EngineX/xc-llm-ascend
3
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

Revert "[Feature][Quant] Auto-detect quantization format from model f… (#6873)

Browse Source 
This reverts commit 3953dcf784. to keep
the basic functions available

---------

Signed-off-by: wangli <wangli858794774@gmail.com>
This commit is contained in:
Li Wang
2026-03-10 11:27:32 +08:00
committed by GitHub
parent 40f7d93f1a
commit 33234aa0c5
7 changed files with 12 additions and 584 deletions
Download Patch File Download Diff File Expand all files Collapse all files

24
tests/ut/test_platform.py
View File

@@ -125,14 +125,13 @@ class TestNPUPlatform(TestBase):
self.assertIsNone(self.platform.inference_mode())
mock_inference_mode.assert_called_once()
@patch("vllm_ascend.quantization.utils.maybe_auto_detect_quantization")
@patch("vllm_ascend.ascend_config.init_ascend_config")
@patch("vllm_ascend.utils.update_aclgraph_sizes")
@patch("vllm_ascend.utils.get_ascend_device_type", return_value=AscendDeviceType.A3)
@patch("os.environ", {})
@patch("vllm_ascend.core.recompute_scheduler.RecomputeSchedulerConfig.initialize_from_config")
def test_check_and_update_config_basic_config_update(
self, mock_init_recompute, mock_soc_version, mock_update_acl, mock_init_ascend, mock_auto_detect
self, mock_init_recompute, mock_soc_version, mock_update_acl, mock_init_ascend
):
mock_init_ascend.return_value = TestNPUPlatform.mock_vllm_ascend_config()
vllm_config = TestNPUPlatform.mock_vllm_config()
@@ -156,12 +155,11 @@ class TestNPUPlatform(TestBase):
mock_init_ascend.assert_called_once_with(vllm_config)
@patch("vllm_ascend.quantization.utils.maybe_auto_detect_quantization")
@patch("vllm_ascend.utils.get_ascend_device_type", return_value=AscendDeviceType.A3)
@patch("vllm_ascend.ascend_config.init_ascend_config")
@patch("vllm_ascend.core.recompute_scheduler.RecomputeSchedulerConfig.initialize_from_config")
def test_check_and_update_config_no_model_config_warning(
self, mock_init_recompute, mock_init_ascend, mock_soc_version, mock_auto_detect
self, mock_init_recompute, mock_init_ascend, mock_soc_version
):
mock_init_ascend.return_value = TestNPUPlatform.mock_vllm_ascend_config()
vllm_config = TestNPUPlatform.mock_vllm_config()
@@ -183,11 +181,10 @@ class TestNPUPlatform(TestBase):
self.assertTrue("Model config is missing" in cm.output[0])
@patch("vllm_ascend.quantization.utils.maybe_auto_detect_quantization")
@patch("vllm_ascend.utils.get_ascend_device_type", return_value=AscendDeviceType.A3)
@patch("vllm_ascend.ascend_config.init_ascend_config")
@patch("vllm_ascend.core.recompute_scheduler.RecomputeSchedulerConfig.initialize_from_config")
def test_check_and_update_config_enforce_eager_mode(self, mock_init_recompute, mock_init_ascend, mock_soc_version, mock_auto_detect):
def test_check_and_update_config_enforce_eager_mode(self, mock_init_recompute, mock_init_ascend, mock_soc_version):
mock_init_ascend.return_value = TestNPUPlatform.mock_vllm_ascend_config()
vllm_config = TestNPUPlatform.mock_vllm_config()
vllm_config.model_config.enforce_eager = True
@@ -218,12 +215,11 @@ class TestNPUPlatform(TestBase):
CUDAGraphMode.NONE,
)
@patch("vllm_ascend.quantization.utils.maybe_auto_detect_quantization")
@patch("vllm_ascend.utils.get_ascend_device_type", return_value=AscendDeviceType.A3)
@patch("vllm_ascend.ascend_config.init_ascend_config")
@patch("vllm_ascend.core.recompute_scheduler.RecomputeSchedulerConfig.initialize_from_config")
