[flang][MLIR][OpenMP] Extend delayed privatization for arrays and characters

flang/include/flang/Optimizer/Builder/HLFIRTools.h

@@ -223,7 +223,7 @@ class EntityWithAttributes : public Entity {
 using CleanupFunction = std::function<void()>;
 std::pair<fir::ExtendedValue, std::optional<CleanupFunction>>
 translateToExtendedValue(mlir::Location loc, fir::FirOpBuilder &builder,
-                         Entity entity);
+                         Entity entity, bool contiguousHint = false);
 
 /// Function to translate FortranVariableOpInterface to fir::ExtendedValue.
 /// It may generates IR to unbox fir.boxchar, but has otherwise no side effects

flang/lib/Lower/OpenMP/DataSharingProcessor.cpp

@@ -15,9 +15,11 @@
 #include "Utils.h"
 #include "flang/Lower/PFTBuilder.h"
 #include "flang/Lower/SymbolMap.h"
+#include "flang/Optimizer/Builder/HLFIRTools.h"
 #include "flang/Optimizer/Builder/Todo.h"
 #include "flang/Semantics/tools.h"
 #include "mlir/Dialect/OpenMP/OpenMPDialect.h"
+#include <variant>
 
 namespace Fortran {
 namespace lower {
@@ -396,8 +398,15 @@ void DataSharingProcessor::doPrivatize(
           &allocRegion, /*insertPt=*/{}, symType, symLoc);
 
       firOpBuilder.setInsertionPointToEnd(allocEntryBlock);
-      symTable->addSymbol(*sym,
-                          fir::substBase(symExV, allocRegion.getArgument(0)));
+
+      fir::ExtendedValue localExV =
+          hlfir::translateToExtendedValue(
+              symLoc, firOpBuilder, hlfir::Entity{allocRegion.getArgument(0)},
+              /*contiguousHint=*/
+              std::holds_alternative<fir::ArrayBoxValue>(symExV.matchee()))
+              .first;
+
+      symTable->addSymbol(*sym, localExV);
       symTable->pushScope();
       cloneSymbol(sym);
       firOpBuilder.create<mlir::omp::YieldOp>(
@@ -414,12 +423,23 @@ void DataSharingProcessor::doPrivatize(
       mlir::Block *copyEntryBlock = firOpBuilder.createBlock(
           &copyRegion, /*insertPt=*/{}, {symType, symType}, {symLoc, symLoc});
       firOpBuilder.setInsertionPointToEnd(copyEntryBlock);
-      symTable->addSymbol(*sym,
-                          fir::substBase(symExV, copyRegion.getArgument(0)),
-                          /*force=*/true);
+
+      auto addSymbol = [&](unsigned argIdx, bool force = false) {
+        symExV.match(
+            [&](const fir::MutableBoxValue &box) {
+              symTable->addSymbol(
+                  *sym, fir::substBase(box, copyRegion.getArgument(argIdx)),
+                  force);
+            },
+            [&](const auto &box) {
+              symTable->addSymbol(*sym, copyRegion.getArgument(argIdx), force);
+            });
+      };
+
+      addSymbol(0, true);
       symTable->pushScope();
-      symTable->addSymbol(*sym,
-                          fir::substBase(symExV, copyRegion.getArgument(1)));
+      addSymbol(1);
+
       auto ip = firOpBuilder.saveInsertionPoint();
       copyFirstPrivateSymbol(sym, &ip);
 

flang/lib/Lower/OpenMP/OpenMP.cpp

@@ -1427,13 +1427,18 @@ genParallelOp(Fortran::lower::AbstractConverter &converter,
         reductionSyms;
     allSymbols.append(privateSyms);
     for (auto [arg, prv] : llvm::zip_equal(allSymbols, region.getArguments())) {
-      converter.bindSymbol(*arg, prv);
+      fir::ExtendedValue hostExV = converter.getSymbolExtendedValue(*arg);
+      converter.bindSymbol(*arg, hlfir::translateToExtendedValue(
+                                     loc, firOpBuilder, hlfir::Entity{prv},
+                                     /*contiguousHint=*/
+                                     std::holds_alternative<fir::ArrayBoxValue>(
+                                         hostExV.matchee()))
+                                     .first);
     }
 
     return allSymbols;
   };
 
