flang/test/Lower/Intrinsics/transpose_opt.f90

   1 ! RUN: bbc -emit-fir -hlfir=false %s -opt-transpose=true -o - | FileCheck %s
   2 ! RUN: bbc -emit-fir -hlfir=false %s -o - | FileCheck %s
   3 ! RUN: %flang_fc1 -emit-fir -flang-deprecated-no-hlfir -O1 %s -o - | FileCheck %s
   4 ! RUN: %flang_fc1 -emit-fir -flang-deprecated-no-hlfir -O2 %s -o - | FileCheck %s
   5 ! RUN: %flang_fc1 -emit-fir -flang-deprecated-no-hlfir -O3 %s -o - | FileCheck %s
   6
   7 ! CHECK-LABEL: func.func @_QPtranspose_test(
   8 ! CHECK-SAME:                               %[[VAL_0:.*]]: !fir.ref<!fir.array<2x3xf32>> {fir.bindc_name = "mat"}) {
   9 subroutine transpose_test(mat)
  10    real :: mat(2,3)
  11    call bar_transpose_test(transpose(mat))
  12 ! CHECK:         %[[VAL_6:.*]] = fir.array_load %[[VAL_0]](%{{.*}}) : (!fir.ref<!fir.array<2x3xf32>>, !fir.shape<2>) -> !fir.array<2x3xf32>
  13 ! CHECK:         %[[VAL_7:.*]] = fir.allocmem !fir.array<3x2xf32>
  14 ! CHECK:         %[[VAL_9:.*]] = fir.array_load %[[VAL_7]](%{{.*}}) : (!fir.heap<!fir.array<3x2xf32>>, !fir.shape<2>) -> !fir.array<3x2xf32>
  15 ! CHECK:         %[[VAL_14:.*]] = fir.do_loop %[[VAL_15:.*]] = %{{.*}} to %{{.*}} step %{{.*}} unordered iter_args(%[[VAL_16:.*]] = %[[VAL_9]]) -> (!fir.array<3x2xf32>) {
  16 ! CHECK:           %[[VAL_17:.*]] = fir.do_loop %[[VAL_18:.*]] = %{{.*}} to %{{.*}} step %{{.*}} unordered iter_args(%[[VAL_19:.*]] = %[[VAL_16]]) -> (!fir.array<3x2xf32>) {
  17 ! CHECK:             %[[VAL_20:.*]] = fir.array_fetch %[[VAL_6]], %[[VAL_15]], %[[VAL_18]] : (!fir.array<2x3xf32>, index, index) -> f32
  18 ! CHECK:             %[[VAL_21:.*]] = fir.array_update %[[VAL_19]], %[[VAL_20]], %[[VAL_18]], %[[VAL_15]] : (!fir.array<3x2xf32>, f32, index, index) -> !fir.array<3x2xf32>
  19 ! CHECK:             fir.result %[[VAL_21]] : !fir.array<3x2xf32>
  20 ! CHECK:           }
  21 ! CHECK:           fir.result %[[VAL_17]] : !fir.array<3x2xf32>
  22 ! CHECK:         }
  23 ! CHECK:         fir.array_merge_store %[[VAL_9]], %[[VAL_14]] to %[[VAL_7]] : !fir.array<3x2xf32>, !fir.array<3x2xf32>, !fir.heap<!fir.array<3x2xf32>>
  24 ! CHECK:         %[[VAL_24:.*]] = fir.convert %[[VAL_7]] : (!fir.heap<!fir.array<3x2xf32>>) -> !fir.ref<!fir.array<3x2xf32>>
  25 ! CHECK:         fir.call @_QPbar_transpose_test(%[[VAL_24]]) {{.*}}: (!fir.ref<!fir.array<3x2xf32>>) -> ()
  26 ! CHECK:         fir.freemem %[[VAL_7]] : !fir.heap<!fir.array<3x2xf32>>
  27
  28 ! CHECK-NOT: @_FortranATranspose
  29 end subroutine
  30
  31 ! CHECK-LABEL: func.func @_QPtranspose_allocatable_test(
  32 ! CHECK-SAME:                                           %[[VAL_0:.*]]: !fir.ref<!fir.box<!fir.heap<!fir.array<?x?xf32>>>> {fir.bindc_name = "mat"}) {
  33 subroutine transpose_allocatable_test(mat)
  34   real, allocatable :: mat(:,:)
  35   mat = transpose(mat)
  36 ! Verify that the "optimized" TRANSPOSE loops are generated
  37 ! three times in each branch checking the status of LHS allocatable.
