diff --git a/flang/include/flang/Lower/AbstractConverter.h b/flang/include/flang/Lower/AbstractConverter.h
index 0ffe27ea038e8..f0fe74af555f3 100644
--- a/flang/include/flang/Lower/AbstractConverter.h
+++ b/flang/include/flang/Lower/AbstractConverter.h
@@ -384,6 +384,10 @@ class AbstractConverter {
 
   virtual mlir::StateStack &getStateStack() = 0;
 
+  virtual void
+  genPermutatedLoops(llvm::ArrayRef<Fortran::lower::pft::Evaluation *> doStmts,
+                     Fortran::lower::pft::Evaluation *innermostDo) = 0;
+
 private:
   /// Options controlling lowering behavior.
   const Fortran::lower::LoweringOptions &loweringOptions;
diff --git a/flang/lib/Lower/Bridge.cpp b/flang/lib/Lower/Bridge.cpp
index 780d56f085f69..b7b4d6c73b9ee 100644
--- a/flang/lib/Lower/Bridge.cpp
+++ b/flang/lib/Lower/Bridge.cpp
@@ -2197,6 +2197,124 @@ class FirConverter : public Fortran::lower::AbstractConverter {
     // so no clean-up needs to be generated for these entities.
   }
 
+  void
+  genPermutatedLoops(llvm::ArrayRef<Fortran::lower::pft::Evaluation *> doStmts,
+                     Fortran::lower::pft::Evaluation *innermostDo) override {
+    // Fortran::lower::pft::Evaluation &eval = getEval();
+    // bool unstructuredContext = eval.lowerAsUnstructured();
+
+    llvm::SmallVector<mlir::Block *> headerBlocks;
+    llvm::SmallVector<IncrementLoopNestInfo, 1> loopInfos;
+
+    auto enterLoop = [&](Fortran::lower::pft::Evaluation &eval) {
+      bool unstructuredContext = eval.lowerAsUnstructured();
+
+      // Collect loop nest information.
+      // Generate begin loop code directly for infinite and while loops.
+      Fortran::lower::pft::Evaluation &doStmtEval =
+          eval.getFirstNestedEvaluation();
+      auto *doStmt = doStmtEval.getIf<Fortran::parser::NonLabelDoStmt>();
+      const auto &loopControl =
+          std::get<std::optional<Fortran::parser::LoopControl>>(doStmt->t);
+      mlir::Block *preheaderBlock = doStmtEval.block;
+      mlir::Block *beginBlock =
+          preheaderBlock ? preheaderBlock : builder->getBlock();
+      auto createNextBeginBlock = [&]() {
+        // Step beginBlock through unstructured preheader, header, and mask
+        // blocks, created in outermost to innermost order.
+        return beginBlock = beginBlock->splitBlock(beginBlock->end());
+      };
+      mlir::Block *headerBlock =
+          unstructuredContext ? createNextBeginBlock() : nullptr;
+      headerBlocks.push_back(headerBlock);
+      mlir::Block *bodyBlock = doStmtEval.lexicalSuccessor->block;
+      mlir::Block *exitBlock = doStmtEval.parentConstruct->constructExit->block;
+      IncrementLoopNestInfo &incrementLoopNestInfo = loopInfos.emplace_back();
+      const Fortran::parser::ScalarLogicalExpr *whileCondition = nullptr;
+      bool infiniteLoop = !loopControl.has_value();
+      if (infiniteLoop) {
+        assert(unstructuredContext && "infinite loop must be unstructured");
+        startBlock(headerBlock);
+      } else if ((whileCondition =
+                      std::get_if<Fortran::parser::ScalarLogicalExpr>(
+                          &loopControl->u))) {
+        assert(unstructuredContext && "while loop must be unstructured");
+        maybeStartBlock(preheaderBlock); // no block or empty block
+        startBlock(headerBlock);
+        genConditionalBranch(*whileCondition, bodyBlock, exitBlock);
+      } else if (const auto *bounds =
+                     std::get_if<Fortran::parser::LoopControl::Bounds>(
+                         &loopControl->u)) {
+        // Non-concurrent increment loop.
+        IncrementLoopInfo &info = incrementLoopNestInfo.emplace_back(
+            *bounds->name.thing.symbol, bounds->lower, bounds->upper,
+            bounds->step);
+        if (unstructuredContext) {
+          maybeStartBlock(preheaderBlock);
+          info.hasRealControl = info.loopVariableSym->GetType()->IsNumeric(
+              Fortran::common::TypeCategory::Real);
+          info.headerBlock = headerBlock;
+          info.bodyBlock = bodyBlock;
+          info.exitBlock = exitBlock;
+        }
+      } else {
+        llvm_unreachable("Cannot permute DO CONCURRENT");
+      }
+
+      // Increment loop begin code. (Infinite/while code was already generated.)
+      if (!infiniteLoop && !whileCondition)
+        genFIRIncrementLoopBegin(incrementLoopNestInfo, doStmtEval.dirs);
+    };
+
+    auto leaveLoop = [&](Fortran::lower::pft::Evaluation &eval,
+                         mlir::Block *headerBlock,
+                         IncrementLoopNestInfo &incrementLoopNestInfo) {
+      bool unstructuredContext = eval.lowerAsUnstructured();
+
+      Fortran::lower::pft::Evaluation &doStmtEval =
+          eval.getFirstNestedEvaluation();
+      auto *doStmt = doStmtEval.getIf<Fortran::parser::NonLabelDoStmt>();
+
+      const auto &loopControl =
+          std::get<std::optional<Fortran::parser::LoopControl>>(doStmt->t);
+      bool infiniteLoop = !loopControl.has_value();
+      const Fortran::parser::ScalarLogicalExpr *whileCondition =
+          std::get_if<Fortran::parser::ScalarLogicalExpr>(&loopControl->u);
+
+      auto iter = std::prev(eval.getNestedEvaluations().end());
+
+      // An EndDoStmt in unstructured code may start a new block.
+      Fortran::lower::pft::Evaluation &endDoEval = *iter;
+      assert(endDoEval.getIf<Fortran::parser::EndDoStmt>() && "no enddo stmt");
+      if (unstructuredContext)
+        maybeStartBlock(endDoEval.block);
+
+      // Loop end code.
+      if (infiniteLoop || whileCondition)
+        genBranch(headerBlock);
+      else
+        genFIRIncrementLoopEnd(incrementLoopNestInfo);
+
+      // This call may generate a branch in some contexts.
+      genFIR(endDoEval, unstructuredContext);
+    };
+
+    for (auto l : doStmts)
+      enterLoop(*l);
+
+    // Loop body code.
+    bool innermostUnstructuredContext = innermostDo->lowerAsUnstructured();
+
+    auto iter = innermostDo->getNestedEvaluations().begin();
+    for (auto end = --innermostDo->getNestedEvaluations().end(); iter != end;
+         ++iter)
+      genFIR(*iter, innermostUnstructuredContext);
+
+    for (auto &&[l, headerBlock, li] :
+         llvm::zip_equal(doStmts, headerBlocks, loopInfos))
+      leaveLoop(*l, headerBlock, li);
+  }
+
   /// Generate FIR for a DO construct. There are six variants:
   ///  - unstructured infinite and while loops
   ///  - structured and unstructured increment loops
diff --git a/flang/lib/Lower/OpenMP/Decomposer.cpp b/flang/lib/Lower/OpenMP/Decomposer.cpp
index 9bfbf67bec88c..bf09bed395285 100644
--- a/flang/lib/Lower/OpenMP/Decomposer.cpp
+++ b/flang/lib/Lower/OpenMP/Decomposer.cpp
@@ -98,14 +98,39 @@ ConstructQueue buildConstructQueue(
   return decompose.output;
 }
 
+// from clang
+static bool isOpenMPLoopTransformationDirective(llvm::omp::Directive DKind) {
+  return DKind == llvm::omp::Directive::OMPD_tile ||
+         DKind == llvm::omp::Directive::OMPD_unroll ||
+         DKind == llvm::omp::Directive::OMPD_reverse ||
+         DKind == llvm::omp::Directive::OMPD_interchange ||
+         DKind == llvm::omp::Directive::OMPD_stripe;
+}
+
+llvm::iterator_range<ConstructQueue::const_iterator> getNonTransformQueue(
+    llvm::iterator_range<ConstructQueue::const_iterator> range) {
+  // remove trailing loop transformations
+  auto b = range.begin();
+  auto e = range.end();
+  while (e != b) {
+    auto e2 = e - 1;
+    if (!isOpenMPLoopTransformationDirective(e2->id))
+      break;
+    e = e2;
+  }
+
+  return llvm::make_range(b, e);
+}
+
 bool matchLeafSequence(ConstructQueue::const_iterator item,
                        const ConstructQueue &queue,
                        llvm::omp::Directive directive) {
   llvm::ArrayRef<llvm::omp::Directive> leafDirs =
       llvm::omp::getLeafConstructsOrSelf(directive);
 
-  for (auto [dir, leaf] :
-       llvm::zip_longest(leafDirs, llvm::make_range(item, queue.end()))) {
+  for (auto [dir, leaf] : llvm::zip_longest(
+           leafDirs,
+           getNonTransformQueue(llvm::make_range(item, queue.end())))) {
     if (!dir.has_value() || !leaf.has_value())
       return false;
 
diff --git a/flang/lib/Lower/OpenMP/Decomposer.h b/flang/lib/Lower/OpenMP/Decomposer.h
index 65492bd76280d..f057009629efc 100644
--- a/flang/lib/Lower/OpenMP/Decomposer.h
+++ b/flang/lib/Lower/OpenMP/Decomposer.h
@@ -57,6 +57,10 @@ bool isLastItemInQueue(ConstructQueue::const_iterator item,
 bool matchLeafSequence(ConstructQueue::const_iterator item,
                        const ConstructQueue &queue,
                        llvm::omp::Directive directive);
+
+llvm::iterator_range<ConstructQueue::const_iterator> getNonTransformQueue(
+    llvm::iterator_range<ConstructQueue::const_iterator> range);
+
 } // namespace Fortran::lower::omp
 
