WO2024041219A1

WO2024041219A1 - Memory management method, electronic device, chip system, and readable storage medium

Info

Publication number: WO2024041219A1
Application number: PCT/CN2023/105048
Authority: WO
Inventors: 刘长红; 邓晓阳; 刘强; 张双武; 林大云; 罗雄
Original assignee: 华为云计算技术有限公司
Priority date: 2022-08-26
Filing date: 2023-06-30
Publication date: 2024-02-29
Also published as: CN117667369A

Abstract

The present application relates to the technical field of communications, and provides a memory management method, an electronic device, a chip system, and a readable storage medium, which solve the problem of inappropriate memory resource allocation in existing RDMA technologies to a certain extent. The method is applied to a first electronic device and comprises: determining the amount of data to be stored of a target application and a valid memory space in a physical address space associated with the target application, the valid memory space comprising memory spaces indicated by N first physical address units, each first physical address unit comprising at least one physical address having a data read-write permission, wherein N is an integer greater than or equal to 0; if the amount of data is greater than the valid memory space, expanding the valid memory space according to a first difference between the amount of data and the valid memory space; and if the amount of data is smaller than the valid memory space, reducing the valid memory space according to a second difference between the valid memory space and the amount of data.

Description

Memory management method, electronic device, chip system and readable storage medium

This application requires the priority of the Chinese patent application submitted to the State Intellectual Property Office on August 26, 2022, with the application number 202211033616.X and the application name "Memory management method, electronic device, chip system and readable storage medium", The entire contents of which are incorporated herein by reference.

Technical field

The present application relates to the field of communication technology, and in particular to a memory management method, electronic equipment, chip system and readable storage medium.

Background technique

Remote Direct Memory Access (RDMA) technology is a network transmission technology that directly accesses the storage area of a remote computer. Compared with traditional network technology, RDMA technology can bypass the operating system protocol stack and reduce data transmission. The number of copies avoids the overhead of user-mode and kernel-mode switching, thereby occupying smaller system resources and reducing latency.

In the process of data transmission based on RDMA technology in applications in electronic equipment, a certain amount of physical address space needs to be pre-registered with the RDMA system for data caching by the RDMA network card during the process of sending and receiving data. When multiple applications are running simultaneously on an electronic device, a corresponding physical address space needs to be pre-allocated for each application. However, the memory resources of the memory in the electronic device are limited, and the pre-allocated fixed-size physical address space may not be sufficient. The memory requirements of some applications may be greater than the memory requirements of some applications, resulting in a waste of resources.

Contents of the invention

This application provides a memory management method, electronic device, chip system and readable storage medium, which solves the problem of unreasonable memory resource allocation in the existing RDMA technology to a certain extent.

In order to achieve the above purpose, this application adopts the following technical solutions:

In a first aspect, the present application provides a memory management method, which is applied to a first electronic device. The method includes: determining the amount of data to be stored by a target application and the effective memory space in the physical address space associated with the target application. The effective memory space Including memory space indicated by N first physical address units, each of the first physical address units includes at least one physical address with data read and write permissions, N is an integer greater than or equal to 0; if the amount of data is greater than the effective memory space , then the effective memory space is expanded according to the first difference between the data amount and the effective memory space; if the data amount is less than the effective memory space, the effective memory space is expanded according to the second difference between the effective memory space and the data amount. Make reductions.

Based on the memory management method provided by this application, based on the difference between the amount of data to be stored by the target application and the effective memory space with data read and write permissions in the physical address space corresponding to the target application, it is determined whether the effective memory space meets the current requirements of the application. The memory space required for the amount of data to be stored, thereby dynamically adjusting the effective memory space of the target application. Among them, if the effective memory space does not meet the memory space required by the amount of data to be stored, the effective memory space can be expanded according to the first difference between the data amount and the effective memory space, so that the expanded effective memory space satisfies The memory space required by the application. If the effective memory space meets the memory space required by the amount of data to be stored, the effective memory space can be reduced according to the second difference between the effective memory space and the amount of data, thereby releasing some unused memory space. Effective memory space occupied to avoid wasting memory resources.

In a possible implementation of the first aspect, the physical address space further includes a second physical address unit, and the second physical address unit includes at least one physical address that does not have data writing permission.

In a possible implementation of the first aspect, the method further includes:

When receiving a memory allocation request sent by the target application, allocate a second physical address unit to the target application; when receiving a registration request based on remote direct data access communication sent by the target application, assign the second physical address unit to the target application The initial mapping table between virtual address spaces is registered.

In a possible implementation of the first aspect, the virtual address space of the target application includes M virtual address units, each virtual address unit includes at least one virtual address, and M is an integer greater than or equal to 1;

The effective memory space is expanded according to the first difference between the data amount and the effective memory space, including:

Allocate m first physical address units to the target application according to the first difference value. The memory space of the m first physical address units is greater than or equal to the first difference value. 1≤m+N<M; divide the M virtual address units into m virtual address units corresponding to the second physical address units are remapped to m first physical address units.

In a possible implementation of the first aspect, the virtual address space of the target application includes M virtual address units, each virtual address unit includes at least one virtual address, M is an integer greater than or equal to 1, and the effective memory space includes N first physical address units corresponding to N virtual address units among M virtual address units, 1≤N≤M;

The effective memory space is reduced according to the second difference between the effective memory space and the amount of data, including:

When the second difference is greater than or equal to the preset threshold, n virtual address units corresponding to the n first physical address units among the N virtual address units are remapped to the second physical address unit, and the n first physical addresses The memory space of the unit is less than or equal to the second difference value, the preset threshold is greater than 0, and 1≤n<N.

In a possible implementation of the first aspect, the target application is an application installed in the first electronic device.

In a possible implementation of the first aspect, the amount of data to be stored is the amount of data of the target application, or the amount of data of the application installed in the second electronic device received by the first electronic device.

In a second aspect, the present application provides an electronic device, including: a processor configured to run a computer program stored in a memory to implement the method in any possible implementation of the first aspect.

In the third aspect, this application provides another electronic device, including: a processor and a remote direct data access network card;

The remote direct data access network card is used to determine the amount of data to be stored by the target application and the effective memory space in the physical address space associated with the target application. The effective memory space includes the memory space indicated by N first physical address units, each of which A physical address unit includes at least one physical address with data read and write permissions, and N is an integer greater than or equal to 0; if the amount of data is greater than the effective memory space, the remote direct data access network card is also used to send the first interrupt information to the processor , the first interrupt information includes the first difference between the data amount and the effective memory space, and the processor is used to expand the effective memory space according to the first difference; if the data amount is less than the effective memory space, the remote direct data access network card also The processor is configured to send second interrupt information to the processor, where the second interrupt information includes a second difference between the effective memory space and the data amount, and the processor is configured to reduce the effective memory space according to the second difference.

In a possible implementation of the third aspect, the physical address space further includes a second physical address unit, and the second physical address unit includes at least one physical address that does not have data writing permission.

In a possible implementation of the third aspect, the processor is also used to:

When receiving a memory allocation request sent by the target application, allocate a second physical address unit to the target application;

Remote direct data access network cards are also used for:

When a registration request based on remote direct data access communication sent by the target application is received, an initial mapping table between the second physical address unit and the virtual address space of the target application is registered.

In a possible implementation of the third aspect, the virtual address space of the target application includes M virtual address units, each virtual address unit includes at least one virtual address, and M is an integer greater than or equal to 1;

The processor is also used for:

In a possible implementation of the third aspect, the virtual address space of the target application includes M virtual address units, each virtual address unit includes at least one virtual address, M is an integer greater than or equal to 1, and the effective memory space includes N first physical address units corresponding to N virtual address units among M virtual address units, 1≤N≤M;

The processor is also configured to: when the second difference is greater than or equal to the preset threshold, remap n virtual address units corresponding to the n first physical address units among the N virtual address units to the second physical address unit, The memory space of n first physical address units is less than or equal to the second difference, the preset threshold is greater than 0, and 1≤n<N.

In a possible implementation of the third aspect, the target application is an application installed in the electronic device.

In a possible implementation of the third aspect, the amount of data to be stored is the amount of data received by the first remote direct data access network card and sent by the second remote direct data access network card.

In a fourth aspect, this application provides a chip system. The chip system includes a processor, and the processor executes the calculation stored in the memory. A computer program to implement the method in any possible implementation manner of the first aspect.

In a fifth aspect, this application provides a computer-readable storage medium. The computer-readable storage medium stores a computer program. When the computer program is executed by a processor, the method in any possible implementation manner of the first aspect is implemented.

In a sixth aspect, the present application provides a computer program product, which when the computer program product is run on an electronic device, causes the electronic device to execute the method in any possible implementation of the first aspect.

The technical effects of the second to sixth aspects provided by this application can be referred to the technical effects of each possible implementation of the above-mentioned first aspect, and will not be described again here.

