diff options
Diffstat (limited to 'test/CodeGen/AMDGPU/add.ll')
| -rw-r--r-- | test/CodeGen/AMDGPU/add.ll | 83 |
1 files changed, 22 insertions, 61 deletions
diff --git a/test/CodeGen/AMDGPU/add.ll b/test/CodeGen/AMDGPU/add.ll index 6a108db879cc7..bce924ec4a081 100644 --- a/test/CodeGen/AMDGPU/add.ll +++ b/test/CodeGen/AMDGPU/add.ll @@ -1,11 +1,8 @@ ; RUN: llc -march=amdgcn -mcpu=verde -verify-machineinstrs < %s | FileCheck -enable-var-scope -check-prefixes=GCN,SIVI,FUNC %s ; RUN: llc -march=amdgcn -mcpu=tonga -mattr=-flat-for-global -verify-machineinstrs < %s | FileCheck -enable-var-scope -check-prefixes=GCN,SIVI,FUNC %s ; RUN: llc -march=amdgcn -mcpu=gfx900 -mattr=-flat-for-global -verify-machineinstrs < %s | FileCheck -enable-var-scope -check-prefixes=GCN,GFX9,FUNC %s -; RUN: llc -march=r600 -mcpu=redwood < %s | FileCheck -enable-var-scope -check-prefix=EG -check-prefix=FUNC %s ; FUNC-LABEL: {{^}}s_add_i32: -; EG: ADD_INT {{[* ]*}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}} - ; GCN: s_add_i32 s[[REG:[0-9]+]], {{s[0-9]+, s[0-9]+}} ; GCN: v_mov_b32_e32 v[[V_REG:[0-9]+]], s[[REG]] ; GCN: buffer_store_dword v[[V_REG]], @@ -19,9 +16,6 @@ define amdgpu_kernel void @s_add_i32(i32 addrspace(1)* %out, i32 addrspace(1)* % } ; FUNC-LABEL: {{^}}s_add_v2i32: -; EG: ADD_INT {{[* ]*}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}} -; EG: ADD_INT {{[* ]*}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}} - ; GCN: s_add_i32 s{{[0-9]+, s[0-9]+, s[0-9]+}} ; GCN: s_add_i32 s{{[0-9]+, s[0-9]+, s[0-9]+}} define amdgpu_kernel void @s_add_v2i32(<2 x i32> addrspace(1)* %out, <2 x i32> addrspace(1)* %in) { @@ -34,11 +28,6 @@ define amdgpu_kernel void @s_add_v2i32(<2 x i32> addrspace(1)* %out, <2 x i32> a } ; FUNC-LABEL: {{^}}s_add_v4i32: -; EG: ADD_INT {{[* ]*}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}} -; EG: ADD_INT {{[* ]*}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}} -; EG: ADD_INT {{[* ]*}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}} -; EG: ADD_INT {{[* ]*}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}} - ; GCN: s_add_i32 s{{[0-9]+, s[0-9]+, s[0-9]+}} ; GCN: s_add_i32 s{{[0-9]+, s[0-9]+, s[0-9]+}} ; GCN: s_add_i32 s{{[0-9]+, s[0-9]+, s[0-9]+}} @@ -53,15 +42,6 @@ define amdgpu_kernel void @s_add_v4i32(<4 x i32> addrspace(1)* %out, <4 x i32> a } ; FUNC-LABEL: {{^}}s_add_v8i32: -; EG: ADD_INT -; EG: ADD_INT -; EG: ADD_INT -; EG: ADD_INT -; EG: ADD_INT -; EG: ADD_INT -; EG: ADD_INT -; EG: ADD_INT - ; GCN: s_add_i32 ; GCN: s_add_i32 ; GCN: s_add_i32 @@ -78,23 +58,6 @@ entry: } ; FUNC-LABEL: {{^}}s_add_v16i32: -; EG: ADD_INT -; EG: ADD_INT -; EG: ADD_INT -; EG: ADD_INT -; EG: ADD_INT -; EG: ADD_INT -; EG: ADD_INT -; EG: ADD_INT -; EG: ADD_INT -; EG: ADD_INT -; EG: ADD_INT -; EG: ADD_INT -; EG: ADD_INT -; EG: ADD_INT -; EG: ADD_INT -; EG: ADD_INT - ; GCN: s_add_i32 ; GCN: s_add_i32 ; GCN: s_add_i32 @@ -124,7 +87,7 @@ entry: ; SIVI: v_add_{{i|u}}32_e32 v{{[0-9]+}}, vcc, [[A]], [[B]] ; GFX9: v_add_u32_e32 