mirror of
https://github.com/NVIDIA/TensorRT-LLM.git
synced 2026-01-27 22:23:25 +08:00
22 lines
1.0 KiB
Plaintext
22 lines
1.0 KiB
Plaintext
/*
|
|
* SPDX-FileCopyrightText: Copyright (c) 2023-2024 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
|
|
* SPDX-License-Identifier: NVIDIA TensorRT Source Code License Agreement
|
|
*
|
|
* NVIDIA CORPORATION, its affiliates and licensors retain all intellectual
|
|
* property and proprietary rights in and to this material, related
|
|
* documentation and any modifications thereto. Any use, reproduction,
|
|
* disclosure or distribution of this material and related documentation
|
|
* without an express license agreement from NVIDIA CORPORATION or
|
|
* its affiliates is strictly prohibited.
|
|
*/
|
|
|
|
#include "softmax_impl.h"
|
|
|
|
void run_softmax_fp32(void* dst, void const* src, void const* mask, void const* attention_sinks, void* softmax_sum_d,
|
|
void* cu_q_seqlens_d, int s_inner, int s_outer, int b, int h, float softcapping_scale_bmm1, int warps_n,
|
|
bool has_alibi)
|
|
{
|
|
run_softmax<fmha::fp16_t, float>(dst, src, mask, attention_sinks, softmax_sum_d, cu_q_seqlens_d, s_inner, s_outer,
|
|
b, h, 0.f, 0.f, softcapping_scale_bmm1, warps_n, has_alibi);
|
|
}
|