Description of drawings

Figure 1 is a schematic structural diagram of an electronic device provided by an embodiment of the present application;

Figure 2 is a schematic diagram of the software structure of an electronic device provided by an embodiment of the present application;

Figure 3 is a schematic diagram of data transmission based on traditional network communication technology provided by an embodiment of the present application;

Figure 4 is a schematic diagram of data transmission based on existing RDMA communication technology provided by an embodiment of the present application;

Figure 5 is a schematic diagram of the mapping relationship between virtual address space, page table and physical address space provided by the embodiment of the present application;

Figure 6 is a schematic flow chart of a memory management method provided by an embodiment of the present application;

Figure 7 is a schematic diagram of the mapping relationship between the virtual address space of a target application and the second physical address unit provided by an embodiment of the present application;

Figure 8 is a schematic diagram of the mapping relationship between the virtual address space and the physical address space of a target application provided by an embodiment of the present application;

Figure 9 is a schematic diagram of the mapping relationship between the virtual address space and the physical address space of another target application provided by an embodiment of the present application;

Figure 10 is a schematic diagram of the interaction between the target application, the processor and the RDMA network card in the electronic device provided by the embodiment of the present application;

Figure 11 is a schematic diagram of interaction between a first electronic device and a second electronic device provided by an embodiment of the present application;

Figure 12 is another schematic diagram of interaction between the target application, the processor and the RDMA network card in the electronic device provided by the embodiment of the present application.

Detailed ways

The technical solutions in the embodiments of the present application will be described below with reference to the drawings and related embodiments in the embodiments of the present application. In the description of the embodiments of the present application, the terms used in the following embodiments are only for the purpose of describing specific embodiments and are not intended to limit the present application. As used in the specification and appended claims of this application, the singular expressions "a," "the," "above," "the" and "the" are intended to also include, for example, "a "or more" unless the context clearly indicates otherwise. It should also be understood that in the following embodiments of this application, "at least one" and "one or more" refer to one or more than two (including two). The term "and/or" is used to describe the relationship between associated objects, indicating that there can be three relationships; for example, A and/or B can mean: A exists alone, A and B exist simultaneously, and B exists alone, Where A and B can be singular or plural. The character "/" generally indicates that the related objects are in an "or" relationship.

Reference in this specification to "one embodiment" or "some embodiments" or the like means that a particular feature, structure or characteristic described in connection with the embodiment is included in one or more embodiments of the application. Therefore, the phrases "in one embodiment", "in some embodiments", "in other embodiments", "in other embodiments", etc. appearing in different places in this specification are not necessarily References are made to the same embodiment, but rather to "one or more but not all embodiments" unless specifically stated otherwise. The terms “including,” “includes,” “having,” and variations thereof all mean “including but not limited to,” unless otherwise specifically emphasized. The term "connected" includes both direct and indirect connections unless otherwise stated. “First” and “second” are used for descriptive purposes only and cannot be understood as indicating or implying relative importance or implicitly indicating the quantity of indicated technical features.

In the embodiments of this application, words such as "exemplarily" or "for example" are used to represent examples, illustrations or explanations. this application Any embodiment or design described in the embodiments as "exemplary" or "such as" is not intended to be construed as preferred or advantageous over other embodiments or designs. Rather, the use of the words "exemplarily" or "for example" is intended to present the relevant concepts in a concrete manner.

The memory management method provided by the embodiment of the present application can be applied to mobile phones, tablet computers, wearable devices, augmented reality (AR)/virtual reality (VR) devices, notebook computers, ultra-mobile personal computers (ultra- On electronic devices such as mobile personal computers (UMPC), netbooks, personal digital assistants (personal digital assistants, PDAs), servers, etc., the embodiments of this application do not place any restrictions on the specific types of electronic devices.

Please refer to Figure 1, which is a schematic structural diagram of an electronic device 100 provided by an embodiment of the present application. The electronic device 100 may include a processor (central processing unit, CPU) 110, an external memory interface 120, an internal memory 121, a universal serial bus ( universal serial bus (USB) interface 130, charging management module 140, power management module 141, battery 142, antenna 1, antenna 2, mobile communication module 150, wireless communication module 160, audio module 170, speaker 170A, receiver 170B, microphone 170C , headphone interface 170D, sensor module 180, button 190, motor 191, indicator 192, camera 193, display screen 194, and subscriber identification module (subscriber identification module, SIM) card interface 195, RDMA network card 196, etc. Among them, the sensor module 180 may include a pressure sensor, a gyroscope sensor, an air pressure sensor, a magnetic sensor, an acceleration sensor, a distance sensor, a proximity light sensor, a fingerprint sensor, a temperature sensor, a touch sensor, an ambient light sensor L, a bone conduction sensor, etc.

It can be understood that the structure illustrated in the embodiment of the present application does not constitute a specific limitation on the electronic device 100 . In other embodiments of the present application, the electronic device 100 may include more or fewer components than shown in the figures, or some components may be combined, some components may be separated, or some components may be arranged differently. The components illustrated may be implemented in hardware, software, or a combination of software and hardware.

The CPU 110 may include one or more processing units. For example, the CPU 110 may include an application processor (application processor, AP), a modem processor, a graphics processing unit (GPU), an image signal processor (image signal processor) , ISP), controller, memory, video codec, digital signal processor (digital signal processor, DSP), baseband processor, and/or neural network processor (neural-network processing unit, NPU), etc. Among them, different processing units can be independent devices or integrated in one or more processors.

The controller may be the nerve center and command center of the electronic device 100 . The controller can generate operation control signals based on the instruction operation code and timing signals to complete the control of fetching and executing instructions.

In some embodiments, CPU 110 may include one or more interfaces. Interfaces may include integrated circuit (inter-integrated circuit, I2C) interface, integrated circuit built-in audio (inter-integrated circuit sound, I2S) interface, pulse code modulation (pulse code modulation, PCM) interface, universal asynchronous receiver transmitter (universal asynchronous receiver) /transmitter, UART) interface, mobile industry processor interface (MIPI), general-purpose input/output (GPIO) interface, subscriber identity module (subscriber identity module, SIM) interface, and/ Or universal serial bus (USB) interface, etc.

The I2C interface is a bidirectional synchronous serial bus, including a serial data line (SDA) and a serial clock line (derail clock line, SCL). In some embodiments, CPU 110 may include multiple sets of I2C buses. The CPU 110 can couple touch sensors, chargers, flashes, cameras 193, etc. respectively through different I2C bus interfaces. For example, the CPU 110 can be coupled to a touch sensor through an I2C interface, so that the CPU 110 and the touch sensor communicate through the I2C bus interface to implement the touch function of the electronic device 100 .

The I2S interface can be used for audio communication. In some embodiments, CPU 110 may include multiple sets of I2S buses. The CPU 110 can be coupled with the audio module 170 through the I2S bus to implement communication between the CPU 110 and the audio module 170 . In some embodiments, the audio module 170 can transmit audio signals to the wireless communication module 160 through the I2S interface to implement the function of answering calls through a Bluetooth headset.

The PCM interface can also be used for audio communications to sample, quantize and encode analog signals. In some embodiments, the audio module 170 and the wireless communication module 160 may be coupled through a PCM bus interface. In some embodiments, the audio module 170 can also transmit audio signals to the wireless communication module 160 through the PCM interface to implement the function of answering calls through a Bluetooth headset. Both the I2S interface and the PCM interface can be used for audio communication.

The UART interface is a universal serial data bus used for asynchronous communication. The bus can be a bidirectional communication bus. It converts the data to be transmitted between serial communication and parallel communication. In some embodiments, the UART interface is typically used to connect CPU110 and wireless communication module 160. For example, the CPU 110 communicates with the Bluetooth module in the wireless communication module 160 through the UART interface to implement the Bluetooth function. In some embodiments, the audio module 170 can transmit audio signals to the wireless communication module 160 through the UART interface to implement the function of playing music through a Bluetooth headset.

The MIPI interface can be used to connect the CPU 110 with peripheral devices such as the display screen 194 and the camera 193 . MIPI interfaces include camera serial interface (CSI), display serial interface (DSI), etc. In some embodiments, the CPU 110 and the camera 193 communicate through the CSI interface to implement the shooting function of the electronic device 100 . The CPU 110 and the display screen 194 communicate through the DSI interface to realize the display function of the electronic device 100 .

The GPIO interface can be configured through software. The GPIO interface can be configured as a control signal or as a data signal. In some embodiments, the GPIO interface can be used to connect the CPU 110 with the camera 193, the display screen 194, the wireless communication module 160, the audio module 170, the sensor module 180, etc. The GPIO interface can also be configured as an I2C interface, I2S interface, UART interface, MIPI interface, etc.

The USB interface 130 is an interface that complies with the USB standard specification, and may be a Mini USB interface, a Micro USB interface, a USB Type C interface, etc. The USB interface 130 can be used to connect a charger to charge the electronic device 100, and can also be used to transmit data between the electronic device 100 and peripheral devices. It can also be used to connect headphones to play audio through them. This interface can also be used to connect other electronic devices 100, such as AR devices, etc.

It can be understood that the interface connection relationships between the modules illustrated in the embodiments of the present application are only schematic illustrations and do not constitute a structural limitation of the electronic device 100 . In other embodiments of the present application, the electronic device 100 may also adopt different interface connection methods in the above embodiments, or a combination of multiple interface connection methods.

The charging management module 140 is used to receive charging input from the charger. Among them, the charger can be a wireless charger or a wired charger. In some wired charging embodiments, the charging management module 140 may receive charging input from the wired charger through the USB interface 130 . In some wireless charging embodiments, the charging management module 140 may receive wireless charging input through the wireless charging coil of the electronic device 100 . While the charging management module 140 charges the battery 142, it can also provide power to the electronic device 100 through the power management module 141.