  38
  39 ! CHECK:         %[[VAL_1:.*]] = fir.load %[[VAL_0]] : !fir.ref<!fir.box<!fir.heap<!fir.array<?x?xf32>>>>
  40 ! CHECK:         %[[VAL_6:.*]] = fir.box_addr %[[VAL_1]] : (!fir.box<!fir.heap<!fir.array<?x?xf32>>>) -> !fir.heap<!fir.array<?x?xf32>>
  41 ! CHECK:         %[[VAL_8:.*]] = fir.array_load %[[VAL_6]](%{{.*}}) : (!fir.heap<!fir.array<?x?xf32>>, !fir.shapeshift<2>) -> !fir.array<?x?xf32>
  42
  43 ! CHECK:         %[[VAL_9:.*]] = fir.load %[[VAL_0]] : !fir.ref<!fir.box<!fir.heap<!fir.array<?x?xf32>>>>
  44 ! CHECK:         %[[VAL_10:.*]] = fir.box_addr %[[VAL_9]] : (!fir.box<!fir.heap<!fir.array<?x?xf32>>>) -> !fir.heap<!fir.array<?x?xf32>>
  45
  46 ! CHECK:         %[[VAL_14:.*]]:2 = fir.if %{{.*}} -> (i1, !fir.heap<!fir.array<?x?xf32>>) {
  47
  48 ! CHECK:           %[[VAL_24:.*]] = fir.if %{{.*}} -> (!fir.heap<!fir.array<?x?xf32>>) {
  49
  50 ! CHECK:             %[[VAL_25:.*]] = fir.allocmem !fir.array<?x?xf32>
  51 ! CHECK:             %[[VAL_27:.*]] = fir.array_load %[[VAL_25]](%{{.*}}) : (!fir.heap<!fir.array<?x?xf32>>, !fir.shape<2>) -> !fir.array<?x?xf32>
  52
  53 ! CHECK:             %[[VAL_32:.*]] = fir.do_loop %[[VAL_33:.*]] = %{{.*}} to %{{.*}} step %{{.*}} unordered iter_args(%[[VAL_34:.*]] = %[[VAL_27]]) -> (!fir.array<?x?xf32>) {
  54 ! CHECK:               %[[VAL_35:.*]] = fir.do_loop %[[VAL_36:.*]] = %{{.*}} to %{{.*}} step %{{.*}} unordered iter_args(%[[VAL_37:.*]] = %[[VAL_34]]) -> (!fir.array<?x?xf32>) {
  55 ! CHECK:                 %[[VAL_38:.*]] = fir.array_fetch %[[VAL_8]], %[[VAL_33]], %[[VAL_36]] : (!fir.array<?x?xf32>, index, index) -> f32
  56 ! CHECK:                 %[[VAL_39:.*]] = fir.array_update %[[VAL_37]], %[[VAL_38]], %[[VAL_36]], %[[VAL_33]] : (!fir.array<?x?xf32>, f32, index, index) -> !fir.array<?x?xf32>
  57 ! CHECK:                 fir.result %[[VAL_39]] : !fir.array<?x?xf32>
  58 ! CHECK:               }
  59 ! CHECK:               fir.result %[[VAL_35]] : !fir.array<?x?xf32>
  60 ! CHECK:             }
  61 ! CHECK:             fir.array_merge_store %[[VAL_27]], %[[VAL_32]] to %[[VAL_25]] : !fir.array<?x?xf32>, !fir.array<?x?xf32>, !fir.heap<!fir.array<?x?xf32>>
  62
  63 ! CHECK:           } else {
  64
  65 ! CHECK:             %[[VAL_43:.*]] = fir.array_load %[[VAL_10]](%{{.*}}) : (!fir.heap<!fir.array<?x?xf32>>, !fir.shape<2>) -> !fir.array<?x?xf32>
  66
  67 ! CHECK:             %[[VAL_48:.*]] = fir.do_loop %[[VAL_49:.*]] = %{{.*}} to %{{.*}} step %{{.*}} unordered iter_args(%[[VAL_50:.*]] = %[[VAL_43]]) -> (!fir.array<?x?xf32>) {
