WO2024051311A1 - 数据处理方法、终端设备和可读存储介质 - Google Patents

数据处理方法、终端设备和可读存储介质 Download PDF

Info

Publication number
WO2024051311A1
WO2024051311A1 PCT/CN2023/103965 CN2023103965W WO2024051311A1 WO 2024051311 A1 WO2024051311 A1 WO 2024051311A1 CN 2023103965 W CN2023103965 W CN 2023103965W WO 2024051311 A1 WO2024051311 A1 WO 2024051311A1
Authority
WO
WIPO (PCT)
Prior art keywords
execution environment
task
thread
shared memory
trusted
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/CN2023/103965
Other languages
English (en)
French (fr)
Inventor
程琨
王宇
罗睿平
王城
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Huawei Technologies Co Ltd
Original Assignee
Huawei Technologies Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Huawei Technologies Co Ltd filed Critical Huawei Technologies Co Ltd
Priority to EP23862012.4A priority Critical patent/EP4575861A4/en
Publication of WO2024051311A1 publication Critical patent/WO2024051311A1/zh
Priority to US19/071,949 priority patent/US20250238269A1/en
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/46—Multiprogramming arrangements
    • G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
    • G06F9/5005—Allocation of resources, e.g. of the central processing unit [CPU] to service a request
    • G06F9/5027—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/46—Multiprogramming arrangements
    • G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F12/00—Accessing, addressing or allocating within memory systems or architectures
    • G06F12/02—Addressing or allocation; Relocation
    • G06F12/08—Addressing or allocation; Relocation in hierarchically structured memory systems, e.g. virtual memory systems
    • G06F12/0802—Addressing of a memory level in which the access to the desired data or data block requires associative addressing means, e.g. caches
    • G06F12/0806—Multiuser, multiprocessor or multiprocessing cache systems
    • G06F12/084—Multiuser, multiprocessor or multiprocessing cache systems with a shared cache
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F12/00—Accessing, addressing or allocating within memory systems or architectures
    • G06F12/02—Addressing or allocation; Relocation
    • G06F12/08—Addressing or allocation; Relocation in hierarchically structured memory systems, e.g. virtual memory systems
    • G06F12/0802—Addressing of a memory level in which the access to the desired data or data block requires associative addressing means, e.g. caches
    • G06F12/0806—Multiuser, multiprocessor or multiprocessing cache systems
    • G06F12/0842—Multiuser, multiprocessor or multiprocessing cache systems for multiprocessing or multitasking
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F12/00—Accessing, addressing or allocating within memory systems or architectures
    • G06F12/14—Protection against unauthorised use of memory or access to memory
    • G06F12/1458—Protection against unauthorised use of memory or access to memory by checking the subject access rights
    • G06F12/1491—Protection against unauthorised use of memory or access to memory by checking the subject access rights in a hierarchical protection system, e.g. privilege levels, memory rings
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
    • G06F21/50—Monitoring users, programs or devices to maintain the integrity of platforms, e.g. of processors, firmware or operating systems
    • G06F21/52—Monitoring users, programs or devices to maintain the integrity of platforms, e.g. of processors, firmware or operating systems during program execution, e.g. stack integrity ; Preventing unwanted data erasure; Buffer overflow
    • G06F21/53—Monitoring users, programs or devices to maintain the integrity of platforms, e.g. of processors, firmware or operating systems during program execution, e.g. stack integrity ; Preventing unwanted data erasure; Buffer overflow by executing in a restricted environment, e.g. sandbox or secure virtual machine
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
    • G06F21/70—Protecting specific internal or peripheral components, in which the protection of a component leads to protection of the entire computer
    • G06F21/71—Protecting specific internal or peripheral components, in which the protection of a component leads to protection of the entire computer to assure secure computing or processing of information
    • G06F21/72—Protecting specific internal or peripheral components, in which the protection of a component leads to protection of the entire computer to assure secure computing or processing of information in cryptographic circuits
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
    • G06F21/70—Protecting specific internal or peripheral components, in which the protection of a component leads to protection of the entire computer
    • G06F21/71—Protecting specific internal or peripheral components, in which the protection of a component leads to protection of the entire computer to assure secure computing or processing of information
    • G06F21/74—Protecting specific internal or peripheral components, in which the protection of a component leads to protection of the entire computer to assure secure computing or processing of information operating in dual or compartmented mode, i.e. at least one secure mode
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/46—Multiprogramming arrangements
    • G06F9/48—Program initiating; Program switching, e.g. by interrupt
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/46—Multiprogramming arrangements
    • G06F9/48—Program initiating; Program switching, e.g. by interrupt
    • G06F9/4806—Task transfer initiation or dispatching
    • G06F9/4843—Task transfer initiation or dispatching by program, e.g. task dispatcher, supervisor, operating system
    • G06F9/4881—Scheduling strategies for dispatcher, e.g. round robin, multi-level priority queues
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/46—Multiprogramming arrangements
    • G06F9/54—Interprogram communication
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/46—Multiprogramming arrangements
    • G06F9/54—Interprogram communication
    • G06F9/544—Buffers; Shared memory; Pipes
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/46—Multiprogramming arrangements
    • G06F9/54—Interprogram communication
    • G06F9/546—Message passing systems or structures, e.g. queues
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
    • G06F21/50—Monitoring users, programs or devices to maintain the integrity of platforms, e.g. of processors, firmware or operating systems
    • G06F21/57—Certifying or maintaining trusted computer platforms, e.g. secure boots or power-downs, version controls, system software checks, secure updates or assessing vulnerabilities
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2209/00—Indexing scheme relating to G06F9/00
    • G06F2209/50—Indexing scheme relating to G06F9/50
    • G06F2209/5011—Pool
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2209/00—Indexing scheme relating to G06F9/00
    • G06F2209/50—Indexing scheme relating to G06F9/50
    • G06F2209/5018—Thread allocation
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2209/00—Indexing scheme relating to G06F9/00
    • G06F2209/54—Indexing scheme relating to G06F9/54
    • G06F2209/548—Queue
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
    • Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management