The power management module 141 is used to connect the battery 142, the charging management module 140 and the CPU 110. The power management module 141 receives input from the battery 142 and/or the charging management module 140, and supplies power to the CPU 110, internal memory 121, external memory, display screen 194, camera 193, wireless communication module 160, etc. The power management module 141 can also be used to monitor battery capacity, battery cycle times, battery health status (leakage, impedance) and other parameters. In some other embodiments, the power management module 141 may also be provided in the CPU 110 . In other embodiments, the power management module 141 and the charging management module 140 may also be provided in the same device.

The wireless communication function of the electronic device 100 can be implemented through the antenna 1, the antenna 2, the mobile communication module 150, the wireless communication module 160, the modem processor and the baseband processor.

Antenna 1 and Antenna 2 are used to transmit and receive electromagnetic wave signals. Each antenna in electronic device 100 may be used to cover a single or multiple communication frequency bands. Different antennas can also be reused to improve antenna utilization. For example: Antenna 1 can be reused as a diversity antenna for a wireless LAN. In other embodiments, antennas may be used in conjunction with tuning switches.

The mobile communication module 150 can provide solutions for wireless communication including 2G/3G/4G/5G applied on the electronic device 100 . The mobile communication module 150 may include at least one filter, switch, power amplifier, low noise amplifier (LNA), etc. The mobile communication module 150 can receive electromagnetic waves through the antenna 1, perform filtering, amplification and other processing on the received electromagnetic waves, and transmit them to the modem processor for demodulation. The mobile communication module 150 can also amplify the signal modulated by the modem processor and convert it into electromagnetic waves through the antenna 1 for radiation. In some embodiments, at least part of the functional modules of the mobile communication module 150 may be disposed in the CPU 110 . In some embodiments, at least part of the functional modules of the mobile communication module 150 may be provided in the same device as at least part of the modules of the CPU 110 .

A modem processor may include a modulator and a demodulator. Among them, the modulator is used to modulate the low-frequency baseband signal to be sent into a medium-high frequency signal. The demodulator is used to demodulate the received electromagnetic wave signal into a low-frequency baseband signal. The demodulator then transmits the demodulated low-frequency baseband signal to the baseband processor for processing. After the low-frequency baseband signal is processed by the baseband processor, it is passed to the application processor. The application processor outputs sound signals through audio devices (not limited to speaker 170A, receiver 170B, etc.), or displays images or videos through display screen 194. In some embodiments, the modem processor may be a stand-alone device. In other embodiments, the modem processor may be independent of the CPU 110 and may be provided in the same device as the mobile communication module 150 or other functional modules.

The wireless communication module 160 can provide applications on the electronic device 100 including wireless local area networks (WLAN) (such as wireless fidelity (Wi-Fi) network), Bluetooth (bluetooth, BT), and global navigation satellites. System (global navigation satellite system, GNSS), frequency modulation (frequency modulation, FM), near field communication technology (near field communication, NFC), infrared technology (infrared, IR) and other wireless communication solutions. The wireless communication module 160 may be one or more devices integrating at least one communication MMU. For example, the wireless communication module 160 may be an RDMA network card. The wireless communication module 160 receives electromagnetic waves via the antenna 2 , frequency-modulates and filters the electromagnetic wave signals, and sends the processed signals to the CPU 110 . The wireless communication module 160 can also receive the signal to be sent from the CPU 110, frequency modulate it, amplify it, and convert it into electromagnetic waves through the antenna 2 for radiation.

In some embodiments, the antenna 1 of the electronic device 100 is coupled to the mobile communication module 150, and the antenna 2 is coupled to the wireless communication module 160, so that the electronic device 100 can communicate with the network and other devices through wireless communication technology. The wireless communication technology may include global system for mobile communications (GSM), general packet radio service (GPRS), code division multiple access (CDMA), broadband Code division multiple access (wideband code division multiple access, WCDMA), time division code division multiple access (time-division code division multiple access, TD-SCDMA), long term evolution (long term evolution, LTE), BT, GNSS, WLAN, NFC , FM, and/or IR technology, etc. The GNSS may include global positioning system (GPS), global navigation satellite system (GLONASS), Beidou navigation satellite system (BDS), quasi-zenith satellite system (quasi) -zenith satellite system (QZSS) and/or satellite based augmentation systems (SBAS).

The electronic device 100 implements display functions through a GPU, a display screen 194, an application processor, and the like. The GPU is an image processing microprocessor and is connected to the display screen 194 and the application processor. GPUs are used to perform mathematical and geometric calculations for graphics rendering. CPU 110 may include one or more GPUs that execute program instructions to generate or change display information.

The display screen 194 is used to display images, videos, etc. Display 194 includes a display panel. The display panel can use a liquid crystal display (LCD), an organic light-emitting diode (OLED), an active matrix organic light emitting diode or an active matrix organic light emitting diode (active-matrix organic light emitting diode). emitting diode (AMOLED), flexible light-emitting diode (FLED), Miniled, MicroLed, Micro-oLed, quantum dot light emitting diode (QLED), etc. In some embodiments, the electronic device 100 may include 1 or N display screens 194, where N is a positive integer greater than 1.

The electronic device 100 can implement the shooting function through an ISP, a camera 193, a video codec, a GPU, a display screen 194, an application processor, and the like.

The ISP is used to process the data fed back by the camera 193. For example, when taking a photo, the shutter is opened, the light is transmitted to the camera sensor through the lens, the optical signal is converted into an electrical signal, and the camera sensor passes the electrical signal to the ISP for processing, and converts it into an image visible to the naked eye. ISP can also perform algorithm optimization on image noise, brightness, and skin color. ISP can also optimize the exposure, color temperature and other parameters of the shooting scene. In some embodiments, the ISP may be provided in the camera 193.

Camera 193 is used to capture still images or video. The object passes through the lens to produce an optical image that is projected onto the photosensitive element. The photosensitive element can be a charge coupled device (CCD) or a complementary metal-oxide-semiconductor (CMOS) phototransistor. The photosensitive element converts the optical signal into an electrical signal, and then passes the electrical signal to the ISP to convert it into a digital image signal. ISP outputs digital image signals to DSP for processing. DSP converts digital image signals into standard RGB, YUV and other format image signals. In some embodiments, the electronic device 100 may include 1 or N cameras 193, where N is a positive integer greater than 1.

Digital signal processors are used to process digital signals. In addition to digital image signals, they can also process other digital signals. For example, when the electronic device 100 selects a frequency point, the digital signal processor is used to perform Fourier transform on the frequency point energy.

Video codecs are used to compress or decompress digital video. Electronic device 100 may support one or more video codecs. In this way, the electronic device 100 can play or record videos in multiple encoding formats, such as moving picture experts group (MPEG) 1, MPEG2, MPEG3, MPEG4, etc.

NPU is a neural network (NN) computing processor. By drawing on the structure of biological neural networks, such as the transmission mode between neurons in the human brain, it can quickly process input information and can continuously learn by itself. Intelligent cognitive applications of the electronic device 100 can be implemented through the NPU, such as image recognition, face recognition, speech recognition, text understanding, etc.

The electronic device 100 can implement audio functions through the audio module 170, the speaker 170A, the receiver 170B, the microphone 170C, the headphone interface 170D, and the application processor. Such as music playback, recording, etc.

The audio module 170 is used to convert digital audio information into analog audio signal output, and is also used to convert analog audio input into digital audio signals. Audio module 170 may also be used to encode and decode audio signals. In some embodiments, the audio module 170 may be disposed in the CPU 110 , or some functional modules of the audio module 170 may be disposed in the CPU 110 .

Speaker 170A, also called "speaker", is used to convert audio electrical signals into sound signals. The electronic device 100 can listen to music through the speaker 170A, or listen to hands-free calls.

Receiver 170B, also called "earpiece", is used to convert audio electrical signals into sound signals. When the electronic device 100 answers a call or a voice message, the voice can be heard by bringing the receiver 170B close to the human ear.

Microphone 170C, also called "microphone" or "microphone", is used to convert sound signals into electrical signals. When making a call or sending a voice message, the user can speak close to the microphone 170C with the human mouth and input the sound signal to the microphone 170C. The electronic device 100 may be provided with at least one microphone 170C. In other embodiments, the electronic device 100 may be provided with two microphones 170C, which in addition to collecting sound signals, may also implement a noise reduction function. In other embodiments, the electronic device 100 can also be provided with three, four or more microphones 170C to collect sound signals, reduce noise, identify sound sources, and implement directional recording functions, etc.

The headphone interface 170D is used to connect wired headphones. The headphone interface 170D can be a USB interface 130, or a 3.5mm open mobile terminal platform (OMTP) standard interface, or a cellular telecommunications industry association of the USA (CTIA) standard interface.

The buttons 190 include a power button, a volume button, etc. Key 190 may be a mechanical key. It can also be a touch button. The electronic device 100 may receive key inputs and generate key signal inputs related to user settings and function control of the electronic device 100 .

The motor 191 can generate vibration prompts. The motor 191 can be used for vibration prompts for incoming calls and can also be used for touch vibration feedback. For example, touch operations for different applications (such as taking pictures, audio playback, etc.) can correspond to different vibration feedback effects. The motor 191 can also respond to different vibration feedback effects for touch operations in different areas of the display screen 194 . Different application scenarios (such as time reminders, receiving information, alarm clocks, games, etc.) can also correspond to different vibration feedback effects. The touch vibration feedback effect can also be customized.

The indicator 192 may be an indicator light, which may be used to indicate charging status, power changes, or may be used to indicate messages, missed calls, notifications, etc.

