TensorRT-LLMs/cpp/include/tensorrt_llm/runtime/memoryCounters.h

/*
 * 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/common/assert.h"
#include "tensorrt_llm/runtime/iBuffer.h"

#include <algorithm>
#include <cstdint>
#include <string>

namespace tensorrt_llm::runtime
{

class MemoryCounters
{
public:
    using SizeType = std::size_t;
    using DiffType = std::ptrdiff_t;

    MemoryCounters() = default;

    [[nodiscard]] SizeType getGpu() const
    {
        return mGpu;
    }

    [[nodiscard]] SizeType getCpu() const
    {
        return mCpu;
    }

    [[nodiscard]] SizeType getPinned() const
    {
        return mPinned;
    }

    [[nodiscard]] SizeType getUVM() const
    {
        return mUVM;
    }

    [[nodiscard]] DiffType getGpuDiff() const
    {
        return mGpuDiff;
    }

    [[nodiscard]] DiffType getCpuDiff() const
    {
        return mCpuDiff;
    }

    [[nodiscard]] DiffType getPinnedDiff() const
    {
        return mPinnedDiff;
    }

    [[nodiscard]] DiffType getUVMDiff() const
    {
        return mUVMDiff;
    }

    template <MemoryType T>
    void allocate(SizeType size)
    {
        auto const sizeDiff = static_cast<DiffType>(size);
        if constexpr (T == MemoryType::kGPU)
        {
            mGpu += size;
            mGpuDiff = sizeDiff;
        }
        else if constexpr (T == MemoryType::kCPU)
        {
            mCpu += size;
            mCpuDiff = sizeDiff;
        }
        else if constexpr (T == MemoryType::kPINNED)
        {
            mPinned += size;
            mPinnedDiff = sizeDiff;
        }
        else if constexpr (T == MemoryType::kUVM)
        {
            mUVM += size;
            mUVMDiff = sizeDiff;
        }
        else
        {
            TLLM_THROW("Unknown memory type: %s", MemoryTypeString<T>::value);
        }
    }

    void allocate(MemoryType memoryType, SizeType size);

    template <MemoryType T>
    void deallocate(SizeType size)
    {
        auto const sizeDiff = -static_cast<DiffType>(size);
        if constexpr (T == MemoryType::kGPU)
        {
            mGpu -= std::min(size, mGpu);
            mGpuDiff = sizeDiff;
        }
        else if constexpr (T == MemoryType::kCPU)
        {
            mCpu -= std::min(size, mCpu);
            mCpuDiff = sizeDiff;
        }
        else if constexpr (T == MemoryType::kPINNED)
        {
            mPinned -= std::min(size, mPinned);
            mPinnedDiff = sizeDiff;
        }
        else if constexpr (T == MemoryType::kUVM)
        {
            mUVM -= std::min(size, mUVM);
            mUVMDiff = sizeDiff;
        }
        else
        {
            TLLM_THROW("Unknown memory type: %s", MemoryTypeString<T>::value);
        }
    }

    void deallocate(MemoryType memoryType, SizeType size);

    static MemoryCounters& getInstance()
    {
        static thread_local MemoryCounters mInstance;
        return mInstance;
    }

    static std::string bytesToString(SizeType bytes, int precision = 2);

    static std::string bytesToString(DiffType bytes, int precision = 2);

    [[nodiscard]] std::string toString() const;

private:
    SizeType mGpu{}, mCpu{}, mPinned{}, mUVM{};
    DiffType mGpuDiff{}, mCpuDiff{}, mPinnedDiff{}, mUVMDiff{};
};

} // namespace tensorrt_llm::runtime