 #endif // FORTRAN_LOWER_OPENMP_DECOMPOSER_H
diff --git a/flang/lib/Lower/OpenMP/OpenMP.cpp b/flang/lib/Lower/OpenMP/OpenMP.cpp
index 9e56c2bfb7e25..805e3dd8aad5e 100644
--- a/flang/lib/Lower/OpenMP/OpenMP.cpp
+++ b/flang/lib/Lower/OpenMP/OpenMP.cpp
@@ -1231,8 +1231,11 @@ static void createBodyOfOp(mlir::Operation &op, const OpWithBodyGenInfo &info,
   }
 
   if (!info.genSkeletonOnly) {
+    // Transforms already processed by getLoopNestOp
+    auto q = getNonTransformQueue(llvm::make_range(item, queue.end()));
+    auto transforms = llvm::make_range(q.end(), queue.end());
     if (ConstructQueue::const_iterator next = std::next(item);
-        next != queue.end()) {
+        next != transforms.begin() && next != queue.end()) {
       genOMPDispatch(info.converter, info.symTable, info.semaCtx, info.eval,
                      info.loc, queue, next);
     } else {
@@ -1551,7 +1554,8 @@ genLoopNestClauses(lower::AbstractConverter &converter,
                    semantics::SemanticsContext &semaCtx,
                    lower::pft::Evaluation &eval, const List<Clause> &clauses,
                    mlir::Location loc, mlir::omp::LoopNestOperands &clauseOps,
-                   llvm::SmallVectorImpl<const semantics::Symbol *> &iv) {
+                   llvm::SmallVectorImpl<const semantics::Symbol *> &iv,
+                   bool enableInterchange = false) {
   ClauseProcessor cp(converter, semaCtx, clauses);
 
   HostEvalInfo *hostEvalInfo = getHostEvalInfoStackTop(converter);
@@ -1559,6 +1563,28 @@ genLoopNestClauses(lower::AbstractConverter &converter,
     cp.processCollapse(loc, eval, clauseOps, clauseOps, iv);
 
   clauseOps.loopInclusive = converter.getFirOpBuilder().getUnitAttr();
+
+  fir::FirOpBuilder &firOpBuilder = converter.getFirOpBuilder();
+  for (auto &clause : clauses) {
+    if (clause.id == llvm::omp::Clause::OMPC_collapse) {
+      const auto &collapse = std::get<clause::Collapse>(clause.u);
+      int64_t collapseValue = evaluate::ToInt64(collapse.v).value();
+      clauseOps.collapseNumLoops =
+          firOpBuilder.getI64IntegerAttr(collapseValue);
+    } else if (clause.id == llvm::omp::Clause::OMPC_sizes) {
+      // This case handles the stand-alone tiling construct
+      const auto &sizes = std::get<clause::Sizes>(clause.u);
+      llvm::SmallVector<int64_t> sizeValues;
+      for (auto &size : sizes.v) {
+        int64_t sizeValue = evaluate::ToInt64(size).value();
+        sizeValues.push_back(sizeValue);
+      }
+      clauseOps.tileSizes = sizeValues;
+    } else if (clause.id == llvm::omp::Clause::OMPC_permutation) {
+      llvm_unreachable("MK: To handle standalone interchange construct");
+    }
+  }
+
   cp.processTileSizes(eval, clauseOps);
 }
 
@@ -1930,7 +1956,9 @@ static mlir::omp::LoopNestOp genLoopNestOp(
     llvm::ArrayRef<
         std::pair<mlir::omp::BlockArgOpenMPOpInterface, const EntryBlockArgs &>>
         wrapperArgs,
-    llvm::omp::Directive directive, DataSharingProcessor &dsp) {
+    llvm::omp::Directive directive, DataSharingProcessor &dsp,
+    std::optional<llvm::iterator_range<ConstructQueue::const_iterator>>
+        transforms = std::nullopt) {
   auto ivCallback = [&](mlir::Operation *op) {
     genLoopVars(op, converter, loc, iv, wrapperArgs);
     return llvm::SmallVector<const semantics::Symbol *>(iv);
@@ -1939,6 +1967,65 @@ static mlir::omp::LoopNestOp genLoopNestOp(
   uint64_t nestValue = getCollapseValue(item->clauses);
   nestValue = nestValue < iv.size() ? iv.size() : nestValue;
   auto *nestedEval = getCollapsedLoopEval(eval, nestValue);
+
+  if (!transforms.has_value()) {
+    // This must be a standalone construct, assume all following actions are
+    // transformations
+    transforms = llvm::make_range(std::next(item), queue.end());
+  }
+
+  for (auto &&transform : llvm::reverse(*transforms)) {
+    auto d = transform.id;
+    auto clauses = transform.clauses;
+
+    switch (d) {
+    case llvm::omp::OMPD_interchange: {
+      llvm::SmallVector<int64_t> permutation;
+
+      auto &&permutationClause = ClauseFinder::findUniqueClause<
+          Fortran::lower::omp::clause::Permutation>(clauses);
+      if (permutationClause) {
+        permutation.reserve(permutationClause->v.size());
+        for (auto &&ts : permutationClause->v) {
+          permutation.push_back(evaluate::ToInt64(ts).value());
+        }
+        //   llvm::append_range( permutation, permutationClause->v);
+
+      } else {
+        permutation = {2, 1};
+      }
+
+      assert(permutation.size() == iv.size() &&
+             "TODO: if permutation is smaller than number of associated loops, "
+             "permute only the first loops");
+      llvm::SmallVector<const semantics::Symbol *> newIVs;
+      llvm::SmallVector<mlir::Value> newLBs;
+      llvm::SmallVector<mlir::Value> newUBs;
+      llvm::SmallVector<mlir::Value> newINCs;
+      llvm::SmallVector<int64_t> newSizes;
+
+      // TODO: Assert this is a valid permution
+      for (auto perm : permutation) {
+        newIVs.push_back(iv[perm - 1]);
+        newLBs.push_back(clauseOps.loopLowerBounds[perm - 1]);
+        newUBs.push_back(clauseOps.loopUpperBounds[perm - 1]);
+        newINCs.push_back(clauseOps.loopSteps[perm - 1]);
+        if (!clauseOps.tileSizes.empty())
+          newSizes.push_back(clauseOps.tileSizes[perm - 1]);
+      }
+
+      iv = newIVs;
+      clauseOps.loopLowerBounds = newLBs;
+      clauseOps.loopUpperBounds = newUBs;
+      clauseOps.loopSteps = newINCs;
+      clauseOps.tileSizes = newSizes;
+
+    } break;
+    default:
+      llvm_unreachable("MK: loop transformation not yet implemented");
+    }
+  }
+
   return genOpWithBody<mlir::omp::LoopNestOp>(
       OpWithBodyGenInfo(converter, symTable, semaCtx, loc, *nestedEval,
                         directive)
@@ -2247,6 +2334,68 @@ static void genUnrollOp(Fortran::lower::AbstractConverter &converter,
   mlir::omp::UnrollHeuristicOp::create(firOpBuilder, loc, cli);
 }
 
+static void
+collectLoops(lower::pft::Evaluation &eval,
+             llvm::SmallVectorImpl<lower::pft::Evaluation *> &result,
+             int numLoops) {
+  lower::pft::Evaluation *doConstructEval = &eval.getFirstNestedEvaluation();
+  for ([[maybe_unused]] auto i : llvm::seq<int>(numLoops)) {
+    lower::pft::Evaluation *doLoop =
+        &doConstructEval->getFirstNestedEvaluation();
+    auto *doStmt = doLoop->getIf<parser::NonLabelDoStmt>();
+    assert(doStmt && "Expected do loop to be in the nested evaluation");
+    const auto &loopControl =
+        std::get<std::optional<parser::LoopControl>>(doStmt->t);
+    const parser::LoopControl::Bounds *bounds =
+        std::get_if<parser::LoopControl::Bounds>(&loopControl->u);
+    assert(bounds && "Expected bounds for worksharing do loop");
+    lower::StatementContext stmtCtx;
+
+    result.push_back(doConstructEval);
+
+    doConstructEval =
+        &*std::next(doConstructEval->getNestedEvaluations().begin());
+  };
+}
+
+static void genStandaloneInterchangeOp(
+    Fortran::lower::AbstractConverter &converter,
+    Fortran::lower::SymMap &symTable, lower::StatementContext &stmtCtx,
+    Fortran::semantics::SemanticsContext &semaCtx,
+    Fortran::lower::pft::Evaluation &eval, mlir::Location loc,
+    const ConstructQueue &queue, ConstructQueue::const_iterator item) {
+  auto q = getNonTransformQueue(llvm::make_range(item, queue.end()));
+  auto transforms = llvm::make_range(q.end(), queue.end());
+  assert(llvm::range_size(transforms) == 1);
+  auto &&transform = *transforms.begin();
+  assert(transform.id == llvm::omp::OMPD_interchange);
+  auto clauses = transform.clauses;
+
+  llvm::SmallVector<int64_t> permutation;
+  auto &&permutationClause =
+      ClauseFinder::findUniqueClause<Fortran::lower::omp::clause::Permutation>(
+          clauses);
+  if (permutationClause) {
+    permutation.reserve(permutationClause->v.size());
+    for (auto &&ts : permutationClause->v) {
+      permutation.push_back(evaluate::ToInt64(ts).value());
+    }
+  } else {
+    permutation = {2, 1};
+  }
+
+  llvm::SmallVector<lower::pft::Evaluation *> loops;
+  collectLoops(eval, loops, permutation.size());
+
+  // TODO: Assert this is a valid permution
+  llvm::SmallVector<lower::pft::Evaluation *> newLoops;
+  for (auto perm : permutation) {
+    newLoops.push_back(loops[perm - 1]);
+  }
+
+  converter.genPermutatedLoops(newLoops, loops.back());
+}
+
 static mlir::omp::MaskedOp
 genMaskedOp(lower::AbstractConverter &converter, lower::SymMap &symTable,
             lower::StatementContext &stmtCtx,
@@ -3022,7 +3171,10 @@ static mlir::omp::DistributeOp genCompositeDistributeParallelDo(
     lower::StatementContext &stmtCtx, semantics::SemanticsContext &semaCtx,
     lower::pft::Evaluation &eval, mlir::Location loc,
     const ConstructQueue &queue, ConstructQueue::const_iterator item) {
-  assert(std::distance(item, queue.end()) == 3 && "Invalid leaf constructs");
+  auto q = getNonTransformQueue(llvm::make_range(item, queue.end()));
+  auto transforms = llvm::make_range(q.end(), queue.end());
+
+  assert(llvm::range_size(q) == 3 && "Invalid leaf constructs");
   ConstructQueue::const_iterator distributeItem = item;
   ConstructQueue::const_iterator parallelItem = std::next(distributeItem);
   ConstructQueue::const_iterator doItem = std::next(parallelItem);
@@ -3076,10 +3228,10 @@ static mlir::omp::DistributeOp genCompositeDistributeParallelDo(
       converter, loc, wsloopClauseOps, wsloopArgs);
   wsloopOp.setComposite(/*val=*/true);
 