The SIM card interface 195 is used to connect a SIM card. The SIM card can be connected to or separated from the electronic device 100 by inserting it into the SIM card interface 195 or pulling it out from the SIM card interface 195 . The electronic device 100 can support 1 or N SIM card interfaces, where N is a positive integer greater than 1. SIM card interface 195 can support Nano SIM card, Micro SIM card, SIM card, etc. Multiple cards can be inserted into the same SIM card interface 195 at the same time. The types of the plurality of cards may be the same or different. The SIM card interface 195 is also compatible with different types of SIM cards. The SIM card interface 195 is also compatible with external memory cards. The electronic device 100 interacts with the network through the SIM card to implement functions such as calls and data communications. In some embodiments, the electronic device 100 uses an eSIM, that is, an embedded SIM card. The eSIM card can be embedded in the electronic device 100 and cannot be separated from the electronic device 100 .

The external memory interface 120 can be used to connect an external memory card, such as a Micro SD card, to expand the storage capacity of the electronic device 100. The external memory card communicates with the CPU 110 through the external memory interface 120 to implement the data storage function. Such as saving music, videos, etc. files in external memory card.

Internal memory 121 may be used to store computer executable program code, which includes instructions. The CPU 110 executes instructions stored in the internal memory 121 to execute various functional applications and data processing of the electronic device 100 . The internal memory 121 may include a program storage area and a data storage area. Among them, the stored program area can store an operating system, at least one application program required for a function (such as a sound playback function, an image playback function, etc.). The storage data area may store data created during use of the electronic device 100 (such as audio data, phone book, etc.). In addition, the internal memory 121 may include high-speed random access memory, and may also include non-volatile memory, such as at least one disk storage device, flash memory device, universal flash storage (UFS), etc.

Refer to Figure 2, which is a schematic diagram of the software structure of the electronic device according to the embodiment of the present application. The operating system in the electronic device can be Android, Microsoft Window System (Windows), Apple Mobile Operating System (iOS) or Harmony OS, etc. Here, the operating system of the electronic device is the Hongmeng system as an example.

In some embodiments, the Hongmeng system can be divided into four layers, including the kernel layer, system service layer, framework layer and application layer. The layers communicate through software interfaces.

As shown in Figure 2, the kernel layer includes the kernel abstract layer (KAL) and driver subsystem. KAL includes multiple kernels, such as the Linux Kernel, the lightweight IoT system kernel LiteOS, etc. The driver subsystem can include the Hardware Driver Foundation (HDF). The hardware driver framework can provide unified peripheral access capabilities and driver development and management framework. The multi-core kernel layer can select the corresponding core for processing according to the needs of the system.

The system service layer is the core capability set of Hongmeng system. The system service layer provides services to applications through the framework layer. This layer may include a system basic capability subsystem set, a basic software service subsystem set, an enhanced software service subsystem set, and a hardware service subsystem set.

The basic capability subsystem set of the system provides basic capabilities for the operation, scheduling, migration and other operations of distributed applications on Hongmeng system devices. Can include distributed soft bus, distributed data management, distributed task scheduling, Ark multi-language runtime, common basic library, multi-mode input, graphics, security, artificial intelligence (AI), user program framework and other subsystems . Among them, the Ark multi-language runtime provides C or C++ or JavaScript (JS) multi-language runtime and basic system class libraries. It can also be used for static Java programs using the Ark compiler (that is, using Java in the application or framework layer). Language development part) provides the runtime.

The basic software service subsystem set provides public and general software services for Hongmeng system. Can include event notification, telephone, multimedia, Design For X (Design For X, DFX), MSDP&DV and other subsystems.

The enhanced software service subsystem set provides Hongmeng system with differentiated capability-enhanced software services for different devices. It can include smart screen proprietary business, wearable proprietary business, and Internet of Things (IoT) proprietary business subsystems.

The hardware service subsystem set provides hardware services for Hongmeng system. It can include subsystems such as location services, biometric identification, wearable proprietary hardware services, and IoT proprietary hardware services.

The framework layer provides multi-language user program frameworks and Ability frameworks such as Java, C, C++, and JS for Hongmeng system application development. Two user interface (UI) frameworks (including Java UI for Java language framework, JS UI framework suitable for JS language), and multi-language framework application program interface (Application Programming Interface, API) that is open to the public for various software and hardware services. Depending on the degree of component tailoring of the system, the APIs supported by Hongmeng system devices will also be different.

The application layer includes system applications and third-party applications (or extended applications). System applications may include applications installed by default on the desktop, control bar, settings, phone and other electronic devices. Extended applications can be non-essential applications developed and designed by manufacturers of electronic devices, such as electronic device manager, replacement migration, notes, weather and other applications. Third-party non-system applications can be developed by other manufacturers, but can run applications in the Hongmeng system, such as games, navigation, social networking or shopping applications.

Provides the ability to run tasks in the background and a unified data access abstraction. PA mainly provides support for FA, such as providing computing power as a background service, or providing data access capabilities as a data warehouse. Applications developed based on FA or PA can implement specific business functions, support cross-device scheduling and distribution, and provide users with a consistent and efficient application experience.

Multiple electronic devices running the Hongmeng system can achieve hardware mutual assistance and resource sharing through distributed soft buses, distributed device virtualization, distributed data management and distributed task scheduling.

The operating system is generally divided into two states: the kernel running state (hereinafter referred to as the kernel state) and the user running state (hereinafter referred to as the user state). Correspondingly, the operating system has operating system applications executed in the kernel mode and applications executed in the user mode. Ordinary applications and user-mode applications have lower execution permissions than kernel-mode applications. Generally, kernel-mode applications can access all hardware resources in electronic devices, such as internal memory, external memory cards, network cards, etc. User-mode applications have limited access to hardware resources, so user-mode applications cannot perform data transmission, data storage, or data reading. When waiting for an operation, you need to fall into the kernel state (that is, call the computer program in the kernel) through system calls and generate interrupt signals, so as to implement the corresponding operations.

In general, an MMU (memory management unit) 111 (as shown in FIG. 1 ) is provided between the CPU 110 of an electronic device, the internal memory 121 and the external memory interface 120 bus. When the CPU 110 needs to access the internal memory 121 (ie, the memory mentioned above), a virtual address can be provided, and the MMU can map the virtual address to a physical address, thereby enabling the CPU 110 to read data from the internal memory 121 based on the physical address. Or write data into the internal memory 121.

The virtual address is an address in the virtual address space that can be recognized or generated by the operating system in the electronic device, and its size range can be determined by the number of bits of the operating system running in the CPU 110 . For example, if the operating system running in CPU110 is 32-bit, the virtual address is also 32-bit, and its address range can be 0-0xFFFFFFFF (4GB); if the operating system running in CPU110 is 64-bit, the virtual address is also 64-bit, Its address space can be 0-0xFFFFFFFFFFFFFFFF (16EB).

Among them, the virtual address can be divided into multiple virtual address spaces according to actual needs, such as user-mode virtual address space and kernel-mode virtual address space. User-mode virtual address space can be accessed by user-mode applications (such as reading, writing, opening, closing, or drawing) and kernel-mode applications (such as process management, storage management, file management, or device management). Kernel-mode virtual address space can only be accessed by Kernel-mode applications access it at runtime. For example, the address range of the virtual address space is [0x00000000, 0xFFFFFFFF], then the address range of the user-mode virtual address space can be [0x00000000, 0x80000000) (that is, the 0th GB-2nd GB), and the address range of the kernel-mode virtual address space can be [0x80000000, 0xFFFFFFFF] (i.e. 2GB-4th GB).

The physical address may be an address in the address space actually owned by a hardware storage device such as the internal memory 121. The address space of the physical address can be smaller than the address space of the virtual address. For example, when the size of the address space of the virtual address can be 4GB, the size of the address space of the physical address can be 256MB.

As shown in Figure 3, during the process of data transmission between the sending end application and the receiving end application based on traditional network communication technology, the sending end CPU needs to copy the application's to-be-sent data to the cache space of the TCP/IP protocol stack in kernel mode. , and then copied to the cache space of the driver layer and the cache space of the traditional network card in sequence. Finally, the traditional network card at the sending end can send the data to be sent to the traditional network card at the receiving end based on the pre-established communication link. In this process, a large amount of space is required. CPU resources are required to copy the data, and conversion between user mode and kernel mode is required, resulting in a long delay during data transmission.

RDMA (Remote Direct Memory Access) technology is a network data transmission technology based on the RDMA network card that allows the sender to directly transmit data to the receiver's memory through the network. Compared with traditional network communication technology, RDMA It can bypass the protocol stack of the operating system and reduce the number of data copies, thereby occupying less CPU resources and reducing latency. Among them, both the receiving end and the sending end can be electronic devices such as servers, clients, and mobile phone terminals. The electronic devices at least include an Ethernet network card (i.e., RDMA network card) with remote direct memory access function and a CPU. The RDMA network card is mainly used in Data reading operations and/or data writing operations are performed during data transmission.

The MMU usually divides the virtual address space and the physical address space into multiple virtual pages with a fixed space size and multiple physical pages with a fixed space size respectively based on the paging management mechanism for management. The number of virtual addresses in the virtual page is related to the number of physical pages. The number of physical addresses in is the same. For example, the fixed space size can be 4kb, 16kb, 64kb, etc. Each virtual page is provided with a corresponding virtual page number. Each virtual page includes at least one virtual address. Each virtual address corresponds to the virtual page number of the virtual page to which it belongs and the page offset in the virtual page to which it belongs. , the so-called intra-page offset is the displacement amount of the virtual address relative to the virtual base address in the virtual page to which it belongs. Each physical page is provided with a corresponding physical page number. Each physical page includes at least one physical address, and each physical address corresponds to the physical page number of the physical page to which it belongs.