-  // TODO Merge with the reduction CB.
   genInfo.setGenRegionEntryCb(genRegionEntryCB).setDataSharingProcessor(&dsp);
   return genOpWithBody<mlir::omp::ParallelOp>(genInfo, clauseOps);
 }

flang/lib/Optimizer/Builder/HLFIRTools.cpp

@@ -844,10 +844,9 @@ hlfir::LoopNest hlfir::genLoopNest(mlir::Location loc,
   return loopNest;
 }
 
-static fir::ExtendedValue
-translateVariableToExtendedValue(mlir::Location loc, fir::FirOpBuilder &builder,
-                                 hlfir::Entity variable,
-                                 bool forceHlfirBase = false) {
+static fir::ExtendedValue translateVariableToExtendedValue(
+    mlir::Location loc, fir::FirOpBuilder &builder, hlfir::Entity variable,
+    bool forceHlfirBase = false, bool contiguousHint = false) {
   assert(variable.isVariable() && "must be a variable");
   /// When going towards FIR, use the original base value to avoid
   /// introducing descriptors at runtime when they are not required.
@@ -858,7 +857,8 @@ translateVariableToExtendedValue(mlir::Location loc, fir::FirOpBuilder &builder,
                                 fir::MutableProperties{});
 
   if (mlir::isa<fir::BaseBoxType>(base.getType())) {
-    if (!variable.isSimplyContiguous() || variable.isPolymorphic() ||
+    bool contiguous = variable.isSimplyContiguous() || contiguousHint;
+    if (!contiguous || variable.isPolymorphic() ||
         variable.isDerivedWithLengthParameters() || variable.isOptional()) {
       llvm::SmallVector<mlir::Value> nonDefaultLbounds =
           getNonDefaultLowerBounds(loc, builder, variable);
@@ -907,9 +907,10 @@ hlfir::translateToExtendedValue(mlir::Location loc, fir::FirOpBuilder &builder,
 
 std::pair<fir::ExtendedValue, std::optional<hlfir::CleanupFunction>>
 hlfir::translateToExtendedValue(mlir::Location loc, fir::FirOpBuilder &builder,
-                                hlfir::Entity entity) {
+                                hlfir::Entity entity, bool contiguousHint) {
   if (entity.isVariable())
-    return {translateVariableToExtendedValue(loc, builder, entity),
+    return {translateVariableToExtendedValue(loc, builder, entity, false,
+                                             contiguousHint),
             std::nullopt};
 