-  genLoopNestOp(converter, symTable, semaCtx, eval, loc, queue, doItem,
-                loopNestClauseOps, iv,
-                {{distributeOp, distributeArgs}, {wsloopOp, wsloopArgs}},
-                llvm::omp::Directive::OMPD_distribute_parallel_do, dsp);
+  genLoopNestOp(
+      converter, symTable, semaCtx, eval, loc, queue, doItem, loopNestClauseOps,
+      iv, {{distributeOp, distributeArgs}, {wsloopOp, wsloopArgs}},
+      llvm::omp::Directive::OMPD_distribute_parallel_do, dsp, transforms);
   return distributeOp;
 }
 
@@ -3088,7 +3240,11 @@ static mlir::omp::DistributeOp genCompositeDistributeParallelDoSimd(
     lower::StatementContext &stmtCtx, semantics::SemanticsContext &semaCtx,
     lower::pft::Evaluation &eval, mlir::Location loc,
     const ConstructQueue &queue, ConstructQueue::const_iterator item) {
-  assert(std::distance(item, queue.end()) == 4 && "Invalid leaf constructs");
+  auto q = getNonTransformQueue(llvm::make_range(item, queue.end()));
+  auto transforms = llvm::make_range(q.end(), queue.end());
+
+  assert(llvm::range_size(q) == 4 && "Invalid leaf constructs");
+
   ConstructQueue::const_iterator distributeItem = item;
   ConstructQueue::const_iterator parallelItem = std::next(distributeItem);
   ConstructQueue::const_iterator doItem = std::next(parallelItem);
@@ -3170,7 +3326,7 @@ static mlir::omp::DistributeOp genCompositeDistributeParallelDoSimd(
                  {wsloopOp, wsloopArgs},
                  {simdOp, simdArgs}},
                 llvm::omp::Directive::OMPD_distribute_parallel_do_simd,
-                simdItemDSP);
+                simdItemDSP, transforms);
   return distributeOp;
 }
 
@@ -3179,7 +3335,11 @@ static mlir::omp::DistributeOp genCompositeDistributeSimd(
     lower::StatementContext &stmtCtx, semantics::SemanticsContext &semaCtx,
     lower::pft::Evaluation &eval, mlir::Location loc,
     const ConstructQueue &queue, ConstructQueue::const_iterator item) {
-  assert(std::distance(item, queue.end()) == 2 && "Invalid leaf constructs");
+  auto q = getNonTransformQueue(llvm::make_range(item, queue.end()));
+  auto transforms = llvm::make_range(q.end(), queue.end());
+
+  assert(llvm::range_size(q) == 2 && "Invalid leaf constructs");
+
   ConstructQueue::const_iterator distributeItem = item;
   ConstructQueue::const_iterator simdItem = std::next(distributeItem);
 
@@ -3231,7 +3391,8 @@ static mlir::omp::DistributeOp genCompositeDistributeSimd(
   genLoopNestOp(converter, symTable, semaCtx, eval, loc, queue, simdItem,
                 loopNestClauseOps, iv,
                 {{distributeOp, distributeArgs}, {simdOp, simdArgs}},
-                llvm::omp::Directive::OMPD_distribute_simd, simdItemDSP);
+                llvm::omp::Directive::OMPD_distribute_simd, simdItemDSP,
+                transforms);
   return distributeOp;
 }
 
@@ -3240,7 +3401,11 @@ static mlir::omp::WsloopOp genCompositeDoSimd(
     lower::StatementContext &stmtCtx, semantics::SemanticsContext &semaCtx,
     lower::pft::Evaluation &eval, mlir::Location loc,
     const ConstructQueue &queue, ConstructQueue::const_iterator item) {
-  assert(std::distance(item, queue.end()) == 2 && "Invalid leaf constructs");
+  auto q = getNonTransformQueue(llvm::make_range(item, queue.end()));
+  auto transforms = llvm::make_range(q.end(), queue.end());
+
+  assert(llvm::range_size(q) == 2 && "Invalid leaf constructs");
+
   ConstructQueue::const_iterator doItem = item;
   ConstructQueue::const_iterator simdItem = std::next(doItem);
 
@@ -3295,7 +3460,7 @@ static mlir::omp::WsloopOp genCompositeDoSimd(
   genLoopNestOp(converter, symTable, semaCtx, eval, loc, queue, simdItem,
                 loopNestClauseOps, iv,
                 {{wsloopOp, wsloopArgs}, {simdOp, simdArgs}},
-                llvm::omp::Directive::OMPD_do_simd, simdItemDSP);
+                llvm::omp::Directive::OMPD_do_simd, simdItemDSP, transforms);
   return wsloopOp;
 }
 
@@ -3304,7 +3469,10 @@ static mlir::omp::TaskloopOp genCompositeTaskloopSimd(
     lower::StatementContext &stmtCtx, semantics::SemanticsContext &semaCtx,
     lower::pft::Evaluation &eval, mlir::Location loc,
     const ConstructQueue &queue, ConstructQueue::const_iterator item) {
-  assert(std::distance(item, queue.end()) == 2 && "Invalid leaf constructs");
+  auto q = getNonTransformQueue(llvm::make_range(item, queue.end()));
+
+  assert(llvm::range_size(q) == 2 && "Invalid leaf constructs");
+
   if (!semaCtx.langOptions().OpenMPSimd)
     TODO(loc, "Composite TASKLOOP SIMD");
   return nullptr;
@@ -3482,6 +3650,10 @@ static void genOMPDispatch(lower::AbstractConverter &converter,
   case llvm::omp::Directive::OMPD_unroll:
     genUnrollOp(converter, symTable, stmtCtx, semaCtx, eval, loc, queue, item);
     break;
+  case llvm::omp::Directive::OMPD_interchange:
+    genStandaloneInterchangeOp(converter, symTable, stmtCtx, semaCtx, eval, loc,
+                               queue, item);
+    break;
   case llvm::omp::Directive::OMPD_workdistribute:
     newOp = genWorkdistributeOp(converter, symTable, semaCtx, eval, loc, queue,
                                 item);
@@ -3926,19 +4098,37 @@ static void genOMP(lower::AbstractConverter &converter, lower::SymMap &symTable,
 
   mlir::Location currentLocation = converter.genLocation(beginSpec.source);
 
+  const parser::OmpDirectiveName &beginName = beginSpec.DirName();
+  ConstructQueue queue{
+      buildConstructQueue(converter.getFirOpBuilder().getModule(), semaCtx,
+                          eval, beginName.source, beginName.v, clauses)};
+
   auto &optLoopCons =
       std::get<std::optional<parser::NestedConstruct>>(loopConstruct.t);
   if (optLoopCons.has_value()) {
     if (auto *ompNestedLoopCons{
             std::get_if<common::Indirection<parser::OpenMPLoopConstruct>>(
                 &*optLoopCons)}) {
+      const Fortran::parser::OpenMPLoopConstruct &x =
+          ompNestedLoopCons->value();
       llvm::omp::Directive nestedDirective =
           parser::omp::GetOmpDirectiveName(*ompNestedLoopCons).v;
+
+      List<Clause> nestedClauses = makeClauses(x.BeginDir().Clauses(), semaCtx);
+
       switch (nestedDirective) {
       case llvm::omp::Directive::OMPD_tile:
         // Skip OMPD_tile since the tile sizes will be retrieved when
         // generating the omp.loop_nest op.
         break;
+      case llvm::omp::Directive::OMPD_interchange: {
+        ConstructQueue nestedQueue{buildConstructQueue(
+            converter.getFirOpBuilder().getModule(), semaCtx, eval,
+            beginName.source, nestedDirective, nestedClauses)};
+        for (auto nl : nestedQueue) {
+          queue.push_back(nl);
+        }
+      } break;
       default: {
         unsigned version = semaCtx.langOptions().OpenMPVersion;
         TODO(currentLocation,
@@ -3950,10 +4140,6 @@ static void genOMP(lower::AbstractConverter &converter, lower::SymMap &symTable,
     }
   }
 
-  const parser::OmpDirectiveName &beginName = beginSpec.DirName();
-  ConstructQueue queue{
-      buildConstructQueue(converter.getFirOpBuilder().getModule(), semaCtx,
-                          eval, beginName.source, beginName.v, clauses)};
   genOMPDispatch(converter, symTable, semaCtx, eval, currentLocation, queue,
                  queue.begin());
 }
diff --git a/flang/lib/Lower/OpenMP/Utils.cpp b/flang/lib/Lower/OpenMP/Utils.cpp
index 29cccbd1bfe5a..10e5114bcb880 100644
--- a/flang/lib/Lower/OpenMP/Utils.cpp
+++ b/flang/lib/Lower/OpenMP/Utils.cpp
@@ -687,14 +687,46 @@ void collectLoopRelatedInfo(
 