Definitions

  • the embodiments of the present application relate to the field of computer technology, and in particular, to a data processing method, terminal device, and readable storage medium.
  • ARM means advanced reduced instruction set processors (advanced RISC machines)
  • RISC means reduced instruction set computer (reduced instruction set computer)
  • TrustZone means trust zone.
  • the software and hardware resources of terminal devices are divided into two worlds, called the secure world and the normal world.
  • the execution environment corresponding to the secure world is called a trusted execution environment (TEE), and a trusted application (TA) runs in the TEE.
  • the execution environment corresponding to the normal world is called rich execution environment (rich execution environment, REE), and common applications (common application, CA) run in REE.
  • REE and TEE are isolated from each other, and TEE has higher safety requirements.
  • Embodiments of the present application provide a data processing method, a terminal device and a readable storage medium, which improve data processing efficiency.
  • a data processing method is provided, applied to a terminal device, and the terminal device deploys a trusted execution environment and a non-trusted execution environment; the method includes: in the first execution environment, the first application program passes the first thread The initiated task is stored in the shared memory; the task is used to call the software function provided by the second application in the second execution environment; the shared memory is pre-applied for the first application in the untrusted execution environment and is used for untrusted The physical memory area accessed by the execution environment and the trusted execution environment; the first execution environment is a non-trusted execution environment and the second execution environment is a trusted execution environment, or the first execution environment is a trusted execution environment and the second execution environment It is a non-trusted execution environment; in the second execution environment, the task is obtained from the shared memory; in the second execution environment, the second thread is assigned to the task; the second thread is created in the second execution environment, and the second thread The processing units bound to the first thread are different; in the second execution environment, the second application program executes the task through
  • the data processing method provided in the first aspect by pre-applying for shared memory for the first application in the first execution environment, and storing relevant data of the execution task in the shared memory, the trusted execution environment and the non-trusted execution environment pass Accessing shared memory completes data transfer without requiring multiple copies of data, shortens the data transmission path between the trusted execution environment and the non-trusted execution environment, and improves the efficiency of data processing.
  • the processing unit bound to the first thread does not need to perform CPU mode switching, reducing the It reduces CPU switching overhead and improves processing efficiency.
  • the processing unit bound to the second thread can execute the task initiated by the first application program in the second execution environment without switching the CPU mode. Reduces CPU switching overhead and improves processing efficiency.
  • storing the task in the shared memory includes: in the first execution environment, serializing the task and obtaining the serialized message of the task; in the first execution environment, storing the task in the shared memory.
  • the serialized messages of the tasks are stored in the message queue; the message queue is stored in the shared memory and is used to store serialized messages corresponding to multiple tasks initiated by the first application program; in the first execution environment, the serialized messages of the tasks are stored in the shared memory.
  • the execution flag bit of is set to the first value; where the first value is used to indicate Indicates that the status of the serialized message is pending execution.
  • the second execution environment can quickly obtain the unprocessed task, thereby improving processing efficiency.
  • obtaining the task from the shared memory includes: in the second execution environment, accessing the message queue to obtain the serialized message of the task whose execution flag bit is the first value; In the second execution environment, the serialized message of the task is parsed to obtain the task.
  • the second execution environment after the second application program executes the task through the second thread, obtains the task result, and stores the task result in the shared memory, it also includes: in the second execution environment , setting the execution flag bit of the serialized message of the task to a second value; wherein the second value is used to indicate that the status of the serialized message is an executed state.
  • the method before storing the task initiated by the first application program through the first thread in the first execution environment into the shared memory, the method further includes: in the first execution environment, by the first application program through the first thread.
  • a thread initiates a task and registers a callback function.
  • the first application by registering the callback function, after the first application initiates a task, it does not need to wait for the task result, but can perform subsequent application business logic processing through the first thread.
  • the waiting time after the first application initiates a task is shortened, and the problem of the second execution environment execution task blocking the execution of the business logic of the first application is solved.
  • the method further includes: in the first execution environment, calling the callback function through a third thread; the third thread is different from the first thread.
  • the number of third threads is less than or equal to the number of tasks initiated by the first application.
  • the task includes incoming parameters and outgoing parameters, and the storage locations of the incoming parameters and outgoing parameters are located in the shared memory; in the first execution environment, the task will be initiated by the first application through the first thread.
  • the task Before the task is stored in the shared memory, it also includes: in the first execution environment, obtaining the first reference address of the shared memory in the first execution environment, the first input address of the incoming parameter in the first execution environment and the outgoing The first output address of the parameter in the first execution environment; in the first execution environment, convert the first input address to a first input offset relative to the first base address, and convert the first output address to a first input offset relative to the first base address. The first output offset of the first base address.
  • the method before the second application program executes the task through the second thread in the second execution environment, the method further includes: in the second execution environment, obtaining the second shared memory in the second execution environment. Base address; in the second execution environment, obtain the second input address of the incoming parameter in the second execution environment according to the second base address and the first input offset, and obtain the second input address of the incoming parameter in the second execution environment according to the second base address and the first output offset Get the second output address of the outgoing parameter in the second execution environment.
  • the first execution environment is a trusted execution environment
  • the second execution environment is a non-trusted execution environment
  • the task is an ocall task.
  • the first execution environment is a non-trusted execution environment
  • the second execution environment is a trusted execution environment
  • the task is an ecall task.
  • a thread pool is established for the first application in the second execution environment.
  • the thread pool includes multiple threads, and the threads in the thread pool are created in the second execution environment; the second thread Is an idle thread in the thread pool.
  • the method further includes: in the second execution environment, obtaining the first number of idle threads in the thread pool; in the second execution environment, obtaining the number of unexecuted tasks in at least one task initiated by the first application program. a second number of tasks; in the second execution environment, adjust the number of threads in the thread pool according to the first number and the second number.
  • the number of threads in the thread pool is adjusted in real time according to the first number of idle threads in the thread pool and the second number of unprocessed tasks, thereby avoiding CPU waste and improving data processing efficiency.
  • the number of threads in the thread pool is adjusted according to the first number and the second number, including: if the first number is greater than the second number, and the total number of threads in the thread pool is greater than the minimum threshold, then in the second execution environment, destroy a thread in the thread pool; or, if the first number is less than the second number, and the total number of threads in the thread pool is less than the maximum threshold, then in the second execution environment, Create at least one thread in the thread pool; at least one thread has a number less than or Equal to the difference between the maximum threshold and the total number of threads in the thread pool.
  • the method further includes: in the first execution environment, sending a minimum value to the second execution environment. threshold, maximum threshold and intermediate threshold; the intermediate threshold is greater than the minimum threshold and less than the maximum threshold; in the second execution environment, a thread pool is established for the first application, and the number of threads in the thread pool is the intermediate threshold.
  • the method further includes: in the first execution environment, obtaining the memory size of the shared memory; In the first execution environment, the shared memory is allocated according to the memory size; in the first execution environment, the related information of the shared memory is sent to the second execution environment.
  • the memory size includes a first memory size occupied by the message queue and a second memory size occupied by the application data; the message queue is stored in the shared memory and is used to store multiple tasks initiated by the first application program respectively.
  • the addresses of the first memory space and the second memory space are consecutive.
  • a terminal device in the second aspect, deploys a trusted execution environment and a non-trusted execution environment.
  • the trusted execution environment deploys trusted applications and trusted processing modules.
  • the untrusted execution environment deploys untrusted applications and untrusted processing modules.
  • the first processing module in the first execution environment is used to store the task initiated by the first application program through the first thread into the shared memory; the task is used to call the software function provided by the second application program in the second execution environment; the shared memory It is a physical memory area that is applied in advance for the first application in the untrusted execution environment and is used for access by the untrusted execution environment and the trusted execution environment; the first execution environment is the untrusted execution environment and the second execution environment is a trusted execution environment, the first processing module is an untrusted processing module, and the second processing module is a trusted processing module, or the first execution environment is a trusted execution environment, the second execution environment is an untrusted execution environment, The first processing module is a trusted processing module and the second processing module is an untrusted processing module; the second processing module in the second execution environment is used to obtain tasks from the shared memory; the second processing module in the second execution environment , used to allocate a second thread to a task; the second thread is created in the second execution environment, and the second thread and the first thread are respectively bound to
  • the first processing module in the first execution environment is used to: serialize the task in the first execution environment and obtain the serialization message of the task; in the first execution environment, convert the task's The serialized message is stored in the message queue; the message queue is stored in the shared memory and is used to store serialized messages corresponding to multiple tasks initiated by the first application program; in the first execution environment, the serialized message of the task is executed
  • the flag bit is set to a first value; wherein the first value is used to indicate that the status of the serialized message is a pending execution status.
  • the second processing module in the second execution environment is used to: in the second execution environment, access the message queue to obtain the serialized message of the task whose execution flag bit is the first value; In the second execution environment, parse the serialized message of the task and obtain the task.
  • the second processing module in the second execution environment is also used to: after the second application executes the task through the second thread, obtains the task result, and stores the task result in the shared memory,
  • the execution flag bit of the serialized message of the task is set to a second value; wherein the second value is used to indicate that the status of the serialized message is an executed status.
  • the first application program in the first execution environment is used to: initiate a task through the first thread in the first execution environment and register a callback function.
  • the first processing module in the first execution environment is also used to: in the first execution environment, call the callback function through a third thread; the third thread is different from the first thread.
  • the number of third threads is less than or equal to the number of tasks initiated by the first application.
  • the task includes incoming parameters and outgoing parameters, and the storage locations of the incoming parameters and outgoing parameters are located in the shared memory; the first processing module in the first execution environment is also used to: Before the first application program is stored in the shared memory through a task initiated by the first thread, in the first execution environment, the first reference address of the shared memory in the first execution environment and the first reference address of the incoming parameter in the first execution environment are obtained. An input address and a first output address of the outgoing parameter in the first execution environment; in the first execution environment, convert the first input address to a first input offset relative to the first base address, convert the first The output address is converted to a first output offset relative to the first base address.
  • the second processing module in the second execution environment is also used to: before the second application program executes the task through the second thread, in the second execution environment, obtain the shared memory for the second execution The second base address in the environment; in the second execution environment, obtain the second input address of the incoming parameter in the second execution environment according to the second base address and the first input offset, and obtain the second input address of the incoming parameter in the second execution environment according to the second base address and the first input offset.
  • An output offset obtains the second output address of the outgoing parameter in the second execution environment.
  • the first execution environment is a trusted execution environment
  • the second execution environment is a non-trusted execution environment
  • the task is an ocall task.
  • the first execution environment is a non-trusted execution environment
  • the second execution environment is a trusted execution environment
  • the task is an ecall task.
  • a thread pool is established for the first application in the second execution environment.
  • the thread pool includes multiple threads, and the threads in the thread pool are created in the second execution environment; the second thread Is an idle thread in the thread pool.
  • the second processing module in the second execution environment is also used to: in the second execution environment, obtain the first number of idle threads in the thread pool; in the second execution environment, obtain the first number of idle threads in the thread pool. A second number of unexecuted tasks in at least one task initiated by an application program; in the second execution environment, adjust the number of threads in the thread pool according to the first number and the second number.
  • the second processing module in the second execution environment is used to: if the first number is greater than the second number, and the total number of threads in the thread pool is greater than the minimum threshold, then in the second execution environment, Destroy one thread in the thread pool; or, if the first number is less than the second number and the total number of threads in the thread pool is less than the maximum threshold, create at least one thread in the thread pool in the second execution environment; at least one thread The number is less than or equal to the difference between the maximum threshold and the total number of threads in the thread pool.
  • the first processing module in the first execution environment is also configured to: before storing the task initiated by the first application program through the first thread into the shared memory, in the first execution environment, The second execution environment sends the minimum threshold, the maximum threshold and the intermediate threshold; the intermediate threshold is greater than the minimum threshold and less than the maximum threshold; the second processing module in the second execution environment is also used to: in the second execution environment, for the first application
  • the program establishes a thread pool, and the number of threads in the thread pool is an intermediate threshold.
  • the first processing module in the first execution environment is also used to: before storing the task initiated by the first application program through the first thread into the shared memory, in the first execution environment, obtain The memory size of the shared memory; in the first execution environment, allocate the shared memory according to the memory size; in the first execution environment, send the relevant information of the shared memory to the second execution environment.
  • the memory size includes a first memory size occupied by the message queue and a second memory size occupied by the application data; the message queue is stored in the shared memory and is used to store multiple tasks initiated by the first application program respectively.
  • the addresses of the first memory space and the second memory space are consecutive.
  • a third aspect provides a terminal device, including a processor.
  • the processor is configured to be coupled to a memory, read instructions in the memory, and cause the terminal device to execute the method provided in the first aspect according to the instructions.
  • a fourth aspect provides a program, which when executed by a processor is used to perform the method provided in the first aspect.
  • a computer-readable storage medium is provided. Instructions are stored in the computer-readable storage medium. When the instructions are run on a computer or processor, the method provided in the first aspect is implemented.
  • a sixth aspect provides a program product, the program product comprising a computer program, the computer program being stored in a readable storage medium, and at least one processor of the device being able to read the computer program from the readable storage medium. , the at least one processor executes the computer program so that the device implements the method provided in the first aspect.
  • 1A to 1B are a set of architectural schematic diagrams of terminal equipment provided by embodiments of the present application.
  • Figure 1C is another architectural schematic diagram of a terminal device provided by an embodiment of the present application.
  • Figure 2 is another architectural schematic diagram of a terminal device provided by an embodiment of the present application.
  • Figures 3A to 3D are a set of schematic diagrams of shared memory provided by embodiments of the present application.
  • Figure 4 is a flow chart of the data processing method provided by the embodiment of the present application.
  • Figure 5 is another flow chart of the data processing method provided by the embodiment of the present application.
  • Figure 6 is another flow chart of the data processing method provided by the embodiment of the present application.
  • Figure 7 is another flow chart of the data processing method provided by the embodiment of the present application.
  • Figure 8 is a schematic structural diagram of a terminal device provided by an embodiment of the present application.
  • the data processing method provided by the embodiment of this application is applied to terminal equipment.
  • the embodiment of the present application does not limit the name of the terminal device.
  • the terminal device is also called: terminal, electronic device, computer device, computing device or computer system, etc.
  • Common terminal devices include, for example: mobile phones, tablets, laptops, desktop computers, servers, wearable devices (such as smart watches or smart bracelets), smart home devices (such as smart speakers or smart TVs), vehicle-mounted smart devices, wireless devices, etc.
  • Human driving equipment virtual reality (VR) equipment, augmented reality (AR) equipment, mixed reality (MR) equipment, and artificial intelligence (artificial intelligence, AI) equipment, etc.
  • OS operating system
  • APP application
  • the OS and APP provide rich functions for the terminal device.
  • types and number of APPs are increasing, and terminal devices support more and more functions, which puts forward higher requirements for user privacy and information security.
  • the software and hardware resources of the terminal device can be divided into two worlds, called the secure world and the non-secure world.
  • the safe world and the non-safe world are relative concepts. Perform operations that require confidentiality in the secure world and store data that require confidentiality, such as fingerprint recognition, face recognition, password processing, data encryption and decryption, security authentication, etc., while the remaining operations are performed in the non-secure world.
  • a security world is also called a security domain.
  • the non-safe world is also called the non-safe domain, normal world or common world.
  • the names of the secure world and the non-secure world may be different.
  • the execution environment corresponding to the secure world is called a trusted execution environment
  • the execution environment corresponding to the non-secure world is called a non-trusted execution environment.
  • Trusted execution environment and non-trusted execution environment are relative concepts. The trusted execution environment and the non-trusted execution environment are isolated from each other. Therefore, operations that need to be kept confidential can be performed in the trusted execution environment, data that needs to be kept confidential can be stored, and key program logic and high-value data can be protected.
  • Trusted applications are deployed in the trusted execution environment, and untrusted applications are deployed in the untrusted execution environment.
  • Trusted applications and untrusted applications are relative concepts.
  • trusted applications include but are not limited to fingerprint recognition programs, face recognition programs, etc.
  • the embodiments of this application do not limit the names of trusted execution environments, non-trusted execution environments, trusted execution programs, and non-trusted execution programs. In different architectures of terminal devices, the names of these concepts may be different. Please refer to the subsequent descriptions of GP architecture and Intel SGX architecture.
  • the behavior initiated in a non-trusted execution environment to call the trusted execution environment function interface is called an entry call (ecall) task or ecall call.
  • the function of the trusted execution environment that is called is called the ecall function. .
  • the behavior initiated in the trusted execution environment to call the non-trusted execution environment function interface is called an out call (ocall) task or ocall call.
  • the function of the called non-trusted execution environment The function is called the ocall function.
  • a program usually refers to a piece of executable code.
  • Application can provide specific functions.
  • a fingerprint recognition application provides fingerprint recognition functionality
  • a face recognition application provides face recognition functionality.
  • Process refers to a running activity of a program and is the basic unit of resource allocation by the operating system. It can be understood that a running program is a process. Processes have their own address space. A process can include a thread, At this point, the process is single-threaded. A process can also include multiple threads, and multiple threads can perform different tasks in parallel. At this time, the process is multi-threaded.
  • Thread included in the process, is the basic unit of operation scheduling by the operating system. It can be understood that a thread is a single sequential control flow in a process. Threads are sometimes also called lightweight processes (LWP).
  • LWP lightweight processes
  • Central processing unit (CPU) binding refers to binding a process or thread to a designated CPU core for execution in a multi-core CPU system.
  • the Global Platform has developed open architecture and application programming interface (API) standards for trusted execution environments.
  • the non-safe world is called the ordinary world or normal world.
  • the trusted execution environment corresponding to the secure world is called TEE, and the untrusted execution environment corresponding to the non-secure world is called REE.
  • the trusted application running in TEE is called TA, and the untrusted application running in REE is called CA.
  • ARM TrustZone is an architecture developed based on GP standards.
  • ARM TrustZone is a hardware-based security function that introduces a secure world and a non-secure world at the processor level.
  • the processor only runs in one of the worlds at any time.
  • When the processor runs in the safe world it runs the code of the safe world; when the processor runs in the non-safe world, it runs the code of the non-safe world.
  • the secure world and the non-secure world are switched through Monitor Mode.
  • Hardware isolation of secure and non-secure worlds provides an effective mechanism for protecting application code and data.
  • FIG. 1A is an architectural schematic diagram of a terminal device provided by an embodiment of the present application.
  • the terminal device adopts GP architecture or ARM TrustZone architecture.
  • the terminal device includes a hardware layer and a software layer.
  • the hardware layer includes but is not limited to: processor and storage media.
  • the software layer includes REE and TEE.
  • the REE operating system and CA are deployed on the REE side.
  • the REE operating system is also called a rich operating system (Rich OS), such as the existing Linux operating system or Windows operating system.
  • the TEE operating system and TA are deployed on the TEE side.
  • the TEE operating system is also called a trusted operating system (Trusted OS).
  • Both the REE side and the TEE side include application state (or user state) and kernel state.
  • the kernel state on the REE side includes the REE operating system, and the application state on the REE side includes CA.
  • the kernel state on the TEE side includes the TEE operating system, and the application state on the TEE side includes TA.
  • the CA can initiate an ecall task and call the API provided by the TEE side to complete certain safe and protected business logic.
  • TA can also initiate ocall tasks and call the API provided by the REE side to complete some common business logic.
  • some application scenarios are: in terminal device unlocking, application login, financial payment and other scenarios, CA is an unlocking application or financial payment application, and TA is a fingerprint recognition application or face recognition application.
  • the CA initiates an ecall task and calls the fingerprint recognition or face recognition function provided by the TA on the TEE side.
  • the CA and the TA collaborate to complete the unlocking of the terminal device, login of the application, fingerprint payment or face payment.
  • REE and TEE run on separate physical memories, that is, memory isolation.
  • the REE side cannot access the physical memory on the TEE side, and conversely, the TEE side cannot access the physical memory on the REE side.
  • the physical memory corresponding to the REE side is called REE memory
  • the physical memory corresponding to the TEE side is called TEE memory.
  • the processor includes multiple processor cores, labeled core 0 to core n. At any time, the processor core is running on the REE side or on the TEE side. For example, core 0 runs in REE and is in REE mode or REE state; core 1 runs in TEE and is in TEE mode or TEE state.
