diff options
Diffstat (limited to 'llvm/test/CodeGen/AMDGPU/max.i16.ll')
-rw-r--r-- | llvm/test/CodeGen/AMDGPU/max.i16.ll | 16 |
1 files changed, 8 insertions, 8 deletions
diff --git a/llvm/test/CodeGen/AMDGPU/max.i16.ll b/llvm/test/CodeGen/AMDGPU/max.i16.ll index d442c8754b7..8fcf85b826f 100644 --- a/llvm/test/CodeGen/AMDGPU/max.i16.ll +++ b/llvm/test/CodeGen/AMDGPU/max.i16.ll @@ -4,7 +4,7 @@ ; FIXME: Need to handle non-uniform case for function below (load without gep). ; GCN-LABEL: {{^}}v_test_imax_sge_i16: ; VIPLUS: v_max_i16_e32 v{{[0-9]+}}, v{{[0-9]+}}, v{{[0-9]+}} -define void @v_test_imax_sge_i16(i16 addrspace(1)* %out, i16 addrspace(1)* %aptr, i16 addrspace(1)* %bptr) nounwind { +define amdgpu_kernel void @v_test_imax_sge_i16(i16 addrspace(1)* %out, i16 addrspace(1)* %aptr, i16 addrspace(1)* %bptr) nounwind { %tid = call i32 @llvm.amdgcn.workitem.id.x() nounwind readnone %gep0 = getelementptr i16, i16 addrspace(1)* %aptr, i32 %tid %gep1 = getelementptr i16, i16 addrspace(1)* %bptr, i32 %tid @@ -23,7 +23,7 @@ define void @v_test_imax_sge_i16(i16 addrspace(1)* %out, i16 addrspace(1)* %aptr ; VI: v_max_i16_e32 v{{[0-9]+}}, v{{[0-9]+}}, v{{[0-9]+}} ; GFX9: v_pk_max_i16 v{{[0-9]+}}, v{{[0-9]+}}, v{{[0-9]+}} -define void @v_test_imax_sge_v2i16(<2 x i16> addrspace(1)* %out, <2 x i16> addrspace(1)* %aptr, <2 x i16> addrspace(1)* %bptr) nounwind { +define amdgpu_kernel void @v_test_imax_sge_v2i16(<2 x i16> addrspace(1)* %out, <2 x i16> addrspace(1)* %aptr, <2 x i16> addrspace(1)* %bptr) nounwind { %tid = call i32 @llvm.amdgcn.workitem.id.x() nounwind readnone %gep0 = getelementptr <2 x i16>, <2 x i16> addrspace(1)* %aptr, i32 %tid %gep1 = getelementptr <2 x i16>, <2 x i16> addrspace(1)* %bptr, i32 %tid @@ -45,7 +45,7 @@ define void @v_test_imax_sge_v2i16(<2 x i16> addrspace(1)* %out, <2 x i16> addrs ; GFX9: v_pk_max_i16 v{{[0-9]+}}, v{{[0-9]+}}, v{{[0-9]+}} ; GFX9: v_pk_max_i16 v{{[0-9]+}}, v{{[0-9]+}}, v{{[0-9]+}} -define void @v_test_imax_sge_v3i16(<3 x i16> addrspace(1)* %out, <3 x i16> addrspace(1)* %aptr, <3 x i16> addrspace(1)* %bptr) nounwind { +define amdgpu_kernel void @v_test_imax_sge_v3i16(<3 x i16> addrspace(1)* %out, <3 x i16> addrspace(1)* %aptr, <3 x i16> addrspace(1)* %bptr) nounwind { %tid = call i32 @llvm.amdgcn.workitem.id.x() nounwind readnone %gep0 = getelementptr <3 x i16>, <3 x i16> addrspace(1)* %aptr, i32 %tid %gep1 = getelementptr <3 x i16>, <3 x i16> addrspace(1)* %bptr, i32 %tid @@ -67,7 +67,7 @@ define void @v_test_imax_sge_v3i16(<3 x i16> addrspace(1)* %out, <3 x i16> addrs ; GFX9: v_pk_max_i16 v{{[0-9]+}}, v{{[0-9]+}}, v{{[0-9]+}} ; GFX9: v_pk_max_i16 v{{[0-9]+}}, v{{[0-9]+}}, v{{[0-9]+}} -define void @v_test_imax_sge_v4i16(<4 x i16> addrspace(1)* %out, <4 x i16> addrspace(1)* %aptr, <4 x i16> addrspace(1)* %bptr) nounwind { +define amdgpu_kernel void @v_test_imax_sge_v4i16(<4 x i16> addrspace(1)* %out, <4 x i16> addrspace(1)* %aptr, <4 x i16> addrspace(1)* %bptr) nounwind { %tid = call i32 @llvm.amdgcn.workitem.id.x() nounwind readnone %gep0 = getelementptr <4 x i16>, <4 x i16> addrspace(1)* %aptr, i32 %tid %gep1 = getelementptr <4 x i16>, <4 x i16> addrspace(1)* %bptr, i32 %tid @@ -83,7 +83,7 @@ define void @v_test_imax_sge_v4i16(<4 x i16> addrspace(1)* %out, <4 x i16> addrs ; FIXME: Need to handle non-uniform case for function below (load without gep). ; GCN-LABEL: {{^}}v_test_imax_sgt_i16: ; VIPLUS: v_max_i16_e32 -define void @v_test_imax_sgt_i16(i16 addrspace(1)* %out, i16 addrspace(1)* %aptr, i16 addrspace(1)* %bptr) nounwind { +define amdgpu_kernel void @v_test_imax_sgt_i16(i16 addrspace(1)* %out, i16 addrspace(1)* %aptr, i16 addrspace(1)* %bptr) nounwind { %tid = call i32 @llvm.amdgcn.workitem.id.x() nounwind readnone %gep0 = getelementptr i16, i16 addrspace(1)* %aptr, i32 %tid %gep1 = getelementptr i16, i16 addrspace(1)* %bptr, i32 %tid @@ -99,7 +99,7 @@ define void @v_test_imax_sgt_i16(i16 addrspace(1)* %out, i16 addrspace(1)* %aptr ; FIXME: Need to handle non-uniform case for function below (load without gep). ; GCN-LABEL: {{^}}v_test_umax_uge_i16: ; VIPLUS: v_max_u16_e32 -define void @v_test_umax_uge_i16(i16 addrspace(1)* %out, i16 addrspace(1)* %aptr, i16 addrspace(1)* %bptr) nounwind { +define amdgpu_kernel void @v_test_umax_uge_i16(i16 addrspace(1)* %out, i16 addrspace(1)* %aptr, i16 addrspace(1)* %bptr) nounwind { %tid = call i32 @llvm.amdgcn.workitem.id.x() nounwind readnone %gep0 = getelementptr i16, i16 addrspace(1)* %aptr, i32 %tid %gep1 = getelementptr i16, i16 addrspace(1)* %bptr, i32 %tid @@ -115,7 +115,7 @@ define void @v_test_umax_uge_i16(i16 addrspace(1)* %out, i16 addrspace(1)* %aptr ; FIXME: Need to handle non-uniform case for function below (load without gep). ; GCN-LABEL: {{^}}v_test_umax_ugt_i16: ; VIPLUS: v_max_u16_e32 -define void @v_test_umax_ugt_i16(i16 addrspace(1)* %out, i16 addrspace(1)* %aptr, i16 addrspace(1)* %bptr) nounwind { +define amdgpu_kernel void @v_test_umax_ugt_i16(i16 addrspace(1)* %out, i16 addrspace(1)* %aptr, i16 addrspace(1)* %bptr) nounwind { %tid = call i32 @llvm.amdgcn.workitem.id.x() nounwind readnone %gep0 = getelementptr i16, i16 addrspace(1)* %aptr, i32 %tid %gep1 = getelementptr i16, i16 addrspace(1)* %bptr, i32 %tid @@ -133,7 +133,7 @@ define void @v_test_umax_ugt_i16(i16 addrspace(1)* %out, i16 addrspace(1)* %aptr ; VI: v_max_u16_e32 ; GFX9: v_pk_max_u16 v{{[0-9]+}}, v{{[0-9]+}}, v{{[0-9]+}} -define void @v_test_umax_ugt_v2i16(<2 x i16> addrspace(1)* %out, <2 x i16> addrspace(1)* %aptr, <2 x i16> addrspace(1)* %bptr) nounwind { +define amdgpu_kernel void @v_test_umax_ugt_v2i16(<2 x i16> addrspace(1)* %out, <2 x i16> addrspace(1)* %aptr, <2 x i16> addrspace(1)* %bptr) nounwind { %tid = call i32 @llvm.amdgcn.workitem.id.x() nounwind readnone %gep0 = getelementptr <2 x i16>, <2 x i16> addrspace(1)* %aptr, i32 %tid %gep1 = getelementptr <2 x i16>, <2 x i16> addrspace(1)* %bptr, i32 %tid |