   // Collect sizes from tile directive if present.
   std::int64_t sizesLengthValue = 0l;
+  std::int64_t permutationLengthValue = 0l;
   if (auto *ompCons{eval.getIf<parser::OpenMPConstruct>()}) {
     processTileSizesFromOpenMPConstruct(
         ompCons, [&](const parser::OmpClause::Sizes *tclause) {
           sizesLengthValue = tclause->v.size();
         });
+
+    if (auto *ompLoop{std::get_if<parser::OpenMPLoopConstruct>(&ompCons->u)}) {
+      const auto &nestedOptional =
+          std::get<std::optional<parser::NestedConstruct>>(ompLoop->t);
+      assert(nestedOptional.has_value() &&
+             "Expected a DoConstruct or OpenMPLoopConstruct");
+      const auto *innerConstruct =
+          std::get_if<common::Indirection<parser::OpenMPLoopConstruct>>(
+              &(nestedOptional.value()));
+      if (innerConstruct) {
+        const auto &innerLoopDirective = innerConstruct->value();
+        const auto &innerBegin =
+            std::get<parser::OmpBeginLoopDirective>(innerLoopDirective.t);
+        const auto &innerDirective =
+            Fortran::parser::omp::GetOmpDirectiveName(innerBegin).v;
+
+        if (innerDirective == llvm::omp::Directive::OMPD_interchange) {
+          // Get the size values from parse tree and convert to a vector
+          const auto &innerClauseList{innerBegin.Clauses()};
+          for (const auto &clause : innerClauseList.v) {
+            if (const auto tclause{
+                    std::get_if<parser::OmpClause::Permutation>(&clause.u)}) {
+              permutationLengthValue = tclause->v.size();
+            }
+          }
+          // default: permution(2,1)
+          if (permutationLengthValue == 0)
+            permutationLengthValue = 2;
+        }
+      }
+    }
   }
 
-  std::int64_t collapseValue = std::max(numCollapse, sizesLengthValue);
+std::int64_t collapseValue = std::max({numCollapse, sizesLengthValue, permutationLengthValue});
   std::size_t loopVarTypeSize = 0;
   do {
     lower::pft::Evaluation *doLoop =
@@ -729,6 +761,44 @@ void collectLoopRelatedInfo(
   convertLoopBounds(converter, currentLocation, result, loopVarTypeSize);
 }
 
+void collectPermutationFromOpenMPConstruct(
+    const parser::OpenMPConstruct *ompCons,
+    llvm::SmallVectorImpl<int64_t> &permutation,
+    Fortran::semantics::SemanticsContext &semaCtx) {
+  if (!ompCons)
+    return;
+
+  if (auto *ompLoop{std::get_if<parser::OpenMPLoopConstruct>(&ompCons->u)}) {
+    const auto &nestedOptional =
+        std::get<std::optional<parser::NestedConstruct>>(ompLoop->t);
+    assert(nestedOptional.has_value() &&
+           "Expected a DoConstruct or OpenMPLoopConstruct");
+    const auto *innerConstruct =
+        std::get_if<common::Indirection<parser::OpenMPLoopConstruct>>(
+            &(nestedOptional.value()));
+    if (innerConstruct) {
+      const auto &innerLoopDirective = innerConstruct->value();
+      const auto &innerBegin = innerLoopDirective.BeginDir();
+      const auto &innerDirective =
+          Fortran::parser::omp::GetOmpDirectiveName(innerLoopDirective).v;
+
+      if (innerDirective == llvm::omp::Directive::OMPD_interchange) {
+        // Get the size values from parse tree and convert to a vector
+        const auto &innerClauseList{innerBegin.Clauses()};
+        // std::get<parser::OmpClauseList>(innerBegin.t)};
+        for (const auto &clause : innerClauseList.v)
+          if (const auto tclause{
+                  std::get_if<parser::OmpClause::Sizes>(&clause.u)}) {
+            for (auto &tval : tclause->v) {
+              if (const auto v{EvaluateInt64(semaCtx, tval)})
+                permutation.push_back(*v);
+            }
+          }
+      }
+    }
+  }
+}
+
 } // namespace omp
 } // namespace lower
 } // namespace Fortran
diff --git a/flang/lib/Lower/OpenMP/Utils.h b/flang/lib/Lower/OpenMP/Utils.h
index 69499f9c7b621..03751dea0e7c8 100644
--- a/flang/lib/Lower/OpenMP/Utils.h
+++ b/flang/lib/Lower/OpenMP/Utils.h
@@ -177,6 +177,11 @@ void collectTileSizesFromOpenMPConstruct(
     llvm::SmallVectorImpl<int64_t> &tileSizes,
     Fortran::semantics::SemanticsContext &semaCtx);
 
+void collectPermutationFromOpenMPConstruct(
+    const parser::OpenMPConstruct *ompCons,
+    llvm::SmallVectorImpl<int64_t> &permuation,
+    Fortran::semantics::SemanticsContext &semaCtx);
+
 } // namespace omp
 } // namespace lower
 } // namespace Fortran
diff --git a/flang/lib/Parser/openmp-parsers.cpp b/flang/lib/Parser/openmp-parsers.cpp
index 9507021057476..cd55b8a296254 100644
--- a/flang/lib/Parser/openmp-parsers.cpp
+++ b/flang/lib/Parser/openmp-parsers.cpp
@@ -2021,6 +2021,7 @@ static constexpr DirectiveSet GetLoopDirectives() {
       unsigned(Directive::OMPD_teams_loop),
       unsigned(Directive::OMPD_tile),
       unsigned(Directive::OMPD_unroll),
+      unsigned(Directive::OMPD_interchange),
   };
   return loopDirectives;
 }
diff --git a/flang/lib/Semantics/canonicalize-omp.cpp b/flang/lib/Semantics/canonicalize-omp.cpp
index c884658bf464a..087a8c53f2afc 100644
--- a/flang/lib/Semantics/canonicalize-omp.cpp
+++ b/flang/lib/Semantics/canonicalize-omp.cpp
@@ -178,7 +178,8 @@ class CanonicalizationOfOmp {
         auto &nestedBeginDirective = ompLoopCons->BeginDir();
         auto &nestedBeginName = nestedBeginDirective.DirName();
         if ((nestedBeginName.v == llvm::omp::Directive::OMPD_unroll ||
-                nestedBeginName.v == llvm::omp::Directive::OMPD_tile) &&
+                nestedBeginName.v == llvm::omp::Directive::OMPD_tile ||
+                nestedBeginName.v == llvm::omp::Directive::OMPD_interchange) &&
             !(nestedBeginName.v == llvm::omp::Directive::OMPD_unroll &&
                 beginName.v == llvm::omp::Directive::OMPD_tile)) {
           // iterate through the remaining block items to find the end directive
diff --git a/flang/lib/Semantics/resolve-directives.cpp b/flang/lib/Semantics/resolve-directives.cpp
index 18fc63814d973..3b9e972771ff8 100644
--- a/flang/lib/Semantics/resolve-directives.cpp
+++ b/flang/lib/Semantics/resolve-directives.cpp
@@ -901,8 +901,8 @@ class OmpAttributeVisitor : DirectiveAttributeVisitor<llvm::omp::Directive> {
   void CollectNumAffectedLoopsFromInnerLoopContruct(
       const parser::OpenMPLoopConstruct &, llvm::SmallVector<std::int64_t> &,
       llvm::SmallVector<const parser::OmpClause *> &);
-  void CollectNumAffectedLoopsFromClauses(const parser::OmpClauseList &,
-      llvm::SmallVector<std::int64_t> &,
+  void CollectNumAffectedLoopsFromClauses(const parser::OpenMPLoopConstruct &x,
+      const parser::OmpClauseList &, llvm::SmallVector<std::int64_t> &,
       llvm::SmallVector<const parser::OmpClause *> &);
 
   Symbol::Flags dataSharingAttributeFlags{Symbol::Flag::OmpShared,
@@ -1975,6 +1975,7 @@ bool OmpAttributeVisitor::Pre(const parser::OpenMPLoopConstruct &x) {
   case llvm::omp::Directive::OMPD_teams_distribute_simd:
   case llvm::omp::Directive::OMPD_teams_loop:
   case llvm::omp::Directive::OMPD_tile:
+  case llvm::omp::Directive::OMPD_interchange:
   case llvm::omp::Directive::OMPD_unroll:
     PushContext(beginName.source, beginName.v);
     break;
@@ -2097,7 +2098,7 @@ bool OmpAttributeVisitor::Pre(const parser::DoConstruct &x) {
 }
 
 static bool isSizesClause(const parser::OmpClause *clause) {
-  return std::holds_alternative<parser::OmpClause::Sizes>(clause->u);
+  return clause && std::holds_alternative<parser::OmpClause::Sizes>(clause->u);
 }
 
 std::int64_t OmpAttributeVisitor::SetAssociatedMaxClause(
@@ -2148,7 +2149,7 @@ void OmpAttributeVisitor::CollectNumAffectedLoopsFromLoopConstruct(
     llvm::SmallVector<const parser::OmpClause *> &clauses) {
   const auto &clauseList{x.BeginDir().Clauses()};
 
-  CollectNumAffectedLoopsFromClauses(clauseList, levels, clauses);
+  CollectNumAffectedLoopsFromClauses(x, clauseList, levels, clauses);
   CollectNumAffectedLoopsFromInnerLoopContruct(x, levels, clauses);
 }
 
