1 ; RUN: llc -amdgpu-scalarize-global-loads=false -mtriple=amdgcn -verify-machineinstrs < %s | FileCheck -check-prefix=SI %s
2 ; RUN: llc -amdgpu-scalarize-global-loads=false -mtriple=amdgcn -mcpu=tonga -mattr=-flat-for-global -verify-machineinstrs < %s | FileCheck -check-prefix=SI %s
4 declare double @llvm.amdgcn.trig.preop.f64(double, i32) nounwind readnone
6 ; SI-LABEL: {{^}}test_trig_preop_f64:
7 ; SI-DAG: buffer_load_dword [[SEG:v[0-9]+]]
8 ; SI-DAG: buffer_load_dwordx2 [[SRC:v\[[0-9]+:[0-9]+\]]],
9 ; SI: v_trig_preop_f64 [[RESULT:v\[[0-9]+:[0-9]+\]]], [[SRC]], [[SEG]]
10 ; SI: buffer_store_dwordx2 [[RESULT]],
12 define amdgpu_kernel void @test_trig_preop_f64(ptr addrspace(1) %out, ptr addrspace(1) %aptr, ptr addrspace(1) %bptr) nounwind {
13 %a = load double, ptr addrspace(1) %aptr, align 8
14 %b = load i32, ptr addrspace(1) %bptr, align 4
15 %result = call double @llvm.amdgcn.trig.preop.f64(double %a, i32 %b) nounwind readnone
16 store double %result, ptr addrspace(1) %out, align 8
20 ; SI-LABEL: {{^}}test_trig_preop_f64_imm_segment:
21 ; SI: buffer_load_dwordx2 [[SRC:v\[[0-9]+:[0-9]+\]]],
22 ; SI: v_trig_preop_f64 [[RESULT:v\[[0-9]+:[0-9]+\]]], [[SRC]], 7
23 ; SI: buffer_store_dwordx2 [[RESULT]],
25 define amdgpu_kernel void @test_trig_preop_f64_imm_segment(ptr addrspace(1) %out, ptr addrspace(1) %aptr) nounwind {
26 %a = load double, ptr addrspace(1) %aptr, align 8
27 %result = call double @llvm.amdgcn.trig.preop.f64(double %a, i32 7) nounwind readnone
28 store double %result, ptr addrspace(1) %out, align 8