mirror of
https://github.com/NVIDIA/TensorRT-LLM.git
synced 2026-01-14 06:27:45 +08:00
d7386d14a8
* refactor: Simplifiy disableLookahead method Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com> * Update DecoderBuffers comments Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com> * refactor: Move numDecodingEngineTokens to DecoderState This commit introduces new methods in the DecoderState class to manage the number of tokens for each request in a batch. The following changes were made: - Added `getNumDecodingEngineTokens()` to retrieve the number of tokens for all requests. - Added `getNumDecodingEngineTokens(SizeType32 batchIdx)` to get the token count for a specific request. - Added `setNumDecodingEngineTokens(SizeType32 batchIdx, SizeType32 numTokens)` to set the token count for a specific request. - Updated the setup method to initialize the token count vector based on the maximum batch size. - Refactored the `CreateNewDecoderRequests` class to utilize the new token management methods, improving clarity and maintainability. Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com> * refactor: Improve shape variables in DecoderState Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com> --------- Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
116 lines
3.9 KiB
C++
116 lines
3.9 KiB
C++
/*
|
|
* Copyright (c) 2022-2024, NVIDIA CORPORATION. All rights reserved.
|
|
*
|
|
* 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.
|
|
*/
|
|
|
|
#pragma once
|
|
|
|
#include "tensorrt_llm/runtime/bufferManager.h"
|
|
#include "tensorrt_llm/runtime/cudaEvent.h"
|
|
#include "tensorrt_llm/runtime/cudaStream.h"
|
|
#include "tensorrt_llm/runtime/decoderState.h"
|
|
#include "tensorrt_llm/runtime/gptDecoder.h"
|
|
#include "tensorrt_llm/runtime/iGptDecoderBatched.h"
|
|
#include "tensorrt_llm/runtime/iTensor.h"
|
|
#include "tensorrt_llm/runtime/worldConfig.h"
|
|
|
|
#include <memory>
|
|
#include <vector>
|
|
|
|
namespace tensorrt_llm::batch_manager
|
|
{
|
|
class LlmRequest;
|
|
} // namespace tensorrt_llm::batch_manager
|
|
|
|
namespace tensorrt_llm::runtime
|
|
{
|
|
|
|
//! GPT decoder class with support for in-flight batching
|
|
class GptDecoderBatched : public IGptDecoderBatched
|
|
{
|
|
public:
|
|
using CudaStreamPtr = std::shared_ptr<CudaStream>;
|
|
using LlmRequestPtr = std::shared_ptr<tensorrt_llm::batch_manager::LlmRequest>;
|
|
using RequestVector = std::vector<LlmRequestPtr>;
|
|
using TensorPtr = ITensor::SharedPtr;
|
|
using SharedConstPtr = ITensor::SharedConstPtr;
|
|
|
|
GptDecoderBatched(
|
|
CudaStreamPtr stream, SpeculativeDecodingMode const& speculativeDecodingMode, nvinfer1::DataType dtype);
|
|
|
|
void setup(executor::DecodingMode const& mode, SizeType32 maxBatchSize, SizeType32 maxBeamWidth,
|
|
SizeType32 maxAttentionWindow, SizeType32 sinkTokenLength, SizeType32 maxSequenceLength,
|
|
SizeType32 maxTokensPerStep, nvinfer1::DataType dtype, ModelConfig const& modelConfig,
|
|
WorldConfig const& worldConfig) override;
|
|
|
|
void disableLookahead(RequestVector const& genRequests, TensorPtr const& batchSlots) override;
|
|
|
|
DecoderFinishedEventPtr forwardAsync(decoder_batch::Output& output, decoder_batch::Input const& input) override;
|
|
void forward(decoder_batch::Output& output, decoder_batch::Input const& input) override;
|
|
|
|
//! @brief Gather final beam search results for request `batchSlot`.
|
|
//! Result will only be available after event returned.
|
|
[[nodiscard]] CudaEvent finalize(decoder::DecoderState const& decoderState, SizeType32 batchSlot,
|
|
SamplingConfig const& samplingConfig, bool streaming) const override;
|
|
|
|
decoder::DecoderState& getDecoderState()
|
|
{
|
|
return *mDecoderState;
|
|
}
|
|
|
|
decoder::DecoderState const& getDecoderState() const
|
|
{
|
|
return *mDecoderState;
|
|
}
|
|
|
|
CudaStreamPtr getDecoderStream() const
|
|
{
|
|
return mDecoderStream;
|
|
}
|
|
|
|
IGptDecoder& getUnderlyingDecoder() const
|
|
{
|
|
return *mDecoder.get();
|
|
}
|
|
|
|
[[nodiscard]] BufferManager const& getBufferManager() const
|
|
{
|
|
return mBufferManager;
|
|
}
|
|
|
|
private:
|
|
//! @brief Sets inputs for explicit draft tokens.
|
|
void setExplicitDraftTokensInputs(decoder_batch::Input const& input);
|
|
|
|
//! @brief Sets inputs for eagle decoding.
|
|
void setEagleInputs(decoder_batch::Input const& input);
|
|
|
|
//! @brief Calls decoders for tokens per engine step
|
|
void forwardDispatch(decoder_batch::Output& output, decoder_batch::Input const& input);
|
|
|
|
//! @brief Prepare Input and Output for decoder step
|
|
void prepareForward(SizeType32 step, decoder_batch::Output& output, decoder_batch::Input const& input);
|
|
|
|
private:
|
|
CudaStreamPtr mRuntimeStream;
|
|
CudaStreamPtr mDecoderStream;
|
|
BufferManager mBufferManager;
|
|
|
|
using GptDecoderPtr = std::unique_ptr<IGptDecoder>;
|
|
GptDecoderPtr mDecoder;
|
|
|
|
std::shared_ptr<decoder::DecoderState> mDecoderState;
|
|
};
|
|
} // namespace tensorrt_llm::runtime
|