   if (entity.isProcedure()) {

flang/test/Lower/OpenMP/delayed-privatization-allocatable-array.f90

@@ -0,0 +1,67 @@
+! Test delayed privatization for allocatable arrays.
+
+! RUN: %flang_fc1 -emit-hlfir -fopenmp -mmlir --openmp-enable-delayed-privatization \
+! RUN:   -o - %s 2>&1 | FileCheck %s
+! RUN: bbc -emit-hlfir -fopenmp --openmp-enable-delayed-privatization -o - %s 2>&1 |\
+! RUN:   FileCheck %s
+
+subroutine delayed_privatization_private(var1, l1)
+  implicit none
+  integer(8):: l1
+  integer, allocatable, dimension(:) :: var1
+
+!$omp parallel firstprivate(var1)
+  var1(l1 + 1) = 10
+!$omp end parallel
+end subroutine
+
+! CHECK-LABEL: omp.private {type = firstprivate}
+! CHECK-SAME: @[[PRIVATIZER_SYM:.*]] : [[TYPE:!fir.ref<!fir.box<!fir.heap<!fir.array<\?xi32>>>>]] alloc {
+
+! CHECK-NEXT: ^bb0(%[[PRIV_ARG:.*]]: [[TYPE]]):
+! CHECK-NEXT:   %[[PRIV_ALLOC:.*]] = fir.alloca !fir.box<!fir.heap<!fir.array<{{\?}}xi32>>> {bindc_name = "var1", pinned, uniq_name = "_QFdelayed_privatization_privateEvar1"}
+
+! CHECK-NEXT:   %[[PRIV_ARG_VAL:.*]] = fir.load %[[PRIV_ARG]]
+! CHECK-NEXT:   %[[PRIV_ARG_BOX:.*]] = fir.box_addr %[[PRIV_ARG_VAL]]
+! CHECK-NEXT:   %[[PRIV_ARG_ADDR:.*]] = fir.convert %[[PRIV_ARG_BOX]]
+! CHECK-NEXT:   %[[C0:.*]] = arith.constant 0 : i64
+! CHECK-NEXT:   %[[ALLOC_COND:.*]] = arith.cmpi ne, %[[PRIV_ARG_ADDR]], %[[C0]] : i64
+
+! CHECK-NEXT:   fir.if %[[ALLOC_COND]] {
+! CHECK-NEXT:     %[[PRIV_ARG_VAL:.*]] = fir.load %[[PRIV_ARG]] : [[TYPE]]
+! CHECK-NEXT:     %[[C0:.*]] = arith.constant 0 : index
+! CHECK-NEXT:     %[[DIMS:.*]]:3 = fir.box_dims %[[PRIV_ARG_VAL]], %[[C0]]
+! CHECK-NEXT:     fir.box_addr %[[PRIV_ARG_VAL]]
+! CHECK-NEXT:     %[[C0_2:.*]] = arith.constant 0 : index 
+! CHECK-NEXT:     %[[CMP:.*]] = arith.cmpi sgt, %[[DIMS]]#1, %[[C0_2]] : index
+! CHECK-NEXT:     %[[SELECT:.*]] = arith.select %[[CMP]], %[[DIMS]]#1, %[[C0_2]] : index
+! CHECK-NEXT:     %[[MEM:.*]] = fir.allocmem !fir.array<?xi32>, %[[SELECT]]
+! CHECK-NEXT:     %[[SHAPE_SHIFT:.*]] = fir.shape_shift %[[DIMS]]#0, %[[SELECT]] : (index, index) -> !fir.shapeshift<1>
+! CHECK-NEXT:     %[[EMBOX:.*]] = fir.embox %[[MEM]](%[[SHAPE_SHIFT]])
+! CHECK-NEXT:     fir.store %[[EMBOX]] to %[[PRIV_ALLOC]]
+! CHECK-NEXT:   } else {
+! CHECK-NEXT:     %[[ZEROS:.*]] = fir.zero_bits
+! CHECK-NEXT:     %[[C0_3:.*]] = arith.constant 0 : index
+! CHECK-NEXT:     %[[SHAPE:.*]] = fir.shape %[[C0_3]] : (index) -> !fir.shape<1>
+! CHECK-NEXT:     %[[EMBOX_2:.*]] = fir.embox %[[ZEROS]](%[[SHAPE]])
+! CHECK-NEXT:     fir.store %[[EMBOX_2]] to %[[PRIV_ALLOC]]
+! CHECK-NEXT:   }
+
+! CHECK-NEXT:   hlfir.declare
+! CHECK-NEXT:   omp.yield
+
+! CHECK-NEXT: } copy {
+! CHECK-NEXT: ^bb0(%[[PRIV_ORIG_ARG:.*]]: [[TYPE]], %[[PRIV_PRIV_ARG:.*]]: [[TYPE]]):
+! CHECK-NEXT:  %[[PRIV_BASE_VAL:.*]] = fir.load %[[PRIV_PRIV_ARG]]
+! CHECK-NEXT:  %[[PRIV_BASE_BOX:.*]] = fir.box_addr %[[PRIV_BASE_VAL]]
+! CHECK-NEXT:  %[[PRIV_BASE_ADDR:.*]] = fir.convert %[[PRIV_BASE_BOX]]
+! CHECK-NEXT:  %[[C0:.*]] = arith.constant 0 : i64
+! CHECK-NEXT:  %[[COPY_COND:.*]] = arith.cmpi ne, %[[PRIV_BASE_ADDR]], %[[C0]] : i64
+
+
+! CHECK-NEXT:  fir.if %[[COPY_COND]] {
+! CHECK-NEXT:    %[[PRIV_ORIG_ARG_VAL:.*]] = fir.load %[[PRIV_ORIG_ARG]]
+! CHECK-NEXT:    hlfir.assign %[[PRIV_ORIG_ARG_VAL]] to %[[PRIV_BASE_VAL]] temporary_lhs
+! CHECK-NEXT:   }
+! CHECK-NEXT:   omp.yield
+! CHECK-NEXT: }