@@ -2172,8 +2173,13 @@ void OmpAttributeVisitor::CollectNumAffectedLoopsFromInnerLoopContruct(
 }
 
 void OmpAttributeVisitor::CollectNumAffectedLoopsFromClauses(
-    const parser::OmpClauseList &x, llvm::SmallVector<std::int64_t> &levels,
+    const parser::OpenMPLoopConstruct &y, const parser::OmpClauseList &x,
+    llvm::SmallVector<std::int64_t> &levels,
     llvm::SmallVector<const parser::OmpClause *> &clauses) {
+  const auto &beginLoopDir{y.BeginDir()};
+  const auto &dirClauses{beginLoopDir.Clauses()};
+  auto ytv = Fortran::parser::omp::GetOmpDirectiveName(y).v;
+
   for (const auto &clause : x.v) {
     if (const auto oclause{
             std::get_if<parser::OmpClause::Ordered>(&clause.u)}) {
@@ -2200,6 +2206,20 @@ void OmpAttributeVisitor::CollectNumAffectedLoopsFromClauses(
       clauses.push_back(&clause);
     }
   }
+
+  if (ytv == llvm::omp::OMPD_interchange) {
+    for (const auto &clause : dirClauses.v) {
+      if (const auto tclause{
+              std::get_if<parser::OmpClause::Permutation>(&clause.u)}) {
+        levels.push_back(tclause->v.size());
+        clauses.push_back(&clause);
+        return;
+      }
+    }
+
+    levels.push_back(2);
+    clauses.push_back(nullptr);
+  }
 }
 
