target datalayout = "e-p:64:64-p1:64:64-p2:32:32-p3:32:32-p4:64:64-p5:32:32-p6:32:32-i64:64-v16:16-v24:32-v32:32-v48:64-v96:128-v192:256-v256:256-v512:512-v1024:1024-v2048:2048-n32:64-S32-A5"
declare void @llvm.amdgcn.s.barrier() #0
declare i32 @llvm.amdgcn.workitem.id.x() #1

define void @__OpenCL_execFFT_reduced_kernel(<2 x float> addrspace(1)* noalias nocapture %in, <2 x float> addrspace(1)* noalias nocapture %out, <2 x float> addrspace(3)* nocapture %data0) #2 {
entry:
  %id = call i32 @llvm.amdgcn.workitem.id.x()
  %arrayidx = getelementptr <2 x float>, <2 x float> addrspace(3)* %data0, i32 %id
  %arrayidx3 = getelementptr <2 x float>, <2 x float> addrspace(1)* %in, i32 %id
  %tmp4 = load <2 x float>, <2 x float> addrspace(1)* %arrayidx3, align 8
  store <2 x float> %tmp4, <2 x float> addrspace(3)* %arrayidx, align 8
  tail call void @llvm.amdgcn.s.barrier() #0
  %cmp = icmp ult i32 %id, 5
  br i1 %cmp, label %if.then, label %if.end

if.then:                                          ; preds = %entry
  %tmp10 = load <2 x float>, <2 x float> addrspace(3)* %arrayidx, align 8
  %tmp16 = mul i32 %id, 2
  %arrayidx17 = getelementptr <2 x float>, <2 x float> addrspace(3)* %data0, i32 %tmp16
  store <2 x float> %tmp10, <2 x float> addrspace(3)* %arrayidx17, align 8
  br label %if.end

if.end:                                           ; preds = %if.then, %entry
  tail call void @llvm.amdgcn.s.barrier() #0
  %arrayidx21 = getelementptr <2 x float>, <2 x float> addrspace(1)* %out, i32 %id
  %tmp25 = load <2 x float>, <2 x float> addrspace(3)* %arrayidx, align 8
  store <2 x float> %tmp25, <2 x float> addrspace(1)* %arrayidx21, align 8
  ret void
}

attributes #0 = { convergent nounwind }
attributes #1 = { nounwind readnone speculatable }
attributes #2 = { nounwind }