[flang-commits] [flang] [mlir] [mlir][IR] Add `SingleBlock` trait to `ModuleOp`/`GPUModuleOp` (PR #110498)

[via flang-commits]

llvmbot wrote:


<!--LLVM PR SUMMARY COMMENT-->

@llvm/pr-subscribers-mlir-gpu

Author: Matthias Springer (matthias-springer)

<details>
<summary>Changes</summary>

The region is already declared as `SizedRegion<1>`, but this will add an additional `getBlock()` convenience function to `ModuleOp`.

---
Full diff: https://github.com/llvm/llvm-project/pull/110498.diff


6 Files Affected:

- (modified) flang/lib/Lower/OpenMP/DataSharingProcessor.cpp (+2-2) 
- (modified) mlir/include/mlir/Dialect/GPU/IR/GPUOps.td (+2-1) 
- (modified) mlir/include/mlir/IR/BuiltinOps.td (+2-2) 
- (modified) mlir/lib/Conversion/BufferizationToMemRef/BufferizationToMemRef.cpp (+1-2) 
- (modified) mlir/lib/Dialect/Bufferization/Transforms/LowerDeallocations.cpp (+1-2) 
- (modified) mlir/lib/Dialect/SparseTensor/Transforms/SparseGPUCodegen.cpp (+2-2) 


``````````diff
diff --git a/flang/lib/Lower/OpenMP/DataSharingProcessor.cpp b/flang/lib/Lower/OpenMP/DataSharingProcessor.cpp
index 5f4138e0f63e73..c3ce9662631503 100644
--- a/flang/lib/Lower/OpenMP/DataSharingProcessor.cpp
+++ b/flang/lib/Lower/OpenMP/DataSharingProcessor.cpp
@@ -478,8 +478,8 @@ void DataSharingProcessor::doPrivatize(const semantics::Symbol *sym,
       return existingPrivatizer;
 
     mlir::OpBuilder::InsertionGuard guard(firOpBuilder);
-    firOpBuilder.setInsertionPoint(&moduleOp.getBodyRegion().front(),
-                                   moduleOp.getBodyRegion().front().begin());
+    firOpBuilder.setInsertionPoint(moduleOp.getBody(),
+                                   moduleOp.getBody()->begin());
     auto result = firOpBuilder.create<mlir::omp::PrivateClauseOp>(
         symLoc, uniquePrivatizerName, symType,
         isFirstPrivate ? mlir::omp::DataSharingClauseType::FirstPrivate
diff --git a/mlir/include/mlir/Dialect/GPU/IR/GPUOps.td b/mlir/include/mlir/Dialect/GPU/IR/GPUOps.td
index 6098eb34d04d52..732391aadab0d4 100644
--- a/mlir/include/mlir/Dialect/GPU/IR/GPUOps.td
+++ b/mlir/include/mlir/Dialect/GPU/IR/GPUOps.td
@@ -1380,7 +1380,8 @@ def GPU_BarrierOp : GPU_Op<"barrier"> {
 
 def GPU_GPUModuleOp : GPU_Op<"module", [
       DataLayoutOpInterface, HasDefaultDLTIDataLayout, IsolatedFromAbove,
-      NoRegionArguments, SymbolTable, Symbol] # GraphRegionNoTerminator.traits> {
+      NoRegionArguments, SingleBlock, SymbolTable, Symbol]
+    # GraphRegionNoTerminator.traits> {
   let summary = "A top level compilation unit containing code to be run on a GPU.";
   let description = [{
     GPU module contains code that is intended to be run on a GPU. A host device
diff --git a/mlir/include/mlir/IR/BuiltinOps.td b/mlir/include/mlir/IR/BuiltinOps.td
index 56edd7519cd67d..63ba914e219569 100644
--- a/mlir/include/mlir/IR/BuiltinOps.td
+++ b/mlir/include/mlir/IR/BuiltinOps.td
@@ -31,8 +31,8 @@ class Builtin_Op<string mnemonic, list<Trait> traits = []> :
 //===----------------------------------------------------------------------===//
 
 def ModuleOp : Builtin_Op<"module", [
-    AffineScope, IsolatedFromAbove, NoRegionArguments, SymbolTable, Symbol,
-    OpAsmOpInterface
