[Mlir-commits] [mlir] [ROCDL] Pass `amd_code_object_version` when serializing ROCDL gpu module (PR #108874)

[Matt Arsenault]

================
@@ -104,6 +104,60 @@ TEST_F(MLIRTargetLLVMROCDL, SKIP_WITHOUT_AMDGPU(SerializeROCDLMToLLVM)) {
     ASSERT_TRUE((*llvmModule)->getFunction("rocdl_kernel") != nullptr);
   }
 }
+// Test ROCDL serialization to ISA with default code object version.
+TEST_F(MLIRTargetLLVMROCDL,
+       SKIP_WITHOUT_AMDGPU(SerializeROCDLToISAWithDefaultCOV)) {
+  MLIRContext context(registry);
+
+  OwningOpRef<ModuleOp> module =
+      parseSourceString<ModuleOp>(moduleStr, &context);
+  ASSERT_TRUE(!!module);
+
+  // Create a ROCDL target.
+  ROCDL::ROCDLTargetAttr target = ROCDL::ROCDLTargetAttr::get(&context);
+
+  // Serialize the module.
+  auto serializer = dyn_cast<gpu::TargetAttrInterface>(target);
+  ASSERT_TRUE(!!serializer);
+  gpu::TargetOptions options("", {}, "", gpu::CompilationTarget::Assembly);
+  for (auto gpuModule : (*module).getBody()->getOps<gpu::GPUModuleOp>()) {
+    std::optional<SmallVector<char, 0>> object =
+        serializer.serializeToObject(gpuModule, options);
+    // Check that the serializer was successful.
+    ASSERT_TRUE(object != std::nullopt);
+    ASSERT_TRUE(!object->empty());
----------------
arsenm wrote:

this is redundant with the following check 

https://github.com/llvm/llvm-project/pull/108874