As shown in Figure 4, before applications in electronic equipment transmit data based on RDMA communication, they need to apply for a physical address space that can be used to store data. Specifically, after receiving the application's memory allocation request, the processor determines the virtual address space corresponding to the application. The MMU can convert and map the application's virtual address space to the physical address space, and create and fill in the application's virtual address space in the local memory. A page table that maps virtual addresses in the address space to physical addresses in the corresponding physical address space. After the application is successful, the application can send a registration request to the RDMA network card, causing the RDMA network card to save the page table used to indicate the mapping relationship between the application's virtual address space and physical address space to the network card memory of the RDMA network card.

Refer to Figure 5, which is a schematic diagram of the mapping relationship between virtual address space, page table and physical address space. The page table includes multiple page table entries. The number of page table entries is the same as the number of virtual pages in the virtual address space. Each page table entry is used to indicate the virtual page number and physical address of the virtual page in the virtual address space. The mapping relationship between physical page numbers in the space. As shown in Figure 5, assume that there are 6 virtual pages in the virtual address space mapped to the page numbers of 6 physical pages in the physical address space. The virtual page numbers of the 6 virtual pages are 0 to 5. The page table records The physical page number of the physical page that corresponds to the virtual page number of each virtual page.

After the RDMA network card receives the virtual address sent by the application sender, it can determine the physical page number of the physical page corresponding to the virtual page number of the virtual page to which the virtual address belongs through the page table. According to the physical base address and virtual address in the physical page The physical address corresponding to the virtual address can be determined by the intra-page offset in the virtual page to which it belongs.

Existing RDMA communication technology generally includes two communication operation modes. The first is a message-type operation mode, such as send operation and receive operation, and the second is a memory-type operation mode, such as RDMA Read operation and RDMA Write operation. base In the message-type operation mode, the second electronic device triggers the receive operation command to determine the physical address used to receive data. The application of the first electronic device can cache the data to be sent in the pre-allocated physical address space and send it to the first electronic device. The first RDMA network card in the device sends a send operation command, and the first RDMA network card caches the data to be sent into the RDMA memory and encapsulates the data to be sent to generate a data packet, so as to communicate through the pre-created communication between the first electronic device and the second electronic device. The link is transmitted to the second RDMA network card in the second electronic device. After decapsulating the data packet, the second RDMA network card can store the data in the data packet into a physical address used to receive the data.

Based on the memory-type operation mode, the second electronic device can provide the first electronic device with the accessible virtual address space and its access rights in the second electronic device through the pre-created communication link. The application in the first electronic device can send an RDMA Read operation instruction to the first RDMA network card in the first electronic device. The RDMA Read operation instruction includes the virtual address where the data to be read is stored in the second electronic device and the username used in the first electronic device. Based on the virtual address of the received data, the first RDMA network card can transmit the Read request packet corresponding to the RDMA Read operation instruction to the second RDMA network card in the second electronic device through the communication link. The second RDMA network card can transmit the read request packet corresponding to the RDMA Read operation instruction to the second RDMA network card in the second electronic device according to the pre-stored page. The table determines the physical address corresponding to the virtual address that stores the data to be read, and encapsulates the data to be read into a data packet and transmits it back to the first RDMA network card through the communication link. The first RDMA network card can query the pre-stored page table to The data to be read is written in the physical address in the first electronic device corresponding to the virtual address used to receive the data.

The application in the first electronic device can send an RDMA Write operation instruction to the first RDMA network card in the first electronic device. The Write operation instruction includes the virtual address and data size used to receive data in the second electronic device. The first RDMA network card is from Obtain the data to be sent from the physical address space corresponding to the application, cache the data to be sent and the virtual address used to receive the data in the second electronic device to the RDMA memory of the first RDMA network card, and then encapsulate it into a data packet to create a data package in advance. The communication link is transmitted to the second RDMA network card. The second RDMA network card can decapsulate the data packet and store the data in the data packet to the second electronic device for receiving data by querying the pre-stored page table. The virtual address corresponds to the physical address.

It should be noted that before data transmission, the first electronic device and the second electronic device can establish a communication link through a Socket connection or a Connection Manager (Connection Manager, CM) connection, and exchange some RDMA communication data through the communication link. For example, the first electronic device can obtain the virtual address of the virtual address space in the second electronic device, and the second electronic device can obtain the virtual address of the virtual address space in the first electronic device.

In the process of data transmission based on the above-mentioned existing RDMA communication technology, the physical address in the physical address space corresponding to the application, and the mapping relationship between the virtual address space and the physical address space of the application are fixed. When multiple applications are running simultaneously on an electronic device, each application needs to be allocated a corresponding physical address space. However, the storage resources of the memory are limited. During the data transmission process, each application can only access the corresponding pre-allocated address space. physical address space, and the size of the physical address space corresponding to each application cannot be changed according to the actual required memory space. As a result, the physical address space of some applications may not meet the actual memory requirements of the application, while the physical address space of some applications may not be able to meet the actual memory requirements of the application. Greater than the actual memory requirements of the application, leaving part of the physical address space unused.

There are methods in the existing technology for expanding the physical address space corresponding to the application. One method is: when the application detects that the current communication memory cannot meet the memory space required for communication data in the current business scenario, it can choose to disconnect Mapping relationship with the current physical address space, re-registering a new physical address space to match the memory space required for communication data in the current business scenario. Although this method can dynamically adjust the memory space corresponding to the application according to actual business needs, However, the method of disconnecting the mapping relationship with the current physical address space and re-registering it is cumbersome and time-consuming. Another method is: the application can register multiple physical address spaces with different fixed space sizes to the RDMA system based on the memory space required for communication data in different business scenarios, but only use one of them during each data transmission. One physical address space, other unused physical address spaces are idle, wasting more memory resources.

Therefore, in order to solve the problem of unreasonable allocation of memory resources in the existing RDMA technology, this application provides a memory management method, electronic device, chip system and readable storage medium, which dynamically adjusts the corresponding data according to the amount of data to be stored in the target application. An effective memory space with data read and write permissions in the physical address space. When the effective memory space is greater than the amount of data to be stored, the unused first physical address unit in the effective memory space is released. When the effective memory space is less than the amount of data to be stored, When measuring, the effective memory space is expanded so that the effective memory space meets the memory requirements of the application in different business scenarios, thereby allocating corresponding memory resources to the application reasonably.

The technical solution of the present application will be described in detail below with specific examples. The following specific embodiments can be mutually In combination, the same or similar concepts or processes may not be described again in some embodiments.

Figure 6 is a schematic flow chart of a memory management method provided by an embodiment of the present application. It mainly involves the target application in the electronic device performing data transmission based on RDMA technology. The electronic device dynamically adjusts the physical address space associated with the target application. of effective memory space. This memory management method can be applied to electronic devices with RDMA communication functions. The electronic devices include internal memory, processors and RDMA network cards. The following is an exemplary description of the memory management method provided by this application with reference to FIG. 6 .

S601. Determine the amount of data to be stored by the target application and the effective memory space in the physical address space associated with the target application. The effective memory space includes memory spaces indicated by N first physical address units, and each first physical address unit includes at least A physical address with data read and write permissions, N is an integer greater than or equal to 0.

It should be noted that after the target application installed in the electronic device registers the corresponding physical address space, the target application will generate data to be stored during subsequent data transmission with applications installed on other electronic devices based on RDMA communication operations. . During the first data transmission process after the target application registers the physical address space, the electronic device can determine whether it is necessary to pre-register the target application based on the relative size between the effective memory space in the pre-registered physical address space and the amount of data to be stored. The registered effective memory space is adjusted so that the data to be stored by the target application can be cached in the effective memory space associated with it.

Among them, the effective memory space is the memory space indicated by all the first physical address units in the physical address space. The memory space of a first physical address unit corresponding to the internal memory is provided with both data writing permission and data reading permission. A physical page, a first physical address unit includes one or more physical addresses with data writing permissions and data reading permissions.

In one example, if there is no valid memory space in the physical address space pre-registered by the target application, then when performing the first data transmission based on the physical address space, it is necessary to allocate valid memory space to the target application, that is, to allocate valid memory space to the target application. The effective memory space in the physical address space is expanded so that the effective memory space meets the memory space required for the first data transfer. For example, the physical address space pre-registered by the target application may include one or more second physical address units with a fixed size, wherein the second physical address unit is configured not to have data writing permission but to have data reading permission. of the physical page, the second physical address unit includes at least one physical address that does not have data writing permission but has data reading permission. For example, as shown in Figure 7, the number of second physical address units in the pre-registered physical address space is 1, and the space size of the second physical address unit can be 4kb, then all the virtual address units in the target application can be The virtual address units are mapped to the second physical address unit.