+    AffineScope, IsolatedFromAbove, NoRegionArguments, SingleBlock,
+    SymbolTable, Symbol, OpAsmOpInterface
   ] # GraphRegionNoTerminator.traits> {
   let summary = "A top level container operation";
   let description = [{
diff --git a/mlir/lib/Conversion/BufferizationToMemRef/BufferizationToMemRef.cpp b/mlir/lib/Conversion/BufferizationToMemRef/BufferizationToMemRef.cpp
index f9903071be0842..06aedc5e139d37 100644
--- a/mlir/lib/Conversion/BufferizationToMemRef/BufferizationToMemRef.cpp
+++ b/mlir/lib/Conversion/BufferizationToMemRef/BufferizationToMemRef.cpp
@@ -134,8 +134,7 @@ struct BufferizationToMemRefPass
 
     bufferization::DeallocHelperMap deallocHelperFuncMap;
     if (auto module = dyn_cast<ModuleOp>(getOperation())) {
-      OpBuilder builder =
-          OpBuilder::atBlockBegin(&module.getBodyRegion().front());
+      OpBuilder builder = OpBuilder::atBlockBegin(module.getBody());
 
       // Build dealloc helper function if there are deallocs.
       getOperation()->walk([&](bufferization::DeallocOp deallocOp) {
diff --git a/mlir/lib/Dialect/Bufferization/Transforms/LowerDeallocations.cpp b/mlir/lib/Dialect/Bufferization/Transforms/LowerDeallocations.cpp
index 9e2c91bad7bfdd..31d165ce154070 100644
--- a/mlir/lib/Dialect/Bufferization/Transforms/LowerDeallocations.cpp
+++ b/mlir/lib/Dialect/Bufferization/Transforms/LowerDeallocations.cpp
@@ -391,8 +391,7 @@ struct LowerDeallocationsPass
 
     bufferization::DeallocHelperMap deallocHelperFuncMap;
     if (auto module = dyn_cast<ModuleOp>(getOperation())) {
-      OpBuilder builder =
-          OpBuilder::atBlockBegin(&module.getBodyRegion().front());
+      OpBuilder builder = OpBuilder::atBlockBegin(module.getBody());
 
       // Build dealloc helper function if there are deallocs.
       getOperation()->walk([&](bufferization::DeallocOp deallocOp) {
diff --git a/mlir/lib/Dialect/SparseTensor/Transforms/SparseGPUCodegen.cpp b/mlir/lib/Dialect/SparseTensor/Transforms/SparseGPUCodegen.cpp
index 8be76cac87f297..b7fac163ba5fe3 100644
--- a/mlir/lib/Dialect/SparseTensor/Transforms/SparseGPUCodegen.cpp
+++ b/mlir/lib/Dialect/SparseTensor/Transforms/SparseGPUCodegen.cpp
@@ -58,7 +58,7 @@ static gpu::GPUModuleOp genGPUModule(OpBuilder &builder, ModuleOp topModule) {
   for (auto op : topModule.getBodyRegion().getOps<gpu::GPUModuleOp>())
     return op; // existing
   markAsGPUContainer(topModule);
-  builder.setInsertionPointToStart(&topModule.getBodyRegion().front());
+  builder.setInsertionPointToStart(topModule.getBody());
   return builder.create<gpu::GPUModuleOp>(topModule->getLoc(),
                                           "sparse_kernels");
 }
@@ -75,7 +75,7 @@ static gpu::GPUFuncOp genGPUFunc(OpBuilder &builder, gpu::GPUModuleOp gpuModule,
     ("kernel" + Twine(kernelNumber++)).toStringRef(kernelName);
   } while (gpuModule.lookupSymbol(kernelName));
   // Then we insert a new kernel with given arguments into the module.
-  builder.setInsertionPointToStart(&gpuModule.getBodyRegion().front());
+  builder.setInsertionPointToStart(gpuModule.getBody());
   SmallVector<Type> argsTp;
   for (auto arg : args)
     argsTp.push_back(arg.getType());

``````````

</details>


https://github.com/llvm/llvm-project/pull/110498