NVIDIA NvNeural SDK  2022.2
GPU inference framework for NVIDIA Nsight Deep Learning Designer
CudnnHelpers.h
Go to the documentation of this file.
1 /*
2 * SPDX-FileCopyrightText: Copyright (c) 2020-2021 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
3 * SPDX-License-Identifier: MIT
4 *
5 * Permission is hereby granted, free of charge, to any person obtaining a
6 * copy of this software and associated documentation files (the "Software"),
7 * to deal in the Software without restriction, including without limitation
8 * the rights to use, copy, modify, merge, publish, distribute, sublicense,
9 * and/or sell copies of the Software, and to permit persons to whom the
10 * Software is furnished to do so, subject to the following conditions:
11 *
12 * The above copyright notice and this permission notice shall be included in
13 * all copies or substantial portions of the Software.
14 *
15 * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
16 * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
17 * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
18 * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
19 * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
20 * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
21 * DEALINGS IN THE SOFTWARE.
22 */
23 
25 
26 #ifndef NVNEURAL_CUDNNHELPERS_H
27 #define NVNEURAL_CUDNNHELPERS_H
28 
29 #include <cudnn.h>
30 
31 #include <nvneural/CoreTypes.h>
32 #include <nvneural/CoreHelpers.h>
33 #include <nvneural/CudaTypes.h>
34 
35 namespace nvneural {
36 
41 
47 cudnnHandle_t GetCudnnHandle(const INetworkBackendCuda* pBackend) noexcept;
48 
49 } // namespace nvneural
50 
51 #endif // NVNEURAL_CUDNNHELPERS_H
Common helper classes and template function implementations.
Fundamental NvNeural data types are declared here.
NeuralResult
NeuralResult is a generic success/failure result type similar to COM HRESULT.
Definition: CoreTypes.h:275
All CUDA-specific NvNeural types are declared here.
NeuralResult InitializeCudnn(INetworkBackendCuda *pBackend) noexcept
Initializes cuDNN and registers it with the backend.
cudnnHandle_t GetCudnnHandle(const INetworkBackendCuda *pBackend) noexcept
Retrieves the cuDNN handle associated with a CUDA backend.
INetworkBackend companion interface with CUDA-specific functionality.
Definition: CudaTypes.h:346