In another example, there is an effective memory space with a fixed space size in the physical address space pre-registered by the target application. For example, the physical address space pre-registered by the target application includes a first physical address unit, and the first physical address unit Corresponds to a virtual address unit in the virtual address space of the target application; or the physical address unit pre-registered by the target application includes a first physical address unit and a second physical address unit, where the first physical address unit is the same as the target application corresponds to one of the virtual address units in the virtual address space, and the remaining virtual address units correspond to the second physical address unit. When performing the first data transmission based on this physical address space, it is necessary to determine whether the pre-registered effective memory space needs to be adjusted based on the relative size between the memory space required for the first data transmission and the pre-registered effective memory space. That is, if the pre-registered effective memory space is If the registered effective memory space is larger than the memory space required for the first data transmission, the first physical address unit used in the pre-registered effective memory space will be recycled and expanded. If the pre-registered effective memory space is smaller than the memory space required for the first data transmission, memory space, expand the pre-registered effective memory space.

It should also be noted that during the subsequent data transmission process after the first data transmission of the electronic device, the physical address space associated with the target application may be the effective memory space in the physical address space adjusted during the last data transmission process. Effective memory space. For example, if the electronic device adjusts the pre-registered effective memory space for the first time during the first data transmission, then in the second data transmission after the first data transmission, the physical address space associated with the target application will The effective memory space is the effective memory space after the first adjustment. The electronic device needs to determine whether the effective memory space after the first adjustment meets the memory space required for the second data transmission, and then determine whether the effective memory space needs to be adjusted again.

In this embodiment of the present application, if the target application is an application installed on the first electronic device, the data to be stored by the target application may be the data of the target application sent by the first electronic device to the application on the second electronic device based on the RDMA communication operation. Data, for example, the RDMA communication operation can be a send operation or an RDMA Write operation. The data to be stored may also be data of an application on the second electronic device received by the first electronic device based on the RDMA communication operation. For example, the RDMA communication operation may Think of receive operation or RDMA Read operation.

S602, if the data amount is larger than the effective memory space, expand the effective memory space according to the first difference between the data amount and the effective memory space.

If the amount of data to be stored by the target application is greater than the effective memory space of the target application, the effective memory space can be expanded according to the first difference between the data amount and the effective memory space. Specifically, m first physical address units may be allocated to the target application from the memory space in the internal memory that is not registered by other applications based on the first difference value, where m is an integer greater than or equal to 1, and m first physical addresses The memory space of the unit is greater than or equal to the first difference value, 1≤m+N<M. That is to say, the expanded effective memory space meets the memory space required by the amount of data to be stored by the target application. In addition, after determining the number of extended first physical address units based on the first difference value, the mapping table between the virtual address space and the physical address space of the target application needs to be updated to compare the M virtual address units with the second The m virtual address units corresponding to the physical address units are remapped to m first physical address units.

For example, refer to the schematic diagram of the mapping relationship between the virtual address space and the physical address space of the target application shown in FIG. 8 . Assume that the amount of data to be stored in the target application is 15kb. The physical address space of the target application includes 3 first physical address units. Each first physical address unit corresponds to a virtual address unit of the target application. In the virtual address space of the target application The other M-3 virtual address units all correspond to the second physical address unit, and the sizes of a first physical address unit, a virtual address unit and a second physical address unit are all 4kb, that is, the effective memory space of the target application is 12kb. , then it is determined that the amount of data stored by the target application is greater than the effective memory space and the first difference between the two is 3kb. Correspondingly, as shown in Figure 8, the effective memory space of the target application can be extended by 1 first physical address unit, that is, allocate a physical page (i.e. page in Figure 8) to the target application from a memory space that has not been registered by other applications, and set the physical page as the first page with data writing permissions and data reading permissions. The physical address unit makes the effective memory space of the target application expand to 16kb, thereby meeting the memory space required for the amount of data to be stored by the target application. Accordingly, the mapping table needs to be updated to remap one of the other M-3 virtual address units corresponding to the second physical address unit in the virtual address space of the target application to the extended first physical address. unit.

S603, if the data amount is less than the effective memory space, reduce the effective memory space according to the second difference between the effective memory space and the data amount.

If the amount of data to be stored by the target application is less than the effective memory space of the target application, the effective memory space can be reduced according to the second difference between the effective memory space and the amount of data. Specifically, assuming that the effective memory space includes N first physical address units corresponding to N virtual address units among the M virtual address units of the target application, then when the second difference is greater than or equal to the preset threshold When the preset threshold is greater than 0, the effective memory space is reduced by n first physical address units, 1≤n<N, and the memory space of n first physical address units is less than or equal to the second difference. In addition, after determining the number of first physical address units that need to be reduced based on the second difference, the mapping table between the virtual address space and the physical address space of the target application needs to be updated, and n of the N virtual address units are The virtual address unit is remapped to a second physical address unit.

For example, the preset threshold may be the size of the memory space indicated by a first physical address unit.

For example, refer to the schematic diagram of the mapping relationship between the virtual address space and the physical address space of the application shown in FIG. 9 . Assume that the sizes of the first physical address unit, virtual address unit and second physical address unit are all 4kb, the preset threshold is 4kb, the amount of data to be stored by the target application is 8kb, and the physical address space of the target application includes 3 first The physical address unit, that is, the effective memory space of the target application is 12kb. Each first physical address unit corresponds to a virtual address unit of the target application. The other M-3 virtual address units in the virtual address space of the target application are all the same as the second If the physical address unit corresponds, the effective memory space is greater than the amount of data stored by the target application, and the second difference between the two is 4kb, which is equal to the preset threshold. As shown in Figure 9, the effective memory space of the target application can be reduced by one first physical address unit, and the first physical address unit can be released, that is, the relationship between the first physical address unit and the virtual address unit in the target application can be released. Mapping relationship, the released first physical address unit is a physical page without read and write permissions. In addition, the mapping table needs to be updated to remap the virtual address unit corresponding to the first physical address unit to the target application. The second physical memory unit.

It can be understood that if the amount of data to be stored by the target application is less than the effective memory space of the target application, and the second difference between the effective memory space and the amount of data is greater than 0 and less than the preset threshold, there is no need to change the target application. The size of the effective memory space.

In the embodiment of the present application, based on different types of RDMA communication operations, the electronic device can use two different triggering methods, namely the active triggering method and the passive triggering method, so that the processor in the electronic device can be stored according to the installed target application. The effective memory space in the physical address space associated with the target application is expanded or reduced based on the difference between the amount of data and the effective memory space of the target application.

In one example, it is assumed that the first electronic device includes a first processor and a first RDMA network card, the first target application is installed in the first electronic device, the second electronic device includes a second processor and a second RDMA network card, and the first electronic device includes a second processor and a second RDMA network card. The second target application is installed in the second electronic device, and the first electronic device and the second electronic device can perform data transmission based on RDMA technology. These two different triggering methods will be exemplified below with reference to the interaction diagrams shown in Figures 10 and 11 .

Method 1, active triggering method.

In this method one, when the first target application in the first electronic device writes the data to be stored into the physical address space associated with the first target application, the first processor can be actively triggered to cause the first processor to write the data to be stored in the physical address space associated with the first target application. The effective memory space in the physical address space is expanded or reduced. Correspondingly, the data to be stored by the first target application is: based on the first type of RDMA communication operation, the first electronic device sends the data of the first target application to the second RDMA network card of the second electronic device through the first RDMA network card. Example Specifically, the first type of RDMA communication operation can be a send operation or an RDMA Write operation.

As an example and not a limitation, refer to the interaction diagram between the target application, the processor and the RDMA network card shown in Figure 10, in which the first processor includes a virtual memory management unit and a physical memory management unit, and the virtual memory management unit runs in user mode. , used to determine the virtual address space corresponding to the application when receiving the application's memory allocation request. The physical memory management unit runs in the kernel mode and is used to allocate the corresponding physical address space to the application and dynamically expand and reduce the effective memory space in the application's physical address space through active triggering. The virtual memory management unit can fall into the kernel state through system calls, allowing the physical memory management module to allocate the corresponding physical address space to the application.

As shown in Figure 10, the active triggering method includes the following steps:

S1001. The first target application sends a query request to the physical memory management unit through the virtual memory management unit.

Before storing the data to be stored in the corresponding physical address space, the first target application may send a query request to the physical memory management unit through the virtual memory management unit. Among them, the first target application can send a query request to the virtual memory management unit by calling the query API (Application Programming Interface, application program interface) interface, and the virtual memory management unit can make the first target application call the function through the ioctl (input/output control) system The physical memory management unit trapped in the kernel state causes the physical memory management unit to receive the query request sent by the first target application.

S1002. The physical memory management unit determines the effective memory space in the physical address space associated with the first target application.

After receiving the query request, the physical memory management unit can determine the effective memory space in the physical address space of the first target application according to the mapping table, and transmit the size of the effective memory space to the first target application through the virtual memory management unit.

Wherein, when the first target application performs the first data transmission, the mapping table is an initial mapping table pre-registered by the first target application. In the subsequent data transmission process, the mapping table may be an updated mapping table corresponding to the adjusted physical address space during the last data transmission.

S1003. The first target application determines whether the amount of data to be stored is greater than the effective memory space.

The first target application can determine whether the amount of data to be stored is greater than the effective memory space based on the obtained size of the effective memory space. Specifically, if the amount of data to be stored is less than the effective memory space, a reduction request can be sent to the physical memory management unit through the virtual memory management unit based on the memory reduction API interface. The reduction request can include the amount of data between the effective memory space and the amount of data to be stored. The second difference; if the amount of data to be stored is greater than the effective memory space, an expansion request can be sent to the physical memory management unit through the virtual memory management unit based on the memory expansion API interface. The expansion request can include the amount of data to be stored and the effective memory. The first difference between the spaces is used to actively trigger the physical memory management unit to expand or reduce the effective memory space.