  • the processor includes multiple processing units.
  • the embodiments of this application do not limit the type and name of the processor, and the type, name, and quantity of the processing units.
  • the processor is a CPU
  • the processing unit is a CPU core (or CPU core, CPU core).
  • the CPU core is an important part of the CPU and is used to complete calculations, accept or store commands, process data, etc.
  • Storage media also known as memory, are used to store executable program code and/or data, and the executable program code may include instructions.
  • the storage medium may include a program storage area and a data storage area.
  • the storage medium may include high-speed random access memory and non-volatile memory, for example, at least one magnetic disk storage device, hard disk drive (HDD), solid-state drive (SS), flash memory device, general-purpose flash memory (universal flash storage, UFS), etc., can also include volatile Memory (volatile memory), such as random-access memory (random-access memory, RAM), etc.
  • the CA and TA exchange data through the REE communication agent, the message channel of the hardware layer, and the TEE communication agent.
  • the REE communication agent, message channel and TEE communication agent establish a secure communication channel for CA and TA.
  • the CA calls the TEE client API to communicate with the corresponding TA, and the TA calls the TEE internal API to use the programming resources provided by the TEE to implement related functions.
  • Intel SGX is a technology that meets the needs of trusted computing. It is similar to ARM TrustZone and can implement a trusted execution environment.
  • FIG. 1C is another architectural schematic diagram of a terminal device provided by an embodiment of the present application.
  • the terminal device adopts Intel SGX architecture or SGX architecture, showing the software layer.
  • the untrusted execution environment includes the operating system and applications.
  • the operating system is, for example, a Linux operating system or a Windows operating system.
  • Applications can be divided into two parts: the secure part (or trusted part, trusted application) and the non-secure part (or untrusted part, untrusted application).
  • an application When an application needs to process confidential data, it creates a secure area (called an enclave, or trusted area, or trusted execution environment) located in trusted memory to store enclave code and enclave data, and then calls trusted functions (Also known as the enclave function, it is a function created by the developer to be executed specifically within the safe zone). Once the trusted function is called, the application will execute in the trusted zone and plainly access the enclave code and enclave data in the safe zone. In addition, any attempt to access the safe zone from outside the safe zone will be rejected by the processor. . After the trusted function is executed, the enclave data in the safe zone still remains in the trusted memory, and the application returns to the untrusted zone (or untrusted execution environment) to continue execution and loses access to the trusted memory. permissions.
  • an enclave or trusted area, or trusted execution environment
  • the trusted execution environment is part of the host process, which means that the application contains its own code (called application code), data (called application data) and enclave, see the left side of Figure 1C part.
  • the enclave also contains its own code (called enclave code) and its own data (called enclave data), see the right part of Figure 1C.
  • SGX protects the confidentiality and integrity of enclave code and enclave data.
  • the embodiment of the present application takes the GP architecture as an example to describe the data processing method provided by the embodiment of the present application.
  • FIG. 2 is another architectural schematic diagram of a terminal device provided by an embodiment of the present application.
  • shared memory is added to the storage medium
  • an untrusted processing module is added to the application state of REE
  • a trusted processing module is added to the application state of TEE.
  • the untrusted processing module is used to process relevant data in the REE or during the task initiated by the application in the TEE.
  • the trusted processing module is used to process relevant data in the TEE during tasks initiated by REE or applications in the TEE.
  • the data processing method provided by the embodiment of this application allows the REE side and the TEE side to access the data in the shared memory by applying for shared memory in advance, completing the data transfer between the REE side and the TEE side. , does not require multiple copies of data, shortens the data transmission path between REE and TEE, and improves the efficiency of data processing. Moreover, threads that perform the same task on the REE side and the TEE side are bound to different processing units. The threads that perform the task on the TEE side are created on the TEE side, which reduces CPU switching overhead and improves data processing efficiency.
  • the first execution environment is an untrusted execution environment and the second execution environment is a trusted execution environment.
  • the first execution environment is a trusted execution environment and the second execution environment is an untrusted execution environment. .
  • the first application program is an application program in the first execution environment
  • the second application program is an application program in the second execution environment.
  • a task refers to a software function initiated by a first application program through a first thread in the first execution environment and used to call a software function provided by a second application program in the second execution environment.
  • the first execution environment is REE
  • the first application is CA
  • the second execution environment is TEE
  • the second application is TA.
  • a user uses a payment APP to pay through fingerprint recognition.
  • CA is a payment APP
  • TA is a fingerprint recognition application.
  • REE when the CA (payment APP) is running, the ecall task is initiated, the fingerprint identification function provided by the TA (fingerprint identification application) in the TEE is called, and the fingerprint payment is completed in the payment APP.
  • Optional tasks include ecall tasks and ocall tasks.
  • shared memory refers to a memory area that is applied in advance for the first application program in the non-trusted execution environment and is used for access by the non-trusted execution environment and the trusted execution environment.
  • the first execution environment is REE
  • the first application is CA
  • CA initiates an ecall task.
  • physical memory management of end devices is performed by an untrusted execution environment.
  • REE a physical memory area is applied for in advance for CA, and both REE and TEE can access this physical memory area.
  • the CA initiates the ecall task
  • the relevant data of the ecall task can be stored in the shared memory. Therefore, the TEE side can obtain the relevant data of the ecall task from the shared memory and then execute the ecall task.
  • the task result of the ecall task can be stored in the shared memory. Therefore, the REE side can obtain the task result of the ecall task from the shared memory and complete the data transmission. Since data transmission is completed without data copying, the efficiency of data processing is improved.
  • the shared memory stores the incoming parameters and outgoing parameters of the function called by the task.
  • the shared memory may be a physical memory area that is pre-applied for tasks initiated by the first application in the untrusted execution environment and used for access by the untrusted execution environment and the trusted execution environment.
  • the embodiment of this application does not limit the memory size and location of the shared memory.
  • the memory size and location of shared memory can be different when the first application is different.
  • the first application includes CA1 and CA2.
  • the shared memory size of CA1 is 5MB, and the starting address is 0x40000000.
  • the shared memory size of CA2 is 1MB, and the starting address is 0x50000000.
  • the memory size of the shared memory includes the first memory size occupied by the message queue and the second memory size occupied by the application data.
  • the message queue is used to store serialized messages corresponding to multiple tasks initiated by the first application program.
  • the application data includes interaction data during the execution of a task initiated by the first application.
  • the embodiment of the present application does not limit the name of application data. For example, it may also be called interaction data, task data, or shared data.
  • the shared memory includes a first memory space and a second memory space.
  • the size of the first memory space is the first memory size and is used to store the message queue.
  • the size of the second memory space is the second memory size and is used to store application data.
  • the embodiments of this application do not limit the names of the first memory space and the second memory space.
  • the second memory space is also called the data buffer.
  • the embodiments of the present application do not limit the values of the first memory size and the second memory size, and do not limit the starting positions of the first memory space and the second memory space.
  • the addresses of the first memory space and the second memory space are consecutive.
  • the first application when it declares the initialization configuration, it may include at least one of the following: the memory size of the shared memory, the first memory size, the second memory size, and the number of tasks initiated by the first application.
  • the first application declares the initialization configuration, it may include at least one of the following: the memory size of the shared memory, the first memory size, the second memory size, and the number of tasks initiated by the first application.
  • the shared memory will be exemplarily explained below with reference to Figures 3A to 3D. Assume that the first application is CA.
  • REE apply for a block according to the memory size of the shared memory.
  • the physical memory area with consecutive addresses is used as the shared memory of CA.
  • in REE apply for a physical memory area with consecutive addresses based on the first memory size as the first memory space in the CA shared memory for storing messages. Queue; apply for a physical memory area with consecutive addresses based on the size of the second memory as the second memory space in the CA shared memory for storing application data. Wherein, the addresses of the first memory space and the second memory space are consecutive, and the first memory space is located after the address of the second memory space.
  • the first memory space and the second memory space are respectively applied for according to the first memory size and the second memory size.
  • the similarity between Figure 3C and Figure 3B is that the addresses of the first memory space and the second memory space are consecutive.
  • the first memory space is located before the address of the second memory space.
  • the first memory space and the second memory space are respectively applied for according to the first memory size and the second memory size.
  • the difference between Figure 3D, Figure 3B and Figure 3C is that in Figure 3D, the addresses of the first memory space and the second memory space are not consecutive.
  • the message queue is stored in the shared memory and is used to store serialized messages corresponding to multiple tasks initiated by the first application program.
  • the embodiments of this application do not limit the number of tasks that the message queue can store or the number of serialized messages that can be stored. When applications are different, the number of tasks initiated by the application program is usually different, and the capacity of the message queue is different.
  • serialization is a data processing mechanism.
  • IO input/output
  • Serialization converts an object into a stream of bytes that a computing device can understand. Through serialization, data can be stored and transmitted in a more comprehensive way.
  • serialization refers to restoring the received byte stream into an object that can be recognized by the program.
  • the task before the task is stored in the message queue, the task needs to be serialized and a serialized message of the task needs to be generated. Accordingly, after retrieving the serialized message corresponding to the task from the message queue, the serialized message corresponding to the task needs to be deserialized or parsed to restore the task.
  • a thread pool is established for untrusted applications in the untrusted execution environment.
  • the thread pool includes multiple threads, and the threads in the thread pool are all created in a trusted execution environment. Threads in the thread pool are used to perform ecall tasks initiated by untrusted applications.
  • the number of threads in the thread pool has a minimum value and a maximum value, which are called the minimum threshold and the maximum threshold respectively.
  • the embodiments of this application do not limit the number of minimum thresholds and maximum thresholds.
  • the minimum threshold it is ensured that there are threads in the trusted execution environment for executing ecall tasks initiated by untrusted applications, avoiding the situation where the ecall task cannot be executed.
  • the maximum threshold the maximum number of threads that can execute ecall tasks initiated by untrusted applications in a trusted execution environment is limited, avoiding resource waste caused by fewer ecall tasks.
  • the number of threads in the thread pool can be adjusted in real time between the minimum threshold and the maximum threshold based on the number of ecall tasks initiated by untrusted applications.
  • the number of threads in the thread pool can be adjusted in real time between the minimum threshold and the maximum threshold based on the number of ecall tasks initiated by untrusted applications.
  • the number of threads in the thread pool can be an intermediate threshold, where minimum threshold ⁇ intermediate threshold ⁇ maximum threshold.
  • At least one of the minimum threshold, the maximum threshold and the intermediate threshold can be set for the trusted execution environment side, or for the non-trusted execution environment side, or for the trusted execution environment side and the non-trusted execution environment side. Determined after mutual negotiation between the trusted execution environment sides.
  • At least one of the minimum threshold, the maximum threshold and the intermediate threshold corresponding to the thread pool may be the same or different.
  • thread pool 1 is established for CA1, and the corresponding minimum threshold, maximum threshold and intermediate threshold are 2, 5 and 10 respectively.
  • Thread pool 2 is established for CA2, and the corresponding minimum threshold, maximum threshold and intermediate threshold are 3, 6 and 10 respectively.
  • 6 threads are created for CA2.
  • a callback function is a function that is passed as a parameter.
  • the process of using a callback function may include: (1) Defining a callback function; (2) The party providing the function implementation registers the function pointer of the callback function to the caller during initialization; (3) When a specific event or condition is met , the caller uses the function pointer to call the callback function to process the event.
  • the pointer of the function to be called can be passed to a function as a parameter, so that the function can flexibly use different methods when handling similar events.
  • a data processing method is provided, and the execution subject is a terminal device.
  • the execution subject is a terminal device.
  • trusted execution environments and non-trusted execution environments deployed in terminal devices. This embodiment does not limit the manner in which the terminal device implements the trusted execution environment.
  • terminal equipment can adopt GP architecture or SGX architecture.
  • Applications and processing modules are deployed in both trusted execution environments and non-trusted execution environments. Among them, the applications deployed in the trusted execution environment are called trusted applications, and the processing modules are called trusted processing modules. Applications deployed in a non-trusted execution environment are called non-trusted applications, and processing modules are called non-trusted processing modules.
  • the data processing method provided by this embodiment is suitable for the scenario where the terminal device initiates a task through the first thread through the first application program in the first execution environment.
  • the application program deployed in the first execution environment is called the first application program, and the deployed processing module is called the first processing module.
  • the application program deployed in the second execution environment is called the second application program, and the deployed processing module is called the second processing module.
  • the first execution environment is a non-trusted execution environment
  • the first application program is an untrusted application program
  • the second execution environment is a trusted execution environment
  • the second application program is a trusted Applications and tasks are ecall tasks.
  • the first execution environment is a trusted execution environment
  • the first application program is a trusted application program
  • the second execution environment is a non-trusted execution environment
  • the second application program is a non-trustable execution environment.
  • Letter applications and tasks are ocall tasks.
  • the data processing method provided in this embodiment can be applied to at least the following scenarios:
  • Scenario 1 The terminal device adopts GP architecture. Please refer to the structure shown in Figure 2.
  • the first execution environment is REE
  • the first application program is CA
  • the first processing module is an untrusted processing module
  • the second execution environment is TEE
  • the second application program is TA
  • the second processing module is a trusted processing module.
  • the terminal device initiates the ecall task through the first thread in the REE by the CA.
  • the terminal device adopts GP architecture. Please refer to the structure shown in Figure 2.
  • the first execution environment is TEE
  • the first application program is TA
  • the first processing module is a trusted processing module
  • the second execution environment is REE
  • the second application program is CA
  • the second processing module is an untrusted processing module.
  • the terminal device initiates the ocall task through the first thread in the TEE.
  • the terminal device adopts the SGX architecture. Please refer to the structure shown in Figure 1C.
  • the first execution environment is a non-trusted execution environment
  • the first application is the non-secure part of the application
  • the first processing module is the non-trusted processing module
  • the second execution environment is the Enclave
  • the second application is the security of the application.
  • Part and the second processing module are trusted processing modules.
  • the terminal device initiates an ecall task through the first thread by the non-secure part of the application in a non-trusted execution environment.
  • Scenario 4 The terminal device adopts SGX architecture, as shown in Figure 1C.
  • the first execution environment is the Enclave
  • the first application is the secure part of the application
  • the first processing module is the trusted processing module
  • the second execution environment is the untrusted execution environment
  • the second application is the non-secure part of the application.
  • the second processing module is an untrusted processing module.
  • the terminal device initiates the ocall task through the first thread in the Enclave by the security part of the application.
  • this embodiment takes scenario 1 as an example to describe the data processing method provided by this embodiment.
  • FIG. 4 is a flow chart of the data processing method provided by the embodiment of the present application.
  • the data processing method provided by this embodiment may include:
  • CA initiates the ecall task through the first thread.
  • CA initiates an ecall task to call the software function provided by TA in TEE.
  • the first thread is CA’s creation in REE
  • the created thread is bound to a CPU core.
  • This embodiment does not limit the CPU core to which the first thread is bound.
  • the CPU includes core 0 to core 31, and the CPU core bound to the first thread is core 0.
  • the ecall task can include but is not limited to the following information: the function name of the called ecall function, incoming parameters, outgoing parameters, and function return value type.
  • the untrusted processing module stores the ecall task in the shared memory.
  • shared memory is a memory area that is pre-applied for CA in REE and used for REE and TEE access.
  • the meaning of shared memory can be found in the previous relevant descriptions of this application and will not be repeated here.
  • the storage locations of the incoming parameters and outgoing parameters of the ecall function are also located in the shared memory.
  • both REE and TEE can access the shared memory, after the untrusted processing module stores the ecall task initiated by the CA into the shared memory, the TEE side can obtain the ecall task by accessing the shared memory, so that no data copy is required. This completes the data transfer from the REE side to the TEE side, shortens the data transmission path under the GP architecture, and improves the efficiency of data processing.
  • the trusted processing module obtains tasks from the shared memory.
  • the trusted processing module on the TEE side accesses the shared memory and obtains the tasks stored in the shared memory.
  • the trusted processing module allocates a second thread to the ecall task.
  • the second thread is created in the TEE, and the second thread and the first thread are respectively bound to different processing units.
  • the processing unit is a CPU core of the processor.
  • This embodiment does not limit the CPU core to which the second thread is bound.
  • the CPU includes cores 0 to 31, and the CPU core bound to the second thread is core 10.
  • the CA on the REE side creates a second thread and binds it to a CPU core, for example, core 10. After that, the core 10 bound to the second thread is switched from the REE side to the TEE side, and the core 10 is switched from the REE mode to the TEE mode.
  • the processing unit (core 10) bound to the second thread is initially in TEE mode and does not need to switch the CPU, thereby improving data processing efficiency.
  • TA executes the ecall task through the second thread, obtains the task result, and stores the task result into the shared memory.
  • the second thread can call the function entry provided by TA, perform the ecall task, and write the outgoing parameters and function return value to the shared area specified by the parameters in the ecall function.
  • the thread that CA initiates the ecall task is the first thread, and the bound processing unit is core 0.
  • the thread in which TA executes the ecall function called by the ecall task is the second thread, which is bound to CPU core 10. Since the CPU cores bound to the second thread and the first thread are different, when TA executes the ecall function, it does not need to switch the CPU core (core 0) bound to the first thread on the REE side to the TEE side, thus reducing CPU switching. overhead, improving processing efficiency.
  • TA since both REE and TEE can access the shared memory, TA stores the task results in the shared memory after executing the ecall task.
  • the REE side can obtain the task results of the ecall task by accessing the shared memory, thus completing the data transfer from the TEE side to the REE side without the need for data copy, shortening the data transmission path under the GP architecture, and improving the efficiency of the ecall task. Data processing efficiency.
  • a data processing method is provided, which is suitable for the scenario where the terminal device initiates a task through the first thread through the first application program in the first execution environment.
  • the relevant data of the execution task such as the incoming and outgoing parameters of the task calling function, task results, etc. are stored in the shared memory.
  • the execution environment and the non-trusted execution environment complete data transfer by accessing shared memory, without requiring multiple copies of data, shortening the data transmission path between REE and TEE, and improving the efficiency of data processing.
  • the processing unit bound to the first thread does not need to perform CPU mode switching, reducing the It reduces CPU switching overhead and improves processing efficiency.
  • the processing unit bound to the second thread can execute the task initiated by the first application program in the second execution environment without switching the CPU mode. Reduces CPU switching overhead and improves processing efficiency.
  • the data processing method provided by this embodiment may also include: before the untrusted processing module stores the ecall task in the shared memory in S402:
  • the untrusted processing module obtains the instruction information of the ecall task.
  • the instruction information is used to determine the ecall task Whether to use the data processing method provided by this embodiment.
  • the untrusted processing module determines whether the ecall task adopts the data processing method provided by this embodiment based on the instruction information of the ecall task.
  • the process will be performed according to the existing ecall task processing flow.
  • the instruction information of the ecall task is used to determine whether the ecall task initiated by the CA executes the data processing method provided in this embodiment, which improves the compatibility between the data processing method provided by this embodiment and the existing ecall task processing flow, and improves the data processing flexibility.
  • the indication information may include at least one of the following: configuration information of the ecall task or the keyword of the ecall function called by the ecall task.
  • the keyword parameters of the function refer to the parameters sent through the syntax when calling the function.
  • the function call keyword method by using the function call keyword method, it can be determined whether the ecall task initiated by the CA executes the data processing method provided by this embodiment.
  • the data processing method provided by this embodiment may also include: before the untrusted processing module stores the ecall task in the shared memory in S402:
  • the untrusted processing module obtains the first base address of the shared memory in the REE, the first input address of the incoming parameter in the REE, and the first output address of the outgoing parameter in the REE.
  • the untrusted processing module converts the first input address into a first input offset relative to the first base address, and converts the first output address into a first output offset relative to the first base address.
  • the first base address of the shared memory in the REE can reflect the layout information of the shared memory in the CA address space, such as the starting position, offset, etc.
  • the ecall task also includes a first input offset and a first output offset.
  • the data processing method provided by this embodiment may also include: before the trusted processing module allocates the second thread to the task in S404:
  • the trusted processing module obtains the second base address of the shared memory in the TEE.
  • the trusted processing module obtains the second input address of the incoming parameter in the TEE based on the second reference address and the first input offset, and obtains the outgoing parameter based on the second reference address and the first output offset. Second output address in TEE.
  • the second base address of the shared memory in the TEE can reflect the layout information of the shared memory in the TA address space, such as the starting position, offset, etc.
  • the address spaces on the TEE side and the REE side are completely isolated.
  • the address mapping of shared memory is taken care of by the REE operating system.
  • the address mapping of shared memory is taken care of by the TEE operating system.
  • the shared memory on the REE side and the TEE side will be mapped to different areas of the CA process address and the TA process address by different operating systems. In this way, when CA initiates an ecall task, the incoming parameters, outgoing parameters, etc. located in the shared memory will not be accessible by TA. Vice versa, the parameters passed by TA and located in shared memory cannot be accessed by CA.
  • the address of the parameter located in the shared memory passed by CA is converted into an offset relative to the shared memory base address, so that on the TEE side, the offset of the parameter address can be converted into the TEE address space
  • the address in TA can successfully access. vice versa.
  • the first base address of the shared memory in the REE is the starting address of the shared memory in the CA address space, which is 0x40000000 and has a length of 1MB.
  • the first input address of the incoming parameter in REE is 0x40001000, and the length is 8 bytes.