v{{[0-9]+}}, [[A]], [[B]] define amdgpu_kernel void @v_add_i32(i32 addrspace(1)* %out, i32 addrspace(1)* %in) #0 { - %tid = call i32 @llvm.r600.read.tidig.x() + %tid = call i32 @llvm.amdgcn.workitem.id.x() %gep = getelementptr inbounds i32, i32 addrspace(1)* %in, i32 %tid %b_ptr = getelementptr i32, i32 addrspace(1)* %gep, i32 1 %a = load volatile i32, i32 addrspace(1)* %gep @@ -139,7 +102,7 @@ define amdgpu_kernel void @v_add_i32(i32 addrspace(1)* %out, i32 addrspace(1)* % ; SIVI: v_add_{{i|u}}32_e32 v{{[0-9]+}}, vcc, 0x7b, [[A]] ; GFX9: v_add_u32_e32 v{{[0-9]+}}, 0x7b, [[A]] define amdgpu_kernel void @v_add_imm_i32(i32 addrspace(1)* %out, i32 addrspace(1)* %in) #0 { - %tid = call i32 @llvm.r600.read.tidig.x() + %tid = call i32 @llvm.amdgcn.workitem.id.x() %gep = getelementptr inbounds i32, i32 addrspace(1)* %in, i32 %tid %b_ptr = getelementptr i32, i32 addrspace(1)* %gep, i32 1 %a = load volatile i32, i32 addrspace(1)* %gep @@ -151,13 +114,6 @@ define amdgpu_kernel void @v_add_imm_i32(i32 addrspace(1)* %out, i32 addrspace(1 ; FUNC-LABEL: {{^}}add64: ; GCN: s_add_u32 ; GCN: s_addc_u32 - -; EG: MEM_RAT_CACHELESS STORE_RAW [[LO:T[0-9]+\.XY]] -; EG-DAG: ADD_INT {{[* ]*}} -; EG-DAG: ADDC_UINT -; EG-DAG: ADD_INT -; EG-DAG: ADD_INT {{[* ]*}} -; EG-NOT: SUB define amdgpu_kernel void @add64(i64 addrspace(1)* %out, i64 %a, i64 %b) { entry: %add = add i64 %a, %b @@ -172,13 +128,6 @@ entry: ; FUNC-LABEL: {{^}}add64_sgpr_vgpr: ; GCN-NOT: v_addc_u32_e32 s - -; EG: MEM_RAT_CACHELESS STORE_RAW [[LO:T[0-9]+\.XY]] -; EG-DAG: ADD_INT {{[* ]*}} -; EG-DAG: ADDC_UINT -; EG-DAG: ADD_INT -; EG-DAG: ADD_INT {{[* ]*}} -; EG-NOT: SUB define amdgpu_kernel void @add64_sgpr_vgpr(i64 addrspace(1)* %out, i64 %a, i64 addrspace(1)* %in) { entry: %0 = load i64, i64 addrspace(1)* %in @@ -191,13 +140,6 @@ entry: ; FUNC-LABEL: {{^}}add64_in_branch: ; GCN: s_add_u32 ; GCN: s_addc_u32 - -; EG: MEM_RAT_CACHELESS STORE_RAW [[LO:T[0-9]+\.XY]] -; EG-DAG: ADD_INT {{[* ]*}} -; EG-DAG: ADDC_UINT -; EG-DAG: ADD_INT -; EG-DAG: ADD_INT {{[* ]*}} -; EG-NOT: SUB define amdgpu_kernel void @add64_in_branch(i64 addrspace(1)* %out, i64 addrspace(1)* %in, i64 %a, i64 %b, i64 %c) { entry: %0 = icmp eq i64 %a, 0 @@ -217,7 +159,26 @@ endif: ret void } -declare i32 @llvm.r600.read.tidig.x() #1 +; Make sure the VOP3 form of add is initially selected. Otherwise pair +; of opies from/to VCC would be necessary + +; GCN-LABEL: {{^}}add_select_vop3: +; SI: v_add_i32_e64 v0, s[0:1], s0, v0 +; VI: v_add_u32_e64 v0, s[0:1], s0, v0 +; GFX9: v_add_u32_e32 v0, s0, v0 + +; GCN: ; def vcc +; GCN: ds_write_b32 +; GCN: ; use vcc +define amdgpu_ps void @add_select_vop3(i32 inreg %s, i32 %v) { + %vcc = call i64 asm sideeffect "; def vcc", "={vcc}"() + %sub = add i32 %v, %s + store i32 %sub, i32 addrspace(3)* undef + call void asm sideeffect "; use vcc", "{vcc}"(i64 %vcc) + ret void +} + +declare i32 @llvm.amdgcn.workitem.id.x() #1 attributes #0 = { nounwind } attributes #1 = { nounwind readnone speculatable } |