S1004. The physical memory management unit expands the effective memory space according to the first difference between the amount of data to be stored and the effective memory space, or expands the effective memory space according to the second difference between the effective memory space and the amount of data to be stored. Memory space is reduced.

When receiving the expansion request and the reduction request, the physical memory management unit can expand and reduce the effective memory space of the first target application according to the method provided in steps S602 and S603 in the above embodiment.

Optionally, after the physical memory management unit expands and reduces the effective memory space of the first target application, the active triggering method also includes the following steps:

S1005, the physical memory management unit can send a signal of successful expansion or successful reduction to the first target application through the virtual memory management unit, so that the first target application can store the data to be stored in the updated physical address space.

S1006, the physical memory management unit can also send an instruction to update the mapping table to the first RDMA network card, so that the first RDMA network card The network card stores the updated mapping table corresponding to the first target application into the network card memory.

Method two, passive triggering method.

In the second method, when the first RDMA network card of the first electronic device receives the data of the second target application sent by the second RDMA network card of the second electronic device, the first processor may receive the data of the second target application from the first RDMA network card. After interrupting the information, the effective memory space in the first physical address space is expanded or reduced through passive triggering. Correspondingly, the data to be stored by the first target application is: Based on the second type of RDMA communication operation, the first electronic device The data of the second target application sent by the second RDMA network card of the second electronic device is received through the first RDMA network card. For example, the second type of RDMA communication operation may be a receive operation or an RDMA Read operation.

It should be noted that in the second method, when the first electronic device performs the second type of RDMA communication operation, the first target application can send an RDMA communication operation instruction to the first RDMA network card on the first electronic device, where the RDMA communication The operation instructions include a virtual base address corresponding to the physical base address in the physical address space used by the first electronic device to store the obtained data. The first RDMA network card obtains the data sent by the second target application on the second electronic device. Cached to the first RDMA network card, the first RDMA network card can determine the physical base address corresponding to the virtual base address according to the pre-stored mapping table between the virtual address space and the physical address space of the first target application, and use the physical base address as The starting address stores the obtained data into an effective memory space in the physical address space of the first target application. Therefore, after the first RDMA network card obtains the data of the second target application sent by the second RDMA network card, it needs to confirm whether the memory address space of the first target application meets the amount of data to be stored. If it does not meet the requirement or there is remaining memory address space, time, thereby adjusting the memory address space of the first target application through passive triggering.

As an example and not a limitation, with reference to the interaction diagram between the first electronic device and the second electronic device shown in Figure 11, the passive triggering method includes the following steps:

S1101. The first RDMA network card determines the amount of data to be stored by the first target application.

The amount of data to be stored by the first target application is the data of the second target application sent by the first RDMA network card of the first electronic device and received by the second RDMA network card of the second electronic device.

S1102. The first RDMA network card determines whether the amount of data to be stored by the first target application is greater than the effective memory space associated with the first target application.

Specifically, the first RDMA network card can determine the effective memory space in the physical address space associated with the first target application based on the pre-stored mapping table corresponding to the first target application, and then determine whether the amount of data to be stored is greater than the effective memory. space. If the amount of data to be stored is greater than the effective memory space, the first RDMA network card can send first interrupt information to the virtual memory management unit in the first processor. The first interrupt information can include the amount of data to be stored and the effective memory space. The first difference; if the amount of data to be stored in the first target application is less than the effective memory space, the RDMA network card can send the second interrupt information to the processor, and the second interrupt information can include the difference between the effective memory space and the amount of data to be stored. the second difference, thereby triggering the physical memory management unit to adjust the effective memory space of the first target application according to the first difference or the second difference through passive triggering.

S1103. The physical memory management unit expands the effective memory space according to the first difference between the amount of data to be stored and the effective memory space, or expands the effective memory space according to the second difference between the effective memory space and the amount of data to be stored. Memory space is reduced.

The physical memory management unit may, after receiving the first interrupt signal sent by the first RDMA network card, expand the effective memory space according to the first difference value, or may, after receiving the second interrupt signal sent by the first RDMA network card, The effective memory space is reduced according to the second difference value.

For the method of expanding and reducing the effective memory space of the first target application, reference may be made to the specific description of steps S602 and S603 in the above method embodiment, which will not be described again here.

Optionally, after the physical memory management unit expands and reduces the effective memory space of the first target application, the passive triggering method also includes the following steps:

S1104. The physical memory management unit may send an instruction to update the mapping table to the first RDMA network card, so that the first RDMA network card stores the updated mapping table corresponding to the first target application into the network card memory.

Based on the above embodiments, this application also provides a physical address space registration method. Referring below to the schematic diagram of the interaction between the target application, the processor and the RDMA network card in the electronic device shown in Figure 12, the specific process of registering the physical address space of the target application installed in the electronic device before data transmission based on RDMA communication technology is exemplified. instruction of.

As shown in Figure 12, the registration method of the physical address space of the target application includes the following steps:

S1201. The target application sends a memory allocation request to the virtual memory management unit. For example, the target application can call the memory Apply for an API interface to send a memory allocation request to the virtual memory management unit.

S1202: After receiving the memory allocation request sent by the target application, the virtual memory management unit allocates the corresponding virtual address space to the target application.

It should be noted that, in one example, when the virtual memory management unit receives a memory allocation request sent by each target application in the electronic device, it can allocate a virtual address space of the same size to each target application, and each The virtual addresses in the target application's virtual address space are the same. In another example, the target application can determine the size of the virtual address space required at runtime, and then the target application can send a memory allocation request containing the required virtual address space size information to the virtual memory management unit. According to the virtual address space size required by each target application, a virtual address space having the same space size as the virtual address space size required by each application may be allocated from the user state virtual address space of the first electronic device.

In this embodiment of the present application, the virtual address space of the target application includes at least one virtual address unit. The virtual address unit is a virtual page with a preset space size that divides the virtual address space based on the paging management mechanism. Each virtual address unit includes at least a virtual address.

S1203, the virtual memory management unit can fall into the kernel state through system calls, causing the physical memory management unit to allocate physical address space for the target application.

In one example, the physical address space allocated by the memory management unit for the target application may include a second physical address unit, and the second physical address unit is a physical page that is set to have data reading permission but not data writing permission. Each second physical address unit includes one or more physical addresses with data writing authority, and the number of physical addresses in the second physical address unit is the same as the number of virtual addresses in the virtual address unit.

For example, if the size of the physical address space in the internal memory of the electronic device is 2GB and the size of the virtual address space of the target application is 4GB, the physical address space can be divided into 524288 physical pages of 4kb based on the paging management mechanism. , Correspondingly, the virtual address space can be divided into 1049576 virtual pages with a size of 4kb. Referring to the schematic diagram of the mapping relationship between the virtual address space of the target application and the second physical address unit shown in FIG. 7 , all virtual address units in the virtual address space of the target application can be mapped to one second physical address unit.

In another example, the physical address space allocated by the memory management unit for the target application may include a second physical address unit and an effective memory space with a smaller space. Among them, the virtual address space of the target application includes M virtual address units, the effective memory space includes N first physical address units, N is an integer greater than or equal to 0, and among the N first physical address units and the M virtual address units There is a one-to-one correspondence between N virtual address units, where the first physical address unit is a physical page set to have data read permission and data write permission. Each first physical address unit includes at least one with data read and write permission. The physical address, that is, the RDMA network card can write data to the physical address in the first physical address unit, and can also read the pre-cached data in the physical address. For example, the number of the first physical address unit and the second physical address unit in the allocated physical address space may be 1, and the first physical address unit may be mapped to one of the virtual address units in the virtual address space of the target application. address unit, and map other virtual address units in the virtual address space of the target application to the second physical address unit. Optionally, the number of physical addresses in the first physical address unit is the same as the number of physical addresses in the second physical address unit.

It can be understood that when the physical address space of the target application includes a second physical address unit that does not have data write permission, the processor and the RDMA network card can determine whether the physical address in the physical address space corresponding to the target application is in the mapping table. Have data read and write permissions, determine the size of the effective memory space in the physical address space of the target application, so that in the subsequent data transmission process, it can be compared based on the difference between the amount of data to be stored by the target application and the size of the effective memory space. Memory space is expanded and reclaimed.

Further, after the physical memory management unit allocates the corresponding physical address space for the target application, it needs to create an initial mapping table between the virtual address space and the physical address space for the target application, and store the initial mapping table in the internal memory. The initial mapping table is used to indicate the mapping relationship between the page number of each virtual address unit in the virtual address space and the page number of the physical address unit in the physical address space, as well as the set read and write permissions of the physical address unit.

S1204: After the physical memory management unit allocates the corresponding physical address space to the target application, the physical memory management unit can send a successful allocation signal to the target application through the virtual memory management unit.

S1205: After receiving the successful allocation signal, the target application sends a registration request to the RDMA network card.

Among them, the registration request contains the identification of the target application.

S1206: After receiving the registration request, the RDMA network card registers the initial mapping table between the physical address space and the virtual address space of the target application.

Specifically, the RDMA network card can store the initial mapping in the memory space into the network card memory according to the identification of the target application, so that the RDMA network card has the permission to read and write the data of the physical address in the physical address space of the target application.