  • the address of the passed parameter determines whether the address of the passed parameter will exceed the range of the memory address. For example, in this example, the length of the incoming parameter is 8 bytes, the first input address is 0x40001000, and the length of the shared memory is 1MB, the starting address is 0x40000000, and the incoming parameter does not exceed the range of the memory address.
  • the TEE side after obtaining the second input address of the incoming parameter in the TEE, it can be determined whether the address of the incoming parameter is Will it exceed the range of the memory address to ensure that the access to the shared memory does not cross the boundary.
  • the data processing method provided by this embodiment may also include: before the CA initiates the ecall task through the first thread in S401:
  • the untrusted processing module obtains the memory size of the shared memory.
  • the untrusted processing module pre-allocates shared memory according to the memory size.
  • the information related to the shared memory includes at least the physical address range of the shared memory.
  • the information transmission channel in the existing technology can be used.
  • the data processing method provided in this embodiment may also include:
  • the untrusted processing module releases the shared memory.
  • Figure 5 is another flow chart of the data processing method provided by the embodiment of the present application, providing the implementation of S402 and S403.
  • S402 includes S4021 ⁇ S4023.
  • S403 includes S4031 ⁇ S4032.
  • S401, S403-S405 may refer to the description in the embodiment shown in Figure 4, and will not be described again here.
  • the data processing method provided by this embodiment may include:
  • CA initiates the ecall task through the first thread.
  • the untrusted processing module serializes the ecall task and obtains the serialized message of the ecall task.
  • the untrusted processing module stores the serialized message of the ecall task into the message queue.
  • the message queue is stored in the shared memory and is used to store serialized messages corresponding to multiple ecall tasks initiated by CA.
  • the untrusted processing module sets the execution flag bit of the serialized message of the ecall task to the first value.
  • the first value is used to indicate that the status of the serialized message in the message queue is the pending execution status, so that the TEE side obtains the unprocessed ecall task from the message queue after accessing the shared memory.
  • the first numerical value and the second numerical value are relative concepts.
  • the second value is used to indicate that the status of the serialized message in the message queue is the executed status, so as to notify the REE side that the ecall task is completed.
  • the first numerical value and the second numerical value have different values.
  • the embodiments of the present application do not limit the values of the first numerical value and the second numerical value.
  • the first value is 0 and the second value is 1; or the first value is 1 and the second value is 0.
  • the trusted processing module accesses the message queue and obtains the serialized message of the ecall task whose execution flag bit is the first value.
  • this embodiment does not limit the rules for the trusted processing module on the TEE side to access the message queue.
  • the unprocessed ecall task whose execution flag bit is the first value in the message queue can be discovered through polling or by configuring a specified thread to monitor the corresponding message queue.
  • the trusted processing module parses the serialized message of the ecall task and obtains the ecall task.
  • parsing and deserialization have the same meaning.
  • the serialized message of the unprocessed ecall task the serialized message is restored to the ecall function called by the ecall task.
  • the trusted processing module allocates a second thread to the ecall task.
  • TA executes the ecall task through the second thread, obtains the task result, and stores the task result into the shared memory.
  • the trusted processing module sets the execution flag bit of the serialized message of the ecall task to the second value.
  • the data processing method provided by this embodiment is suitable for the scenario where the terminal device initiates a task through the first thread through the first application program in the first execution environment.
  • the non-trusted execution environment apply for shared memory in advance for the first application.
  • Both the non-trusted execution environment and the trusted execution environment can access the shared memory.
  • the shared memory stores message queues and related data during task execution. Set the task's execution flag to a different value to indicate whether the task has been processed.
  • the CPU switching overhead is reduced. Improved data processing efficiency.
  • the second thread is created in the second execution environment, CPU switching overhead is reduced and data processing efficiency is improved.
  • a data processing method is provided, and the execution subject is a terminal device.
  • the execution subject is a terminal device.
  • trusted execution environments and non-trusted execution environments deployed in terminal devices. This embodiment does not limit the manner in which the terminal device implements the trusted execution environment.
  • terminal equipment can adopt GP architecture or SGX architecture.
  • Applications and processing modules are deployed in both trusted execution environments and non-trusted execution environments. Among them, the applications deployed in the trusted execution environment are called trusted applications, and the processing modules are called trusted processing modules. Applications deployed in a non-trusted execution environment are called non-trusted applications, and processing modules are called non-trusted processing modules.
  • the data processing method provided by this embodiment is suitable for the scenario where the terminal device initiates a task through the first thread through the first application program in the first execution environment.
  • the data processing method provided in this embodiment can be applied to at least the above scenarios one to four.
  • this embodiment takes scenario 1 as an example to describe the data processing method provided by this embodiment.
  • FIG. 6 is another flow chart of the data processing method provided by the embodiment of the present application.
  • the data processing method provided by this embodiment may include:
  • CA initiates the ecall task through the first thread and registers the callback function.
  • CA initiates an ecall task to call the software function provided by TA in TEE.
  • the first thread is the thread created by CA in REE.
  • CA initiates an ecall task, it provides the address of the callback function.
  • the ecall task can include but is not limited to the following information: the function name of the called ecall function, incoming parameters, outgoing parameters, and function return value type.
  • the ecall task also includes callback functions.
  • the untrusted processing module calls the callback function through the third thread.
  • the third thread is different from the first thread.
  • the third thread is a thread created by CA in REE.
  • the REE side determines that the ecall task executed by the TEE side has ended, calls the callback function through the third thread, and then automatically completes the processing of the outgoing parameters and function return values returned by the TEE side.
  • the CA initiates the ecall task through the first thread and registers the callback function at the same time. After the CA initiates the ecall task, it does not need to wait for the task result of the ecall task executed by the TEE side. Instead, it can perform the ecall task through the first thread. Perform subsequent application business logic processing. After the ecall task is executed on the TEE side, the REE side calls the callback function through the third thread to automatically complete the processing of the outgoing parameters and function return values returned by the TEE side. By registering the callback function, the waiting time after the REE side initiates the ecall task is shortened, and the problem that the TEE side ecall task execution blocks the execution of the REE side application business logic is solved.
  • the number of third threads is less than or equal to the number of ecall tasks initiated by the CA.
  • each third thread can execute one ecall task, which improves data processing efficiency.
  • CA initiates three ecall tasks, namely: ecall task 1 to ecall task 3.
  • thread 31 calls the callback function corresponding to ecall task 1
  • thread 32 calls the callback function corresponding to ecall task 2
  • thread 33 calls the callback function corresponding to ecall task 3. This embodiment does not limit the correspondence between the third thread and the callback function of the ecall task.
  • third threads If the number of third threads is less than the number of ecall tasks initiated by CA, then there is at least one third thread, and the third thread calls callback functions corresponding to at least two ecall tasks.
  • CA initiates three ecall tasks, namely: ecall task 1 to ecall task 3.
  • thread 31 and thread 32 There are two third threads, namely thread 31 and thread 32.
  • thread 31 calls the callback functions corresponding to ecall task 1 and ecall task 2
  • thread 32 calls the callback function corresponding to ecall task 3.
  • thread 31 calls the callback function corresponding to ecall task 1
  • thread 32 calls the callback functions corresponding to ecall task 2 and ecall task 3.
  • This embodiment does not limit the correspondence between the third thread and the callback function of the ecall task.
  • the callback functions can be called sequentially in chronological order.
  • a data processing method is provided, and the execution subject is a terminal device.
  • the execution subject is a terminal device.
  • trusted execution environments and non-trusted execution environments deployed in terminal devices. This embodiment does not limit the manner in which the terminal device implements the trusted execution environment.
  • terminal equipment can adopt GP architecture or SGX architecture.
  • Applications and processing modules are deployed in both trusted execution environments and non-trusted execution environments. Among them, the applications deployed in the trusted execution environment are called trusted applications, and the processing modules are called trusted processing modules. Applications deployed in a non-trusted execution environment are called non-trusted applications, and processing modules are called non-trusted processing modules.
  • the data processing method provided by this embodiment is suitable for the scenario where the terminal device initiates a task through the first thread through the first application program in the first execution environment.
  • the first execution environment is an untrusted execution environment
  • the first application program is an untrusted application program
  • the second execution environment is a trusted execution environment
  • the second application program is a trusted application program
  • the task is an ecall task.
  • the data processing method provided in this embodiment can be applied to at least the above scenario one and scenario three.
  • this embodiment takes scenario 1 as an example to describe the data processing method provided by this embodiment.
  • FIG. 7 is another flow chart of the data processing method provided by the embodiment of the present application.
  • the data processing method provided by this embodiment may include:
  • the trusted processing module establishes a thread pool for the CA in the REE, and the thread pool includes multiple threads.
  • each thread in the thread pool is created in TEE.
  • CA initiates the ecall task through the first thread.
  • the trusted processing module allocates a second thread to the ecall task.
  • the second thread is an idle thread in the thread pool.
  • the trusted processing module obtains the first number of idle threads in the thread pool.
  • the trusted processing module obtains the second number of unexecuted ecall tasks in at least one ecall task initiated by the CA.
  • the trusted processing module adjusts the number of threads in the thread pool based on the first number and the second number.
  • a thread pool is established for the CA in the REE, and the thread pool includes multiple threads.
  • the number of threads in the thread pool is adjusted in real time to avoid having no threads available on the TEE side due to too many ecall tasks initiated by CA. situation, it also avoids CPU waste on the TEE side due to fewer ecall tasks initiated by the CA, improves the flexibility of thread management on the TEE side, improves resource utilization on the TEE side, and improves the processing efficiency of ecall tasks.
  • the trusted processing module adjusts the number of threads in the thread pool based on the first number and the second number, which may include:
  • the trusted processing module destroys a thread in the thread pool.
  • the trusted processing module creates at least one thread in the thread pool.
  • the number of at least one thread is less than or equal to the maximum threshold minus the total number of threads in the thread pool.
  • this embodiment does not limit the number of at least one thread created. However, after creating at least one thread, you need to ensure that the total number of all threads in the thread pool is less than or equal to the maximum threshold.
  • the number of at least one thread is 1. That is, a new thread is created each time.
  • the number of at least one thread is a minimum threshold.
  • the number of threads in the thread pool has a minimum value and a maximum value, which are called the minimum threshold and the maximum threshold respectively.
  • a thread is destroyed in the thread pool until the number of threads in the thread pool reaches the minimum threshold, ensuring that There are threads in the TEE for executing ecall tasks initiated by CA, which avoids the situation where the ecall task cannot be executed.
  • the number of idle threads in the thread pool is less than the number of unprocessed ecall tasks initiated by CA, at least one new thread is created in the thread pool, and the number of threads in the thread pool is gradually or rapidly increased until the thread pool The number of threads in the thread reaches the maximum threshold.
  • the trusted processing module establishes a thread pool for the CA in the REE, which may include:
  • REE send minimum threshold, maximum threshold and intermediate threshold to TEE.
  • the intermediate threshold is greater than the minimum threshold and less than the maximum threshold.
  • the trusted processing module establishes a thread pool for CA, and the number of threads in the thread pool is an intermediate threshold.
  • the information transmission channel in the existing technology can be used.
  • the TEE side establishes a thread pool for CA based on the minimum threshold, maximum threshold, and intermediate threshold set by the REE side, which improves the rationality and accuracy of managing the thread pool.
  • the data processing method provided in this embodiment may also include:
  • the untrusted processing module calls the destruction interface function.
  • the destruction interface function is used to instruct the destruction of all threads in the thread pool in TEE.
  • the trusted processing module destroys all threads in the thread pool according to the destruction interface function.
  • the various embodiments of the present application can be combined with each other, and the combination method is not limited.
  • the embodiment shown in Figure 4 and the embodiment shown in Figure 6 can be combined with each other.
  • the embodiment shown in Figure 4 and the embodiment shown in Figure 7 can be combined with each other.
  • the terminal device includes hardware and/or software modules corresponding to each function.
  • the present application can be implemented in the form of hardware or a combination of hardware and computer software. Whether a function is performed by hardware or by computer software driving the hardware depends on the specific application and design constraints of the technical solution. Those skilled in the art may use different methods to implement the described functions in conjunction with the embodiments for each specific application, but such implementations should not be considered beyond the scope of this application.
  • Embodiments of the present application can divide the terminal device into functional modules according to the above method examples. For example, each functional module can be divided corresponding to each function, or two or more functions can be integrated into one module. It should be noted that the division of modules in the embodiment of the present application is schematic and is only a logical function division. In actual implementation, there may be other division methods. It should be noted that the names of the modules in the embodiments of this application are schematic, and there are no restrictions on the names of the modules during actual implementation.
  • an embodiment of the present application provides a terminal device.
  • the terminal device deploys a trusted execution environment and a non-trusted execution environment.
  • the trusted execution environment deploys trusted applications and trusted processing modules.
  • the untrusted execution environment deploys untrusted applications and untrusted processing modules.
  • the first processing module in the first execution environment is used to store tasks initiated by the first application program through the first thread into the shared memory; the tasks are used to call software functions provided by the second application program in the second execution environment;
  • the shared memory is a physical memory area that is applied in advance for the first application program in the non-trusted execution environment and is used for access by the non-trusted execution environment and the trusted execution environment;
  • the third An execution environment is the non-trusted execution environment, the second execution environment is the trusted execution environment, the first processing module is the non-trusted processing module, and the second processing module is the trusted processing module, or the first execution environment is the trusted execution environment, the second execution environment is the non-trusted execution environment, the first processing module is the trusted processing module, and the second execution environment is the trusted execution environment.
  • the processing module is the untrusted processing module;
  • the second processing module in the second execution environment is used to obtain the task from the shared memory
  • the second processing module in the second execution environment is used to allocate a second thread to the task; the second thread is created in the second execution environment, and the second thread and the The first threads are respectively bound to different processing units;
  • the second application program in the second execution environment is configured to execute the task through the second thread, obtain a task result, and store the task result in the shared memory.
  • the first processing module in the first execution environment is used for:
  • the serialized message of the task is stored in a message queue; the message queue is stored in the shared memory and is used to store the corresponding corresponding tasks of multiple tasks initiated by the first application program. serialized message;
  • the execution flag bit of the serialized message of the task is set to a first value; wherein the first value is used to indicate that the status of the serialized message is a pending execution state.
  • the second processing module in the second execution environment is used for:
  • the serialized message of the task is parsed to obtain the task.
  • the second processing module in the second execution environment is also used to:
  • the execution flag bit of the serialized message is set to a second value; wherein the second value is used to indicate that the status of the serialized message is an executed status.
  • the first application program in the first execution environment is used for:
  • the task is initiated through the first thread, and a callback function is registered.
  • the first processing module in the first execution environment is also used to:
  • the callback function is called through a third thread; the third thread is different from the first thread.
  • the number of the third threads is less than or equal to the number of tasks initiated by the first application program.
  • the task includes incoming parameters and outgoing parameters, and the storage locations of the incoming parameters and outgoing parameters are located in the shared memory;
  • the first processing module in the first execution environment is also used to:
  • the first execution environment Before storing the task initiated by the first application program through the first thread into the shared memory, in the first execution environment, obtain the first reference address and the first reference address of the shared memory in the first execution environment. The first input address of the incoming parameter in the first execution environment and the first output address of the outgoing parameter in the first execution environment;
  • the first output offset of the address In the first execution environment, convert the first input address to a first input offset relative to the first base address, and convert the first output address to a first input offset relative to the first base address. The first output offset of the address.
  • the second processing module in the second execution environment is also used to:
  • the second execution environment obtain the second input address of the incoming parameter in the second execution environment according to the second base address and the first input offset.
  • the base address and the first output offset obtain the second output address of the outgoing parameter in the second execution environment.
  • the first execution environment is the trusted execution environment
  • the second execution environment is the non-trusted execution environment
  • the task is an ocall task.
  • the first execution environment is the untrusted execution environment
  • the second execution environment is the trusted execution environment
  • the task is an ecall task.
  • a thread pool is established for the first application in the second execution environment, the thread pool includes multiple threads, and the threads in the thread pool are in the second execution environment. created;
  • the second thread is an idle thread in the thread pool.
  • the second processing module in the second execution environment is also used to:
  • the number of threads in the thread pool is adjusted according to the first number and the second number.
  • the second processing module in the second execution environment is used for:
  • first number is greater than the second number, and the total number of threads in the thread pool is greater than the minimum threshold, then in the second execution environment, one thread is destroyed in the thread pool; or,
  • the first number is less than the second number, and the total number of threads in the thread pool is less than the maximum threshold, then in the second execution environment, at least one thread is created in the thread pool; the at least The number of one thread is less than or equal to the difference between the maximum threshold and the total number of threads in the thread pool.
  • the first processing module in the first execution environment is also used to:
  • the first execution environment Before storing the task initiated by the first application program through the first thread into the shared memory, in the first execution environment, send a minimum threshold, a maximum threshold and an intermediate threshold to the second execution environment; the intermediate The threshold is greater than the minimum threshold and less than the maximum threshold;
  • the second processing module in the second execution environment is also used to:
  • the thread pool is established for the first application program, and the number of threads in the thread pool is the intermediate threshold.
  • the first processing module in the first execution environment is also used to:
  • the related information of the shared memory is sent to the second execution environment.
  • the memory size includes a first memory size occupied by a message queue and a second memory size occupied by application data; the message queue is stored in the shared memory and is used to store messages initiated by the first application program. Serialized messages corresponding to multiple tasks;
  • the shared memory includes a first memory space of the first memory size and a second memory space of the second memory size.
  • the addresses of the first memory space and the second memory space are consecutive.
  • FIG. 8 shows another structure of a terminal device provided by an embodiment of the present application.
  • the terminal device includes: processor 801, memory 804 and bus 805.
  • a receiver 802 and a transmitter 803 may also be included.
  • the processor 801 includes one or more processing cores.
  • the processor 801 executes various functional applications and information processing by running software programs and modules.
  • the memory 804 can be used to store at least one program instruction, and the processor 801 is used to execute at least one program instruction to implement the technical solutions of the above embodiments.
  • the receiver 802 and the transmitter 803 can be implemented as a communication component, and the communication component can be a baseband chip.
  • the memory 804 is connected to the processor 801 through a bus 805. The implementation principles and technical effects are similar to the above-mentioned method-related embodiments, and will not be described again here.
  • FIG. 8 only shows one memory and processor. In an actual terminal device, there may be multiple processors and memories.
  • the memory may also be called a storage medium or a storage device, which is not limited in the embodiments of the present application.
  • the processor may be a general-purpose processor, a digital signal processor, an application-specific integrated circuit, a field programmable gate array or other programmable logic device, a discrete gate or transistor logic device, or a discrete hardware component, which may implement or Execute each method, step and logical block diagram disclosed in the embodiment of this application.
  • a general-purpose processor may be a microprocessor or any conventional processor, etc. The steps of the methods disclosed in conjunction with the embodiments of the present application can be directly implemented by a hardware processor, or executed by a combination of hardware and software modules in the processor.
  • the memory may be a non-volatile memory, such as a hard disk drive (HDD) or a solid-state drive (SS), etc., or it may be a volatile memory (volatile memory), such as Random-access memory (RAM).
  • Memory is any other medium that can be used to carry or store desired program code in the form of instructions or data structures and that can be accessed by the computer, without limitation.
  • the memory in the embodiment of the present application can also be a circuit or any other device capable of realizing the storage function, used for storing programs. sequence instructions and/or data.
  • the methods provided by the embodiments of this application can be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented using software, it may be implemented in whole or in part in the form of a computer program product.
  • the computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the processes or functions described in the embodiments of the present application are generated in whole or in part.
  • the computer may be a general-purpose computer, a special-purpose computer, a computer network, a network device, a user equipment, or other programmable device.
  • the computer instructions may be stored in or transmitted from one computer-readable storage medium to another, e.g., the computer instructions may be transferred from a website, computer, server, or data center Transmit to another website, computer, server or data center through wired (such as coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (such as infrared, wireless, microwave, etc.) means.
  • the readable storage medium can be any available media that can be accessed by a computer or a data storage device such as a server, data center, etc. that contains one or more available media integrated.
  • the available media can be magnetic media (for example, floppy disks, hard disks, tapes ), optical media (for example, digital video disc (DWD), or semiconductor media (for example, SSD), etc.).
  • Embodiments of the present application provide a computer program product.
  • the computer program product When the computer program product is run on a terminal device, it causes the terminal device to execute the technical solutions in the above embodiments.
  • the implementation principles and technical effects are similar to the above-mentioned related embodiments and will not be described again here.
  • Embodiments of the present application provide a computer-readable storage medium on which program instructions are stored.
  • program instructions When the program instructions are executed by a terminal device, they cause the terminal device to execute the technical solutions of the above embodiments.
  • the implementation principles and technical effects are similar to the above-mentioned related embodiments and will not be described again here.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Software Systems (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Computer Security & Cryptography (AREA)
  • Computer Hardware Design (AREA)
  • Mathematical Physics (AREA)
  • Storage Device Security (AREA)
  • Stored Programmes (AREA)