flang/test/Lower/OpenMP/delayed-privatization-array.f90

@@ -0,0 +1,100 @@
+! Test delayed privatization for arrays.
+
+! RUN: split-file %s %t
+
+! RUN: %flang_fc1 -emit-hlfir -fopenmp -mmlir --openmp-enable-delayed-privatization \
+! RUN:   -o - %t/one_dim_array.f90 2>&1 | FileCheck %s --check-prefix=ONE_DIM
+! RUN: bbc -emit-hlfir -fopenmp --openmp-enable-delayed-privatization -o - \
+! RUN:   %t/one_dim_array.f90 2>&1 | FileCheck %s --check-prefix=ONE_DIM
+
+! RUN: %flang_fc1 -emit-hlfir -fopenmp -mmlir --openmp-enable-delayed-privatization \
+! RUN:   -o - %t/two_dim_array.f90 2>&1 | FileCheck %s --check-prefix=TWO_DIM
+! RUN: bbc -emit-hlfir -fopenmp --openmp-enable-delayed-privatization -o - \
+! RUN:   %t/two_dim_array.f90 2>&1 | FileCheck %s --check-prefix=TWO_DIM
+
+! RUN: %flang_fc1 -emit-hlfir -fopenmp -mmlir --openmp-enable-delayed-privatization \
+! RUN:   -o - %t/one_dim_array_default_lb.f90 2>&1 | FileCheck %s --check-prefix=ONE_DIM_DEFAULT_LB
+! RUN: bbc -emit-hlfir -fopenmp --openmp-enable-delayed-privatization -o - \
+! RUN:   %t/one_dim_array_default_lb.f90 2>&1 | FileCheck %s --check-prefix=ONE_DIM_DEFAULT_LB
+
+!--- one_dim_array.f90
+subroutine delayed_privatization_private_1d(var1, l1, u1)
+  implicit none
+  integer(8):: l1, u1
+  integer, dimension(l1:u1) :: var1
+
+!$omp parallel firstprivate(var1)
+  var1(l1 + 1) = 10
+!$omp end parallel
+end subroutine
+
+! ONE_DIM-LABEL: omp.private {type = firstprivate}
+! ONE_DIM-SAME: @[[PRIVATIZER_SYM:.*]] : [[TYPE:!fir.box<!fir.array<\?xi32>>]] alloc {
+
+! ONE_DIM-NEXT: ^bb0(%[[PRIV_ARG:.*]]: [[TYPE]]):
+
+! ONE_DIM:   %[[C0:.*]] = arith.constant 0 : index
+! ONE_DIM-NEXT:   %[[DIMS:.*]]:3 = fir.box_dims %[[PRIV_ARG]], %[[C0]] : ([[TYPE]], index) -> (index, index, index)
+! ONE_DIM:   %[[PRIV_ALLOCA:.*]] = fir.alloca !fir.array<{{\?}}xi32>
+! ONE_DIM-NEXT:   %[[SHAPE_SHIFT:.*]] = fir.shape_shift %[[DIMS]]#0, %[[DIMS]]#1 : (index, index) -> !fir.shapeshift<1>
+! ONE_DIM-NEXT:   %[[PRIV_DECL:.*]]:2 = hlfir.declare %[[PRIV_ALLOCA]](%[[SHAPE_SHIFT]]) {uniq_name = "_QFdelayed_privatization_private_1dEvar1"}
+! ONE_DIM-NEXT:  omp.yield(%[[PRIV_DECL]]#0 : [[TYPE]])
+
+! ONE_DIM-NEXT: } copy {
+! ONE_DIM-NEXT: ^bb0(%[[PRIV_ORIG_ARG:.*]]: [[TYPE]], %[[PRIV_PRIV_ARG:.*]]: [[TYPE]]):
+! ONE_DIM-NEXT:  hlfir.assign %[[PRIV_ORIG_ARG]] to %[[PRIV_PRIV_ARG]] temporary_lhs
+! ONE_DIM-NEXT:   omp.yield(%[[PRIV_PRIV_ARG]] : [[TYPE]])