 void OmpAttributeVisitor::CheckPerfectNestAndRectangularLoop(
diff --git a/openmp/runtime/test/transform/interchange/distribute-parallel-wsloop-intdo.f90 b/openmp/runtime/test/transform/interchange/distribute-parallel-wsloop-intdo.f90
new file mode 100644
index 0000000000000..b5ef5214f5064
--- /dev/null
+++ b/openmp/runtime/test/transform/interchange/distribute-parallel-wsloop-intdo.f90
@@ -0,0 +1,37 @@
+
+! RUN: %flang %flags %openmp_flags -fopenmp-version=60 %s -o %t.exe
+! RUN: %t.exe | FileCheck %s --match-full-lines
+
+program interchange_wsloop_intdo
+  integer :: i, j
+  print *, 'do'
+
+  !$OMP TEAMS NUM_TEAMS(1)
+
+  !$OMP DISTRIBUTE PARALLEL DO SCHEDULE(static,2) NUM_THREADS(1)
+  !$OMP INTERCHANGE
+  do i = 7, 15, 3
+    do j = -1, 1
+      print '("i=", I0, " j=", I0)', i, j
+    end do
+  end do
+  !$OMP END INTERCHANGE
+  !$OMP END DISTRIBUTE PARALLEL DO
+
+  !$OMP END TEAMS
+
+  print *, 'done'
+end program
+
+
+! CHECK:      do
+! CHECK-NEXT: i=7 j=-1
+! CHECK-NEXT: i=10 j=-1
+! CHECK-NEXT: i=13 j=-1
+! CHECK-NEXT: i=7 j=0
+! CHECK-NEXT: i=10 j=0
+! CHECK-NEXT: i=13 j=0
+! CHECK-NEXT: i=7 j=1
+! CHECK-NEXT: i=10 j=1
+! CHECK-NEXT: i=13 j=1
+! CHECK-NEXT: done
diff --git a/openmp/runtime/test/transform/interchange/distribute-parallel-wsloop-simd-intdo.f90 b/openmp/runtime/test/transform/interchange/distribute-parallel-wsloop-simd-intdo.f90
new file mode 100644
index 0000000000000..2a192cad017a6
--- /dev/null
+++ b/openmp/runtime/test/transform/interchange/distribute-parallel-wsloop-simd-intdo.f90
@@ -0,0 +1,37 @@
+
+! RUN: %flang %flags %openmp_flags -fopenmp-version=60 %s -o %t.exe
+! RUN: %t.exe | FileCheck %s --match-full-lines
+
+program interchange_wsloop_intdo
+  integer :: i, j
+  print *, 'do'
+
+  !$OMP TEAMS NUM_TEAMS(1)
+
+  !$OMP DISTRIBUTE PARALLEL DO SIMD SCHEDULE(static,2) NUM_THREADS(1)
+  !$OMP INTERCHANGE
+  do i = 7, 15, 3
+    do j = -1, 1
+      print '("i=", I0, " j=", I0)', i, j
+    end do
+  end do
+  !$OMP END INTERCHANGE
+  !$OMP END DISTRIBUTE PARALLEL DO SIMD
+
+  !$OMP END TEAMS
+
+  print *, 'done'
+end program
+
+
+! CHECK:      do
+! CHECK-NEXT: i=7 j=-1
+! CHECK-NEXT: i=10 j=-1
+! CHECK-NEXT: i=13 j=-1
+! CHECK-NEXT: i=7 j=0
+! CHECK-NEXT: i=10 j=0
+! CHECK-NEXT: i=13 j=0
+! CHECK-NEXT: i=7 j=1
+! CHECK-NEXT: i=10 j=1
+! CHECK-NEXT: i=13 j=1
+! CHECK-NEXT: done
diff --git a/openmp/runtime/test/transform/interchange/distribute-simd-intdo.f90 b/openmp/runtime/test/transform/interchange/distribute-simd-intdo.f90
new file mode 100644
index 0000000000000..fce62b7f3ccda
--- /dev/null
+++ b/openmp/runtime/test/transform/interchange/distribute-simd-intdo.f90
@@ -0,0 +1,37 @@
+
+! RUN: %flang %flags %openmp_flags -fopenmp-version=60 %s -o %t.exe
+! RUN: %t.exe | FileCheck %s --match-full-lines
+
+program interchange_wsloop_intdo
+  integer :: i, j
+  print *, 'do'
+
+  !$OMP TEAMS NUM_TEAMS(1)
+
+  !$OMP DISTRIBUTE SIMD
+  !$OMP INTERCHANGE
+  do i = 7, 15, 3
+    do j = -1, 1
+      print '("i=", I0, " j=", I0)', i, j
+    end do
+  end do
+  !$OMP END INTERCHANGE
+  !$OMP END DISTRIBUTE SIMD
+
+  !$OMP END TEAMS
+
+  print *, 'done'
+end program
+
+
+! CHECK:      do
+! CHECK-NEXT: i=7 j=-1
+! CHECK-NEXT: i=10 j=-1
+! CHECK-NEXT: i=13 j=-1
+! CHECK-NEXT: i=7 j=0
+! CHECK-NEXT: i=10 j=0
+! CHECK-NEXT: i=13 j=0
+! CHECK-NEXT: i=7 j=1
+! CHECK-NEXT: i=10 j=1
+! CHECK-NEXT: i=13 j=1
+! CHECK-NEXT: done
diff --git a/openmp/runtime/test/transform/interchange/intdo-permutation.f90 b/openmp/runtime/test/transform/interchange/intdo-permutation.f90
new file mode 100644
index 0000000000000..a8a8e7f35d018
--- /dev/null
+++ b/openmp/runtime/test/transform/interchange/intdo-permutation.f90
@@ -0,0 +1,42 @@
+
+! RUN: %flang %flags %openmp_flags -fopenmp-version=60 %s -o %t.exe
+! RUN: %t.exe | FileCheck %s --match-full-lines
+
+program interchange_intdo
+  integer :: i, j
+  print *, 'do'
+
+  !$OMP INTERCHANGE PERMUTATION(2,3,1)
+  do i = 7, 15, 3
+    do j = -1, 1, 2
+      do k = 3, 1, -1
+        print '("i=", I0, " j=", I0, " k=", I0)', i, j, k
+      end do
+    end do
+  end do
+  !$OMP END INTERCHANGE
+
+  print *, 'done'
+end program
+
+
+! CHECK:      do
+! CHECK-NEXT: i=7 j=-1 k=3
+! CHECK-NEXT: i=10 j=-1 k=3
+! CHECK-NEXT: i=13 j=-1 k=3
+! CHECK-NEXT: i=7 j=-1 k=2
+! CHECK-NEXT: i=10 j=-1 k=2
+! CHECK-NEXT: i=13 j=-1 k=2
+! CHECK-NEXT: i=7 j=-1 k=1
+! CHECK-NEXT: i=10 j=-1 k=1
+! CHECK-NEXT: i=13 j=-1 k=1
+! CHECK-NEXT: i=7 j=1 k=3
+! CHECK-NEXT: i=10 j=1 k=3
+! CHECK-NEXT: i=13 j=1 k=3
+! CHECK-NEXT: i=7 j=1 k=2
+! CHECK-NEXT: i=10 j=1 k=2
+! CHECK-NEXT: i=13 j=1 k=2
+! CHECK-NEXT: i=7 j=1 k=1
+! CHECK-NEXT: i=10 j=1 k=1
+! CHECK-NEXT: i=13 j=1 k=1
+! CHECK-NEXT: done
diff --git a/openmp/runtime/test/transform/interchange/intdo.f90 b/openmp/runtime/test/transform/interchange/intdo.f90
new file mode 100644
index 0000000000000..fe6820f41dba6
--- /dev/null
+++ b/openmp/runtime/test/transform/interchange/intdo.f90
@@ -0,0 +1,31 @@
+
+! RUN: %flang %flags %openmp_flags -fopenmp-version=60 %s -o %t.exe
+! RUN: %t.exe | FileCheck %s --match-full-lines
+
+program interchange_intdo
+  integer :: i, j
+  print *, 'do'
+
+  !$OMP INTERCHANGE
+  do i = 7, 15, 3
+    do j = -1, 1
+      print '("i=", I0, " j=", I0)', i, j
+    end do
+  end do
+  !$OMP END INTERCHANGE
+
+  print *, 'done'
+end program
+
+
+! CHECK:      do
+! CHECK-NEXT: i=7 j=-1
+! CHECK-NEXT: i=10 j=-1
+! CHECK-NEXT: i=13 j=-1
+! CHECK-NEXT: i=7 j=0
+! CHECK-NEXT: i=10 j=0
+! CHECK-NEXT: i=13 j=0
+! CHECK-NEXT: i=7 j=1
+! CHECK-NEXT: i=10 j=1
+! CHECK-NEXT: i=13 j=1
+! CHECK-NEXT: done
diff --git a/openmp/runtime/test/transform/interchange/parallel-wsloop-collapse-intdo.f90 b/openmp/runtime/test/transform/interchange/parallel-wsloop-collapse-intdo.f90
new file mode 100644
index 0000000000000..498534374ea30
--- /dev/null
+++ b/openmp/runtime/test/transform/interchange/parallel-wsloop-collapse-intdo.f90
@@ -0,0 +1,33 @@
+
+! RUN: %flang %flags %openmp_flags -fopenmp-version=60 %s -o %t.exe
+! RUN: %t.exe | FileCheck %s --match-full-lines
+
+program interchange_wsloop_collapse_intdo
+  integer :: i, j
+  print *, 'do'
+
+  !$OMP PARALLEL DO COLLAPSE(2) NUM_THREADS(1)
+  !$OMP INTERCHANGE
+  do i = 7, 15, 3
+    do j = -1, 1
+      print '("i=", I0, " j=", I0)', i, j
+    end do
+  end do
+  !$OMP END INTERCHANGE
+  !$OMP END PARALLEL DO
+
+  print *, 'done'
+end program
+
+
+! CHECK:      do
+! CHECK-NEXT: i=7 j=-1
+! CHECK-NEXT: i=10 j=-1
+! CHECK-NEXT: i=13 j=-1
+! CHECK-NEXT: i=7 j=0
+! CHECK-NEXT: i=10 j=0
+! CHECK-NEXT: i=13 j=0
+! CHECK-NEXT: i=7 j=1
+! CHECK-NEXT: i=10 j=1
+! CHECK-NEXT: i=13 j=1
+! CHECK-NEXT: done
diff --git a/openmp/runtime/test/transform/interchange/parallel-wsloop-collapse-intdo.o b/openmp/runtime/test/transform/interchange/parallel-wsloop-collapse-intdo.o
new file mode 100644
index 0000000000000..a0abcfdf74fda
Binary files /dev/null and b/openmp/runtime/test/transform/interchange/parallel-wsloop-collapse-intdo.o differ
diff --git a/openmp/runtime/test/transform/interchange/parallel-wsloop-intdo-collapse.f90 b/openmp/runtime/test/transform/interchange/parallel-wsloop-intdo-collapse.f90
new file mode 100644
index 0000000000000..4285edaa775b8
--- /dev/null
+++ b/openmp/runtime/test/transform/interchange/parallel-wsloop-intdo-collapse.f90
@@ -0,0 +1,33 @@
+
+! RUN: %flang %flags %openmp_flags -fopenmp-version=60 %s -o %t.exe
+! RUN: %t.exe | FileCheck %s --match-full-lines
+
+program interchange_wsloop_intdo
+  integer :: i, j
+  print *, 'do'
+
+  !$OMP PARALLEL DO COLLAPSE(2) NUM_THREADS(1)
+  !$OMP INTERCHANGE
+  do i = 7, 15, 3
+    do j = -1, 1
+      print '("i=", I0, " j=", I0)', i, j
+    end do
+  end do
+  !$OMP END INTERCHANGE
+  !$OMP END PARALLEL DO
+
+  print *, 'done'
+end program
+
+
+! CHECK:      do
+! CHECK-NEXT: i=7 j=-1
+! CHECK-NEXT: i=10 j=-1
+! CHECK-NEXT: i=13 j=-1
+! CHECK-NEXT: i=7 j=0
+! CHECK-NEXT: i=10 j=0
+! CHECK-NEXT: i=13 j=0
+! CHECK-NEXT: i=7 j=1
+! CHECK-NEXT: i=10 j=1
+! CHECK-NEXT: i=13 j=1
+! CHECK-NEXT: done
diff --git a/openmp/runtime/test/transform/interchange/parallel-wsloop-intdo-firstprivate.f90 b/openmp/runtime/test/transform/interchange/parallel-wsloop-intdo-firstprivate.f90
new file mode 100644
index 0000000000000..e53bb107bad2b
--- /dev/null
+++ b/openmp/runtime/test/transform/interchange/parallel-wsloop-intdo-firstprivate.f90
@@ -0,0 +1,35 @@
+
