/* * SPDX-FileCopyrightText: Copyright (c) 2021-2022 NVIDIA CORPORATION & AFFILIATES. All rights reserved. * SPDX-License-Identifier: LicenseRef-NvidiaProprietary * * 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 "filtered_lrelu.cu" // Template/kernel specializations for sign write mode. // Full op, 32-bit indexing. template filtered_lrelu_kernel_spec choose_filtered_lrelu_kernel(const filtered_lrelu_kernel_params& p, int sharedKB); template filtered_lrelu_kernel_spec choose_filtered_lrelu_kernel(const filtered_lrelu_kernel_params& p, int sharedKB); // Full op, 64-bit indexing. template filtered_lrelu_kernel_spec choose_filtered_lrelu_kernel(const filtered_lrelu_kernel_params& p, int sharedKB); template filtered_lrelu_kernel_spec choose_filtered_lrelu_kernel(const filtered_lrelu_kernel_params& p, int sharedKB); // Activation/signs only for generic variant. 64-bit indexing. template void* choose_filtered_lrelu_act_kernel(void); template void* choose_filtered_lrelu_act_kernel(void); template void* choose_filtered_lrelu_act_kernel(void); // Copy filters to constant memory. template cudaError_t copy_filters(cudaStream_t stream);