+! ONE_DIM-NEXT: }
+
+!--- two_dim_array.f90
+subroutine delayed_privatization_private_2d(var1, l1, u1, l2, u2)
+  implicit none
+  integer(8):: l1, u1, l2, u2
+  integer, dimension(l1:u1, l2:u2) :: var1
+
+!$omp parallel firstprivate(var1)
+  var1(l1 + 1, u2) = 10
+!$omp end parallel
+end subroutine
+
+! TWO_DIM-LABEL: omp.private {type = firstprivate}
+! TWO_DIM-SAME: @[[PRIVATIZER_SYM:.*]] : [[TYPE:!fir.box<!fir.array<\?x\?xi32>>]] alloc {
+
+! TWO_DIM-NEXT: ^bb0(%[[PRIV_ARG:.*]]: [[TYPE]]):
+! TWO_DIM:        %[[C0:.*]] = arith.constant 0 : index
+! TWO_DIM-NEXT:   %[[DIMS0:.*]]:3 = fir.box_dims %[[PRIV_ARG]], %[[C0]] : ([[TYPE]], index) -> (index, index, index)
+
+! TWO_DIM-NEXT:   %[[C1:.*]] = arith.constant 1 : index
+! TWO_DIM-NEXT:   %[[DIMS1:.*]]:3 = fir.box_dims %[[PRIV_ARG]], %[[C1]] : ([[TYPE]], index) -> (index, index, index)
+
+! TWO_DIM-NEXT:   %[[PRIV_ALLOCA:.*]] = fir.alloca !fir.array<{{\?}}x{{\?}}xi32>, %[[DIMS0]]#1, %[[DIMS1]]#1 {bindc_name = "var1", pinned, uniq_name = "_QFdelayed_privatization_private_2dEvar1"}
+! TWO_DIM-NEXT:   %[[SHAPE_SHIFT:.*]] = fir.shape_shift %[[DIMS0]]#0, %[[DIMS0]]#1, %[[DIMS1]]#0, %[[DIMS1]]#1 : (index, index, index, index) -> !fir.shapeshift<2>
+
+! TWO_DIM-NEXT:   %[[PRIV_DECL:.*]]:2 = hlfir.declare %[[PRIV_ALLOCA]](%[[SHAPE_SHIFT]]) {uniq_name = "_QFdelayed_privatization_private_2dEvar1"}
+! TWO_DIM-NEXT:  omp.yield(%[[PRIV_DECL]]#0 : [[TYPE]])
+
+! TWO_DIM-NEXT: } copy {
+! TWO_DIM-NEXT: ^bb0(%[[PRIV_ORIG_ARG:.*]]: [[TYPE]], %[[PRIV_PRIV_ARG:.*]]: [[TYPE]]):
+! TWO_DIM-NEXT:  hlfir.assign %[[PRIV_ORIG_ARG]] to %[[PRIV_PRIV_ARG]] temporary_lhs
+! TWO_DIM-NEXT:   omp.yield(%[[PRIV_PRIV_ARG]] : [[TYPE]])
+! TWO_DIM-NEXT: }
+
+!--- one_dim_array_default_lb.f90
+program main
+  implicit none
+  integer, dimension(10) :: var1
+
+!$omp parallel private(var1)
+  var1(1) = 10
+!$omp end parallel
+end program
+
+! ONE_DIM_DEFAULT_LB-LABEL: omp.private {type = private}
+! ONE_DIM_DEFAULT_LB-SAME: @[[PRIVATIZER_SYM:.*]] : [[TYPE:!fir.ref<!fir.array<10xi32>>]] alloc {
+
+! ONE_DIM_DEFAULT_LB-NEXT: ^bb0(%[[PRIV_ARG:.*]]: [[TYPE]]):
+
+! ONE_DIM_DEFAULT_LB:   %[[C10:.*]] = arith.constant 10 : index
+! ONE_DIM_DEFAULT_LB:   %[[PRIV_ALLOCA:.*]] = fir.alloca !fir.array<10xi32>
+! ONE_DIM_DEFAULT_LB:   %[[SHAPE:.*]] = fir.shape %[[C10]] : (index) -> !fir.shape<1>
+! ONE_DIM_DEFAULT_LB:   hlfir.declare %[[PRIV_ALLOCA]](%[[SHAPE]])

