Add WOQ int8 test with Inductor Freeze (#362)

Author: leslie-fang-intel

test/integration/test_integration.py

@@ -798,6 +798,14 @@ def test_int8_weight_only_quant_subclass_api(self, device, dtype):
             _int8wo_api, device, 40, test_dtype=dtype
         )
 
+    @parameterized.expand(COMMON_DEVICE_DTYPE)
+    @torch._inductor.config.patch({"freezing": True})
+    @unittest.skipIf(not TORCH_VERSION_AFTER_2_4, "freeze requires torch 2.4 and after.")
+    def test_int8_weight_only_quant_with_freeze(self, device, dtype):
+        self._test_lin_weight_subclass_api_impl(
+            _int8wo_api, device, 40, test_dtype=dtype
+        )
+
     @parameterized.expand(COMMON_DEVICE_DTYPE)
     @unittest.skipIf(not TORCH_VERSION_AFTER_2_3, "int4 requires torch nightly.")
     def test_int4_weight_only_quant_subclass_api(self, device, dtype):