Arm backend: Enable mypy lintrunner for backends/arm/test

.lintrunner.toml

@@ -320,7 +320,6 @@ exclude_patterns = [
     '**/third-party/**',
     'scripts/check_binary_dependencies.py',
     'profiler/test/test_profiler_e2e.py',
-    'backends/arm/test/**',
 ]
 command = [
     'python',

backends/arm/test/common.py

@@ -13,14 +13,20 @@
 from typing import Any
 
 import pytest
-from executorch.backends.arm.arm_backend import ArmCompileSpecBuilder
-from executorch.backends.arm.test.runner_utils import (
+from executorch.backends.arm.arm_backend import (  # type: ignore[import-not-found]
+    ArmCompileSpecBuilder,
+)
+from executorch.backends.arm.test.runner_utils import (  # type: ignore[import-not-found]
     arm_executor_runner_exists,
     corstone300_installed,
     corstone320_installed,
 )
-from executorch.backends.arm.tosa_specification import TosaSpecification
-from executorch.exir.backend.compile_spec_schema import CompileSpec
+from executorch.backends.arm.tosa_specification import (  # type: ignore[import-not-found]
+    TosaSpecification,
+)
+from executorch.exir.backend.compile_spec_schema import (  # type: ignore[import-not-found]
+    CompileSpec,
+)
 
 
 def get_time_formatted_path(path: str, log_prefix: str) -> str:
@@ -193,7 +199,7 @@ def get_u85_compile_spec_unbuilt(
 
 
 def parametrize(
-    arg_name: str, test_data: dict[str, Any], xfails: dict[str, str] = None
+    arg_name: str, test_data: dict[str, Any], xfails: dict[str, str] | None = None
 ):
     """
     Custom version of pytest.mark.parametrize with some syntatic sugar and added xfail functionality

backends/arm/test/conftest.py

@@ -15,7 +15,7 @@
 import pytest
 
 try:
-    import tosa_reference_model
+    import tosa_reference_model  # type: ignore[import-untyped]
 except ImportError:
     logging.warning("tosa_reference_model not found, can't run reference model tests")
     tosa_reference_model = None

backends/arm/test/runner_utils.py

@@ -20,22 +20,36 @@
 
 logger = logging.getLogger(__name__)
 try:
-    import tosa_reference_model
+    import tosa_reference_model  # type: ignore[import-untyped]
 except ImportError:
     tosa_reference_model = None
-from executorch.backends.arm.arm_backend import get_tosa_spec, is_tosa
-
-from executorch.backends.arm.test.conftest import is_option_enabled
-from executorch.backends.arm.tosa_specification import TosaSpecification
-from executorch.exir import ExecutorchProgramManager, ExportedProgram
-from executorch.exir.backend.compile_spec_schema import CompileSpec
-from executorch.exir.lowered_backend_module import LoweredBackendModule
+from executorch.backends.arm.arm_backend import (  # type: ignore[import-not-found]
+    get_tosa_spec,
+    is_tosa,
+)
+
+from executorch.backends.arm.test.conftest import (  # type: ignore[import-not-found]
+    is_option_enabled,
+)
+from executorch.backends.arm.tosa_specification import (  # type: ignore[import-not-found]
+    TosaSpecification,
+)
+from executorch.exir import (  # type: ignore[import-not-found]
+    ExecutorchProgramManager,
+    ExportedProgram,
+)
+from executorch.exir.backend.compile_spec_schema import (  # type: ignore[import-not-found]
+    CompileSpec,
+)
+from executorch.exir.lowered_backend_module import (  # type: ignore[import-not-found]
+    LoweredBackendModule,
+)
 from packaging.version import Version
 from torch.fx.node import Node
 
 from torch.overrides import TorchFunctionMode
 from torch.testing._internal.common_utils import torch_to_numpy_dtype_dict
-from tosa import TosaGraph
+from tosa import TosaGraph  # type: ignore[import-untyped]
 
 logger = logging.getLogger(__name__)
 logger.setLevel(logging.CRITICAL)
@@ -148,16 +162,16 @@ def get_output_quantization_params(
     for node in output_nodes:
         if node.target == torch.ops.quantized_decomposed.dequantize_per_tensor.default:
             quant_params[node] = QuantizationParams(
-                node_name=node.args[0].name,
-                scale=node.args[1],
-                zp=node.args[2],
-                qmin=node.args[3],
-                qmax=node.args[4],
-                dtype=node.args[5],
+                node_name=node.args[0].name,  # type: ignore
+                scale=node.args[1],  # type: ignore
+                zp=node.args[2],  # type: ignore
+                qmin=node.args[3],  # type: ignore
+                qmax=node.args[4],  # type: ignore
+                dtype=node.args[5],  # type: ignore
             )
         else:
-            quant_params[node] = None
-    return quant_params
+            quant_params[node] = None  # type: ignore[assignment]
+    return quant_params  # type: ignore[return-value]
 
 
 class TosaReferenceModelDispatch(TorchFunctionMode):
@@ -243,7 +257,7 @@ def run_corstone(
     input_names = get_input_names(exported_program)
     input_paths = []
     for input_name, input_ in zip(input_names, inputs):
-        input_path = save_bytes(intermediate_path, input_, input_name)
+        input_path = save_bytes(intermediate_path.as_posix(), input_, input_name)
         input_paths.append(input_path)
 
     out_path = os.path.join(intermediate_path, "out")
@@ -348,7 +362,7 @@ def prep_data_for_save(
     quant_param: Optional[QuantizationParams] = None,
 ):
     if isinstance(data, torch.Tensor):
-        data_np = np.array(data.detach(), order="C").astype(
+        data_np = np.array(data.detach(), order="C").astype(  # type: ignore[var-annotated]
             torch_to_numpy_dtype_dict[data.dtype]
         )
     else:

backends/arm/test/test_model.py

@@ -83,7 +83,7 @@ def get_args():
     return args
 
 
-def run_external_cmd(cmd: []):
+def run_external_cmd(cmd: list[str]):
     print("CALL:", *cmd, sep=" ")
     try:
         subprocess.check_call(cmd)