dd622aa6a6
### What this PR does / why we need it?
We introduced the npugraph_ex backend through the vllm's adaptor
dispatch mechanism to accelerate aclgraph. This solution is based on
torch.compile and uses torchair to optimize the fx.graph. The
performance gains are mainly obtained from the static kernel. We
conducted tests on Qwen3-30B and achieved over 5% performance
optimization.
### Does this PR introduce _any_ user-facing change?
Yes, we add a new switch named"enable_npugraph_ex" in additional_config,
default is False.
We also add an example to show how to register custom replacement pass
### More information about this PR
This feature depends on the release of CANN and torch_npu in Q4.
We tested it on a package that has not been publicly released yet and
verified that the functionality works.
This feature is still experimental at the moment; setting the config
true will directly raise error.
Merging into the main branch initially involves some preliminary commits
to facilitate subsequent development and testing of the feature, as well
as to avoid submitting an excessively large PR at once.
- vLLM version: v0.12.0
- vLLM main:
ad32e3e19c
---------
Signed-off-by: chencangtao <chencangtao@huawei.com>
Signed-off-by: ChenCangtao <50493711+ChenCangtao@users.noreply.github.com>
Co-authored-by: chencangtao <chencangtao@huawei.com>
Co-authored-by: panchao-hub <315134829@qq.com>
Co-authored-by: wbigat <wbigat@163.com>
Co-authored-by: Mengqing Cao <cmq0113@163.com>
94 lines
3.5 KiB
Python
94 lines
3.5 KiB
Python
#
|
|
# Licensed under the Apache License, Version 2.0 (the "License");
|
|
# you may not use this file except in compliance with the License.
|
|
# You may obtain a copy of the License at
|
|
#
|
|
# http://www.apache.org/licenses/LICENSE-2.0
|
|
#
|
|
# Unless required by applicable law or agreed to in writing, software
|
|
# distributed under the License is distributed on an "AS IS" BASIS,
|
|
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
|
# See the License for the specific language governing permissions and
|
|
# limitations under the License.
|
|
# This file is a part of the vllm-ascend project.
|
|
#
|
|
|
|
from vllm.config import VllmConfig
|
|
|
|
from tests.ut.base import TestBase
|
|
from vllm_ascend.ascend_config import (clear_ascend_config, get_ascend_config,
|
|
init_ascend_config)
|
|
|
|
|
|
class TestAscendConfig(TestBase):
|
|
|
|
@staticmethod
|
|
def _clean_up_ascend_config(func):
|
|
|
|
def wrapper(*args, **kwargs):
|
|
clear_ascend_config()
|
|
func(*args, **kwargs)
|
|
clear_ascend_config()
|
|
|
|
return wrapper
|
|
|
|
@_clean_up_ascend_config
|
|
def test_init_ascend_config_without_additional_config(self):
|
|
test_vllm_config = VllmConfig()
|
|
# No additional config given, check the default value here.
|
|
ascend_config = init_ascend_config(test_vllm_config)
|
|
self.assertIsNone(ascend_config.expert_map_path)
|
|
self.assertFalse(ascend_config.multistream_overlap_shared_expert)
|
|
|
|
ascend_compilation_config = ascend_config.ascend_compilation_config
|
|
self.assertTrue(ascend_compilation_config.enable_quantization_fusion)
|
|
|
|
@_clean_up_ascend_config
|
|
def test_init_ascend_config_with_additional_config(self):
|
|
test_vllm_config = VllmConfig()
|
|
test_vllm_config.additional_config = {
|
|
"ascend_compilation_config": {
|
|
"enable_quantization_fusion": False,
|
|
},
|
|
"multistream_overlap_shared_expert": True,
|
|
"expert_map_path": "test_expert_map_path",
|
|
"refresh": True,
|
|
}
|
|
ascend_config = init_ascend_config(test_vllm_config)
|
|
self.assertEqual(ascend_config.expert_map_path, "test_expert_map_path")
|
|
self.assertTrue(ascend_config.multistream_overlap_shared_expert)
|
|
self.assertFalse(ascend_config.enable_npugraph_ex)
|
|
|
|
ascend_compilation_config = ascend_config.ascend_compilation_config
|
|
self.assertFalse(ascend_compilation_config.enable_quantization_fusion)
|
|
|
|
@_clean_up_ascend_config
|
|
def test_init_ascend_config_enable_npugraph_ex(self):
|
|
with self.assertRaises(NotImplementedError):
|
|
test_vllm_config = VllmConfig()
|
|
test_vllm_config.additional_config = {
|
|
"enable_npugraph_ex": True,
|
|
"refresh": True,
|
|
}
|
|
init_ascend_config(test_vllm_config)
|
|
|
|
@_clean_up_ascend_config
|
|
def test_get_ascend_config(self):
|
|
test_vllm_config = VllmConfig()
|
|
ascend_config = init_ascend_config(test_vllm_config)
|
|
self.assertEqual(get_ascend_config(), ascend_config)
|
|
|
|
@_clean_up_ascend_config
|
|
def test_get_ascend_config_without_init(self):
|
|
with self.assertRaises(RuntimeError):
|
|
get_ascend_config()
|
|
|
|
@_clean_up_ascend_config
|
|
def test_clear_ascend_config(self):
|
|
test_vllm_config = VllmConfig()
|
|
ascend_config = init_ascend_config(test_vllm_config)
|
|
self.assertEqual(get_ascend_config(), ascend_config)
|
|
clear_ascend_config()
|
|
with self.assertRaises(RuntimeError):
|
|
get_ascend_config()
|