  68 ! CHECK:               %[[VAL_51:.*]] = fir.do_loop %[[VAL_52:.*]] = %{{.*}} to %{{.*}} step %{{.*}} unordered iter_args(%[[VAL_53:.*]] = %[[VAL_50]]) -> (!fir.array<?x?xf32>) {
  69 ! CHECK:                 %[[VAL_54:.*]] = fir.array_fetch %[[VAL_8]], %[[VAL_49]], %[[VAL_52]] : (!fir.array<?x?xf32>, index, index) -> f32
  70 ! CHECK:                 %[[VAL_55:.*]] = fir.array_update %[[VAL_53]], %[[VAL_54]], %[[VAL_52]], %[[VAL_49]] : (!fir.array<?x?xf32>, f32, index, index) -> !fir.array<?x?xf32>
  71 ! CHECK:                 fir.result %[[VAL_55]] : !fir.array<?x?xf32>
  72 ! CHECK:               }
  73 ! CHECK:               fir.result %[[VAL_51]] : !fir.array<?x?xf32>
  74 ! CHECK:             }
  75 ! CHECK:             fir.array_merge_store %[[VAL_43]], %[[VAL_48]] to %[[VAL_10]] : !fir.array<?x?xf32>, !fir.array<?x?xf32>, !fir.heap<!fir.array<?x?xf32>>
  76
  77 ! CHECK:             fir.result %[[VAL_10]] : !fir.heap<!fir.array<?x?xf32>>
  78 ! CHECK:           }
  79
  80 ! CHECK:         } else {
  81
  82 ! CHECK:           %[[VAL_60:.*]] = fir.allocmem !fir.array<?x?xf32>
  83 ! CHECK:           %[[VAL_62:.*]] = fir.array_load %[[VAL_60]](%{{.*}}) : (!fir.heap<!fir.array<?x?xf32>>, !fir.shape<2>) -> !fir.array<?x?xf32>
  84
  85 ! CHECK:           %[[VAL_67:.*]] = fir.do_loop %[[VAL_68:.*]] = %{{.*}} to %{{.*}} step %{{.*}} unordered iter_args(%[[VAL_69:.*]] = %[[VAL_62]]) -> (!fir.array<?x?xf32>) {
  86 ! CHECK:             %[[VAL_70:.*]] = fir.do_loop %[[VAL_71:.*]] = %{{.*}} to %{{.*}} step %{{.*}} unordered iter_args(%[[VAL_72:.*]] = %[[VAL_69]]) -> (!fir.array<?x?xf32>) {
  87 ! CHECK:               %[[VAL_73:.*]] = fir.array_fetch %[[VAL_8]], %[[VAL_68]], %[[VAL_71]] : (!fir.array<?x?xf32>, index, index) -> f32
  88 ! CHECK:               %[[VAL_74:.*]] = fir.array_update %[[VAL_72]], %[[VAL_73]], %[[VAL_71]], %[[VAL_68]] : (!fir.array<?x?xf32>, f32, index, index) -> !fir.array<?x?xf32>
  89 ! CHECK:               fir.result %[[VAL_74]] : !fir.array<?x?xf32>
  90 ! CHECK:             }
  91 ! CHECK:             fir.result %[[VAL_70]] : !fir.array<?x?xf32>
  92 ! CHECK:           }
  93 ! CHECK:           fir.array_merge_store %[[VAL_62]], %[[VAL_67]] to %[[VAL_60]] : !fir.array<?x?xf32>, !fir.array<?x?xf32>, !fir.heap<!fir.array<?x?xf32>>
  94
  95 ! CHECK:         }
  96
  97 ! CHECK-NOT: @_FortranATranspose
  98 end subroutine