66b60c9440
### What this PR does / why we need it?
1. [Refact] Refact MLA/SFA weight prefetch to consist with moe weight
prefetch
2. Remove duplicated o_proj weight prefetch in forward for MLA/SFA
### Does this PR introduce _any_ user-facing change?
NA
### How was this patch tested?
1) Performance result:
Perf test data:
*) MLA:
| | 1st test | 2nd test | Output Token Throughput(Avg) | Performance
improvement percentage |
| --- | --- | --- | --- | --- |
| o_proj duplicate prefetch | 11.9669 token/s | 12.0287 token/s |
11.9978 |
| o_proj no duplicate prefetch | 12.5594 token/s | 12.6216 token/s |
12.5905 | 4.94%| |
single layer performace improve: 5%~8%
*) SFA:
| | 1st test | 2nd test | Output Token Throughput(Avg) | Performance
improvement percentage |
| --- | --- | --- | --- | --- |
| o_proj duplicate prefetch | 13.0523 token/s | 13.1084 token/s |
13.08035 | |
| o_proj no duplicate prefetch | 13.9844 token/s | 14.1678 token/s |
14.0761 | 7.6% |
- vLLM version: v0.15.0
- vLLM main:
d7e17aaacd
---------
Signed-off-by: leo-pony <nengjunma@outlook.com>
41 lines
1.4 KiB
Python
41 lines
1.4 KiB
Python
#
|
|
# Copyright (c) 2025 Huawei Technologies Co., Ltd. All Rights Reserved.
|
|
#
|
|
# 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.
|
|
#
|
|
|
|
import torch
|
|
from vllm.model_executor.layers.activation import QuickGELU, SiluAndMul
|
|
|
|
from vllm_ascend.utils import get_weight_prefetch_method
|
|
|
|
|
|
class AscendQuickGELU(QuickGELU):
|
|
def forward_oot(self, x: torch.tensor) -> torch.Tensor:
|
|
import torch_npu
|
|
|
|
out = torch_npu.npu_fast_gelu(x)
|
|
return out
|
|
|
|
|
|
class AscendSiluAndMul(SiluAndMul):
|
|
def forward_oot(self, x: torch.Tensor) -> torch.Tensor:
|
|
import torch_npu
|
|
|
|
weight_prefetch_method = get_weight_prefetch_method()
|
|
weight_prefetch_method.maybe_prefetch_mlp_weight_preprocess(weight_prefetch_method.MLP_DOWN, x)
|
|
out = torch_npu.npu_swiglu(x)
|
|
weight_prefetch_method.maybe_prefetch_mlp_weight_postprocess(out)
|
|
return out
|