1; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=tonga -verify-machineinstrs < %s | FileCheck -enable-var-scope -check-prefix=GCN -check-prefix=UNPACKED %s 2; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx810 -verify-machineinstrs < %s | FileCheck -enable-var-scope -check-prefix=GCN -check-prefix=PACKED -check-prefix=GFX81 %s 3; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx900 -verify-machineinstrs < %s | FileCheck -enable-var-scope -check-prefix=GCN -check-prefix=PACKED -check-prefix=GFX9 %s 4 5 6; GCN-LABEL: {{^}}tbuffer_store_d16_x: 7; GCN: s_load_dwordx4 8; GCN: s_load_dword s[[S_LO:[0-9]+]] 9; GCN: v_mov_b32_e32 v[[V_LO:[0-9]+]], s[[S_LO]] 10; GCN: tbuffer_store_format_d16_x v[[V_LO]], off, s[{{[0-9]+:[0-9]+}}], dfmt:1, nfmt:2, 0 11define amdgpu_kernel void @tbuffer_store_d16_x(<4 x i32> %rsrc, half %data) { 12main_body: 13 call void @llvm.amdgcn.raw.tbuffer.store.f16(half %data, <4 x i32> %rsrc, i32 0, i32 0, i32 33, i32 0) 14 ret void 15} 16 17; GCN-LABEL: {{^}}tbuffer_store_d16_xy: 18; GCN: s_load_dword [[S_DATA:s[0-9]+]], s{{\[[0-9]+:[0-9]+\]}}, 0x10 19; UNPACKED-DAG: s_lshr_b32 [[SHR:s[0-9]+]], [[S_DATA]], 16 20; UNPACKED-DAG: s_and_b32 [[MASKED:s[0-9]+]], [[S_DATA]], 0xffff{{$}} 21; UNPACKED-DAG: v_mov_b32_e32 v[[V_LO:[0-9]+]], [[MASKED]] 22; UNPACKED-DAG: v_mov_b32_e32 v[[V_HI:[0-9]+]], [[SHR]] 23; UNPACKED: tbuffer_store_format_d16_xy v{{\[}}[[V_LO]]:[[V_HI]]{{\]}}, off, s[{{[0-9]+:[0-9]+}}], dfmt:1, nfmt:2, 0 24 25; PACKED: tbuffer_store_format_d16_xy v{{[0-9]+}}, off, s[{{[0-9]+:[0-9]+}}], dfmt:1, nfmt:2, 0 26define amdgpu_kernel void @tbuffer_store_d16_xy(<4 x i32> %rsrc, <2 x half> %data) { 27main_body: 28 call void @llvm.amdgcn.raw.tbuffer.store.v2f16(<2 x half> %data, <4 x i32> %rsrc, i32 0, i32 0, i32 33, i32 0) 29 ret void 30} 31 32; GCN-LABEL: {{^}}tbuffer_store_d16_xyzw: 33; GCN-DAG: s_load_dwordx2 s{{\[}}[[S_DATA_0:[0-9]+]]:[[S_DATA_1:[0-9]+]]{{\]}}, s{{\[[0-9]+:[0-9]+\]}}, 0x10 34 35; UNPACKED-DAG: s_mov_b32 [[K:s[0-9]+]], 0xffff{{$}} 36; UNPACKED-DAG: s_lshr_b32 [[SHR0:s[0-9]+]], s[[S_DATA_0]], 16 37; UNPACKED-DAG: s_and_b32 [[MASKED0:s[0-9]+]], s[[S_DATA_0]], [[K]] 38; UNPACKED-DAG: s_lshr_b32 [[SHR1:s[0-9]+]], s[[S_DATA_1]], 16 39; UNPACKED-DAG: s_and_b32 [[MASKED1:s[0-9]+]], s[[S_DATA_1]], [[K]] 40 41; UNPACKED-DAG: v_mov_b32_e32 v[[LO:[0-9]+]], [[MASKED0]] 42; UNPACKED-DAG: v_mov_b32_e32 v[[HI:[0-9]+]], [[SHR1]] 43; UNPACKED: tbuffer_store_format_d16_xyzw v{{\[}}[[LO]]:[[HI]]{{\]}}, off, s[{{[0-9]+:[0-9]+}}], dfmt:1, nfmt:2, 0 44 45 46; PACKED-DAG: v_mov_b32_e32 v[[LO:[0-9]+]], s[[S_DATA_0]] 47; PACKED-DAG: v_mov_b32_e32 v[[HI:[0-9]+]], s[[S_DATA_1]] 48; PACKED: tbuffer_store_format_d16_xyzw v{{\[}}[[LO]]:[[HI]]{{\]}}, off, s[{{[0-9]+:[0-9]+}}], dfmt:1, nfmt:2, 0 49define amdgpu_kernel void @tbuffer_store_d16_xyzw(<4 x i32> %rsrc, <4 x half> %data) { 50main_body: 51 call void @llvm.amdgcn.raw.tbuffer.store.v4f16(<4 x half> %data, <4 x i32> %rsrc, i32 0, i32 0, i32 33, i32 0) 52 ret void 53} 54 55declare void @llvm.amdgcn.raw.tbuffer.store.f16(half, <4 x i32>, i32, i32, i32, i32) 56declare void @llvm.amdgcn.raw.tbuffer.store.v2f16(<2 x half>, <4 x i32>, i32, i32, i32, i32) 57declare void @llvm.amdgcn.raw.tbuffer.store.v4f16(<4 x half>, <4 x i32>, i32, i32, i32, i32) 58