[flang-commits] [flang] [flang][cuda] Avoid inserting GetDeviceAddress call in offload regions (PR #175225)

[Razvan Lupusoru via flang-commits]

================
@@ -49,9 +50,9 @@ namespace {
 static bool inDeviceContext(mlir::Operation *op) {
   if (op->getParentOfType<cuf::KernelOp>())
     return true;
-  if (auto funcOp = op->getParentOfType<mlir::gpu::GPUFuncOp>())
+  if (op->getParentOfType<mlir::acc::OffloadRegionOpInterface>())
----------------
razvanlupusoru wrote:

The diffs look a bit weird - so to explain what I did:
- I added a new check for acc::OffloadRegionOpInterface which includes both acc regions and gpu.launch
- I removed the gpu.launch case since it is already covered.

https://github.com/llvm/llvm-project/pull/175225