Abstract

本申请实施例涉及计算机技术领域，提供一种数据处理方法、终端设备和可读存储介质。终端设备部署REE和TEE。数据处理方法包括：在REE中，将由CA通过第一线程发起的ecall任务存入共享内存；在TEE中，从共享内存获取ecall任务，由TA通过第二线程执行ecall任务，将任务结果存入共享内存。共享内存是在REE中针对CA预先申请的可被REE和TEE访问，第二线程在TEE中被创建，第二线程和第一线程绑定的处理单元不同。通过预先申请共享内存，避免了数据拷贝，通过REE和TEE中执行ecall任务的线程绑定的处理单元不同，且在TEE中创建第二线程，减小了切换开销，提高了数据处理效率。

Description

数据处理方法、终端设备和可读存储介质
本申请要求于2022年09月07日提交国家知识产权局、申请号为202211091500.1、申请名称为“数据处理方法、终端设备和可读存储介质”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本申请实施例涉及计算机技术领域，尤其涉及一种数据处理方法、终端设备和可读存储介质。
背景技术
随着移动互联网的发展，终端设备的应用越来越广泛。为了保证终端设备中数据的安全性，出现了以ARM TrustZone为代表的安全框架。其中，ARM的含义为高级精简指令集处理器(advanced RISC machines)，RISC的含义为精简指令集计算机(reduced instruction set computer)，TrustZone的含义为信任区。
在ARM TrustZone框架下，终端设备的软硬件资源划分到两个世界中，称为安全世界(secure world)和正常世界(normal world)。安全世界对应的执行环境称为可信执行环境(trusted execution environment，TEE)，TEE中运行可信应用(trusted application，TA)。正常世界对应的执行环境称为富执行环境(rich execution environment，REE)，REE中运行普通应用(common application，CA)。REE和TEE相互隔离，TEE的安全要求更高。
目前，REE与TEE之间进行数据传递的开销很大，数据处理效率很低。
发明内容
本申请实施例提供一种数据处理方法、终端设备和可读存储介质，提高了数据处理效率。
第一方面，提供了一种数据处理方法，应用于终端设备，终端设备部署可信执行环境和非可信执行环境；方法包括：在第一执行环境中，将由第一应用程序通过第一线程发起的任务存入共享内存；任务用于调用第二执行环境中第二应用程序提供的软件功能；共享内存是在非可信执行环境中针对第一应用程序预先申请的且用于非可信执行环境和可信执行环境访问的物理内存区域；第一执行环境为非可信执行环境、第二执行环境为可信执行环境，或者，第一执行环境为可信执行环境、第二执行环境为非可信执行环境；在第二执行环境中，从共享内存获取任务；在第二执行环境中，为任务分配第二线程；第二线程是在第二执行环境中创建的，第二线程和第一线程分别绑定的处理单元不同；在第二执行环境中，由第二应用程序通过第二线程执行任务，得到任务结果，并将任务结果存入共享内存。
第一方面提供的数据处理方法，通过在第一执行环境中针对第一应用程序预先申请共享内存，将执行任务的相关数据等存储在共享内存中，可信执行环境和非可信执行环境通过访问共享内存完成了数据传递，不需要数据的多次拷贝，缩短了可信执行环境和非可信执行环境之间的数据传输路径，提高了数据处理的效率。而且，由于第一应用程序发起任务的第一线程和第二应用程序执行任务的第二线程分别绑定的处理单元不同，使得第一线程绑定的处理单元不需要进行CPU模式切换，减小了CPU的切换开销，提高了处理效率。而且，由于第二线程本身就在第二执行环境中被创建，使得第二线程绑定的处理单元不需要进行CPU模式切换就可以在第二执行环境中执行第一应用程序发起的任务，减小了CPU的切换开销，提高了处理效率。
一种可能的实现方式中，在第一执行环境中，将任务存入共享内存，包括：在第一执行环境中，序列化任务，得到任务的序列化消息；在第一执行环境中，将任务的序列化消息存入消息队列；消息队列存储在共享内存中，用于存储第一应用程序发起的多个任务分别对应的序列化消息；在第一执行环境中，将任务的序列化消息的执行标志位设置为第一数值；其中，第一数值用于指 示序列化消息的状态为待执行状态。
在该实现方式中，通过将任务的执行标志位设置为第一数值，指示任务尚未处理，使得第二执行环境可以快速获得尚未处理的任务，提高了处理效率。
一种可能的实现方式中，在第二执行环境中，从共享内存获取任务，包括：在第二执行环境中，访问消息队列，获取执行标志位为第一数值的任务的序列化消息；在第二执行环境中，解析任务的序列化消息，得到任务。
一种可能的实现方式中，在第二执行环境中，由第二应用程序通过第二线程执行任务，得到任务结果，并将任务结果存入共享内存之后，还包括：在第二执行环境中，将任务的序列化消息的执行标志位设置为第二数值；其中，第二数值用于指示序列化消息的状态为已执行状态。
在该实现方式中，通过将任务的执行标志位设置为第二数值，指示任务已经处理，使得第一执行环境可以快速得知任务的状态，提高了处理效率。
一种可能的实现方式中，在第一执行环境中，将由第一应用程序通过第一线程发起的任务存入共享内存之前，还包括：在第一执行环境中，由第一应用程序通过第一线程发起任务，并注册回调函数。
在该实现方式中，通过注册回调函数，第一应用程序发起任务后，不需要等待任务结果，而是可以通过第一线程进行后续应用业务逻辑的处理。缩短了第一应用程序发起任务后的等待时间，解决了第二执行环境执行任务阻塞第一应用程序业务逻辑执行的问题。
一种可能的实现方式中，方法还包括：在第一执行环境中，通过第三线程调用回调函数；第三线程和第一线程不同。
在该实现方式中，通过第三线程调用回调函数，可以自动完成对TEE侧返回的传出参数和函数返回值的处理。
一种可能的实现方式中，第三线程的数量小于或等于第一应用程序发起的任务的数量。
一种可能的实现方式中，任务包括传入参数和传出参数，传入参数和传出参数的存储位置位于共享内存中；在第一执行环境中，将由第一应用程序通过第一线程发起的任务存入共享内存之前，还包括：在第一执行环境中，获取共享内存在第一执行环境中的第一基准地址、传入参数在第一执行环境中的第一输入地址和传出参数在第一执行环境中的第一输出地址；在第一执行环境中，将第一输入地址转换为相对于第一基准地址的第一输入偏移量，将第一输出地址转换为相对于第一基准地址的第一输出偏移量。
一种可能的实现方式中，在第二执行环境中，由第二应用程序通过第二线程执行任务之前，还包括：在第二执行环境中，获取共享内存在第二执行环境中的第二基准地址；在第二执行环境中，根据第二基准地址和第一输入偏移量获取传入参数在第二执行环境中的第二输入地址，根据第二基准地址和第一输出偏移量获取传出参数在第二执行环境中的第二输出地址。
一种可能的实现方式中，第一执行环境为可信执行环境，第二执行环境为非可信执行环境，任务为ocall任务。
一种可能的实现方式中，第一执行环境为非可信执行环境，第二执行环境为可信执行环境，任务为ecall任务。
一种可能的实现方式中，在第二执行环境中针对第一应用程序建立有线程池，线程池中包括多个线程，线程池中的线程是在第二执行环境中创建的；第二线程为线程池中的空闲线程。
一种可能的实现方式中，方法还包括：在第二执行环境中，获取线程池中空闲线程的第一数量；在第二执行环境中，获取第一应用程序发起的至少一个任务中未执行的任务的第二数量；在第二执行环境中，根据第一数量和第二数量，调整线程池中线程的数量。
在该实现方式中，根据线程池中空闲线程的第一数量和尚未处理的任务的第二数量，实时调整线程池中线程的数量，避免了CPU浪费，提高了数据处理效率。
一种可能的实现方式中，在第二执行环境中，根据第一数量和第二数量，调整线程池中线程的数量，包括：若第一数量大于第二数量，且线程池中线程的总数大于最小阈值，则在第二执行环境中，在线程池中销毁一个线程；或者，若第一数量小于第二数量，且线程池中线程的总数小于最大阈值，则在第二执行环境中，在线程池中创建至少一个线程；至少一个线程的数量小于或 等于最大阈值与线程池中线程的总数之差。
一种可能的实现方式中，在第一执行环境中，将由第一应用程序通过第一线程发起的任务存入共享内存之前，还包括：在第一执行环境中，向第二执行环境发送最小阈值、最大阈值和中间阈值；中间阈值大于最小阈值且小于最大阈值；在第二执行环境中，针对第一应用程序建立线程池，线程池中线程的数量为中间阈值。
一种可能的实现方式中，在第一执行环境中，将由第一应用程序通过第一线程发起的任务存入共享内存之前，还包括：在第一执行环境中，获取共享内存的内存大小；在第一执行环境中，根据内存大小分配共享内存；在第一执行环境中，向第二执行环境发送共享内存的相关信息。
一种可能的实现方式中，内存大小包括消息队列占用的第一内存大小和应用数据占用的第二内存大小；消息队列存储在共享内存中，用于存储第一应用程序发起的多个任务分别对应的序列化消息；共享内存包括第一内存大小的第一内存空间和第二内存大小的第二内存空间。
一种可能的实现方式中，第一内存空间和第二内存空间的地址连续。
第二方面，提供了一种终端设备，终端设备部署可信执行环境和非可信执行环境。可信执行环境部署可信应用程序和可信处理模块。非可信执行环境部署非可信应用程序和非可信处理模块。第一执行环境中的第一处理模块，用于将第一应用程序通过第一线程发起的任务存入共享内存；任务用于调用第二执行环境中第二应用程序提供的软件功能；共享内存是在非可信执行环境中针对第一应用程序预先申请的且用于非可信执行环境和可信执行环境访问的物理内存区域；第一执行环境为非可信执行环境、第二执行环境为可信执行环境、第一处理模块为非可信处理模块、第二处理模块为可信处理模块，或者，第一执行环境为可信执行环境、第二执行环境为非可信执行环境、第一处理模块为可信处理模块、第二处理模块为非可信处理模块；第二执行环境中的第二处理模块，用于从共享内存获取任务；第二执行环境中的第二处理模块，用于为任务分配第二线程；第二线程是在第二执行环境中创建的，第二线程和第一线程分别绑定的处理单元不同；第二执行环境中的第二应用程序，用于通过第二线程执行任务，得到任务结果，并将任务结果存入共享内存。
一种可能的实现方式中，第一执行环境中的第一处理模块，用于：在第一执行环境中，序列化任务，得到任务的序列化消息；在第一执行环境中，将任务的序列化消息存入消息队列；消息队列存储在共享内存中，用于存储第一应用程序发起的多个任务分别对应的序列化消息；在第一执行环境中，将任务的序列化消息的执行标志位设置为第一数值；其中，第一数值用于指示序列化消息的状态为待执行状态。
一种可能的实现方式中，第二执行环境中的第二处理模块，用于：在第二执行环境中，访问消息队列，获取执行标志位为第一数值的任务的序列化消息；在第二执行环境中，解析任务的序列化消息，得到任务。
一种可能的实现方式中，第二执行环境中的第二处理模块，还用于：在第二应用程序通过第二线程执行任务，得到任务结果，并将任务结果存入共享内存之后，在第二执行环境中，将任务的序列化消息的执行标志位设置为第二数值；其中，第二数值用于指示序列化消息的状态为已执行状态。
一种可能的实现方式中，第一执行环境中的第一应用程序，用于：在第一执行环境中，通过第一线程发起任务，并注册回调函数。
一种可能的实现方式中，第一执行环境中的第一处理模块，还用于：在第一执行环境中，通过第三线程调用回调函数；第三线程和第一线程不同。
一种可能的实现方式中，第三线程的数量小于或等于第一应用程序发起的任务的数量。
一种可能的实现方式中，任务包括传入参数和传出参数，传入参数和传出参数的存储位置位于共享内存中；第一执行环境中的第一处理模块，还用于：在将第一应用程序通过第一线程发起的任务存入共享内存之前，在第一执行环境中，获取共享内存在第一执行环境中的第一基准地址、传入参数在第一执行环境中的第一输入地址和传出参数在第一执行环境中的第一输出地址；在第一执行环境中，将第一输入地址转换为相对于第一基准地址的第一输入偏移量，将第一输出地址转换为相对于第一基准地址的第一输出偏移量。
一种可能的实现方式中，第二执行环境中的第二处理模块，还用于：在第二应用程序通过第二线程执行任务之前，在第二执行环境中，获取共享内存在第二执行环境中的第二基准地址；在第二执行环境中，根据第二基准地址和第一输入偏移量获取传入参数在第二执行环境中的第二输入地址，根据第二基准地址和第一输出偏移量获取传出参数在第二执行环境中的第二输出地址。
一种可能的实现方式中，第一执行环境为可信执行环境，第二执行环境为非可信执行环境，任务为ocall任务。
一种可能的实现方式中，第一执行环境为非可信执行环境，第二执行环境为可信执行环境，任务为ecall任务。
一种可能的实现方式中，在第二执行环境中针对第一应用程序建立有线程池，线程池中包括多个线程，线程池中的线程是在第二执行环境中创建的；第二线程为线程池中的空闲线程。
一种可能的实现方式中，第二执行环境中的第二处理模块，还用于：在第二执行环境中，获取线程池中空闲线程的第一数量；在第二执行环境中，获取第一应用程序发起的至少一个任务中未执行的任务的第二数量；在第二执行环境中，根据第一数量和第二数量，调整线程池中线程的数量。
一种可能的实现方式中，第二执行环境中的第二处理模块，用于：若第一数量大于第二数量，且线程池中线程的总数大于最小阈值，则在第二执行环境中，在线程池中销毁一个线程；或者，若第一数量小于第二数量，且线程池中线程的总数小于最大阈值，则在第二执行环境中，在线程池中创建至少一个线程；至少一个线程的数量小于或等于最大阈值与线程池中线程的总数之差。
一种可能的实现方式中，第一执行环境中的第一处理模块，还用于：在将第一应用程序通过第一线程发起的任务存入共享内存之前，在第一执行环境中，向第二执行环境发送最小阈值、最大阈值和中间阈值；中间阈值大于最小阈值且小于最大阈值；第二执行环境中的第二处理模块，还用于：在第二执行环境中，针对第一应用程序建立线程池，线程池中线程的数量为中间阈值。
一种可能的实现方式中，第一执行环境中的第一处理模块，还用于：在将第一应用程序通过第一线程发起的任务存入共享内存之前，在第一执行环境中，获取共享内存的内存大小；在第一执行环境中，根据内存大小分配共享内存；在第一执行环境中，向第二执行环境发送共享内存的相关信息。
一种可能的实现方式中，内存大小包括消息队列占用的第一内存大小和应用数据占用的第二内存大小；消息队列存储在共享内存中，用于存储第一应用程序发起的多个任务分别对应的序列化消息；共享内存包括第一内存大小的第一内存空间和第二内存大小的第二内存空间。
一种可能的实现方式中，第一内存空间和第二内存空间的地址连续。
第三方面，提供一种终端设备，包括处理器，处理器用于与存储器耦合，并读取存储器中的指令并根据指令使得终端设备执行第一方面提供的方法。
第四方面，提供一种程序，该程序在被处理器执行时用于执行第一方面提供的方法。
第五方面，提供一种计算机可读存储介质，计算机可读存储介质中存储有指令，当指令在计算机或处理器上运行时，实现第一方面提供的方法。
第六方面，提供一种程序产品，所述程序产品包括计算机程序，所述计算机程序存储在可读存储介质中，设备的至少一个处理器可以从所述可读存储介质读取所述计算机程序，所述至少一个处理器执行所述计算机程序使得该设备实施第一方面提供的方法。
附图说明
图1A～图1B为本申请实施例提供的终端设备的一组架构示意图；
图1C为本申请实施例提供的终端设备的另一种架构示意图；
图2为本申请实施例提供的终端设备的又一种架构示意图；
图3A～图3D为本申请实施例提供的共享内存的一组示意图；
图4为本申请实施例提供的数据处理方法的一种流程图；
图5为本申请实施例提供的数据处理方法的另一种流程图；
图6为本申请实施例提供的数据处理方法的又一种流程图；
图7为本申请实施例提供的数据处理方法的又一种流程图；
图8为本申请实施例提供的终端设备的一种结构示意图。
具体实施方式
下面结合附图描述本申请实施例。
本申请实施例提供的数据处理方法，应用于终端设备。本申请实施例对终端设备的名称不做限定，例如，终端设备也称为：终端、电子设备、计算机设备、计算设备或计算机系统等。常见的终端设备例如包括：手机、平板电脑、笔记本电脑、台式电脑、服务器、可穿戴设备(例如智能手表或智能手环)、智能家庭设备(例如智能音响或智能电视)、车载智能设备、无人驾驶设备、虚拟现实(virtual reality，VR)设备、增强现实(augmented reality，AR)设备、混合现实(mix reality，MR)设备以及人工智能(artificial intelligence，AI)设备等。
首先，为了方便理解本申请，对相关概念进行说明。
1.安全世界、非安全世界
终端设备上运行有操作系统(operating system，OS)，OS上部署有应用程序(application，APP)，OS和APP为终端设备提供了丰富的功能。随着移动互联网的发展，APP的类型和数量越来越多，终端设备支持的功能也越来越多，对用户隐私和信息安全提出了更高的要求。
为了提高终端设备运行的安全性，可以将终端设备的软硬件资源划分到两个世界中，称为安全世界和非安全世界。安全世界和非安全世界为相对概念。在安全世界中执行需要保密的操作，存储需要保密的数据，例如，进行指纹识别、人脸识别、密码处理、数据加解密、安全认证等，而其余操作在非安全世界中执行。
需要说明，本申请实施例对安全世界、非安全世界的名称不做限定。例如，安全世界也称为安全域。非安全世界也称为非安全域、正常世界(normal world)或普通世界(common world)。在终端设备不同的体系架构中，安全世界、非安全世界的名称可能不同。
2.可信执行环境、非可信执行环境、可信应用程序、非可信应用程序
安全世界对应的执行环境称为可信执行环境，非安全世界对应的执行环境称为非可信执行环境。可信执行环境和非可信执行环境为相对概念。可信执行环境和非可信执行环境相互隔离，从而，可以在可信执行环境中执行需要保密的操作，存储需要保密的数据，保护关键程序逻辑、高价值数据等。
可信执行环境中部署有可信应用程序，非可信执行环境中部署有非可信应用程序。可信应用程序和非可信应用程序为相对概念。比如，可信应用程序包括但不限于指纹识别程序、人脸识别程序等。
需要说明，本申请实施例对可信执行环境、非可信执行环境、可信执行程序、非可信执行程序的名称不做限定。在终端设备不同的体系架构中，这些概念的名称可能不同，可以参见后续关于GP架构和Intel SGX架构的描述。
3.ecall任务、ocall任务
在非可信执行环境中发起的、用于调用可信执行环境函数接口的行为，称为入口调用(entry call，ecall)任务或ecall调用，被调用的可信执行环境的函数称为ecall函数。
相对应的，在可信执行环境中发起的、用于调用非可信执行环境函数接口的行为，称为出口调用(out call，ocall)任务或ocall调用，被调用的非可信执行环境的函数称为ocall函数。
4.程序、应用程序、进程、线程
程序(program)，通常指一段可执行的代码。
应用程序(application)，包括多个程序，可以提供特定的功能。例如，指纹识别应用程序提供指纹识别功能，人脸识别应用程序提供人脸识别功能。
进程(process)，指程序的一次运行活动，是操作系统进行资源分配的基本单位。可以理解为，一个运行起来的程序就是一个进程。进程具有自己的地址空间。一个进程可以包括一个线程， 此时，进程是单线程的。一个进程也可以包括多个线程，多个线程可以并行执行不同的任务，此时，进程是多线程的。
线程(thread)，包含在进程之中，是操作系统进行运算调度的基本单位。可以理解为，一条线程是进程中一个单一顺序的控制流。线程有时也称为轻量级进程(light weight process，LWP)。
5.CPU绑定
中央处理器(central processing unit，CPU)绑定，是指在多核CPU的系统中将进程或线程绑定到指定的CPU核上去执行。
通过CPU绑定，确保了进程或线程在指定的CPU核上运行，避免了操作系统在多个CPU核之间频繁调度切换应用程序而导致的CPU缓存失效、降低缓存命中率的问题，提高了应用程序对CPU的使用效率。
6.GP、ARM TrustZone、TEE、REE、TA、CA
全球平台组织(GlobalPlatform，GP)制定了可信执行环境的开放架构和应用编程接口(application programming interface，API)标准。在GP架构中，非安全世界称为普通世界或正常世界。安全世界对应的可信执行环境称为TEE，非安全世界对应的非可信执行环境称为REE。TEE中运行的可信应用程序称为TA，REE中运行的非可信应用程序称为CA。
其中，ARM TrustZone是基于GP标准开发的一种架构。ARM TrustZone基于硬件的安全功能，在处理器层次引入了安全世界和非安全世界，任何时刻处理器仅在其中的一个世界内运行。处理器运行在安全世界时，运行安全世界的代码；处理器运行在非安全世界时，运行非安全世界的代码。安全世界和非安全世界通过监控模式(Monitor Mode)进行转换。安全世界和非安全世界的硬件隔离，为保护应用程序的代码和数据提供了有效机制。
示例性的，图1A为本申请实施例提供的终端设备的一种架构示意图，终端设备采用GP架构或ARM TrustZone架构。如图1A所示，终端设备包括硬件层和软件层。硬件层包括但不限于：处理器和存储介质。软件层包括REE和TEE。REE侧部署有REE操作系统和CA，REE操作系统也称为富操作系统(rich operating system，Rich OS)，比如，现有的Linux操作系统或Windows操作系统。TEE侧部署有TEE操作系统和TA，TEE操作系统也称为可信操作系统(trusted operating system，Trusted OS)。
REE侧和TEE侧均包括应用态(或用户态)和内核态。REE侧的内核态包括REE操作系统，REE侧的应用态包括CA。TEE侧的内核态包括TEE操作系统，TEE侧的应用态包括TA。
CA可以发起ecall任务，调用TEE侧提供的API来完成某些安全的、受保护的业务逻辑。相反的，TA也可以发起ocall任务，调用REE侧提供的API来完成一些通常的业务逻辑。示例性的，一些应用场景为：在终端设备解锁、应用程序登陆、金融支付等场景中，CA为解锁应用或金融支付类应用，TA为指纹识别应用或者人脸识别应用。CA发起ecall任务，调用TEE侧的TA提供的指纹识别或者人脸识别功能，CA和TA二者协同，完成终端设备的解锁、应用程序的登陆、指纹支付或者人脸支付。
其中，REE和TEE运行在各自独立的物理内存上，即内存隔离。REE侧不能访问TEE侧的物理内存，反过来，TEE侧不能访问REE侧的物理内存。比如，在图1A中，REE侧对应的物理内存称为REE内存，TEE侧对应的物理内存称为TEE内存。
其中，处理器包括多个处理器核心，标记为核心0～核心n。任何时刻，处理器核心运行在REE侧或者运行在TEE侧。比如，核心0在REE中运行，处于REE模式或REE状态；核心1在TEE中运行，处于TEE模式或TEE状态。
需要说明，处理器包括多个处理单元。本申请实施例对处理器的类型和名称，处理单元的类型、名称和数量不作限定。例如，处理器为CPU，处理单元为CPU核心(或CPU核、CPU内核)。CPU核心是CPU的重要组成部分，用来完成计算、接受或存储命令、处理数据等。
存储介质，也称为存储器，用于存储可执行的程序代码和/或数据，可执行的程序代码可以包括指令。存储介质可以包括存储程序区和存储数据区。存储介质可以包括高速随机存取存储器和非易失性存储器，例如，至少一个磁盘存储器件，硬盘(hard disk drive，HDD)，固态硬盘(solid-state drive，SS)，闪存器件，通用闪存存储器(universal flash storage，UFS)等，还可以包括易失性 存储器(volatile memory)，例如随机存取存储器(random-access memory，RAM)等。
