[Mlir-commits] [clang] [llvm] [mlir] [clang][NVPTX][MLIR][NVVM] Add overloaded fadd intrinsics (PR #217336)

[Durgadoss R]

================
@@ -448,6 +448,22 @@ getFenceProxySyncRestrictID(NVVM::MemOrderKind order) {
                    nvvm_fence_proxy_async_generic_release_sync_restrict_space_cta_scope_cluster;
 }
 
+static llvm::RoundingMode getLLVMRoundingMode(NVVM::FPRoundingMode rndMode) {
+  switch (rndMode) {
+  case NVVM::FPRoundingMode::NONE:
----------------
durga4github wrote:

This is here with RN because RN is the default mode?
I wonder if we should just let it be handled by the `default` case 

https://github.com/llvm/llvm-project/pull/217336