1// RUN: mlir-translate -mlir-to-llvmir %s | FileCheck %s 2 3// The aim of the test is to check the GPU LLVM IR codegen 4// for nested omp do loop with collapse clause inside omp target region 5 6module attributes {dlti.dl_spec = #dlti.dl_spec<#dlti.dl_entry<"dlti.alloca_memory_space", 5 : ui32>>, llvm.data_layout = "e-p:64:64-p1:64:64-p2:32:32-p3:32:32-p4:64:64-p5:32:32-p6:32:32-p7:160:256:256:32-p8:128:128-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-G1-ni:7:8", llvm.target_triple = "amdgcn-amd-amdhsa", omp.is_gpu = true, omp.is_target_device = true } { 7 llvm.func @target_collapsed_wsloop(%arg0: !llvm.ptr) attributes {omp.declare_target = #omp.declaretarget<device_type = (any), capture_clause = (to)>} { 8 %loop_ub = llvm.mlir.constant(99 : i32) : i32 9 %loop_lb = llvm.mlir.constant(0 : i32) : i32 10 %loop_step = llvm.mlir.constant(1 : index) : i32 11 omp.wsloop { 12 omp.loop_nest (%arg1, %arg2) : i32 = (%loop_lb, %loop_lb) to (%loop_ub, %loop_ub) inclusive step (%loop_step, %loop_step) { 13 %1 = llvm.add %arg1, %arg2 : i32 14 %2 = llvm.mul %arg2, %loop_ub overflow<nsw> : i32 15 %3 = llvm.add %arg1, %2 :i32 16 %4 = llvm.getelementptr %arg0[%3] : (!llvm.ptr, i32) -> !llvm.ptr, i32 17 llvm.store %1, %4 : i32, !llvm.ptr 18 omp.yield 19 } 20 } 21 llvm.return 22 } 23} 24 25// CHECK: define void @[[FUNC_COLLAPSED_WSLOOP:.*]](ptr %[[ARG0:.*]]) 26// CHECK: call void @__kmpc_for_static_loop_4u(ptr addrspacecast (ptr addrspace(1) @[[GLOB2:[0-9]+]] to ptr), 27// CHECK-SAME: ptr @[[COLLAPSED_WSLOOP_BODY_FN:.*]], ptr %[[STRUCT_ARG:.*]], i32 10000, 28// CHECK-SAME: i32 %[[NUM_THREADS:.*]], i32 0) 29 30// CHECK: define internal void @[[COLLAPSED_WSLOOP_BODY_FN]](i32 %[[LOOP_CNT:.*]], ptr %[[LOOP_BODY_ARG:.*]]) 31// CHECK: %[[TMP0:.*]] = urem i32 %[[LOOP_CNT]], 100 32// CHECK: %[[TMP1:.*]] = udiv i32 %[[LOOP_CNT]], 100 33// CHECK: %[[TMP2:.*]] = mul i32 %[[TMP1]], 1 34// CHECK: %[[TMP3:.*]] = add i32 %[[TMP2]], 0 35// CHECK: %[[TMP4:.*]] = mul i32 %[[TMP0]], 1 36// CHECK: %[[TMP5:.*]] = add i32 %[[TMP4]], 0 37// CHECK: %[[TMP6:.*]] = add i32 %[[TMP3]], %[[TMP5]] 38// CHECK: %[[TMP7:.*]] = mul nsw i32 %[[TMP5]], 99 39// CHECK: %[[TMP8:.*]] = add i32 %[[TMP3]], %[[TMP7]] 40// CHECK: %[[TMP9:.*]] = getelementptr i32, ptr %[[ARRAY:.*]], i32 %[[TMP8]] 41// CHECK: store i32 %[[TMP6]], ptr %[[TMP9]], align 4 42