flang/test/Lower/OpenMP/delayed-privatization-character-array.f90

@@ -0,0 +1,67 @@
+! Test delayed privatization for the `CHARACTER` array type.
+
+! RUN: split-file %s %t
+
+! RUN: %flang_fc1 -emit-hlfir -fopenmp -mmlir --openmp-enable-delayed-privatization \
+! RUN:   -o - %t/static_len.f90 2>&1 | FileCheck %s --check-prefix=STATIC_LEN
+! RUN: bbc -emit-hlfir -fopenmp --openmp-enable-delayed-privatization -o - %t/static_len.f90 2>&1 \
+! RUN:   | FileCheck %s --check-prefix=STATIC_LEN
+
+! RUN: %flang_fc1 -emit-hlfir -fopenmp -mmlir --openmp-enable-delayed-privatization \
+! RUN:   -o - %t/dyn_len.f90 2>&1 | FileCheck %s --check-prefix=DYN_LEN
+! RUN: bbc -emit-hlfir -fopenmp --openmp-enable-delayed-privatization -o - %t/dyn_len.f90 2>&1 \
+! RUN:   | FileCheck %s --check-prefix=DYN_LEN
+
+!--- static_len.f90
+subroutine delayed_privatization_character_array_static_len(var1)
+  implicit none
+  character(len = 10)  :: var1(5)
+
+!$omp parallel firstprivate(var1)
+  var1(1) = "test"
+!$omp end parallel
+end subroutine
+
+! STATIC_LEN-LABEL: omp.private {type = firstprivate}
+! STATIC_LEN-SAME: @[[PRIVATIZER_SYM:.*]] : [[TYPE:!fir.ref<!fir.array<5x!fir.char<1,10>>>]] alloc {
+
+! STATIC_LEN-NEXT: ^bb0(%[[PRIV_ARG:.*]]: [[TYPE]]):
+! STATIC_LEN-DAG:    %[[C5:.*]] = arith.constant 5 : index
+! STATIC_LEN-DAG:    %[[C10:.*]] = arith.constant 10 : index
+! STATIC_LEN-NEXT:   %[[PRIV_ALLOC:.*]] = fir.alloca !fir.array<5x!fir.char<1,10>>
+! STATIC_LEN-NEXT:   %[[ARRAY_SHAPE:.*]] = fir.shape %[[C5]]
+! STATIC_LEN-NEXT:   %[[PRIV_DECL:.*]]:2 = hlfir.declare %[[PRIV_ALLOC]](%[[ARRAY_SHAPE]]) typeparams %[[C10]]
+! STATIC_LEN-NEXT:   omp.yield(%[[PRIV_DECL]]#0
+
+! STATIC_LEN-NEXT: } copy {
+! STATIC_LEN-NEXT: ^bb0(%[[PRIV_ORIG_ARG:.*]]: [[TYPE]], %[[PRIV_PRIV_ARG:.*]]: [[TYPE]]):
+! STATIC_LEN-NEXT:   hlfir.assign %[[PRIV_ORIG_ARG]] to %[[PRIV_PRIV_ARG]]
+
+! STATIC_LEN-NEXT:   omp.yield(%[[PRIV_PRIV_ARG]]
+! STATIC_LEN-NEXT: }
+
+!--- dyn_len.f90
+subroutine delayed_privatization_character_array_dynamic_len(var1, char_len, array_len)
+  implicit none
+  integer(8):: char_len
+  integer(8):: array_len
+  character(len = char_len)  :: var1(array_len)
+
+!$omp parallel private(var1)
+  var1(1) = "test"
+!$omp end parallel
+end subroutine
+
+! DYN_LEN-LABEL: omp.private {type = private}
+! DYN_LEN-SAME: @[[PRIVATIZER_SYM:.*]] : [[TYPE:!fir.box<!fir.array<\?x!fir.char<1,\?>>>]] alloc {
+
+! DYN_LEN-NEXT: ^bb0(%[[PRIV_ARG:.*]]: [[TYPE]]):
+
+! DYN_LEN:        %[[C0:.*]] = arith.constant 0 : index
+! DYN_LEN-NEXT:   %[[BOX_DIM:.*]]:3 = fir.box_dims %[[PRIV_ARG]], %[[C0]]
+! DYN_LEN:        %[[CHAR_LEN:.*]] = fir.box_elesize %[[PRIV_ARG]]
+! DYN_LEN-NEXT:   %[[PRIV_ALLOC:.*]] = fir.alloca !fir.array<?x!fir.char<1,?>>(%[[CHAR_LEN]] : index)
+! DYN_LEN-NEXT:   %[[ARRAY_SHAPE:.*]] = fir.shape
+! DYN_LEN-NEXT:   %[[PRIV_DECL:.*]]:2 = hlfir.declare %[[PRIV_ALLOC]](%[[ARRAY_SHAPE]]) typeparams %[[CHAR_LEN]]
+
+! DYN_LEN-NEXT:   omp.yield(%[[PRIV_DECL]]#0