* Copyright (c) 2026 Huawei Technologies Co., Ltd.
* This program is free software, you can redistribute it and/or modify it under the terms and conditions of
* CANN Open Software License Agreement Version 2.0 (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 swiglu_group_quant_def.cpp
* \brief
*/
#include "register/op_def_registry.h"
namespace ops {
namespace {
constexpr float DEFAULT_CLAMP_LIMIT = -1.0f;
constexpr float DEFAULT_DST_TYPE_MAX = 15.0f;
}
class SwigluGroupQuant : public OpDef {
public:
explicit SwigluGroupQuant(const char* name) : OpDef(name)
{
this->Input("x")
.ParamType(REQUIRED)
.DataType({ge::DT_FLOAT16, ge::DT_BF16, ge::DT_FLOAT16, ge::DT_BF16, ge::DT_FLOAT16, ge::DT_BF16,
ge::DT_FLOAT16, ge::DT_BF16, ge::DT_FLOAT16, ge::DT_BF16, ge::DT_FLOAT16, ge::DT_BF16,
ge::DT_FLOAT, ge::DT_FLOAT16, ge::DT_BF16})
.Format({ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND,
ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND,
ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND})
.UnknownShapeFormat({ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND,
ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND,
ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND});
this->Input("weight")
.ParamType(OPTIONAL)
.DataType({ge::DT_FLOAT, ge::DT_FLOAT, ge::DT_FLOAT, ge::DT_FLOAT, ge::DT_FLOAT, ge::DT_FLOAT, ge::DT_FLOAT,
ge::DT_FLOAT, ge::DT_FLOAT, ge::DT_FLOAT, ge::DT_FLOAT, ge::DT_FLOAT, ge::DT_FLOAT, ge::DT_FLOAT,
ge::DT_FLOAT})
.Format({ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND,
ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND,
ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND})
.UnknownShapeFormat({ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND,
ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND,
ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND});
this->Input("group_index")
.ParamType(OPTIONAL)
.DataType({ge::DT_INT64, ge::DT_INT64, ge::DT_INT64, ge::DT_INT64, ge::DT_INT64, ge::DT_INT64, ge::DT_INT64,
ge::DT_INT64, ge::DT_INT64, ge::DT_INT64, ge::DT_INT64, ge::DT_INT64, ge::DT_INT64, ge::DT_INT64,
ge::DT_INT64})
.Format({ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND,
ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND,
ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND})
.UnknownShapeFormat({ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND,
ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND,
ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND});
this->Input("scale")
.ParamType(OPTIONAL)
.DataType({ge::DT_FLOAT, ge::DT_FLOAT, ge::DT_FLOAT, ge::DT_FLOAT, ge::DT_FLOAT, ge::DT_FLOAT, ge::DT_FLOAT,
ge::DT_FLOAT, ge::DT_FLOAT, ge::DT_FLOAT, ge::DT_FLOAT, ge::DT_FLOAT, ge::DT_FLOAT, ge::DT_FLOAT,
ge::DT_FLOAT})
.Format({ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND,
ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND,
ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND})
.UnknownShapeFormat({ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND,
ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND,
ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND});
this->Output("y")
.ParamType(REQUIRED)
.DataType({ge::DT_FLOAT8_E4M3FN, ge::DT_FLOAT8_E4M3FN, ge::DT_FLOAT8_E5M2, ge::DT_FLOAT8_E5M2,
ge::DT_FLOAT8_E4M3FN, ge::DT_FLOAT8_E4M3FN, ge::DT_FLOAT8_E5M2, ge::DT_FLOAT8_E5M2,
ge::DT_FLOAT4_E2M1, ge::DT_FLOAT4_E2M1, ge::DT_FLOAT4_E1M2, ge::DT_FLOAT4_E1M2, ge::DT_HIFLOAT8,
ge::DT_HIFLOAT8, ge::DT_HIFLOAT8})
.Format({ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND,
ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND,
ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND})
.UnknownShapeFormat({ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND,
ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND,
ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND});
this->Output("y_scale")
.ParamType(REQUIRED)
.DataType({ge::DT_FLOAT, ge::DT_FLOAT, ge::DT_FLOAT, ge::DT_FLOAT, ge::DT_FLOAT8_E8M0, ge::DT_FLOAT8_E8M0,
ge::DT_FLOAT8_E8M0, ge::DT_FLOAT8_E8M0, ge::DT_FLOAT8_E8M0, ge::DT_FLOAT8_E8M0,
ge::DT_FLOAT8_E8M0, ge::DT_FLOAT8_E8M0, ge::DT_FLOAT, ge::DT_FLOAT, ge::DT_FLOAT})
.Format({ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND,
ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND,
ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND})
.UnknownShapeFormat({ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND,
ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND,
ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND});
this->Output("y_origin")
.ParamType(REQUIRED)
.DataType({ge::DT_FLOAT16, ge::DT_BF16, ge::DT_FLOAT16, ge::DT_BF16, ge::DT_FLOAT16, ge::DT_BF16,
ge::DT_FLOAT16, ge::DT_BF16, ge::DT_FLOAT16, ge::DT_BF16, ge::DT_FLOAT16, ge::DT_BF16,
ge::DT_FLOAT, ge::DT_FLOAT16, ge::DT_BF16})
.Format({ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND,
ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND,
ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND})
.UnknownShapeFormat({ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND,
ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND,
ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND, ge::FORMAT_ND});
this->Attr("dst_type").AttrType(OPTIONAL).Int(ge::DT_FLOAT8_E4M3FN);
this->Attr("quant_mode").AttrType(OPTIONAL).Int(0);
this->Attr("block_size").AttrType(OPTIONAL).Int(0);
this->Attr("round_scale").AttrType(OPTIONAL).Bool(false);
this->Attr("clamp_limit").AttrType(OPTIONAL).Float(DEFAULT_CLAMP_LIMIT);
this->Attr("dst_type_max").AttrType(OPTIONAL).Float(DEFAULT_DST_TYPE_MAX);
this->Attr("output_origin").AttrType(OPTIONAL).Bool(false);
OpAICoreConfig aicoreConfig;
aicoreConfig.DynamicCompileStaticFlag(true)
.DynamicFormatFlag(false)
.DynamicRankSupportFlag(true)
.DynamicShapeSupportFlag(true)
.NeedCheckSupportFlag(false)
.PrecisionReduceFlag(true)
.ExtendCfgInfo("opFile.value", "swiglu_group_quant");
this->AICore().AddConfig("ascend950", aicoreConfig);
}
};
OP_ADD(SwigluGroupQuant);
}