示例性的，参见图1B。当REE侧CA发起ecall任务时，CA和TA之间通过REE通信代理、硬件层的消息通道以及TEE通信代理交互数据。REE通信代理、消息通道以及TEE通信代理为CA和TA建立了安全的通信通道。具体的，CA调用TEE客户端API来和对应的TA实现通信，TA调用TEE内部API来使用TEE提供的编程资源实现相关功能。
但是，由于REE与TEE的内存隔离，导致数据传递必须依赖操作系统与内存进行拷贝传递，使得CA和TA之间的数据传递经历多次拷贝，交互开销较大。而且，发起ecall任务的CA所在的处理器核心(比如，核心0)要从REE模式切换到TEE模式，在TEE中执行TA中的ecall函数，执行完毕后，核心0再从TEE模式切换回REE模式，在REE中继续执行CA应用，CPU切换的开销较大，数据处理效率很低。
7.Intel(因特尔)软件保护执行模式(software guard eXecution，SGX)架构
Intel SGX是一种满足可信计算需求的技术，它与ARM TrustZone类似，可以实现可信执行环境。
示例性的，图1C为本申请实施例提供的终端设备的另一种架构示意图，终端设备采用Intel SGX架构或SGX架构，示出了软件层。如图1C所示，非可信执行环境包括操作系统和应用程序。操作系统比如为Linux操作系统或Windows操作系统。应用程序可以分为两个部分：安全部分(或可信部分、可信应用程序)和非安全部分(或不可信部分、非可信应用程序)。当应用程序需要处理机密数据时，它会创建一个位于可信内存的安全区(称为enclave，或可信区域、可信执行环境)，用来存放enclave代码和enclave数据，然后调用可信函数(也称为enclave函数，它是由开发者创建的、专门在安全区内执行的函数)。一旦可信函数被调用，应用程序就会在可信区域执行并明文访问安全区内的enclave代码和enclave数据，除此之外，试图从安全区外访问安全区内的行为都会被处理器拒绝。当可信函数执行完毕后，安全区内的enclave数据依然保留在可信内存中，而应用程序返回到不可信区域(或非可信执行环境)继续执行，并且失去了对可信内存的访问权限。
可见，在SGX架构中，可信执行环境是宿主进程的一部分，这意味着，应用程序包含自己的代码(称为应用代码)、数据(称为应用数据)和enclave，参见图1C中左侧部分。而enclave也包含自己的代码(称为enclave代码)和自己的数据(称为enclave数据)，参见图1C中右侧部分。SGX可以保护enclave代码和enclave数据的机密性和完整性。
为了方便说明，本申请实施例以GP架构作为示例，对本申请实施例提供的数据处理方法进行说明。
本申请实施例在图1A和图1B所示GP架构的基础上，提供了终端设备的另一种架构图。示例性的，图2为本申请实施例提供的终端设备的又一种架构示意图。在图2中，存储介质中增加了共享内存，REE的应用态增加了非可信处理模块，TEE的应用态增加了可信处理模块。其中，非可信处理模块用于在REE中处理REE或TEE中应用程序发起任务过程中的相关数据。可信处理模块用于在TEE中处理REE或TEE中应用程序发起任务过程中的相关数据。
基于图2所示的GP架构，本申请实施例提供的数据处理方法，通过预先申请共享内存，REE侧和TEE侧可以访问共享内存中的数据，完成了REE侧和TEE侧之间的数据传递，不需要数据的多次拷贝，缩短了REE和TEE之间数据传输的路径，提高了数据处理的效率。而且，REE侧和TEE侧中执行同一任务的线程绑定的处理单元不同，TEE侧中执行任务的线程在TEE侧中被创建，减小了CPU的切换开销，提高了数据处理效率。
下面，对本申请实施例中的概念进行说明。
1.第一执行环境、第一应用程序、第二执行环境、第二应用程序
在本申请实施例中，第一执行环境为非可信执行环境、第二执行环境为可信执行环境，或者，第一执行环境为可信执行环境、第二执行环境为非可信执行环境。
第一应用程序为第一执行环境中的应用程序，第二应用程序为第二执行环境中的应用程序。
2.任务
在本申请实施例中，任务是指在第一执行环境中由第一应用程序通过第一线程发起的、用于调用第二执行环境中第二应用程序提供的软件功能。
举例说明。假设，第一执行环境为REE，第一应用程序为CA，第二执行环境为TEE，第二应用程序为TA。在一个场景中，用户使用支付类APP通过指纹识别进行支付。在该场景中，CA为支付类APP，TA为指纹识别应用程序。在REE中，CA(支付类APP)运行的过程中，发起ecall任务，调用TEE中的TA(指纹识别应用程序)提供的指纹识别功能，在支付类APP中完成指纹支付。
可选的，任务包括ecall任务和ocall任务。
3.共享内存、第一内存空间、第二内存空间
在本申请实施例中，共享内存是指，在非可信执行环境中针对第一应用程序预先申请的、且用于非可信执行环境和可信执行环境访问的内存区域。
举例说明。假设，第一执行环境为REE，第一应用程序为CA，CA发起ecall任务。通常，终端设备的物理内存管理由非可信执行环境执行。在REE中，针对CA预先申请一块物理内存区域，REE和TEE均可以访问该物理内存区域。那么，当CA发起ecall任务时，可以将ecall任务的相关数据存入共享内存，从而，TEE侧可以从共享内存获取该ecall任务的相关数据，之后执行该ecall任务。相应的，在TEE侧执行ecall任务结束后，可以将ecall任务的任务结果存储在共享内存中，从而，REE侧可以从共享内存中获取该ecall任务的任务结果，完成数据传输。由于不需要数据拷贝的情况下完成了数据的传输，因此提高了数据处理的效率。
其中，共享内存中存储有任务调用的函数的传入参数和传出参数。
可选的，共享内存可以为，在非可信执行环境中针对第一应用程序发起的任务预先申请的、且用于非可信执行环境和可信执行环境访问的物理内存区域。
在该实现方式中，由于共享内存的内存大小用于执行第一应用程序发起的任务，根据第一应用程序发起的任务预先申请共享内存，提高了预先申请共享内存的合理性和准确性，提高了内存资源的利用率。
需要说明，本申请实施例对共享内存的内存大小和位置不作限定。当第一应用程序不同时，共享内存的内存大小和位置可以不同。比如，第一应用程序包括CA1和CA2。CA1的共享内存大小为5MB，起始地址为0x40000000。CA2的共享内存大小为1MB，起始地址为0x50000000。
可选的，共享内存的内存大小包括消息队列占用的第一内存大小和应用数据占用的第二内存大小。其中，消息队列用于存储第一应用程序发起的多个任务分别对应的序列化消息。应用数据包括执行第一应用程序发起的任务的过程中的交互数据。本申请实施例对应用数据的名称不作限定，例如，也可以称为交互数据、任务数据或共享数据等。
可选的，第一内存大小可以根据第一应用程序发起的任务的总数量进行估算。举例说明。假设，CA发起100个任务，每个任务在消息队列中占用的内存大小为0.05MB，那么，第一内存大小可以为100*0.05MB＝5MB。可选的，第一内存大小还可以包括防溢容量值。比如，防溢容量值为1MB，那么，CA发起100个任务时，第一内存大小为：100*0.05MB+1MB＝6MB。
可选的，如图2所示，共享内存包括第一内存空间和第二内存空间。第一内存空间的大小为第一内存大小，用于存储消息队列。第二内存空间的大小为第二内存大小，用于存储应用数据。本申请实施例对第一内存空间和第二内存空间的名称不作限定。例如，第二内存空间也称为数据缓冲区。
本申请实施例对第一内存大小和第二内存大小的取值不作限定，对第一内存空间和第二内存空间的起始位置不作限定。
可选的，为了便于内存管理，第一内存空间和第二内存空间的地址连续。
可选的，第一应用程序声明初始化配置时，可以包括下列中的至少一项：共享内存的内存大小、第一内存大小、第二内存大小和第一应用程序发起的任务的数量。在非可信执行环境中，调用非可信操作系统提供的API可以申请分配共享内存。
下面结合图3A～图3D，对共享内存进行示例性说明。假设，第一应用程序为CA。
可选的，在一种实现方式中，如图3A所示，在REE中，根据共享内存的内存大小申请一块 地址连续的物理内存区域，作为CA的共享内存。
可选的，在另一种实现方式中，如图3B所示，在REE中，根据第一内存大小申请一块地址连续的物理内存区域作为CA共享内存中的第一内存空间，用于存储消息队列；根据第二内存大小申请一块地址连续的物理内存区域作为CA共享内存中的第二内存空间，用于存储应用数据。其中，第一内存空间和第二内存空间的地址连续，且第一内存空间位于第二内存空间的地址之后。
可选的，在又一种实现方式中，如图3C所示，根据第一内存大小和第二内存大小分别申请第一内存空间和第二内存空间。图3C与图3B的相同点在于第一内存空间和第二内存空间的地址连续，区别在于，在图3C中，第一内存空间位于第二内存空间的地址之前。
可选的，在又一种实现方式中，如图3D所示，根据第一内存大小和第二内存大小分别申请第一内存空间和第二内存空间。图3D与图3B、图3C的区别在于：在图3D中，第一内存空间和第二内存空间的地址不连续。
4.消息队列、序列化、反序列化
消息队列存储在共享内存中，用于存储第一应用程序发起的多个任务分别对应的序列化消息。
本申请实施例对消息队列可以存储的任务的数量或者存储的序列化消息的数量不作限定，应用程序不同时，应用程序发起的任务的数量通常不同，消息队列的容量不同。
其中，序列化是一种数据处理机制。计算设备执行输入/输出(input/output，IO)操作时，数据要以流的形式进行读写。序列化可以将对象转换成计算设备可以识别的字节流。通过序列化，可以更方面的进行数据的存储和传输。
与序列化相反的操作称为反序列化，是指将接收到的字节流还原成程序能识别的对象。
在本申请实施例中，任务存入消息队列之前，需要对任务进行序列化，生成任务的序列化消息。相应的，从消息队列中取出任务对应的序列化消息后，需要对任务对应的序列化消息进行反序列化或者进行解析，从而恢复出任务。
5.线程池、最小阈值、中间阈值、最大阈值
在本申请实施例中，在可信执行环境中，针对非可信执行环境中的非可信应用程序建立有线程池。线程池中包括多个线程，线程池中的线程均是在可信执行环境中创建的。线程池中的线程用于执行非可信应用程序发起的ecall任务。
可选的，线程池中线程的数量具有最小值和最大值，分别称为最小阈值和最大阈值。本申请实施例对最小阈值和最大阈值的数量不作限定。通过最小阈值，确保了在可信执行环境中存在线程用于执行非可信应用程序发起的ecall任务，避免了ecall任务无法执行的情况。通过最大阈值，限制了在可信执行环境中执行非可信应用程序发起的ecall任务的线程最大值，避免了因为ecall任务较少而造成的资源浪费。
可选的，线程池中线程的数量可以根据非可信应用程序发起的ecall任务的数量在最小阈值和最大阈值之间实时调整。通过线程池中线程数量的实时调整，避免了由于非可信应用程序发起的ecall任务较多而造成的没有线程可用的情况，也避免了由于非可信应用程序发起的ecall任务较少而造成的资源浪费，提高了可信执行环境侧的资源利用率，提高了ecall任务的处理效率。
可选的，在可信执行环境中初始建立线程池时，线程池中线程的数量可以为中间阈值，其中，最小阈值<中间阈值<最大阈值。
可选的，最小阈值、最大阈值和中间阈值中的至少一项，可以为可信执行环境侧设置的，或者，为非可信执行环境侧设置的，或者，为可信执行环境侧和非可信执行环境侧相互协商后确定的。
可选的，非可信执行环境中的非可信应用程序不同时，线程池对应的最小阈值、最大阈值和中间阈值中的至少一项，可以相同，也可以不同。
下面举例说明。
假设，非可信应用程序包括CA1和CA2。在可信执行环境中，针对CA1建立线程池1，对应的最小阈值、最大阈值和中间阈值分别为2、5和10。初始建立线程时，针对CA1建立5个线程。针对CA2建立线程池2，对应的最小阈值、最大阈值和中间阈值分别为3、6和10。初始建立线程时，针对CA2建立6个线程。
6.回调函数
回调函数是指一个被作为参数传递的函数。
使用回调函数的过程可以包括：(1)定义一个回调函数；(2)提供函数实现的一方在初始化的时候，将回调函数的函数指针注册给调用者；(3)当特定事件或条件满足时，调用者使用函数指针调用回调函数对事件进行处理。
通过回调函数，可以把需要调用的函数的指针作为参数传递给一个函数，以便该函数在处理相似事件的时候可以灵活的使用不同的方法。
下面通过具体的实施例对本申请的技术方案进行详细说明。下面的实施例可以相互结合，对于相同或相似的概念或过程可能在某些实施例中不再赘述。
本申请实施例中的术语“第一”、“第二”、“第三”、“第四”等(如果存在)是用于区别类似的对象，而不必用于描述特定的顺序或先后次序。
在本申请的一个实施例中，提供了一种数据处理方法，执行主体为终端设备。终端设备中部署有可信执行环境和非可信执行环境。本实施例对终端设备实现可信执行环境的方式不作限定。比如，终端设备可以采用GP架构，也可以采用SGX架构。可信执行环境和非可信执行环境中均部署有应用程序和处理模块。其中，可信执行环境中部署的应用程序称为可信应用程序，处理模块称为可信处理模块。非可信执行环境中部署的应用程序称为非可信应用程序，处理模块称为非可信处理模块。
本实施例提供的数据处理方法，适用于终端设备在第一执行环境中由第一应用程序通过第一线程发起任务的场景。为了方便说明，第一执行环境中部署的应用程序称为第一应用程序，部署的处理模块称为第一处理模块。第二执行环境中部署的应用程序称为第二应用程序，部署的处理模块称为第二处理模块。
可选的，在一种实现方式中，第一执行环境为非可信执行环境、第一应用程序为非可信应用程序、第二执行环境为可信执行环境、第二应用程序为可信应用程序、任务为ecall任务。
可选的，在另一种实现方式中，第一执行环境为可信执行环境、第一应用程序为可信应用程序、第二执行环境为非可信执行环境、第二应用程序为非可信应用程序、任务为ocall任务。
可以理解，终端设备采用的架构不同时，第一执行环境和第二执行环境不同时，本实施例提供的数据处理方法，至少可以应用于如下场景：
场景一：终端设备采用GP架构，可以参见图2所示结构。第一执行环境为REE、第一应用程序为CA、第一处理模块为非可信处理模块、第二执行环境为TEE、第二应用程序为TA、第二处理模块为可信处理模块。终端设备在REE中由CA通过第一线程发起ecall任务。
场景二：终端设备采用GP架构，可以参见图2所示结构。第一执行环境为TEE、第一应用程序为TA、第一处理模块为可信处理模块、第二执行环境为REE、第二应用程序为CA、第二处理模块为非可信处理模块。终端设备在TEE中由TA通过第一线程发起ocall任务。
场景三：终端设备采用SGX架构，可以参见图1C所示结构。第一执行环境为非可信执行环境、第一应用程序为应用程序的非安全部分、第一处理模块为非可信处理模块、第二执行环境为Enclave、第二应用程序为应用程序的安全部分、第二处理模块为可信处理模块。终端设备在非可信执行环境中由应用程序的非安全部分通过第一线程发起ecall任务。
场景四：终端设备采用SGX架构，可以参见图1C所示结构。第一执行环境为Enclave、第一应用程序为应用程序的安全部分、第一处理模块为可信处理模块，第二执行环境为非可信执行环境、第二应用程序为应用程序的非安全部分、第二处理模块为非可信处理模块。终端设备在Enclave中由应用程序的安全部分通过第一线程发起ocall任务。
为了方便说明，本实施例以场景一作为示例，对本实施例提供的数据处理方法进行说明。
示例性的，图4为本申请实施例提供的数据处理方法的一种流程图。如图4所示，本实施例提供的数据处理方法，可以包括：
S401、在REE中，CA通过第一线程发起ecall任务。
具体的，CA发起ecall任务，调用TEE中TA提供的软件功能。第一线程是CA在REE中创 建的线程，绑定有CPU核。本实施例对第一线程绑定的CPU核不作限定。比如，CPU包括核心0～核心31，第一线程绑定的CPU核是核心0。
ecall任务可以包括但不限于下列信息：调用的ecall函数的函数名、传入参数、传出参数和函数返回值类型。
S402、在REE中，非可信处理模块将ecall任务存入共享内存。
其中，共享内存是在REE中针对CA预先申请的且用于REE和TEE访问的内存区域。共享内存的含义可以参见本申请前面的相关描述，此处不再赘述。
其中，ecall函数的传入参数和传出参数的存储位置也位于共享内存中。
由于REE和TEE都可以访问共享内存，非可信处理模块将CA发起的ecall任务存入共享内存后，TEE侧通过对共享内存进行访问，可以获取该ecall任务，从而在不需要数据拷贝的情况下完成了REE侧向TEE侧的数据传递，缩短了GP架构下的数据传输路径，提高了数据处理的效率。
S403、在TEE中，可信处理模块从共享内存获取任务。
具体的，TEE侧的可信处理模块访问共享内存，获取共享内存中存储的任务。
S404、在TEE中，可信处理模块为ecall任务分配第二线程。
其中，第二线程是在TEE中创建的，第二线程和第一线程分别绑定的处理单元不同。
可选的，处理单元为处理器的CPU核。本实施例对第二线程绑定的CPU核不作限定。比如，CPU包括核心0～核心31，第二线程绑定的CPU核是核心10。
在相关技术的一种实现方式中，由REE侧的CA创建第二线程，绑定CPU核，比如，核心10。之后，将第二线程绑定的核心10从REE侧切换到TEE侧，核心10从REE模式切换到TEE模式。
而在本实施例中，由于第二线程是在TEE侧被创建的，第二线程绑定的处理单元(核心10)初始就为TEE模式，不需要进行CPU切换，因而提高了数据处理效率。
S405、在TEE中，TA通过第二线程执行ecall任务，得到任务结果，并将任务结果存入共享内存。
具体的，第二线程可以调用TA提供的函数入口，执行ecall任务，将传出参数和函数返回值写入到ecall函数中的参数指定的共享区域。
下面对本步骤举例说明。在REE侧，CA发起ecall任务的线程为第一线程，绑定的处理单元为核心0。在TEE侧，TA执行ecall任务调用的ecall函数的线程为第二线程，绑定CPU核10。由于第二线程和第一线程绑定的CPU核不同，TA执行ecall函数时，不需要将REE侧的第一线程绑定的CPU核(核心0)切换到TEE侧，因此减少了CPU的切换开销，提高了处理效率。
而且，由于REE和TEE都可以访问共享内存，TA执行ecall任务后将任务结果存入共享内存。REE侧通过对共享内存进行访问，可以获取该ecall任务的任务结果，从而在不需要数据拷贝的情况下完成了TEE侧向REE侧的数据传递，缩短了GP架构下的数据传输路径，提高了数据处理的效率。
可见，在本实施例中，以GP架构作为示例，提供了一种数据处理方法，适用于终端设备在第一执行环境中由第一应用程序通过第一线程发起任务的场景。通过在非可信执行环境中针对第一应用程序预先申请共享内存，将执行任务的相关数据，例如，任务调用函数的传入参数和传出参数，任务结果等存储在共享内存中，可信执行环境和非可信执行环境通过访问共享内存完成了数据传递，不需要数据的多次拷贝，缩短了REE和TEE之间的数据传输路径，提高了数据处理的效率。而且，由于第一应用程序发起任务的第一线程和第二应用程序执行任务的第二线程分别绑定的处理单元不同，使得第一线程绑定的处理单元不需要进行CPU模式切换，减小了CPU的切换开销，提高了处理效率。而且，由于第二线程本身就在第二执行环境中被创建，使得第二线程绑定的处理单元不需要进行CPU模式切换就可以在第二执行环境中执行第一应用程序发起的任务，减小了CPU的切换开销，提高了处理效率。
可选的，本实施例提供的数据处理方法，在S402中非可信处理模块将ecall任务存入共享内存之前，还可以包括：
在REE中，非可信处理模块获取ecall任务的指示信息。其中，指示信息用于确定ecall任务 是否采用本实施例提供的数据处理方法。
在REE中，非可信处理模块根据ecall任务的指示信息，确定ecall任务是否采用本实施例提供的数据处理方法。
若确定ecall任务采用本实施例提供的数据处理方法，则执行后续的S403～S405。
若确定ecall任务不采用本实施例提供的数据处理方法，则根据现有的ecall任务处理流程进行处理。
可见，通过ecall任务的指示信息判断CA发起的ecall任务是否执行本实施例提供的数据处理方法，提高了本实施例提供的数据处理方法与现有ecall任务处理流程的兼容性，提高了数据处理的灵活性。
可选的，指示信息可以包括下列中的至少一项：ecall任务的配置信息或者ecall任务调用的ecall函数的关键字。
其中，函数的关键字参数是指在调用函数时，通过语法发送的参数。在本申请实施例中，通过函数调用关键字方法，可以判断CA发起的ecall任务是否执行本实施例提供的数据处理方法。
可选的，本实施例提供的数据处理方法，在S402中非可信处理模块将ecall任务存入共享内存之前，还可以包括：
在REE中，非可信处理模块获取共享内存在REE中的第一基准地址、传入参数在REE中的第一输入地址和传出参数在REE中的第一输出地址。
在REE中，非可信处理模块将第一输入地址转换为相对于第一基准地址的第一输入偏移量，将第一输出地址转换为相对于第一基准地址的第一输出偏移量。
其中，共享内存在REE中的第一基准地址，可以反映共享内存在CA地址空间的布局信息，例如，起始位置、偏移量等。ecall任务还包括第一输入偏移量和第一输出偏移量。
可选的，本实施例提供的数据处理方法，在S404中可信处理模块为任务分配第二线程之前，还可以包括：
在TEE中，可信处理模块获取共享内存在TEE中的第二基准地址。
在TEE中，可信处理模块根据第二基准地址和第一输入偏移量获取传入参数在TEE中的第二输入地址，根据第二基准地址和第一输出偏移量获取传出参数在TEE中的第二输出地址。
其中，共享内存在TEE中的第二基准地址，可以反映共享内存在TA地址空间的布局信息，例如，起始位置、偏移量等。
具体的，在GP架构中，TEE侧和REE侧的地址空间是完全隔离的。在REE侧，共享内存的地址映射由REE操作系统负责。在TEE侧，共享内存的地址映射由TEE操作系统负责。共享内存在REE侧和TEE侧会被不同的操作系统映射到CA进程地址和TA进程地址的不同区域中。这样，当CA发起ecall任务时，位于共享内存中的传入参数、传出参数等将无法被TA访问。反之亦然，TA传递的位于共享内存中的参数，也无法被CA访问。因此，在REE侧，将CA传递的位于共享内存中的参数的地址，转换为相对于共享内存基准地址的偏移量，从而在TEE侧，可以根据参数地址的偏移量转换成TEE地址空间中的地址，TA可以成功访问。反之亦然。
下面举例说明。
假设，共享内存在REE中的第一基准地址是共享内存在CA地址空间的起始地址，为0x40000000，长度为1MB。传入参数在REE中的第一输入地址为0x40001000，长度为8字节。那么，传入参数在REE中的第一输入地址相对于共享内存在REE中的第一基准地址的偏移量为：0x40001000-0x40000000＝0x1000，即，第一输入偏移量为0x1000。
可选的，在计算地址的偏移量之前，判断传入参数的地址是否会超过内存地址的范围。比如，在该示例中，传入参数的长度为8字节，第一输入地址为0x40001000，而共享内存的长度为1MB，起始地址为0x40000000，传入参数没有超过内存地址的范围。
相应的，假设，共享内存在TEE中的第二基准地址为0x32000000，长度为1MB。根据第二基准地址0x32000000以及传入参数的第一输入偏移量0x1000，可以计算得到，传入参数在TEE中的第二输入地址为：0x32000000+0x1000＝0x32001000。
可选的，在TEE侧，得到传入参数在TEE中的第二输入地址后，可以判断传入参数的地址是 否会超过内存地址的范围，确保对共享内存的访问不越界。
可选的，本实施例提供的数据处理方法，在S401中CA通过第一线程发起ecall任务之前，还可以包括：
在REE中，非可信处理模块获取共享内存的内存大小。
