[Mlir-commits] [clang] [flang] [mlir] [flang][flang-driver][mlir][OpenMP] atomic control support (PR #143441)

[Sergio Afonso]

================
@@ -88,6 +88,36 @@ void fir::setTuneCPU(mlir::ModuleOp mod, llvm::StringRef cpu) {
   mod->setAttr(tuneCpuName, mlir::StringAttr::get(ctx, cpu));
 }
 
+static constexpr const char *ignoreDenormalModeName =
+    "fir.ignore.denormal.mode";
+void fir::setIgnoreDenormalMode(mlir::ModuleOp mod) {
+  auto *ctx = mod.getContext();
+  mod->setAttr(ignoreDenormalModeName, mlir::UnitAttr::get(ctx));
+}
+
+bool fir::getIgnoreDenormalMode(mlir::ModuleOp mod) {
+  return mod->hasAttrOfType<mlir::UnitAttr>(ignoreDenormalModeName);
+}
+
+static constexpr const char *fineGrainedMemoryName = "fir.fine.grained.memory";
----------------
skatrak wrote:

```suggestion
static constexpr const char *fineGrainedMemoryName = "fir.atomic_fine_grained_memory";
```

https://github.com/llvm/llvm-project/pull/143441