Based on the memory management method provided by this application, the electronic device can pre-allocate the second physical address unit without data writing permission to the target application, so that the subsequent data transmission process based on RDMA communication technology can be based on the amount of data to be stored by the target application. The difference between the effective memory space with data read and write permissions in the physical address space corresponding to the target application determines whether the effective memory space meets the memory space required for the amount of data to be stored by the application, thereby dynamically adjusting the effective memory space of the target application. Memory space. If the effective memory space does not meet the memory space required by the amount of data to be stored, the effective memory space can be expanded according to the first difference between the data amount and the effective memory space, so that the expanded effective memory space Meet the memory space required by the application. If the effective memory space meets the memory space required by the amount of data to be stored, the effective memory space can be reduced according to the second difference between the effective memory space and the amount of data, thereby releasing some Use unoccupied effective memory space to avoid wasting memory resources, thereby rationally allocating memory resources to target applications and improving memory resource utilization.

Based on the memory management method provided by the above embodiment, the embodiment of this application also provides the following content:

An embodiment of the present application provides a computer program product. The program product includes a program. When the program is run by an electronic device, the electronic device causes the memory management method shown in the above embodiments.

Embodiments of the present application provide a computer-readable storage medium. The computer-readable storage medium stores a computer program. When the computer program is executed by a processor, the memory management method shown in the above embodiments is implemented.

An embodiment of the present application provides a chip, which includes a memory and a processor. The processor executes a computer program stored in the memory to control the above-mentioned electronic device to execute the memory management method shown in the above-mentioned embodiments.

If the above-mentioned integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, this application can implement all or part of the processes in the methods of the above embodiments by instructing relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium. The computer program When executed by a processor, the steps of each of the above method embodiments may be implemented. Wherein, the computer program includes computer program code, which may be in the form of source code, object code, executable file or some intermediate form. The computer-readable storage medium may at least include: any entity or device capable of carrying computer program code to the camera device/terminal device, recording media, computer memory, read-only memory (ROM), random access Memory (random access memory, RAM), electrical carrier signals, telecommunications signals, and software distribution media. For example, U disk, mobile hard disk, magnetic disk or CD, etc. In some jurisdictions, subject to legislation and patent practice, computer-readable media may not be electrical carrier signals and telecommunications signals.

In the above embodiments, each embodiment is described with its own emphasis. For parts that are not detailed or documented in a certain embodiment, please refer to the relevant descriptions of other embodiments.

Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented with electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Skilled artisans may implement the described functionality using different methods for each specific application, but such implementations should not be considered beyond the scope of this application.

In the embodiments provided in this application, it should be understood that the disclosed devices/devices and methods can be implemented in other ways. For example, the apparatus/equipment embodiments described above are only illustrative. For example, the division of modules or units is only a logical function division. In actual implementation, there may be other division methods, such as multiple units or units. Components may be combined or may be integrated into another system, or some features may be ignored, or not implemented. On the other hand, the coupling or direct coupling or communication connection between each other shown or discussed may be through some interfaces, indirect coupling or communication connection of devices or units, which may be in electrical, mechanical or other forms.

It will be understood that, when used in this specification and the appended claims, the term "comprising" indicates the presence of the described features, integers, steps, operations, elements and/or components but does not exclude one or more other The presence or addition of features, integers, steps, operations, elements, components and/or collections thereof.

It will also be understood that the term "and/or" as used in this specification and the appended claims refers to and includes any and all possible combinations of one or more of the associated listed items.

As used in this specification and the appended claims, the term "if" may be interpreted as "when" or "once" or "in response to determining" or "in response to detecting" depending on the context. ". Similarly, the phrase "if determined" or "if [the described condition or event] is detected" may be interpreted, depending on the context, to mean "once determined" or "in response to a determination" or "once the [described condition or event] is detected ]" or "in response to detection of [the described condition or event]".

In addition, in the description of this application and the appended claims, the terms "first", "second", "third", etc. are only used to distinguish the description, and cannot be understood as indicating or implying relative importance.

Finally, it should be noted that the above are only specific implementation modes of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions within the technical scope disclosed in the present application shall be covered by this application. within the scope of protection applied for. Therefore, the protection scope of this application should be subject to the protection scope of the claims.

Claims

A memory management method, applied to a first electronic device, characterized in that the method includes:

Determine the amount of data to be stored by the target application and the effective memory space in the physical address space associated with the target application. The effective memory space includes memory spaces indicated by N first physical address units, each of the first physical address units. The address unit includes at least one physical address with data read and write permissions, and N is an integer greater than or equal to 0;

If the amount of data is greater than the effective memory space, the effective memory space is expanded according to the first difference between the amount of data and the effective memory space;

If the data amount is less than the effective memory space, the effective memory space is reduced according to the second difference between the effective memory space and the data amount.
The method according to claim 1, characterized in that the physical address space further includes a second physical address unit, and the second physical address unit includes at least one physical address that does not have data writing permission.
The method of claim 2, further comprising:

When receiving the memory allocation request sent by the target application, allocate the second physical address unit to the target application;

When a registration request based on remote direct data access communication sent by the target application is received, an initial mapping table between the second physical address unit and the virtual address space of the target application is registered.
The method according to claim 2 or 3, characterized in that the virtual address space of the target application includes M virtual address units, each of the virtual address units includes at least one virtual address, and M is greater than or equal to 1. integer;

Expanding the effective memory space according to the first difference between the data amount and the effective memory space includes:

Allocate m first physical address units to the target application according to the first difference, and the memory space of the m first physical address units is greater than or equal to the first difference, 1≤m+N <M;

Remap m virtual address units corresponding to the second physical address units among the M virtual address units to m first physical address units.
The method according to any one of claims 2 to 4, characterized in that the virtual address space of the target application includes M virtual address units, each of the virtual address units includes at least one virtual address, and M is greater than or An integer equal to 1, the effective memory space includes N first physical address units corresponding to N virtual address units among the M virtual address units, 1≤N≤M;

The reducing the effective memory space according to the second difference between the effective memory space and the data amount includes:

When the second difference is greater than or equal to the preset threshold, n virtual address units corresponding to n first physical address units among the N virtual address units are remapped to the second Physical address units, the memory spaces of n first physical address units are less than or equal to the second difference, the preset threshold is greater than 0, and 1≤n<N.
The method according to any one of claims 1 to 5, wherein the target application is an application installed in the first electronic device.
The method according to any one of claims 1 to 6, characterized in that the amount of data to be stored is the amount of data of the target application, or the amount of data received by the first electronic device is installed in the second electronic device. The amount of data used by the application.
An electronic device, characterized in that the electronic device includes a processor, and the processor is used to run a computer program stored in a memory to implement the method according to any one of claims 1 to 7.
An electronic device, characterized in that the electronic device includes a processor and a first remote direct data access network card;

The first remote direct data access network card is used to determine the amount of data to be stored by the target application and the effective memory space in the physical address space associated with the target application. The effective memory space includes N first physical address units. In the indicated memory space, each first physical address unit includes at least one physical address with data read and write permissions, and N is an integer greater than or equal to 0;

If the data amount is greater than the effective memory space, the first remote direct data access network card is also used to send first interrupt information to the processor, where the first interrupt information includes the data amount and the A first difference between effective memory spaces, the processor is configured to expand the effective memory space according to the first difference;

If the amount of data is less than the effective memory space, the first remote direct data access network card is also used to send second interrupt information to the processor, where the second interrupt information includes the effective memory space and all the memory space. The processor is configured to reduce the effective memory space according to the second difference between the data amounts.
The electronic device according to claim 9, wherein the physical address space further includes a second physical address unit, and the second physical address unit includes at least one physical address that does not have data writing permission.
The electronic device according to claim 10, wherein the processor is further configured to:

When receiving the memory allocation request sent by the target application, allocate the second physical address unit to the target application;

The first remote direct data access network card is also used for:

When a registration request based on remote direct data access communication sent by the target application is received, an initial mapping table between the second physical address unit and the virtual address space of the target application is registered.
The electronic device according to claim 10 or 11, characterized in that the virtual address space of the target application includes M virtual address units, each of the virtual address units includes at least one virtual address, and M is greater than or equal to 1 an integer;

The processor is also used to:

Allocate m first physical address units to the target application according to the first difference, and the memory space of the m first physical address units is greater than or equal to the first difference, 1≤m+N <M;

Remap m virtual address units corresponding to the second physical address units among the M virtual address units to m first physical address units.
The electronic device according to any one of claims 10 to 12, wherein the virtual address space of the target application includes M virtual address units, each of the virtual address units includes at least one virtual address, and M is greater than Or an integer equal to 1, the effective memory space includes N first physical address units corresponding to N virtual address units among the M virtual address units, 1≤N≤M;

The processor is also used to:

When the second difference is greater than or equal to the preset threshold, n virtual address units corresponding to n first physical address units among the N virtual address units are remapped to the second Physical address units, the memory space of n first physical address units is less than or equal to the second difference, the preset threshold is greater than 0, 1≤n<N;

Remap n virtual address units among the N virtual address units to the second physical address unit.
The electronic device according to any one of claims 9 to 13, wherein the target application is an application installed in the electronic device.
The electronic device according to any one of claims 9 to 14, wherein the amount of data to be stored is the amount of data received by the first remote direct data access network card and sent by the second remote direct data access network card. The amount of data.
A chip system, characterized in that the chip system includes a processor, and the processor executes a computer program stored in a memory to implement the method according to any one of claims 1 to 7.
A computer-readable storage medium, characterized in that the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.