def test_check_and_update_config_unsupported_compilation_level(
self, mock_init_recompute, mock_init_ascend, mock_soc_version, mock_auto_detect
self, mock_init_recompute, mock_init_ascend, mock_soc_version
):
mock_init_ascend.return_value = TestNPUPlatform.mock_vllm_ascend_config()
vllm_config = TestNPUPlatform.mock_vllm_config()
@@ -257,10 +253,9 @@ class TestNPUPlatform(TestBase):
)
@pytest.mark.skip("Revert me when vllm support setting cudagraph_mode on oot platform")
@patch("vllm_ascend.quantization.utils.maybe_auto_detect_quantization")
@patch("vllm_ascend.utils.get_ascend_device_type", return_value=AscendDeviceType.A3)
@patch("vllm_ascend.ascend_config.init_ascend_config")
def test_check_and_update_config_unsupported_cudagraph_mode(self, mock_init_ascend, mock_soc_version, mock_auto_detect):
def test_check_and_update_config_unsupported_cudagraph_mode(self, mock_init_ascend, mock_soc_version):
mock_init_ascend.return_value = TestNPUPlatform.mock_vllm_ascend_config()
vllm_config = TestNPUPlatform.mock_vllm_config()
vllm_config.model_config.enforce_eager = False
@@ -282,12 +277,11 @@ class TestNPUPlatform(TestBase):
CUDAGraphMode.NONE,
)
@patch("vllm_ascend.quantization.utils.maybe_auto_detect_quantization")
@patch("vllm_ascend.utils.get_ascend_device_type", return_value=AscendDeviceType.A3)
@patch("vllm_ascend.ascend_config.init_ascend_config")
@patch("vllm_ascend.core.recompute_scheduler.RecomputeSchedulerConfig.initialize_from_config")
def test_check_and_update_config_cache_config_block_size(
self, mock_init_recompute, mock_init_ascend, mock_soc_version, mock_auto_detect
self, mock_init_recompute, mock_init_ascend, mock_soc_version
):
mock_init_ascend.return_value = TestNPUPlatform.mock_vllm_ascend_config()
vllm_config = TestNPUPlatform.mock_vllm_config()
@@ -307,12 +301,11 @@ class TestNPUPlatform(TestBase):
self.assertEqual(vllm_config.cache_config.block_size, 128)
@patch("vllm_ascend.quantization.utils.maybe_auto_detect_quantization")
@patch("vllm_ascend.utils.get_ascend_device_type", return_value=AscendDeviceType.A3)
@patch("vllm_ascend.ascend_config.init_ascend_config")
@patch("vllm_ascend.core.recompute_scheduler.RecomputeSchedulerConfig.initialize_from_config")
def test_check_and_update_config_v1_worker_class_selection(
self, mock_init_recompute, mock_init_ascend, mock_soc_version, mock_auto_detect
self, mock_init_recompute, mock_init_ascend, mock_soc_version
):
mock_init_ascend.return_value = TestNPUPlatform.mock_vllm_ascend_config()
vllm_config = TestNPUPlatform.mock_vllm_config()
@@ -343,11 +336,10 @@ class TestNPUPlatform(TestBase):
"vllm_ascend.xlite.xlite_worker.XliteWorker",
)
@patch("vllm_ascend.quantization.utils.maybe_auto_detect_quantization")
@patch("vllm_ascend.ascend_config.init_ascend_config")
@patch("vllm_ascend.utils.get_ascend_device_type", return_value=AscendDeviceType._310P)
@patch("vllm_ascend.core.recompute_scheduler.RecomputeSchedulerConfig.initialize_from_config")
def test_check_and_update_config_310p_no_custom_ops(self, mock_init_recompute, mock_soc_version, mock_init_ascend, mock_auto_detect):
def test_check_and_update_config_310p_no_custom_ops(self, mock_init_recompute, mock_soc_version, mock_init_ascend):
mock_init_ascend.return_value = TestNPUPlatform.mock_vllm_ascend_config()
vllm_config = TestNPUPlatform.mock_vllm_config()
vllm_config.compilation_config.custom_ops = []