在REE中，非可信处理模块根据内存大小预先分配共享内存。
在REE中，向TEE发送共享内存的相关信息。
可选的，共享内存的相关信息至少包括共享内存的物理地址范围。
可选的，向TEE发送共享内存的相关信息，可以采用现有技术中的信息传输通道。例如，图1B所示架构中的REE通信代理和TEE通信代理。
可选的，本实施例提供的数据处理方法，还可以包括：
在CA完成与TA的交互后，在REE中，非可信处理模块释放共享内存。
可选的，在图4所示实施例的基础上，图5为本申请实施例提供的数据处理方法的另一种流程图，提供了S402和S403的实现方式。其中，S402包括S4021～S4023。S403包括S4031～S4032。S401、S403～S405可以参见图4所示实施例中的描述，此处不再赘述。
如图5所示，本实施例提供的数据处理方法，可以包括：
S401、在REE中，CA通过第一线程发起ecall任务。
S4021、在REE中，非可信处理模块序列化ecall任务，得到ecall任务的序列化消息。
其中，序列化的含义可以参见本申请前面的相关描述，此处不再赘述。
S4022、在REE中，非可信处理模块将ecall任务的序列化消息存入消息队列。其中，消息队列存储在共享内存中，用于存储CA发起的多个ecall任务分别对应的序列化消息。
S4023、在REE中，非可信处理模块将ecall任务的序列化消息的执行标志位设置为第一数值。
其中，第一数值用于指示消息队列中序列化消息的状态为待执行状态，以使TEE侧访问共享内存后从消息队列中获取尚未处理的ecall任务。
第一数值和第二数值为相对概念。第二数值用于指示消息队列中序列化消息的状态为已执行状态，以向REE侧通知该ecall任务执行完成。
第一数值和第二数值的取值不同。本申请实施例对第一数值和第二数值的取值不作限定。例如，第一数值为0，第二数值为1；或者，第一数值为1，第二数值为0。
S4031、在TEE中，可信处理模块访问消息队列，获取执行标志位为第一数值的ecall任务的序列化消息。
其中，本实施例对TEE侧的可信处理模块访问消息队列的规则不作限定。例如，可以通过轮询的方式，或者通过配置指定线程监控对应消息队列的方式，发现消息队列中执行标志位为第一数值的尚未处理的ecall任务。
S4032、在TEE中，可信处理模块解析ecall任务的序列化消息，得到ecall任务。
其中，解析和反序列化具有相同含义。通过对尚未处理的ecall任务的序列化消息进行解析，将序列化消息还原为ecall任务调用的ecall函数。
S404、在TEE中，可信处理模块为ecall任务分配第二线程。
S405、在TEE中，TA通过第二线程执行ecall任务，得到任务结果，并将任务结果存入共享内存。
S406、在TEE中，可信处理模块将ecall任务的序列化消息的执行标志位设置为第二数值。
可见，本实施例提供的数据处理方法，以GP架构作为示例，适用于终端设备在第一执行环境中由第一应用程序通过第一线程发起任务的场景。在非可信执行环境中针对第一应用程序预先申请共享内存，非可信执行环境和可信执行环境都可以访问共享内存，共享内存中存储有消息队列和执行任务过程中的相关数据，通过将任务的执行标志位设置为不同的数值，指示任务是否已经处理。通过预先申请共享内存，实现了非可信执行环境和可信执行环境之间的数据传递，不需要数据拷贝，缩短了数据传递路径，提高了数据处理的效率。通过设置第一应用程序发起任务的第一线程和第二应用程序执行任务的第二线程分别绑定的处理单元不同，减小了CPU的切换开销， 提高了数据处理效率。而且，由于第二线程在第二执行环境中被创建，减小了CPU的切换开销，提高了数据处理效率。
在本申请的又一个实施例中，提供了一种数据处理方法，执行主体为终端设备。终端设备中部署有可信执行环境和非可信执行环境。本实施例对终端设备实现可信执行环境的方式不作限定。比如，终端设备可以采用GP架构，也可以采用SGX架构。可信执行环境和非可信执行环境中均部署有应用程序和处理模块。其中，可信执行环境中部署的应用程序称为可信应用程序，处理模块称为可信处理模块。非可信执行环境中部署的应用程序称为非可信应用程序，处理模块称为非可信处理模块。
本实施例提供的数据处理方法，适用于终端设备在第一执行环境中由第一应用程序通过第一线程发起任务的场景。
终端设备采用的架构不同时，第一执行环境和第二执行环境不同时，本实施例提供的数据处理方法，至少可以应用于上述的场景一～场景四。
为了方便说明，本实施例以场景一作为示例，对本实施例提供的数据处理方法进行说明。
示例性的，图6为本申请实施例提供的数据处理方法的又一种流程图。如图6所示，本实施例提供的数据处理方法，可以包括：
S601、在REE中，CA通过第一线程发起ecall任务，并注册回调函数。
具体的，CA发起ecall任务，调用TEE中TA提供的软件功能。第一线程是CA在REE中创建的线程。CA在发起ecall任务时，提供回调函数的地址。
ecall任务可以包括但不限于下列信息：调用的ecall函数的函数名、传入参数、传出参数和函数返回值类型。ecall任务还包括回调函数。
S602、在REE中，非可信处理模块通过第三线程调用回调函数。其中，第三线程和第一线程不同。
具体的，第三线程是CA在REE中创建的线程。REE侧确定TEE侧执行ecall任务结束，通过第三线程调用回调函数，进而自动完成对TEE侧返回的传出参数和函数返回值的处理。
可见，本实施例提供的数据处理方法，CA通过第一线程发起ecall任务的同时注册回调函数，CA发起ecall任务后，不需要等待TEE侧执行ecall任务的任务结果，而是可以通过第一线程进行后续应用业务逻辑的处理。在TEE侧执行ecall任务结束后，REE侧通过第三线程调用回调函数，自动完成对TEE侧返回的传出参数和函数返回值的处理。通过注册回调函数，缩短了REE侧发起ecall任务后的等待时间，解决了TEE侧ecall任务执行阻塞REE侧应用业务逻辑执行的问题。
可选的，第三线程的数量小于或等于CA发起的ecall任务的数量。
下面举例说明。
如果第三线程的数量等于CA发起的ecall任务的数量，由于二者数量相等，每个第三线程可以执行一个ecall任务，提高了数据处理效率。假设，CA发起3个ecall任务，分别为：ecall任务1～ecall任务3。第三线程为3个，分别为线程31～线程33。在一种实现方式中，线程31调用ecall任务1对应的回调函数、线程32调用ecall任务2对应的回调函数、线程33调用ecall任务3对应的回调函数。本实施例对第三线程和ecall任务的回调函数之间的对应关系不作限定。
如果第三线程的数量小于CA发起的ecall任务的数量，那么，存在至少一个第三线程，该第三线程调用至少2个ecall任务对应的回调函数。假设，CA发起3个ecall任务，分别为：ecall任务1～ecall任务3。第三线程为2个，分别为线程31、线程32。在一种实现方式中，线程31调用ecall任务1和ecall任务2对应的回调函数、线程32调用ecall任务3对应的回调函数。在另一种实现方式中，线程31调用ecall任务1对应的回调函数、线程32调用ecall任务2和ecall任务3对应的回调函数。本实施例对第三线程和ecall任务的回调函数之间的对应关系不作限定。可选的，当一个第三线程调用至少2个ecall任务对应的回调函数时，可以按照时间顺序依次调用回调函数。
在本申请的又一个实施例中，提供了一种数据处理方法，执行主体为终端设备。终端设备中部署有可信执行环境和非可信执行环境。本实施例对终端设备实现可信执行环境的方式不作限定。比如，终端设备可以采用GP架构，也可以采用SGX架构。可信执行环境和非可信执行环境中均部署有应用程序和处理模块。其中，可信执行环境中部署的应用程序称为可信应用程序，处理模块称为可信处理模块。非可信执行环境中部署的应用程序称为非可信应用程序，处理模块称为非可信处理模块。
本实施例提供的数据处理方法，适用于终端设备在第一执行环境中由第一应用程序通过第一线程发起任务的场景。其中，第一执行环境为非可信执行环境、第一应用程序为非可信应用程序、第二执行环境为可信执行环境、第二应用程序为可信应用程序、任务为ecall任务。
终端设备采用的架构不同时，本实施例提供的数据处理方法，至少可以应用于上述的场景一和场景三。
为了方便说明，本实施例以场景一作为示例，对本实施例提供的数据处理方法进行说明。
示例性的，图7为本申请实施例提供的数据处理方法的又一种流程图。如图7所示，本实施例提供的数据处理方法，可以包括：
S701、在TEE中，可信处理模块针对REE中的CA建立线程池，线程池中包括多个线程。
其中，线程池中的每个线程均是在TEE中创建的。
其中，线程池的含义可以参见本申请前面的相关描述，此处不再赘述。
S401、在REE中，CA通过第一线程发起ecall任务。
S404、在TEE中，可信处理模块为ecall任务分配第二线程。
其中，第二线程为线程池中的空闲线程。
其中，S401和S404可以参见图4所示实施例，此处不再赘述。
S702、在TEE中，可信处理模块获取线程池中空闲线程的第一数量。
S703、在TEE中，可信处理模块获取CA发起的至少一个ecall任务中未执行的ecall任务的第二数量。
S704、在TEE中，可信处理模块根据第一数量和第二数量，调整线程池中线程的数量。
本实施例提供的数据处理方法，在TEE中，针对REE中的CA建立有线程池，线程池中包括多个线程。根据线程池中空闲线程的第一数量和CA发起的尚未处理的ecall任务的第二数量，实时调整线程池中线程的数量，避免了由于CA发起的ecall任务较多而造成TEE侧没有线程可用的情况，也避免了由于CA发起的ecall任务较少而造成TEE侧的CPU浪费，提高了TEE侧线程管理的灵活性，提高了TEE侧的资源利用率，也提高了ecall任务的处理效率。
可选的，S704中，在TEE中，可信处理模块根据第一数量和第二数量，调整线程池中线程的数量，可以包括：
若第一数量大于第二数量，且线程池中线程的总数大于最小阈值，则在TEE中，可信处理模块在线程池中销毁一个线程。或者，
若第一数量小于第二数量，且线程池中线程的总数小于最大阈值，则在TEE中，可信处理模块在线程池中创建至少一个线程。至少一个线程的数量小于或等于最大阈值与线程池中线程的总数之差。
其中，本实施例对创建的至少一个线程的数量不做限定。但是，创建至少一个线程后需要保证线程池中所有线程的总数小于或等于最大阈值。
可选的，在一种实现方式中，至少一个线程的数量为1。即，每次新创建1个线程。
可选的，在另一种实现方式中，至少一个线程的数量为最小阈值。
举例说明。假设，第一数量为3、第二数量为5、最大阈值为10、最小阈值为2、线程池中线程的总数当前为7。那么，可信处理模块可以在线程池中创建最小阈值个线程，即创建2个线程。创建2个线程后，线程池中线程的总数更新为7+2＝9个。之后继续判断第一数量和第二数量的大小。假设，第一数量为2、第二数量为3、线程池中线程的总数当前为9。那么，如果可信处理模块在线程池中创建最小阈值个线程，线程池中线程的总数将更新为9+2＝11，超过了最大阈值10。所以，可信处理模块可以在线程池中创建1个线程。创建1个线程后，线程池中线程的总数更新 为9+1＝10个，达到最大阈值。
其中，线程池中线程的数量具有最小值和最大值，分别称为最小阈值和最大阈值。
具体的，当线程池中空闲线程的数量较多，大于CA发起的尚未处理的ecall任务的数量时，则在线程池中销毁一个线程，直至线程池中线程的数量达到最小阈值，确保了在TEE中存在线程用于执行CA发起的ecall任务，避免了ecall任务无法执行的情况。
当线程池中空闲线程的数量较少，小于CA发起的尚未处理的ecall任务的数量时，则在线程池中新创建至少一个线程，逐步或快速的增加线程池中线程的数量，直至线程池中线程的数量达到最大阈值。通过限制TEE中执行CA发起的ecall任务的线程最大值，避免了TEE侧的资源浪费。
可选的，S701中，可信处理模块针对REE中的CA建立线程池，可以包括：
在REE中，向TEE发送最小阈值、最大阈值和中间阈值。中间阈值大于最小阈值且小于最大阈值。
在TEE中，可信处理模块针对CA建立线程池，线程池中线程的数量为中间阈值。
可选的，向TEE发送最小阈值、最大阈值和中间阈值，可以采用现有技术中的信息传输通道。例如，图1B所示架构中的REE通信代理和TEE通信代理。
在本实施方式中，TEE侧根据REE侧设置的最小阈值、最大阈值和中间阈值建立针对CA的线程池，提高了管理线程池的合理性和准确性。
可选的，本实施例提供的数据处理方法，还可以包括：
在CA完成与TA的交互后，在REE中，非可信处理模块调用销毁接口函数。销毁接口函数用于指示在TEE中销毁线程池中的所有线程。
在TEE中，可信处理模块根据销毁接口函数，销毁线程池中的所有线程。
需要说明的是，本申请的各个实施例之间可以相互结合，对结合方式不作限定。比如，图4所示的实施例和图6所示的实施例可以相互结合。又比如，当第一执行环境为非可信执行环境、第二执行环境为可信执行环境时，图4所示实施例和图7所示实施例可以相互结合。
可以理解的是，终端设备为了实现上述功能，其包含了执行各个功能相应的硬件和/或软件模块。结合本文中所公开的实施例描述的各示例的算法步骤，本申请能够以硬件或硬件和计算机软件的结合形式来实现。某个功能究竟以硬件还是计算机软件驱动硬件的方式来执行，取决于技术方案的特定应用和设计约束条件。本领域技术人员可以结合实施例对每个特定的应用来使用不同方法来实现所描述的功能，但是这种实现不应认为超出本申请的范围。
本申请实施例可以根据上述方法示例对终端设备进行功能模块的划分，例如，可以对应各个功能划分各个功能模块，也可以将两个或两个以上的功能集成在一个模块中。需要说明的是，本申请实施例中对模块的划分是示意性的，仅仅为一种逻辑功能划分，实际实现时可以有另外的划分方式。需要说明的是，本申请实施例中模块的名称是示意性的，实际实现时对模块的名称不做限定。
如图2所示，本申请实施例提供一种终端设备，终端设备部署可信执行环境和非可信执行环境。可信执行环境部署可信应用程序和可信处理模块。非可信执行环境部署非可信应用程序和非可信处理模块。
第一执行环境中的第一处理模块，用于将第一应用程序通过第一线程发起的任务存入共享内存；所述任务用于调用第二执行环境中第二应用程序提供的软件功能；所述共享内存是在所述非可信执行环境中针对所述第一应用程序预先申请的且用于所述非可信执行环境和所述可信执行环境访问的物理内存区域；所述第一执行环境为所述非可信执行环境、所述第二执行环境为所述可信执行环境、所述第一处理模块为所述非可信处理模块、第二处理模块为所述可信处理模块，或者，所述第一执行环境为所述可信执行环境、所述第二执行环境为所述非可信执行环境、所述第一处理模块为所述可信处理模块、第二处理模块为所述非可信处理模块；
所述第二执行环境中的所述第二处理模块，用于从所述共享内存获取所述任务；
所述第二执行环境中的所述第二处理模块，用于为所述任务分配第二线程；所述第二线程是在所述第二执行环境中创建的，所述第二线程和所述第一线程分别绑定的处理单元不同；
所述第二执行环境中的所述第二应用程序，用于通过所述第二线程执行所述任务，得到任务结果，并将所述任务结果存入所述共享内存。
可选的，所述第一执行环境中的所述第一处理模块，用于：
在所述第一执行环境中，序列化所述任务，得到所述任务的序列化消息；
在所述第一执行环境中，将所述任务的序列化消息存入消息队列；所述消息队列存储在所述共享内存中，用于存储所述第一应用程序发起的多个任务分别对应的序列化消息；
在所述第一执行环境中，将所述任务的序列化消息的执行标志位设置为第一数值；其中，第一数值用于指示序列化消息的状态为待执行状态。
可选的，所述第二执行环境中的所述第二处理模块，用于：
在所述第二执行环境中，访问所述消息队列，获取执行标志位为所述第一数值的所述任务的序列化消息；
在所述第二执行环境中，解析所述任务的序列化消息，得到所述任务。
可选的，所述第二执行环境中的所述第二处理模块，还用于：
在所述第二应用程序通过所述第二线程执行所述任务，得到任务结果，并将所述任务结果存入所述共享内存之后，在所述第二执行环境中，将所述任务的序列化消息的所述执行标志位设置为第二数值；其中，第二数值用于指示序列化消息的状态为已执行状态。
可选的，所述第一执行环境中的所述第一应用程序，用于：
在所述第一执行环境中，通过所述第一线程发起所述任务，并注册回调函数。
可选的，所述第一执行环境中的所述第一处理模块，还用于：
在所述第一执行环境中，通过第三线程调用所述回调函数；所述第三线程和所述第一线程不同。
可选的，所述第三线程的数量小于或等于所述第一应用程序发起的任务的数量。
可选的，所述任务包括传入参数和传出参数，所述传入参数和所述传出参数的存储位置位于所述共享内存中；
所述第一执行环境中的所述第一处理模块，还用于：
在将所述第一应用程序通过第一线程发起的任务存入共享内存之前，在所述第一执行环境中，获取所述共享内存在所述第一执行环境中的第一基准地址、所述传入参数在所述第一执行环境中的第一输入地址和所述传出参数在所述第一执行环境中的第一输出地址；
在所述第一执行环境中，将所述第一输入地址转换为相对于所述第一基准地址的第一输入偏移量，将所述第一输出地址转换为相对于所述第一基准地址的第一输出偏移量。
可选的，所述第二执行环境中的所述第二处理模块，还用于：
在所述第二应用程序通过所述第二线程执行所述任务之前，在所述第二执行环境中，获取所述共享内存在所述第二执行环境中的第二基准地址；
在所述第二执行环境中，根据所述第二基准地址和所述第一输入偏移量获取所述传入参数在所述第二执行环境中的第二输入地址，根据所述第二基准地址和所述第一输出偏移量获取所述传出参数在所述第二执行环境中的第二输出地址。
可选的，所述第一执行环境为所述可信执行环境，所述第二执行环境为所述非可信执行环境，所述任务为ocall任务。
可选的，所述第一执行环境为所述非可信执行环境，所述第二执行环境为所述可信执行环境，所述任务为ecall任务。
可选的，在所述第二执行环境中针对所述第一应用程序建立有线程池，所述线程池中包括多个线程，所述线程池中的线程是在所述第二执行环境中创建的；
所述第二线程为所述线程池中的空闲线程。
可选的，所述第二执行环境中的所述第二处理模块，还用于：
在所述第二执行环境中，获取所述线程池中空闲线程的第一数量；
在所述第二执行环境中，获取所述第一应用程序发起的至少一个任务中未执行的任务的第二数量；
在所述第二执行环境中，根据所述第一数量和所述第二数量，调整所述线程池中线程的数量。
可选的，所述第二执行环境中的所述第二处理模块，用于：
若所述第一数量大于所述第二数量，且所述线程池中线程的总数大于最小阈值，则在所述第二执行环境中，在所述线程池中销毁一个线程；或者，
若所述第一数量小于所述第二数量，且所述线程池中线程的总数小于最大阈值，则在所述第二执行环境中，在所述线程池中创建至少一个线程；所述至少一个线程的数量小于或等于所述最大阈值与所述线程池中线程的总数之差。
可选的，所述第一执行环境中的所述第一处理模块，还用于：
在将所述第一应用程序通过第一线程发起的任务存入共享内存之前，在所述第一执行环境中，向所述第二执行环境发送最小阈值、最大阈值和中间阈值；所述中间阈值大于所述最小阈值且小于所述最大阈值；
所述第二执行环境中的所述第二处理模块，还用于：
在所述第二执行环境中，针对所述第一应用程序建立所述线程池，所述线程池中线程的数量为所述中间阈值。
可选的，所述第一执行环境中的所述第一处理模块，还用于：
在将所述第一应用程序通过第一线程发起的任务存入共享内存之前，在所述第一执行环境中，获取所述共享内存的内存大小；
在所述第一执行环境中，根据所述内存大小分配所述共享内存；
在所述第一执行环境中，向所述第二执行环境发送所述共享内存的相关信息。
可选的，所述内存大小包括消息队列占用的第一内存大小和应用数据占用的第二内存大小；所述消息队列存储在所述共享内存中，用于存储所述第一应用程序发起的多个任务分别对应的序列化消息；
所述共享内存包括所述第一内存大小的第一内存空间和所述第二内存大小的第二内存空间。
可选的，所述第一内存空间和所述第二内存空间的地址连续。
请参考图8，其示出了本申请实施例提供的终端设备的另一种结构。终端设备包括：处理器801、存储器804和总线805。可选的，还可以包括接收器802和发射器803。处理器801包括一个或者多个处理核心，处理器801通过运行软件程序以及模块，从而执行各种功能的应用以及信息处理。存储器804可用于存储至少一个程序指令，处理器801用于执行至少一个程序指令，以实现上述实施例的技术方案。接收器802和发射器803可以实现为一个通信组件，该通信组件可以是一块基带芯片。存储器804通过总线805和处理器801相连。其实现原理和技术效果与上述方法相关实施例类似，此处不再赘述。
本领域技术人员可以理解，为了便于说明，图8仅示出了一个存储器和处理器。在实际的终端设备中，可以存在多个处理器和存储器。存储器也可以称为存储介质或者存储设备等，本申请实施例对此不做限制。
在本申请实施例中，处理器可以是通用处理器、数字信号处理器、专用集成电路、现场可编程门阵列或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件，可以实现或者执行本申请实施例中的公开的各方法、步骤及逻辑框图。通用处理器可以是微处理器或者任何常规的处理器等。结合本申请实施例所公开的方法的步骤可以直接体现为硬件处理器执行完成，或者用处理器中的硬件及软件模块组合执行完成。
在本申请实施例中，存储器可以是非易失性存储器，比如硬盘(hard disk drive，HDD)或固态硬盘(solid-state drive，SS)等，还可以是易失性存储器(volatile memory)，例如随机存取存储器(random-access memory，RAM)。存储器是能够用于携带或存储具有指令或数据结构形式的期望的程序代码并能够由计算机存取的任何其他介质，不限于此。
本申请实施例中的存储器还可以是电路或者其它任意能够实现存储功能的装置，用于存储程 序指令和/或数据。本申请各实施例提供的方法中，可以全部或部分地通过软件、硬件、固件或者其任意组合来实现。当使用软件实现时，可以全部或部分地以计算机程序产品的形式实现。所述计算机程序产品包括一个或多个计算机指令。在计算机上加载和执行所述计算机程序指令时，全部或部分地产生按照本申请实施例所述的流程或功能。所述计算机可以是通用计算机、专用计算机、计算机网络、网络设备、用户设备或者其他可编程装置。所述计算机指令可以存储在计算机可读存储介质中，或者从一个计算机可读存储介质向另一个计算机可读存储介质传输，例如，所述计算机指令可以从一个网站站点、计算机、服务器或数据中心通过有线(例如同轴电缆、光纤、数字用户线(digital subscriber line，DSL)或无线(例如红外、无线、微波等)方式向另一个网站站点、计算机、服务器或数据中心进行传输。所述计算机可读存储介质可以是计算机可以存取的任何可用介质或者是包含一个或多个可用介质集成的服务器、数据中心等数据存储设备。所述可用介质可以是磁性介质(例如，软盘、硬盘、磁带)、光介质(例如，数字视频光盘(digital video disc，DWD)、或者半导体介质(例如，SSD)等。
本申请实施例提供一种计算机程序产品，当所述计算机程序产品在终端设备运行时，使得所述终端设备执行上述实施例中的技术方案。其实现原理和技术效果与上述相关实施例类似，此处不再赘述。
本申请实施例提供一种计算机可读存储介质，其上存储有程序指令，所述程序指令被终端设备执行时，使得所述终端设备执行上述实施例的技术方案。其实现原理和技术效果与上述相关实施例类似，此处不再赘述。
综上所述，以上实施例仅用以说明本申请的技术方案，而非对其限制；尽管参照前述实施例对本申请进行了详细的说明，本领域的普通技术人员应当理解：其依然可以对前述各实施例所记载的技术方案进行修改，或者对其中部分技术特征进行等同替换；而这些修改或者替换，并不使相应技术方案的本质脱离本申请各实施例技术方案的范围。

