diff options
Diffstat (limited to 'llvm/test/CodeGen/AMDGPU/uint_to_fp.i64.ll')
| -rw-r--r-- | llvm/test/CodeGen/AMDGPU/uint_to_fp.i64.ll | 16 |
1 files changed, 8 insertions, 8 deletions
diff --git a/llvm/test/CodeGen/AMDGPU/uint_to_fp.i64.ll b/llvm/test/CodeGen/AMDGPU/uint_to_fp.i64.ll index cd816b27fce..4168326e14c 100644 --- a/llvm/test/CodeGen/AMDGPU/uint_to_fp.i64.ll +++ b/llvm/test/CodeGen/AMDGPU/uint_to_fp.i64.ll @@ -4,7 +4,7 @@ ; FIXME: This should be merged with uint_to_fp.ll, but s_uint_to_fp_v2i64 crashes on r600 ; FUNC-LABEL: {{^}}s_uint_to_fp_i64_to_f16: -define void @s_uint_to_fp_i64_to_f16(half addrspace(1)* %out, i64 %in) #0 { +define amdgpu_kernel void @s_uint_to_fp_i64_to_f16(half addrspace(1)* %out, i64 %in) #0 { %result = uitofp i64 %in to half store half %result, half addrspace(1)* %out ret void @@ -24,7 +24,7 @@ define void @s_uint_to_fp_i64_to_f16(half addrspace(1)* %out, i64 %in) #0 { ; GCN: v_add_i32_e32 [[VR:v[0-9]+]] ; GCN: v_cvt_f16_f32_e32 [[VR_F16:v[0-9]+]], [[VR]] ; GCN: {{buffer|flat}}_store_short {{.*}}[[VR_F16]] -define void @v_uint_to_fp_i64_to_f16(half addrspace(1)* %out, i64 addrspace(1)* %in) #0 { +define amdgpu_kernel void @v_uint_to_fp_i64_to_f16(half addrspace(1)* %out, i64 addrspace(1)* %in) #0 { %tid = call i32 @llvm.r600.read.tidig.x() %in.gep = getelementptr i64, i64 addrspace(1)* %in, i32 %tid %out.gep = getelementptr half, half addrspace(1)* %out, i32 %tid @@ -35,7 +35,7 @@ define void @v_uint_to_fp_i64_to_f16(half addrspace(1)* %out, i64 addrspace(1)* } ; FUNC-LABEL: {{^}}s_uint_to_fp_i64_to_f32: -define void @s_uint_to_fp_i64_to_f32(float addrspace(1)* %out, i64 %in) #0 { +define amdgpu_kernel void @s_uint_to_fp_i64_to_f32(float addrspace(1)* %out, i64 %in) #0 { %result = uitofp i64 %in to float store float %result, float addrspace(1)* %out ret void @@ -54,7 +54,7 @@ define void @s_uint_to_fp_i64_to_f32(float addrspace(1)* %out, i64 %in) #0 { ; GCN: v_add_i32_e32 [[VR:v[0-9]+]] ; GCN: {{buffer|flat}}_store_dword {{.*}}[[VR]] -define void @v_uint_to_fp_i64_to_f32(float addrspace(1)* %out, i64 addrspace(1)* %in) #0 { +define amdgpu_kernel void @v_uint_to_fp_i64_to_f32(float addrspace(1)* %out, i64 addrspace(1)* %in) #0 { %tid = call i32 @llvm.r600.read.tidig.x() %in.gep = getelementptr i64, i64 addrspace(1)* %in, i32 %tid %out.gep = getelementptr float, float addrspace(1)* %out, i32 %tid @@ -65,14 +65,14 @@ define void @v_uint_to_fp_i64_to_f32(float addrspace(1)* %out, i64 addrspace(1)* } ; FUNC-LABEL: {{^}}s_uint_to_fp_v2i64_to_v2f32: -define void @s_uint_to_fp_v2i64_to_v2f32(<2 x float> addrspace(1)* %out, <2 x i64> %in) #0{ +define amdgpu_kernel void @s_uint_to_fp_v2i64_to_v2f32(<2 x float> addrspace(1)* %out, <2 x i64> %in) #0{ %result = uitofp <2 x i64> %in to <2 x float> store <2 x float> %result, <2 x float> addrspace(1)* %out ret void } ; FUNC-LABEL: {{^}}v_uint_to_fp_v4i64_to_v4f32: -define void @v_uint_to_fp_v4i64_to_v4f32(<4 x float> addrspace(1)* %out, <4 x i64> addrspace(1)* %in) #0 { +define amdgpu_kernel void @v_uint_to_fp_v4i64_to_v4f32(<4 x float> addrspace(1)* %out, <4 x i64> addrspace(1)* %in) #0 { %tid = call i32 @llvm.r600.read.tidig.x() %in.gep = getelementptr <4 x i64>, <4 x i64> addrspace(1)* %in, i32 %tid %out.gep = getelementptr <4 x float>, <4 x float> addrspace(1)* %out, i32 %tid @@ -83,14 +83,14 @@ define void @v_uint_to_fp_v4i64_to_v4f32(<4 x float> addrspace(1)* %out, <4 x i6 } ; FUNC-LABEL: {{^}}s_uint_to_fp_v2i64_to_v2f16: -define void @s_uint_to_fp_v2i64_to_v2f16(<2 x half> addrspace(1)* %out, <2 x i64> %in) #0{ +define amdgpu_kernel void @s_uint_to_fp_v2i64_to_v2f16(<2 x half> addrspace(1)* %out, <2 x i64> %in) #0{ %result = uitofp <2 x i64> %in to <2 x half> store <2 x half> %result, <2 x half> addrspace(1)* %out ret void } ; FUNC-LABEL: {{^}}v_uint_to_fp_v4i64_to_v4f16: -define void @v_uint_to_fp_v4i64_to_v4f16(<4 x half> addrspace(1)* %out, <4 x i64> addrspace(1)* %in) #0 { +define amdgpu_kernel void @v_uint_to_fp_v4i64_to_v4f16(<4 x half> addrspace(1)* %out, <4 x i64> addrspace(1)* %in) #0 { %tid = call i32 @llvm.r600.read.tidig.x() %in.gep = getelementptr <4 x i64>, <4 x i64> addrspace(1)* %in, i32 %tid %out.gep = getelementptr <4 x half>, <4 x half> addrspace(1)* %out, i32 %tid |
