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

[MOE] commit GMM custom operator (#7010)

Browse Source 
### What this PR does / why we need it?
GMM custom operator optimization in small batch scenarios

### How was this patch tested?
Submit the GMM custom operator for subsequent integration into the MOE
process.


- vLLM version: v0.16.0
- vLLM main:
15d76f74e2

---------

Signed-off-by: chenxi-hh <chen464822955@163.com>
Signed-off-by: chenxi-hh <32731611+chenxi-hh@users.noreply.github.com>
This commit is contained in:
chenxi-hh
2026-03-09 09:56:31 +08:00
committed by GitHub
parent 01d3515dcf
commit 737dfcf638
16 changed files with 1214 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files

71
csrc/moe_grouped_matmul/op_host/CMakeLists.txt Normal file
View File

@@ -0,0 +1,71 @@
# Copyright (c) 2026 Huawei Technologies Co., Ltd.
# This file is a part of the CANN Open Software.
# Licensed under CANN Open Software License Agreement Version 1.0 (the "License").
# Please refer to the License for details. You may not use this file except in compliance with the License.
# THIS SOFTWARE IS PROVIDED ON AN "AS IS" BASIS, WITHOUT WARRANTIES OF ANY KIND, EITHER EXPRESS OR IMPLIED,
# INCLUDING BUT NOT LIMITED TO NON-INFRINGEMENT, MERCHANTABILITY, OR FITNESS FOR A PARTICULAR PURPOSE.
# See LICENSE in the root of the software repository for the full text of the License.
# ======================================================================================================================
add_ops_compile_options(
OP_NAME MoeGroupedMatmulCustom
OPTIONS --cce-auto-sync=off
-Wno-deprecated-declarations
-Werror
)
target_sources(optiling PRIVATE
moe_grouped_matmul_cpu.cpp
)
target_include_directories(optiling PRIVATE
${CMAKE_CURRENT_SOURCE_DIR}
${ASCEND_CANN_PACKAGE_PATH}/include
${ASCEND_CANN_PACKAGE_PATH}/include/external
${ASCEND_CANN_PACKAGE_PATH}/include/experiment
${ASCEND_CANN_PACKAGE_PATH}/include/experiment/platform
${ASCEND_CANN_PACKAGE_PATH}/include/experiment/metadef
${ASCEND_CANN_PACKAGE_PATH}/include/experiment/runtime
${ASCEND_CANN_PACKAGE_PATH}/include/experiment/msprof
)
target_sources(opsproto PRIVATE
moe_grouped_matmul_infershape.cpp
)
target_include_directories(opsproto PRIVATE
${CMAKE_CURRENT_SOURCE_DIR}
${ASCEND_CANN_PACKAGE_PATH}/include
${ASCEND_CANN_PACKAGE_PATH}/include/external
${ASCEND_CANN_PACKAGE_PATH}/include/experiment
${ASCEND_CANN_PACKAGE_PATH}/include/experiment/platform
${ASCEND_CANN_PACKAGE_PATH}/include/experiment/metadef
${ASCEND_CANN_PACKAGE_PATH}/include/experiment/runtime
${ASCEND_CANN_PACKAGE_PATH}/include/experiment/msprof
)
target_sources(op_host_aclnnInner PRIVATE
moe_grouped_matmul_def.cpp
)
target_include_directories(op_host_aclnnInner PRIVATE
${CMAKE_CURRENT_SOURCE_DIR}
${ASCEND_CANN_PACKAGE_PATH}/include
${ASCEND_CANN_PACKAGE_PATH}/include/external
${ASCEND_CANN_PACKAGE_PATH}/include/experiment
${ASCEND_CANN_PACKAGE_PATH}/include/experiment/platform
${ASCEND_CANN_PACKAGE_PATH}/include/experiment/metadef
${ASCEND_CANN_PACKAGE_PATH}/include/experiment/runtime
${ASCEND_CANN_PACKAGE_PATH}/include/experiment/msprof
)
target_sources(opapi PRIVATE
moe_grouped_matmul_l0.cpp
aclnn_moe_grouped_matmul.cpp
)
install(FILES "${CMAKE_CURRENT_SOURCE_DIR}/aclnn_moe_grouped_matmul.h"
DESTINATION ${ACLNN_INC_INSTALL_DIR} OPTIONAL)
install(FILES "${CMAKE_CURRENT_SOURCE_DIR}/aclnn_moe_grouped_matmul_weight_nz.h"
DESTINATION ${ACLNN_INC_INSTALL_DIR} OPTIONAL)