+! RUN: %flang %flags %openmp_flags -fopenmp-version=60 %s -o %t.exe
+! RUN: %t.exe | FileCheck %s --match-full-lines
+
+program interchange_wsloop_intdo
+  integer :: i, j, k
+  print *, 'do'
+
+  k = 1
+  !$OMP PARALLEL DO NUM_THREADS(3) FIRSTPRIVATE(k)
+  !$OMP INTERCHANGE
+  do i = 7, 15, 3
+    do j = -1, 1
+      k = k + 1
+      print '("i=", I0, " j=", I0, " k=", I0)', i, j, k
+    end do
+  end do
+  !$OMP END INTERCHANGE
+  !$OMP END PARALLEL DO
+
+  print *, 'done'
+end program
+
+
+! CHECK:      do
+! CHECK-DAG: i=7 j=-1 k=2
+! CHECK-DAG: i=10 j=-1 k=3
+! CHECK-DAG: i=13 j=-1 k=4
+! CHECK-DAG: i=7 j=0 k=2
+! CHECK-DAG: i=10 j=0 k=3
+! CHECK-DAG: i=13 j=0 k=4
+! CHECK-DAG: i=7 j=1 k=2
+! CHECK-DAG: i=10 j=1 k=3
+! CHECK-DAG: i=13 j=1 k=4
+! CHECK-NEXT: done
diff --git a/openmp/runtime/test/transform/interchange/parallel-wsloop-intdo-lastprivate.f90 b/openmp/runtime/test/transform/interchange/parallel-wsloop-intdo-lastprivate.f90
new file mode 100644
index 0000000000000..42d7032bd2184
--- /dev/null
+++ b/openmp/runtime/test/transform/interchange/parallel-wsloop-intdo-lastprivate.f90
@@ -0,0 +1,28 @@
+
+! RUN: %flang %flags %openmp_flags -fopenmp-version=60 %s -o %t.exe
+! RUN: %t.exe | FileCheck %s --match-full-lines
+! XFAIL: *
+
+program interchange_wsloop_intdo
+  integer :: i, j, k
+  print *, 'do'
+
+
+  !$OMP PARALLEL DO LASTPRIVATE(k)
+  !$OMP INTERCHANGE
+  do i = 7, 15, 3
+    do j = -1, 1
+      k = i*10 + j
+    end do
+  end do
+  !$OMP END INTERCHANGE
+  !$OMP END PARALLEL DO
+
+  print *, 'done'
+  print '("k=", I0)', k
+end program
+
+
+! CHECK:      do
+! CHECK-NEXT: done
+! CHECK-NEXT: k=131
diff --git a/openmp/runtime/test/transform/interchange/parallel-wsloop-intdo-permutation.f90 b/openmp/runtime/test/transform/interchange/parallel-wsloop-intdo-permutation.f90
new file mode 100644
index 0000000000000..e52389f2448e4
--- /dev/null
+++ b/openmp/runtime/test/transform/interchange/parallel-wsloop-intdo-permutation.f90
@@ -0,0 +1,33 @@
+
+! RUN: %flang %flags %openmp_flags -fopenmp-version=60 %s -o %t.exe
+! RUN: %t.exe | FileCheck %s --match-full-lines
+
+program interchange_wsloop_intdo
+  integer :: i, j
+  print *, 'do'
+
+  !$OMP PARALLEL DO NUM_THREADS(1)
+  !$OMP INTERCHANGE PERMUTATION(2,1)
+  do i = 7, 15, 3
+    do j = -1, 1
+      print '("i=", I0, " j=", I0)', i, j
+    end do
+  end do
+  !$OMP END INTERCHANGE
+  !$OMP END PARALLEL DO
+
+  print *, 'done'
+end program
+
+
+! CHECK:      do
+! CHECK-NEXT: i=7 j=-1
+! CHECK-NEXT: i=10 j=-1
+! CHECK-NEXT: i=13 j=-1
+! CHECK-NEXT: i=7 j=0
+! CHECK-NEXT: i=10 j=0
+! CHECK-NEXT: i=13 j=0
+! CHECK-NEXT: i=7 j=1
+! CHECK-NEXT: i=10 j=1
+! CHECK-NEXT: i=13 j=1
+! CHECK-NEXT: done
diff --git a/openmp/runtime/test/transform/interchange/parallel-wsloop-intdo-private-i.f90 b/openmp/runtime/test/transform/interchange/parallel-wsloop-intdo-private-i.f90
new file mode 100644
index 0000000000000..76928ce93577e
--- /dev/null
+++ b/openmp/runtime/test/transform/interchange/parallel-wsloop-intdo-private-i.f90
@@ -0,0 +1,33 @@
+
+! RUN: %flang %flags %openmp_flags -fopenmp-version=60 %s -o %t.exe
+! RUN: %t.exe | FileCheck %s --match-full-lines
+
+program interchange_wsloop_intdo
+  integer :: i, j
+  print *, 'do'
+
+  !$OMP PARALLEL DO NUM_THREADS(1) PRIVATE(i)
+  !$OMP INTERCHANGE
+  do i = 7, 15, 3
+    do j = -1, 1
+      print '("i=", I0, " j=", I0)', i, j
+    end do
+  end do
+  !$OMP END INTERCHANGE
+  !$OMP END PARALLEL DO
+
+  print *, 'done'
+end program
+
+
+! CHECK:      do
+! CHECK-NEXT: i=7 j=-1
+! CHECK-NEXT: i=10 j=-1
+! CHECK-NEXT: i=13 j=-1
+! CHECK-NEXT: i=7 j=0
+! CHECK-NEXT: i=10 j=0
+! CHECK-NEXT: i=13 j=0
+! CHECK-NEXT: i=7 j=1
+! CHECK-NEXT: i=10 j=1
+! CHECK-NEXT: i=13 j=1
+! CHECK-NEXT: done
diff --git a/openmp/runtime/test/transform/interchange/parallel-wsloop-intdo-private-j.f90 b/openmp/runtime/test/transform/interchange/parallel-wsloop-intdo-private-j.f90
new file mode 100644
index 0000000000000..a679c921e9660
--- /dev/null
+++ b/openmp/runtime/test/transform/interchange/parallel-wsloop-intdo-private-j.f90
@@ -0,0 +1,33 @@
+
+! RUN: %flang %flags %openmp_flags -fopenmp-version=60 %s -o %t.exe
+! RUN: %t.exe | FileCheck %s --match-full-lines
+
+program interchange_wsloop_intdo
+  integer :: i, j
+  print *, 'do'
+
+  !$OMP PARALLEL DO NUM_THREADS(1) PRIVATE(j)
+  !$OMP INTERCHANGE
+  do i = 7, 15, 3
+    do j = -1, 1
+      print '("i=", I0, " j=", I0)', i, j
+    end do
+  end do
+  !$OMP END INTERCHANGE
+  !$OMP END PARALLEL DO
+
+  print *, 'done'
+end program
+
+
+! CHECK:      do
+! CHECK-NEXT: i=7 j=-1
+! CHECK-NEXT: i=10 j=-1
+! CHECK-NEXT: i=13 j=-1
+! CHECK-NEXT: i=7 j=0
+! CHECK-NEXT: i=10 j=0
+! CHECK-NEXT: i=13 j=0
+! CHECK-NEXT: i=7 j=1
+! CHECK-NEXT: i=10 j=1
+! CHECK-NEXT: i=13 j=1
+! CHECK-NEXT: done
diff --git a/openmp/runtime/test/transform/interchange/parallel-wsloop-intdo-private.f90 b/openmp/runtime/test/transform/interchange/parallel-wsloop-intdo-private.f90
new file mode 100644
index 0000000000000..372ff573a10d2
--- /dev/null
+++ b/openmp/runtime/test/transform/interchange/parallel-wsloop-intdo-private.f90
@@ -0,0 +1,34 @@
+
+! RUN: %flang %flags %openmp_flags -fopenmp-version=60 %s -o %t.exe
+! RUN: %t.exe | FileCheck %s --match-full-lines
+
+program interchange_wsloop_intdo
+  integer :: i, j, k
+  print *, 'do'
+
+  !$OMP PARALLEL DO NUM_THREADS(4) PRIVATE(k)
+  !$OMP INTERCHANGE
+  do i = 7, 15, 3
+    do j = -1, 1
+      k = i + j
+      print '("i=", I0, " j=", I0, " k=", I0)', i, j, k
+    end do
+  end do
+  !$OMP END INTERCHANGE
+  !$OMP END PARALLEL DO
+
+  print *, 'done'
+end program
+
+
+! CHECK:      do
+! CHECK-DAG: i=7 j=-1 k=6
+! CHECK-DAG: i=10 j=-1 k=9
+! CHECK-DAG: i=13 j=-1 k=12
+! CHECK-DAG: i=7 j=0 k=7
+! CHECK-DAG: i=10 j=0 k=10
+! CHECK-DAG: i=13 j=0 k=13
+! CHECK-DAG: i=7 j=1 k=8
+! CHECK-DAG: i=10 j=1 k=11
+! CHECK-DAG: i=13 j=1 k=14
+! CHECK-NEXT: done
diff --git a/openmp/runtime/test/transform/interchange/parallel-wsloop-intdo-reduction.f90 b/openmp/runtime/test/transform/interchange/parallel-wsloop-intdo-reduction.f90
new file mode 100644
index 0000000000000..8d313becef862
--- /dev/null
+++ b/openmp/runtime/test/transform/interchange/parallel-wsloop-intdo-reduction.f90
@@ -0,0 +1,27 @@
+
+! RUN: %flang %flags %openmp_flags -fopenmp-version=60 %s -o %t.exe
+! RUN: %t.exe | FileCheck %s --match-full-lines
+
+program interchange_wsloop_intdo
+  integer :: i, j, k
+  print *, 'do'
+
+  k = 1
+  !$OMP PARALLEL DO REDUCTION(+:k)
+  !$OMP INTERCHANGE
+  do i = 7, 15, 3
+    do j = -1, 1
+      k = k + 1
+    end do
+  end do
+  !$OMP END INTERCHANGE
+  !$OMP END PARALLEL DO
+
+  print *, 'done'
+  print '("k=", I0)', k
+end program
+
+
+! CHECK:      do
+! CHECK-NEXT: done
+! CHECK-NEXT: k=10
diff --git a/openmp/runtime/test/transform/interchange/parallel-wsloop-intdo.f90 b/openmp/runtime/test/transform/interchange/parallel-wsloop-intdo.f90
new file mode 100644
index 0000000000000..cfa3bddf5c8d5
--- /dev/null
+++ b/openmp/runtime/test/transform/interchange/parallel-wsloop-intdo.f90
@@ -0,0 +1,33 @@
+
+! RUN: %flang %flags %openmp_flags -fopenmp-version=60 %s -o %t.exe
+! RUN: %t.exe | FileCheck %s --match-full-lines
+
+program interchange_wsloop_intdo
+  integer :: i, j
+  print *, 'do'
+
+  !$OMP PARALLEL DO NUM_THREADS(1)
+  !$OMP INTERCHANGE
+  do i = 7, 15, 3
+    do j = -1, 1
+      print '("i=", I0, " j=", I0)', i, j
+    end do
+  end do
+  !$OMP END INTERCHANGE
+  !$OMP END PARALLEL DO
+
+  print *, 'done'
+end program
+
+
+! CHECK:      do
+! CHECK-NEXT: i=7 j=-1
+! CHECK-NEXT: i=10 j=-1
+! CHECK-NEXT: i=13 j=-1
+! CHECK-NEXT: i=7 j=0
+! CHECK-NEXT: i=10 j=0
+! CHECK-NEXT: i=13 j=0
+! CHECK-NEXT: i=7 j=1
+! CHECK-NEXT: i=10 j=1
+! CHECK-NEXT: i=13 j=1
+! CHECK-NEXT: done
diff --git a/openmp/runtime/test/transform/interchange/target-teams-distribute-parallel-wsloop-intdo.f90 b/openmp/runtime/test/transform/interchange/target-teams-distribute-parallel-wsloop-intdo.f90
new file mode 100644
index 0000000000000..81e681b55eb1d
--- /dev/null
+++ b/openmp/runtime/test/transform/interchange/target-teams-distribute-parallel-wsloop-intdo.f90
@@ -0,0 +1,33 @@
+
+! RUN: %flang %flags %openmp_flags -fopenmp-version=60 %s -o %t.exe
+! RUN: %t.exe | FileCheck %s --match-full-lines
+
+program target_teams_distribute_parallel_do
+  integer :: i, j
+  print *, 'do'
+
