-9]+}}] +; GCN-DAG: v_mov_b32_e32 v0, v2 +; GCN-DAG: v_mov_b32_e32 v1, v3 +; GCN: v_mqsad_u32_u8 v[2:5], v[0:1], v6, v[{{[0-9]+:[0-9]+}}] define amdgpu_kernel void @v_mqsad_u32_u8_use_sgpr_vgpr(<4 x i32> addrspace(1)* %out, i64 %src, i32 %a, <4 x i32> addrspace(1)* %input) { %in = load <4 x i32>, <4 x i32> addrspace(1) * %input - - %result = call <4 x i32> @llvm.amdgcn.mqsad.u32.u8(i64 %src, i32 %a, <4 x i32> %in) #0 - store <4 x i32> %result, <4 x i32> addrspace(1)* %out, align 4 + %tmp = call i64 asm "v_lsrlrev_b64 $0, $1, 1", "={v[2:3]},v"(i64 %src) #0 + %tmp1 = call i32 asm "v_mov_b32 $0, $1", "={v4},v"(i32 %a) #0 + %tmp2 = call <4 x i32> @llvm.amdgcn.mqsad.u32.u8(i64 %tmp, i32 %tmp1, <4 x i32> %in) #0 + %tmp3 = call <4 x i32> asm ";; force constraint", "=v,{v[2:5]}"(<4 x i32> %tmp2) #0 + store <4 x i32> %tmp3, <4 x i32> addrspace(1)* %out, align 4 ret void } |