[Kernel] add custom op GmmSwigluQuantWeightNzTensorList (#3804)
### What this PR does / why we need it? This PR introduces support for adding custom CANN `aclnn` ops to `vllm-ascend`, allowing users to define and use their own custom operators. Key changes include: - Building and installing custom ops into the `vllm-ascend`-specified directory - Binding the `aclnn` op interface to the `torch.ops._C_ascend` module - Enabling invocation of these ops within `vllm-ascend` This PR includes a sample custom op: `aclnnGroupedMatmulSwigluQuantWeightNzTensorList`, which is adapted from the CANN operator [`aclnnGroupedMatmulSwigluQuantWeightNZ`](https://www.hiascend.com/document/detail/zh/canncommercial/83RC1/API/aolapi/context/aclnnGroupedMatmulSwigluQuantWeightNZ.md). Its input parameters `weight` and `weight_scale` now accept `list[torch.Tensor]` (i.e., `at::TensorList`). ### Does this PR introduce _any_ user-facing change? No. - vLLM version: v0.11.2 --------- Signed-off-by: QianChenxi <chenxi.qian.cq@outlook.com>
This commit is contained in:
47
csrc/utils/inc/tiling/data_copy_transpose_tiling.h
Normal file
47
csrc/utils/inc/tiling/data_copy_transpose_tiling.h
Normal file
@@ -0,0 +1,47 @@
|
||||
/**
|
||||
* Copyright (c) 2023-2024 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.
|
||||
*/
|
||||
|
||||
/*!
|
||||
* \file data_copy_transpose_tiling.h
|
||||
* \brief
|
||||
*/
|
||||
|
||||
#pragma once
|
||||
|
||||
#include <vector>
|
||||
#include <graph/tensor.h>
|
||||
#include "data_copy_transpose_tiling_def.h"
|
||||
|
||||
namespace optiling {
|
||||
|
||||
inline void GetDataCopyTransposeTiling(const ge::Shape &dstShape, const ge::Shape &srcShape, const uint32_t typeSize,
|
||||
optiling::CopyTransposeTiling &tiling)
|
||||
{
|
||||
std::vector<int64_t> dstShapeInfo = dstShape.GetDims();
|
||||
std::vector<int64_t> srcShapeInfo = srcShape.GetDims();
|
||||
|
||||
tiling.set_dstShapeB(dstShapeInfo[0]);
|
||||
tiling.set_dstShapeN(dstShapeInfo[1]);
|
||||
tiling.set_dstShapeS(dstShapeInfo[2]);
|
||||
tiling.set_dstShapeH(dstShapeInfo[3]);
|
||||
tiling.set_dstShapeHN(tiling.get_dstShapeH() / tiling.get_dstShapeN());
|
||||
|
||||
tiling.set_srcShapeB(srcShapeInfo[0]);
|
||||
tiling.set_srcShapeN(srcShapeInfo[1]);
|
||||
tiling.set_srcShapeS(srcShapeInfo[2]);
|
||||
tiling.set_srcShapeHN(srcShapeInfo[3]);
|
||||
tiling.set_originalShapeNLen(tiling.get_srcShapeHN() * typeSize);
|
||||
tiling.set_shapeSHValue(tiling.get_dstShapeS() * tiling.get_dstShapeH());
|
||||
tiling.set_shapeNsValue(tiling.get_dstShapeN() * tiling.get_dstShapeS());
|
||||
tiling.set_shapeNsnValue(tiling.get_dstShapeN() * tiling.get_srcShapeS() * tiling.get_srcShapeN());
|
||||
tiling.set_shapeBHValue(tiling.get_dstShapeB() * tiling.get_dstShapeH());
|
||||
}
|
||||
|
||||
} // namespace optiling
|
||||
Reference in New Issue
Block a user