+  !$OMP TARGET TEAMS DISTRIBUTE PARALLEL DO SCHEDULE(static,2) NUM_TEAMS(1) NUM_THREADS(1)
+  !$OMP INTERCHANGE
+  do i = 7, 15, 3
+    do j = -1, 1
+      print '("i=", I0, " j=", I0)', i, j
+    end do
+  end do
+  !$OMP END INTERCHANGE
+  !$OMP END TARGET TEAMS DISTRIBUTE PARALLEL DO
+
+  print *, 'done'
+end program
+
+
+! CHECK:      do
+! CHECK-NEXT: i=7 j=-1
+! CHECK-NEXT: i=10 j=-1
+! CHECK-NEXT: i=13 j=-1
+! CHECK-NEXT: i=7 j=0
+! CHECK-NEXT: i=10 j=0
+! CHECK-NEXT: i=13 j=0
+! CHECK-NEXT: i=7 j=1
+! CHECK-NEXT: i=10 j=1
+! CHECK-NEXT: i=13 j=1
+! CHECK-NEXT: done
diff --git a/openmp/runtime/test/transform/interchange/taskloop-intdo.f90 b/openmp/runtime/test/transform/interchange/taskloop-intdo.f90
new file mode 100644
index 0000000000000..d79f92d2ad074
--- /dev/null
+++ b/openmp/runtime/test/transform/interchange/taskloop-intdo.f90
@@ -0,0 +1,35 @@
+
+! XFAIL: *
+! RUN: %flang %flags %openmp_flags -fopenmp-version=60 %s -o %t.exe
+! RUN: %t.exe | FileCheck %s --match-full-lines
+
+
+program interchange_wsloop_intdo
+  integer :: i, j
+  print *, 'do'
+
+  !$OMP TASKLOOP
+  !$OMP INTERCHANGE
+  do i = 7, 15, 3
+    do j = -1, 1
+      print '("i=", I0, " j=", I0)', i, j
+    end do
+  end do
+  !$OMP END INTERCHANGE
+  !$OMP END TASKLOOP
+
+  print *, 'done'
+end program
+
+
+! CHECK:      do
+! CHECK-NEXT: i=7 j=-1
+! CHECK-NEXT: i=10 j=-1
+! CHECK-NEXT: i=13 j=-1
+! CHECK-NEXT: i=7 j=0
+! CHECK-NEXT: i=10 j=0
+! CHECK-NEXT: i=13 j=0
+! CHECK-NEXT: i=7 j=1
+! CHECK-NEXT: i=10 j=1
+! CHECK-NEXT: i=13 j=1
+! CHECK-NEXT: done
diff --git a/openmp/runtime/test/transform/interchange/taskloop-simd-intdo.f90 b/openmp/runtime/test/transform/interchange/taskloop-simd-intdo.f90
new file mode 100644
index 0000000000000..d84be9d1d7a96
--- /dev/null
+++ b/openmp/runtime/test/transform/interchange/taskloop-simd-intdo.f90
@@ -0,0 +1,34 @@
+
+! RUN: %flang %flags %openmp_flags -fopenmp-version=60 %s -o %t.exe
+! RUN: %t.exe | FileCheck %s --match-full-lines
+! XFAIL: *
+
+program interchange_wsloop_intdo
+  integer :: i, j
+  print *, 'do'
+
+  !$OMP TASKLOOP SIMD
+  !$OMP INTERCHANGE
+  do i = 7, 15, 3
+    do j = -1, 1
+      print '("i=", I0, " j=", I0)', i, j
+    end do
+  end do
+  !$OMP END INTERCHANGE
+  !$OMP END TASKLOOP SIMD
+
+  print *, 'done'
+end program
+
+
+! CHECK:      do
+! CHECK-NEXT: i=7 j=-1
+! CHECK-NEXT: i=10 j=-1
+! CHECK-NEXT: i=13 j=-1
+! CHECK-NEXT: i=7 j=0
+! CHECK-NEXT: i=10 j=0
+! CHECK-NEXT: i=13 j=0
+! CHECK-NEXT: i=7 j=1
+! CHECK-NEXT: i=10 j=1
+! CHECK-NEXT: i=13 j=1
+! CHECK-NEXT: done
diff --git a/openmp/runtime/test/transform/interchange/teams-distribute-parallel-wsloop-intdo.f90 b/openmp/runtime/test/transform/interchange/teams-distribute-parallel-wsloop-intdo.f90
new file mode 100644
index 0000000000000..6d7fe1afdcdd5
--- /dev/null
+++ b/openmp/runtime/test/transform/interchange/teams-distribute-parallel-wsloop-intdo.f90
@@ -0,0 +1,33 @@
+
+! RUN: %flang %flags %openmp_flags -fopenmp-version=60 %s -o %t.exe
+! RUN: %t.exe | FileCheck %s --match-full-lines
+
+program teams_distribute_parallel_do
+  integer :: i, j
+  print *, 'do'
+
+  !$OMP TEAMS DISTRIBUTE PARALLEL DO SCHEDULE(static,2) NUM_TEAMS(1) NUM_THREADS(1)
+  !$OMP INTERCHANGE
+  do i = 7, 15, 3
+    do j = -1, 1
+      print '("i=", I0, " j=", I0)', i, j
+    end do
+  end do
+  !$OMP END INTERCHANGE
+  !$OMP END TEAMS DISTRIBUTE PARALLEL DO
+
+  print *, 'done'
+end program
+
+
+! CHECK:      do
+! CHECK-NEXT: i=7 j=-1
+! CHECK-NEXT: i=10 j=-1
+! CHECK-NEXT: i=13 j=-1
+! CHECK-NEXT: i=7 j=0
+! CHECK-NEXT: i=10 j=0
+! CHECK-NEXT: i=13 j=0
+! CHECK-NEXT: i=7 j=1
+! CHECK-NEXT: i=10 j=1
+! CHECK-NEXT: i=13 j=1
+! CHECK-NEXT: done
diff --git a/openmp/runtime/test/transform/interchange/wsloop-intdo.f90 b/openmp/runtime/test/transform/interchange/wsloop-intdo.f90
new file mode 100644
index 0000000000000..32b1b87a9e859
--- /dev/null
+++ b/openmp/runtime/test/transform/interchange/wsloop-intdo.f90
@@ -0,0 +1,33 @@
+
+! RUN: %flang %flags %openmp_flags -fopenmp-version=60 %s -o %t.exe
+! RUN: %t.exe | FileCheck %s --match-full-lines
+
+program interchange_wsloop_intdo
+  integer :: i, j
+  print *, 'do'
+
+  !$OMP DO SCHEDULE(static,2)
+  !$OMP INTERCHANGE
+  do i = 7, 15, 3
+    do j = -1, 1
+      print '("i=", I0, " j=", I0)', i, j
+    end do
+  end do
+  !$OMP END INTERCHANGE
+  !$OMP END DO
+
+  print *, 'done'
+end program
+
+
+! CHECK:      do
+! CHECK-NEXT: i=7 j=-1
+! CHECK-NEXT: i=10 j=-1
+! CHECK-NEXT: i=13 j=-1
+! CHECK-NEXT: i=7 j=0
+! CHECK-NEXT: i=10 j=0
+! CHECK-NEXT: i=13 j=0
+! CHECK-NEXT: i=7 j=1
+! CHECK-NEXT: i=10 j=1
+! CHECK-NEXT: i=13 j=1
+! CHECK-NEXT: done
diff --git a/openmp/runtime/test/transform/interchange/wsloop-simd-intdo.f90 b/openmp/runtime/test/transform/interchange/wsloop-simd-intdo.f90
new file mode 100644
index 0000000000000..56ed14b165fa3
--- /dev/null
+++ b/openmp/runtime/test/transform/interchange/wsloop-simd-intdo.f90
@@ -0,0 +1,33 @@
+
+! RUN: %flang %flags %openmp_flags -fopenmp-version=60 %s -o %t.exe
+! RUN: %t.exe | FileCheck %s --match-full-lines
+
+program interchange_wsloop_intdo
+  integer :: i, j
+  print *, 'do'
+
+  !$OMP DO SIMD SCHEDULE(static,2)
+  !$OMP INTERCHANGE
+  do i = 7, 15, 3
+    do j = -1, 1
+      print '("i=", I0, " j=", I0)', i, j
+    end do
+  end do
+  !$OMP END INTERCHANGE
+  !$OMP END DO SIMD
+
+  print *, 'done'
+end program
+
+
+! CHECK:      do
+! CHECK-NEXT: i=7 j=-1
+! CHECK-NEXT: i=10 j=-1
+! CHECK-NEXT: i=13 j=-1
+! CHECK-NEXT: i=7 j=0
+! CHECK-NEXT: i=10 j=0
+! CHECK-NEXT: i=13 j=0
+! CHECK-NEXT: i=7 j=1
+! CHECK-NEXT: i=10 j=1
+! CHECK-NEXT: i=13 j=1
+! CHECK-NEXT: done
diff --git a/openmp/runtime/test/transform/tile/intdo.f90 b/openmp/runtime/test/transform/tile/intdo.f90
new file mode 100644
index 0000000000000..27cd383a69fea
--- /dev/null
+++ b/openmp/runtime/test/transform/tile/intdo.f90
@@ -0,0 +1,57 @@
+! This test checks lowering of OpenMP tile directive
+
+! RUN: %flang %flags %openmp_flags -fopenmp-version=51 %s -o %t.exe
+! RUN: %t.exe | FileCheck %s --match-full-lines
+
+
+program tile_intdo
+  integer i, j
+  print *, 'do'
+
+  !$OMP TILE SIZES(2,2)
+  do i=7, 18, 3
+    do j=7, 19, 3
+      print '("i=", I0, " j=", I0)', i, j
+    end do
+  end do
+  !$OMP END TILE
+
+  print *, 'done'
+end program
+
+
+! CHECK:      do
+
+! Complete tile
+! CHECK-NEXT: i=7 j=7
+! CHECK-NEXT: i=7 j=10
+! CHECK-NEXT: i=10 j=7
+! CHECK-NEXT: i=10 j=10
+
+! Complete tile
+! CHECK-NEXT: i=7 j=13
+! CHECK-NEXT: i=7 j=16
+! CHECK-NEXT: i=10 j=13
+! CHECK-NEXT: i=10 j=16
+
+! Partial tile
+! CHECK-NEXT: i=7 j=19
+! CHECK-NEXT: i=10 j=19
+
+! Complete tile
+! CHECK-NEXT: i=13 j=7
+! CHECK-NEXT: i=13 j=10
+! CHECK-NEXT: i=16 j=7
+! CHECK-NEXT: i=16 j=10
+
+! Complete tile
+! CHECK-NEXT: i=13 j=13
+! CHECK-NEXT: i=13 j=16
+! CHECK-NEXT: i=16 j=13
+! CHECK-NEXT: i=16 j=16
+
+! Partial tile
+! CHECK-NEXT: i=13 j=19
+! CHECK-NEXT: i=16 j=19
+
+! CHECK-NEXT: done
diff --git a/openmp/runtime/test/transform/tile/parallel-wsloop-collapse-intdo.f90 b/openmp/runtime/test/transform/tile/parallel-wsloop-collapse-intdo.f90
new file mode 100644
index 0000000000000..66bddf30e045a
--- /dev/null
+++ b/openmp/runtime/test/transform/tile/parallel-wsloop-collapse-intdo.f90
@@ -0,0 +1,59 @@
+! This test checks lowering of OpenMP tile directive
+
+! RUN: %flang %flags %openmp_flags -fopenmp-version=51 %s -o %t.exe
+! RUN: %t.exe | FileCheck %s --match-full-lines
+
+
+program tile_wsloop_collapse_intdo
+  integer i, j
+  print *, 'do'
+
+  !$OMP PARALLEL DO COLLAPSE(2) NUM_THREADS(1)
+  !$OMP TILE SIZES(2,2)
+  do i=7, 18, 3
+    do j=7, 19, 3
+      print '("i=", I0, " j=", I0)', i, j
+    end do
+  end do
+  !$OMP END TILE
+  !$OMP END PARALLEL DO
+
+  print *, 'done'
+end program
+
+
+! CHECK:      do
+
+! Complete tile
+! CHECK-NEXT: i=7 j=7
+! CHECK-NEXT: i=7 j=10
+! CHECK-NEXT: i=10 j=7
+! CHECK-NEXT: i=10 j=10
+
+! Complete tile
+! CHECK-NEXT: i=7 j=13
+! CHECK-NEXT: i=7 j=16
+! CHECK-NEXT: i=10 j=13
+! CHECK-NEXT: i=10 j=16
+
+! Partial tile
+! CHECK-NEXT: i=7 j=19
+! CHECK-NEXT: i=10 j=19
+
+! Complete tile
+! CHECK-NEXT: i=13 j=7
+! CHECK-NEXT: i=13 j=10
+! CHECK-NEXT: i=16 j=7
+! CHECK-NEXT: i=16 j=10
+
+! Complete tile
+! CHECK-NEXT: i=13 j=13
+! CHECK-NEXT: i=13 j=16
+! CHECK-NEXT: i=16 j=13
+! CHECK-NEXT: i=16 j=16
+
+! Partial tile
+! CHECK-NEXT: i=13 j=19
+! CHECK-NEXT: i=16 j=19
+
+! CHECK-NEXT: done