mirror of
https://github.com/NVIDIA/TensorRT-LLM.git
synced 2026-01-14 06:27:45 +08:00
104 lines
4.2 KiB
C++
104 lines
4.2 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/common.h"
|
|
#include "tensorrt_llm/runtime/generationConfig.h"
|
|
#include "tensorrt_llm/runtime/iTensor.h"
|
|
#include "tensorrt_llm/runtime/modelConfig.h"
|
|
#include "tensorrt_llm/runtime/tllmRuntime.h"
|
|
#include "tensorrt_llm/runtime/worldConfig.h"
|
|
|
|
namespace tensorrt_llm::batch_manager::kv_cache_manager
|
|
{
|
|
class KVCacheManager;
|
|
} // namespace tensorrt_llm::batch_manager::kv_cache_manager
|
|
|
|
namespace tensorrt_llm::runtime
|
|
{
|
|
|
|
class RuntimeBuffers;
|
|
|
|
class TransformerBuffers
|
|
{
|
|
public:
|
|
using TensorPtr = ITensor::SharedPtr;
|
|
using KvCacheManager = batch_manager::kv_cache_manager::KVCacheManager;
|
|
using TensorMap = StringPtrMap<ITensor>;
|
|
|
|
TransformerBuffers();
|
|
|
|
TransformerBuffers(
|
|
TllmRuntime const& runtime, runtime::ModelConfig const& modelConfig, runtime::WorldConfig const& worldConfig);
|
|
|
|
void reshape(
|
|
GenerationConfig const& generationConfig, ModelConfig const& modelConfig, WorldConfig const& worldConfig);
|
|
|
|
void reshapeKvTensors(SizeType32 maxBatchSize, SizeType32 maxBeamWidth, SizeType32 maxBlocksPerSeq,
|
|
runtime::TllmRuntime const& runtime);
|
|
|
|
void setKvPoolPointers(KvCacheManager const* kvCacheManager);
|
|
void setKvPoolMapping(KvCacheManager const* kvCacheManager);
|
|
|
|
void reset(BufferManager& manager){};
|
|
|
|
TransformerBuffers sliceTo(GenerationConfig const& generationConfig, ModelConfig const& modelConfig,
|
|
SizeType32 offset, SizeType32 batchSize);
|
|
|
|
void prepareContextStep(RuntimeBuffers* runtimeBuffers, TensorPtr const& inputIds, TokenIdType padId,
|
|
BufferManager& manager, KvCacheManager const* kvCacheManager, SizeType32 firstBatchSlotIdx,
|
|
ModelConfig const& modelConfig, WorldConfig const& worldConfig);
|
|
|
|
void postContextStep(RuntimeBuffers* runtimeBuffers, std::vector<RuntimeBuffers> const& contextBuffers,
|
|
BufferManager& manager, ModelConfig const& modelConfig, WorldConfig const& worldConfig);
|
|
|
|
void prepareNextStep(RuntimeBuffers* runtimeBuffers, SizeType32 step, BufferManager& manager,
|
|
KvCacheManager* kvCacheManager, SizeType32 firstBatchSlotIdx, ModelConfig const& modelConfig,
|
|
WorldConfig const& worldConfig);
|
|
|
|
void getRuntimeBuffers(RuntimeBuffers const* runtimeBuffers, TensorMap& inputBuffers, TensorMap& outputBuffers,
|
|
SizeType32 step, TensorPtr const& inputIds, ModelConfig const& modelConfig,
|
|
WorldConfig const& worldConfig) const;
|
|
|
|
protected:
|
|
void copyAttentionMasks(
|
|
RuntimeBuffers* runtimeBuffers, std::vector<RuntimeBuffers> const& contextBatches, BufferManager& manager);
|
|
|
|
void tile(RuntimeBuffers* runtimeBuffers, BufferManager& manager, ModelConfig const& modelConfig,
|
|
WorldConfig const& worldConfig);
|
|
|
|
public:
|
|
// engine
|
|
TensorPtr pastKeyValueLengths; // with attention plugin, host tensor
|
|
TensorPtr attentionMask; // without attention plugin
|
|
TensorPtr positionIds;
|
|
|
|
std::vector<TensorPtr> presentKeysVals;
|
|
std::vector<TensorPtr> presentKeysValsAlt; // without attention plugin
|
|
TensorPtr maxAttentionWindows; // with attention plugin, host tensor
|
|
TensorPtr sinkTokenLengths; // with attention plugin, host tensor
|
|
TensorPtr kvCacheBlockPoolPointers;
|
|
TensorPtr kvCacheBlockPoolMapping;
|
|
TensorPtr kvCacheBlockOffsetsHost; // [numPools, batchSize * beamWidth, 2, maxBlocksPerSeq * 2]
|
|
TensorPtr kvCacheBlockOffsetsDevice; // [numPools, batchSize * beamWidth, 2, maxBlocksPerSeq * 2]
|
|
TensorPtr runtimePerfKnobsHost; // can hold max 16 perf knobs
|
|
TensorPtr contextProgressHost;
|
|
};
|
|
|
|
} // namespace tensorrt_llm::runtime
|