Update the OpenXLA pin (#9216)

Co-authored-by: Haifeng Jin <[email protected]>

Author: haifeng-jin

WORKSPACE

@@ -46,7 +46,7 @@ new_local_repository(
 
 # To build PyTorch/XLA with a new revison of OpenXLA, update the xla_hash to
 # the openxla git commit hash and note the date of the commit.
-xla_hash = '37b580d0dd7b52e611430d33424159719e363ffc'  # Committed on 2025/5/13.
+xla_hash = '30ebb24dd16a31b613623d9f2f513d31914cb6f7'  # Committed on 2025/5/19.
 
 http_archive(
     name = "xla",

torch_xla/csrc/runtime/ifrt_computation_client.cpp

@@ -502,7 +502,7 @@ std::vector<ComputationClient::ComputationPtr> IfrtComputationClient::Compile(
     torch_xla::ConvertHloToStableHlo(instance.computation.mutable_proto(),
                                      &mlir_module);
     std::shared_ptr<xla::ifrt::LoadedExecutable> executable =
-        ConsumeValue(client_->GetDefaultCompiler()->Compile(
+        ConsumeValue(client_->GetDefaultCompiler()->CompileAndLoad(
             std::make_unique<xla::ifrt::HloProgram>(mlir_module),
             std::make_unique<xla::ifrt::XlaCompileOptions>(compile_options,
                                                            devices_list)));