mirror of
https://github.com/NVIDIA/TensorRT-LLM.git
synced 2026-01-14 06:27:45 +08:00
90 lines
2.7 KiB
C++
90 lines
2.7 KiB
C++
/*
|
|
* SPDX-FileCopyrightText: Copyright (c) 2025 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
|
|
* SPDX-License-Identifier: Apache-2.0
|
|
*
|
|
* Licensed under the Apache License, Version 2.0 (the "License");
|
|
* you may not use this file except in compliance with the License.
|
|
* You may obtain a copy of the License at
|
|
*
|
|
* http://www.apache.org/licenses/LICENSE-2.0
|
|
*
|
|
* Unless required by applicable law or agreed to in writing, software
|
|
* distributed under the License is distributed on an "AS IS" BASIS,
|
|
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
|
* See the License for the specific language governing permissions and
|
|
* limitations under the License.
|
|
*/
|
|
|
|
#include "tensorrt_llm/common/logger.h"
|
|
#include "tensorrt_llm/executor/executor.h"
|
|
|
|
namespace
|
|
{
|
|
void validateCudaGraphCacheSize(bool cudaGraphMode, tensorrt_llm::executor::SizeType32 cudaGraphCacheSize)
|
|
{
|
|
TLLM_CHECK_WITH_INFO(cudaGraphCacheSize >= 0, "CUDA graph cache size must be greater or equal to 0.");
|
|
if (!cudaGraphMode && cudaGraphCacheSize > 0)
|
|
{
|
|
TLLM_LOG_WARNING(
|
|
"Setting cudaGraphCacheSize to a value greater than 0 without enabling cudaGraphMode has no effect.");
|
|
}
|
|
}
|
|
|
|
} // namespace
|
|
|
|
namespace tensorrt_llm::executor
|
|
{
|
|
|
|
ExtendedRuntimePerfKnobConfig::ExtendedRuntimePerfKnobConfig(
|
|
bool multiBlockMode, bool enableContextFMHAFP32Acc, bool cudaGraphMode, SizeType32 cudaGraphCacheSize)
|
|
: mMultiBlockMode(multiBlockMode)
|
|
, mEnableContextFMHAFP32Acc(enableContextFMHAFP32Acc)
|
|
, mCudaGraphMode(cudaGraphMode)
|
|
, mCudaGraphCacheSize(cudaGraphCacheSize)
|
|
{
|
|
validateCudaGraphCacheSize(mCudaGraphMode, mCudaGraphCacheSize);
|
|
}
|
|
|
|
bool ExtendedRuntimePerfKnobConfig::getMultiBlockMode() const
|
|
{
|
|
return mMultiBlockMode;
|
|
}
|
|
|
|
bool ExtendedRuntimePerfKnobConfig::getEnableContextFMHAFP32Acc() const
|
|
{
|
|
return mEnableContextFMHAFP32Acc;
|
|
}
|
|
|
|
bool ExtendedRuntimePerfKnobConfig::getCudaGraphMode() const
|
|
{
|
|
return mCudaGraphMode;
|
|
}
|
|
|
|
SizeType32 ExtendedRuntimePerfKnobConfig::getCudaGraphCacheSize() const
|
|
{
|
|
return mCudaGraphCacheSize;
|
|
}
|
|
|
|
void ExtendedRuntimePerfKnobConfig::setMultiBlockMode(bool multiBlockMode)
|
|
{
|
|
mMultiBlockMode = multiBlockMode;
|
|
}
|
|
|
|
void ExtendedRuntimePerfKnobConfig::setEnableContextFMHAFP32Acc(bool enableContextFMHAFP32Acc)
|
|
{
|
|
mEnableContextFMHAFP32Acc = enableContextFMHAFP32Acc;
|
|
}
|
|
|
|
void ExtendedRuntimePerfKnobConfig::setCudaGraphMode(bool cudaGraphMode)
|
|
{
|
|
mCudaGraphMode = cudaGraphMode;
|
|
}
|
|
|
|
void ExtendedRuntimePerfKnobConfig::setCudaGraphCacheSize(SizeType32 cudaGraphCacheSize)
|
|
{
|
|
mCudaGraphCacheSize = cudaGraphCacheSize;
|
|
validateCudaGraphCacheSize(mCudaGraphMode, mCudaGraphCacheSize);
|
|
}
|
|
|
|
} // namespace tensorrt_llm::executor
|