Claims (20)

  1. 一种数据处理方法，其特征在于，应用于终端设备，所述终端设备部署可信执行环境和非可信执行环境；所述方法包括：
    在第一执行环境中，将由第一应用程序通过第一线程发起的任务存入共享内存；所述任务用于调用第二执行环境中第二应用程序提供的软件功能；所述共享内存是在所述非可信执行环境中针对所述第一应用程序预先申请的且用于所述非可信执行环境和所述可信执行环境访问的物理内存区域；所述第一执行环境为所述非可信执行环境、所述第二执行环境为所述可信执行环境，或者，所述第一执行环境为所述可信执行环境、所述第二执行环境为所述非可信执行环境；
    在所述第二执行环境中，从所述共享内存获取所述任务；
    在所述第二执行环境中，为所述任务分配第二线程；所述第二线程是在所述第二执行环境中创建的，所述第二线程和所述第一线程分别绑定的处理单元不同；
    在所述第二执行环境中，由所述第二应用程序通过所述第二线程执行所述任务，得到任务结果，并将所述任务结果存入所述共享内存。
  2. 根据权利要求1所述的方法，其特征在于，所述在所述第一执行环境中，将所述任务存入共享内存，包括：
    在所述第一执行环境中，序列化所述任务，得到所述任务的序列化消息；
    在所述第一执行环境中，将所述任务的序列化消息存入消息队列；所述消息队列存储在所述共享内存中，用于存储所述第一应用程序发起的多个任务分别对应的序列化消息；
    在所述第一执行环境中，将所述任务的序列化消息的执行标志位设置为第一数值；其中，第一数值用于指示序列化消息的状态为待执行状态。
  3. 根据权利要求2所述的方法，其特征在于，所述在所述第二执行环境中，从所述共享内存获取所述任务，包括：
    在所述第二执行环境中，访问所述消息队列，获取执行标志位为所述第一数值的所述任务的序列化消息；
    在所述第二执行环境中，解析所述任务的序列化消息，得到所述任务。
  4. 根据权利要求2所述的方法，其特征在于，所述在所述第二执行环境中，由所述第二应用程序通过所述第二线程执行所述任务，得到任务结果，并将所述任务结果存入所述共享内存之后，还包括：
    在所述第二执行环境中，将所述任务的序列化消息的所述执行标志位设置为第二数值；其中，第二数值用于指示序列化消息的状态为已执行状态。
  5. 根据权利要求1-4中任一项所述的方法，其特征在于，所述在第一执行环境中，将由第一应用程序通过第一线程发起的任务存入共享内存之前，还包括：
    在所述第一执行环境中，由所述第一应用程序通过所述第一线程发起所述任务，并注册回调函数。
  6. 根据权利要求5所述的方法，其特征在于，所述方法还包括：
    在所述第一执行环境中，通过第三线程调用所述回调函数；所述第三线程和所述第一线程不同。
  7. 根据权利要求6所述的方法，其特征在于，所述第三线程的数量小于或等于所述第一应用程序发起的任务的数量。
  8. 根据权利要求1-7中任一项所述的方法，其特征在于，所述任务包括传入参数和传出参数，所述传入参数和所述传出参数的存储位置位于所述共享内存中；
    所述在第一执行环境中，将由第一应用程序通过第一线程发起的任务存入共享内存之前，还包括：
    在所述第一执行环境中，获取所述共享内存在所述第一执行环境中的第一基准地址、所述传入参数在所述第一执行环境中的第一输入地址和所述传出参数在所述第一执行环境中的第一输出地址；
    在所述第一执行环境中，将所述第一输入地址转换为相对于所述第一基准地址的第一输入偏 移量，将所述第一输出地址转换为相对于所述第一基准地址的第一输出偏移量。
  9. 根据权利要求8所述的方法，其特征在于，所述在所述第二执行环境中，由所述第二应用程序通过所述第二线程执行所述任务之前，还包括：
    在所述第二执行环境中，获取所述共享内存在所述第二执行环境中的第二基准地址；
    在所述第二执行环境中，根据所述第二基准地址和所述第一输入偏移量获取所述传入参数在所述第二执行环境中的第二输入地址，根据所述第二基准地址和所述第一输出偏移量获取所述传出参数在所述第二执行环境中的第二输出地址。
  10. 根据权利要求1-9中任一项所述的方法，其特征在于，所述第一执行环境为所述可信执行环境，所述第二执行环境为所述非可信执行环境，所述任务为ocall任务。
  11. 根据权利要求1-9中任一项所述的方法，其特征在于，所述第一执行环境为所述非可信执行环境，所述第二执行环境为所述可信执行环境，所述任务为ecall任务。
  12. 根据权利要求11所述的方法，其特征在于，在所述第二执行环境中针对所述第一应用程序建立有线程池，所述线程池中包括多个线程，所述线程池中的线程是在所述第二执行环境中创建的；
    所述第二线程为所述线程池中的空闲线程。
  13. 根据权利要求12所述的方法，其特征在于，所述方法还包括：
    在所述第二执行环境中，获取所述线程池中空闲线程的第一数量；
    在所述第二执行环境中，获取所述第一应用程序发起的至少一个任务中未执行的任务的第二数量；
    在所述第二执行环境中，根据所述第一数量和所述第二数量，调整所述线程池中线程的数量。
  14. 根据权利要求13所述的方法，其特征在于，所述在所述第二执行环境中，根据所述第一数量和所述第二数量，调整所述线程池中线程的数量，包括：
    若所述第一数量大于所述第二数量，且所述线程池中线程的总数大于最小阈值，则在所述第二执行环境中，在所述线程池中销毁一个线程；或者，
    若所述第一数量小于所述第二数量，且所述线程池中线程的总数小于最大阈值，则在所述第二执行环境中，在所述线程池中创建至少一个线程；所述至少一个线程的数量小于或等于所述最大阈值与所述线程池中线程的总数之差。
  15. 根据权利要求12所述的方法，其特征在于，所述在第一执行环境中，将由第一应用程序通过第一线程发起的任务存入共享内存之前，还包括：
    在所述第一执行环境中，向所述第二执行环境发送最小阈值、最大阈值和中间阈值；所述中间阈值大于所述最小阈值且小于所述最大阈值；
    在所述第二执行环境中，针对所述第一应用程序建立所述线程池，所述线程池中线程的数量为所述中间阈值。
  16. 根据权利要求11所述的方法，其特征在于，所述在第一执行环境中，将由第一应用程序通过第一线程发起的任务存入共享内存之前，还包括：
    在所述第一执行环境中，获取所述共享内存的内存大小；
    在所述第一执行环境中，根据所述内存大小分配所述共享内存；
    在所述第一执行环境中，向所述第二执行环境发送所述共享内存的相关信息。
  17. 根据权利要求16所述的方法，其特征在于，所述内存大小包括消息队列占用的第一内存大小和应用数据占用的第二内存大小；所述消息队列存储在所述共享内存中，用于存储所述第一应用程序发起的多个任务分别对应的序列化消息；
    所述共享内存包括所述第一内存大小的第一内存空间和所述第二内存大小的第二内存空间。
  18. 根据权利要求17所述的方法，其特征在于，所述第一内存空间和所述第二内存空间的地址连续。
  19. 一种终端设备，其特征在于，包括处理器，所述处理器用于与存储器耦合，并读取存储器中的指令并根据所述指令使得所述终端设备执行如权利要求1-18中任一项所述的方法。
  20. 一种计算机可读存储介质，其特征在于，所述计算机可读存储介质存储有计算机指令，当 所述计算机指令在终端设备上运行时，使得所述终端设备执行如权利要求1-18中任一项所述的方法。
PCT/CN2023/103965 2022-09-07 2023-06-29 数据处理方法、终端设备和可读存储介质 Ceased WO2024051311A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
EP23862012.4A EP4575861A4 (en) 2022-09-07 2023-06-29 DATA PROCESSING METHOD, TERMINAL DEVICE AND READABLE STORAGE MEDIA
US19/071,949 US20250238269A1 (en) 2022-09-07 2025-03-06 Data processing method, terminal device, and readable storage medium

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202211091500.1 2022-09-07
CN202211091500.1A CN117707799A (zh) 2022-09-07 2022-09-07 数据处理方法、终端设备和可读存储介质

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US19/071,949 Continuation US20250238269A1 (en) 2022-09-07 2025-03-06 Data processing method, terminal device, and readable storage medium

Publications (1)

Publication Number Publication Date
WO2024051311A1 true WO2024051311A1 (zh) 2024-03-14

Family

ID=90157553

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2023/103965 Ceased WO2024051311A1 (zh) 2022-09-07 2023-06-29 数据处理方法、终端设备和可读存储介质

Country Status (4)

Country Link
US (1) US20250238269A1 (zh)
EP (1) EP4575861A4 (zh)
CN (1) CN117707799A (zh)
WO (1) WO2024051311A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN118409751A (zh) * 2024-04-12 2024-07-30 北京天数智芯半导体科技有限公司 Ai加速卡计算误差自动分析方法、系统、装置及设备

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20180288095A1 (en) * 2017-03-29 2018-10-04 At&T Intellectual Property I, L.P. Method and system to secure and dynamically share iot information cross multiple platforms in 5g network
CN109460373A (zh) * 2017-09-06 2019-03-12 阿里巴巴集团控股有限公司 一种数据共享方法、终端设备和存储介质
CN110442462A (zh) * 2019-07-16 2019-11-12 阿里巴巴集团控股有限公司 Tee系统中的多线程数据传输方法和装置
CN111585999A (zh) * 2020-04-28 2020-08-25 深圳震有科技股份有限公司 一种数据转换方法、系统及存储介质
CN111859395A (zh) * 2020-07-21 2020-10-30 中国人民解放军国防科技大学 具备tee扩展的计算平台上的通信优化方法及系统
CN111858004A (zh) * 2020-07-21 2020-10-30 中国人民解放军国防科技大学 基于tee扩展的计算机安全世界实时应用动态加载方法及系统

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109960582B (zh) * 2018-06-19 2020-04-28 华为技术有限公司 在tee侧实现多核并行的方法、装置及系统
DE102018132970A1 (de) * 2018-10-10 2020-04-16 Bayerische Motoren Werke Aktiengesellschaft Verfahren und Vorrichtung zur Isolation von sensiblem nichtvertrauenswürdigem Programmcode auf mobilen Endgeräten
CN114661380B (zh) * 2020-12-31 2024-10-01 支付宝(杭州)信息技术有限公司 面向可信执行环境的系统调用处理方法及装置
CN113138845B (zh) * 2021-04-25 2024-10-22 北京小米移动软件有限公司 指纹识别方法、装置、终端及存储介质

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20180288095A1 (en) * 2017-03-29 2018-10-04 At&T Intellectual Property I, L.P. Method and system to secure and dynamically share iot information cross multiple platforms in 5g network
CN109460373A (zh) * 2017-09-06 2019-03-12 阿里巴巴集团控股有限公司 一种数据共享方法、终端设备和存储介质
CN110442462A (zh) * 2019-07-16 2019-11-12 阿里巴巴集团控股有限公司 Tee系统中的多线程数据传输方法和装置
CN111585999A (zh) * 2020-04-28 2020-08-25 深圳震有科技股份有限公司 一种数据转换方法、系统及存储介质
CN111859395A (zh) * 2020-07-21 2020-10-30 中国人民解放军国防科技大学 具备tee扩展的计算平台上的通信优化方法及系统
CN111858004A (zh) * 2020-07-21 2020-10-30 中国人民解放军国防科技大学 基于tee扩展的计算机安全世界实时应用动态加载方法及系统

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP4575861A4

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN118409751A (zh) * 2024-04-12 2024-07-30 北京天数智芯半导体科技有限公司 Ai加速卡计算误差自动分析方法、系统、装置及设备

Also Published As

Publication number Publication date
EP4575861A4 (en) 2025-11-26
EP4575861A1 (en) 2025-06-25
US20250238269A1 (en) 2025-07-24
CN117707799A (zh) 2024-03-15

Similar Documents

Publication Publication Date Title
US7200695B2 (en) Method, system, and program for processing packets utilizing descriptors
US20220091911A1 (en) Method and apparatus for inter-process communication, and computer device
US11507285B1 (en) Systems and methods for providing high-performance access to shared computer memory via different interconnect fabrics
US11847225B2 (en) Blocking access to firmware by units of system on chip
CN113495795B (zh) 一种进程间通信方法及相关设备
US8244826B2 (en) Providing a memory region or memory window access notification on a system area network
US8131814B1 (en) Dynamic pinning remote direct memory access
US10104005B2 (en) Data buffering
EP1861778B1 (en) Data processing system
US20050038941A1 (en) Method and apparatus for accessing a memory
WO2022228485A1 (zh) 一种数据传输方法、数据处理方法及相关产品
US20120131124A1 (en) Rdma read destination buffers mapped onto a single representation
EP4421631A1 (en) Data migration method and apparatus, and electronic device
US11010309B2 (en) Computer system and method for executing one or more software applications, host computer device and method for a host computer device, memory device and method for a memory device and non-transitory computer readable medium
CN105009134B (zh) 提供安全操作的方法、装置、系统和计算机可读介质
US20230168953A1 (en) Inter-process communication method and apparatus
CN113591091A (zh) 应用程序的处理方法、装置、云环境及存储介质
CN113302613A (zh) 旁路保护
WO2025161674A1 (zh) 网络报文处理方法、装置、计算机设备和存储介质
US20250238269A1 (en) Data processing method, terminal device, and readable storage medium
US20080080715A1 (en) Apparatus and method for high-speed, large-volume data encryption using secure memory
US20070169120A1 (en) Mechanism to transition control between components in a virtual machine environment
CN112231124B (zh) 基于隐私保护的应用间通信方法及装置
US7089378B2 (en) Shared receive queues
US20040019882A1 (en) Scalable data communication model

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 23862012

Country of ref document: EP

Kind code of ref document: A1

WWE Wipo information: entry into national phase

Ref document number: 2023862012

Country of ref document: EP

ENP Entry into the national phase

Ref document number: 2023862012

Country of ref document: EP

Effective date: 20250320

NENP Non-entry into the national phase

Ref country code: DE

WWP Wipo information: published in national office

Ref document number: 2023862012

Country of ref document: EP