1; RUN: llc < %s -march=amdgcn -mcpu=tonga -verify-machineinstrs -show-mc-encoding | FileCheck -enable-var-scope -check-prefix=GCN -check-prefix=UNPACKED %s
2; RUN: llc < %s -march=amdgcn -mcpu=gfx810 -verify-machineinstrs | FileCheck -enable-var-scope -check-prefix=GCN -check-prefix=PACKED -check-prefix=GFX81 %s
3; RUN: llc < %s -march=amdgcn -mcpu=gfx900 -verify-machineinstrs | FileCheck -enable-var-scope -check-prefix=GCN -check-prefix=PACKED -check-prefix=GFX9 %s
4
5; GCN-LABEL: {{^}}buffer_store_format_d16_x:
6; GCN: v_trunc_f16_e32 v[[LO:[0-9]+]], s{{[0-9]+}}
7; GCN: buffer_store_format_d16_x v[[LO]], v{{[0-9]+}}, s[{{[0-9]+:[0-9]+}}], 0 idxen
8define amdgpu_kernel void @buffer_store_format_d16_x(<4 x i32> %rsrc, half %data, i32 %index) {
9main_body:
10  call void @llvm.amdgcn.buffer.store.format.f16(half %data, <4 x i32> %rsrc, i32 %index, i32 0, i1 0, i1 0)
11  ret void
12}
13
14; GCN-LABEL: {{^}}buffer_store_format_d16_xy:
15
16; UNPACKED: flat_load_ushort v[[HI:[0-9]+]], v[{{[0-9]+:[0-9]+}}] glc slc
17; UNPACKED: flat_load_ushort v[[LO:[0-9]+]], v[{{[0-9]+:[0-9]+}}] glc slc
18; UNPACKED: buffer_store_format_d16_xy v{{\[}}[[LO]]:[[HI]]{{\]}}, v{{[0-9]+}}, s[{{[0-9]+:[0-9]+}}], 0 idxen
19
20; PACKED: buffer_store_format_d16_xy v{{[0-9]+}}, v{{[0-9]+}}, s[{{[0-9]+:[0-9]+}}], 0 idxen
21define amdgpu_kernel void @buffer_store_format_d16_xy(<4 x i32> %rsrc, <2 x half> %data, i32 %index) {
22main_body:
23  call void @llvm.amdgcn.buffer.store.format.v2f16(<2 x half> %data, <4 x i32> %rsrc, i32 %index, i32 0, i1 0, i1 0)
24  ret void
25}
26
27; GCN-LABEL: {{^}}buffer_store_format_d16_xyzw:
28
29; UNPACKED: flat_load_ushort v[[HI:[0-9]+]], v[{{[0-9]+:[0-9]+}}] glc slc
30; UNPACKED: flat_load_ushort v{{[0-9]+}}, v[{{[0-9]+:[0-9]+}}] glc slc
31; UNPACKED: flat_load_ushort v{{[0-9]+}}, v[{{[0-9]+:[0-9]+}}] glc slc
32; UNPACKED: flat_load_ushort v[[LO:[0-9]+]], v[{{[0-9]+:[0-9]+}}] glc slc
33; UNPACKED: buffer_store_format_d16_xyzw v{{\[}}[[LO]]:[[HI]]{{\]}}, v{{[0-9]+}}, s[{{[0-9]+:[0-9]+}}], 0 idxen
34
35; GFX81: v_or_b32_e32 v[[HI:[0-9]+]]
36; GFX81: v_or_b32_e32 v[[LO:[0-9]+]]
37
38; GFX9: v_mov_b32_e32 v[[LO:[0-9]+]]
39; GFX9: v_mov_b32_e32 v[[HI:[0-9]+]]
40
41; PACKED: buffer_store_format_d16_xyzw v{{\[}}[[LO]]:[[HI]]{{\]}}, v{{[0-9]+}}, s[{{[0-9]+:[0-9]+}}], 0 idxen
42define amdgpu_kernel void @buffer_store_format_d16_xyzw(<4 x i32> %rsrc, <4 x half> %data, i32 %index) {
43main_body:
44  call void @llvm.amdgcn.buffer.store.format.v4f16(<4 x half> %data, <4 x i32> %rsrc, i32 %index, i32 0, i1 0, i1 0)
45  ret void
46}
47
48declare void @llvm.amdgcn.buffer.store.format.f16(half, <4 x i32>, i32, i32, i1, i1)
49declare void @llvm.amdgcn.buffer.store.format.v2f16(<2 x half>, <4 x i32>, i32, i32, i1, i1)
50declare void @llvm.amdgcn.buffer.store.format.v4f16(<4 x half>, <4 x i32>, i32, i32, i1, i1)
51