Adjustable aerospace hardware acceleration command system, configuration method and storage medium

The adjustable aerospace hardware acceleration command system solves the problems of scalability and programming convenience of aerospace data processing equipment platforms, and realizes flexible hardware acceleration command configuration and efficient information processing.

CN116009962BActive Publication Date: 2026-11-17SHANGHAI SPACEFLIGHT INST OF TT&C & TELECOMM
View PDF 3 Cites 0 Cited by

Patent Information

Application Number
CN202211325432.0
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-10-27
Publication Date
2026-11-17
Estimated Expiration
2042-10-27

AI Technical Summary

Technical Problem

Traditional aerospace data processing equipment platforms are unable to meet the high-speed processing capabilities and diverse command requirements of onboard real-time information processing, resulting in insufficient scalability and ease of program design.

Method used

An adjustable aerospace hardware acceleration command system is provided, including an adjustable hardware acceleration command module, an command library, and an acceleration module. Hardware acceleration commands can be customized through adjustable parameters, supporting flexible configuration and management of functional requirements. It has parallel, serial, and out-of-order execution capabilities, and prevents command incompatibility through a management module.

Benefits of technology

It enhances the scalability of onboard real-time information processing in the aerospace field, improves the convenience of program design and execution efficiency, and reduces the resource consumption of invalid instructions.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN116009962B_ABST
    Figure CN116009962B_ABST
Patent Text Reader

Abstract

This invention relates to an adjustable aerospace hardware acceleration command system, configuration method, and storage medium, comprising: configuring a plurality of hardware acceleration commands corresponding to functional requirements through an adjustable hardware acceleration command module; storing the configured plurality of hardware acceleration commands; and calling corresponding data and acceleration cores for performing the same function according to the hardware acceleration commands for accelerated processing. This application utilizes custom hardware acceleration commands for storage and retrieval, enhancing the scalability of onboard real-time information processing in the aerospace field and improving the convenience of program design.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of hardware acceleration instruction architecture technology, specifically to an adjustable aerospace hardware acceleration instruction system, configuration method, and storage medium. Background Technology

[0002] In recent years, the demand for real-time onboard information processing in the spaceflight (hereinafter referred to as aerospace) field has become extremely urgent. Technologies such as artificial intelligence, deep learning, and data analysis and mining are gradually being applied in various aerospace fields. This necessitates that onboard data processing equipment platforms possess high-speed processing capabilities for massive payload data, and that high-speed processing commands are becoming increasingly diverse. Traditional processing methods are difficult to fully meet these demands. This application, through an adjustable aerospace hardware acceleration command architecture, allows for the customization, storage, and retrieval of hardware acceleration commands according to the needs of data processing functions, enhancing the scalability of real-time onboard information processing in the aerospace field and improving the convenience of program design. Summary of the Invention

[0003] This invention provides an adjustable aerospace hardware acceleration command system, configuration method, and storage medium thereof. This adjustable aerospace hardware acceleration command system, configuration method, and storage medium enhance the scalability of onboard real-time information processing in the aerospace field.

[0004] One aspect of the present invention provides an adjustable aerospace hardware acceleration command system, which includes an adjustable hardware acceleration command module, an command library, and an acceleration module.

[0005] The adjustable hardware acceleration instruction module includes several adjustable parameters;

[0006] The adjustable parameters are used to configure hardware acceleration instructions according to functional requirements;

[0007] The instruction library is used to store the pre-set rules for hardware acceleration instructions and the configured hardware acceleration instructions.

[0008] The acceleration module includes several acceleration cores, which are used to call corresponding data and acceleration cores that perform the same function to accelerate processing according to the hardware acceleration instructions.

[0009] According to one embodiment of the present invention, the adjustable parameters configured according to functional requirements follow rules preset by hardware acceleration instructions.

[0010] The acceleration module includes an adjustable acceleration core module, which is used to add or delete acceleration cores according to functional requirements.

[0011] According to an embodiment of the present invention, the plurality of adjustable parameters include instruction name, instruction opcode, instruction operation type, instruction acceleration mode, instruction priority, instruction lifecycle, instruction operand type, instruction operand number, and instruction parameters;

[0012] The instruction name is used by the host computer of the upper-layer software to identify specific functions, and a unique corresponding instruction name is configured according to different functional requirements;

[0013] The instruction opcode is used to configure according to the pre-set rules of hardware acceleration instructions based on functional requirements. Different functional requirements have a unique corresponding instruction opcode.

[0014] The instruction operation type is used to set the instruction mode of this instruction; the instruction mode includes the direct instruction mode, which allows the remote end to directly call the accelerated computing core, and the indirect instruction mode, which allows the local end to enter the instruction sequence for unified scheduling.

[0015] The instruction acceleration mode is used to set the operating mode of the control acceleration core; the operating mode includes acceleration core reset, acceleration core initialization, acceleration core start, acceleration core stop, and acceleration core data transfer;

[0016] The instruction priority is used to configure the priority execution level of instructions that arrive at the acceleration core simultaneously; wherein;

[0017] The instruction lifecycle is used to set whether the currently received instruction needs to be executed, and the number of times the instruction that needs to be executed can be delayed.

[0018] The instruction operand type is used to set the type of data required for accelerated processing;

[0019] The instruction operand count is used to set the number of instructions required to process the data according to functional requirements;

[0020] The instruction operands are used to represent the data required for accelerated processing.

[0021] According to an embodiment of the present invention, the instruction name and the instruction opcode have a unique correspondence, which is used to unify the same functional requirement to uniquely correspond to the same instruction at different layers;

[0022] The corresponding data is the data pointed to by the instruction operand address pointer in the configured hardware acceleration instruction;

[0023] The acceleration cores with the same function are acceleration cores with the same function as the instruction opcodes in the configured hardware acceleration instructions.

[0024] According to one embodiment of the present invention, the hardware acceleration instructions include parallel execution, serial execution, and out-of-order execution, which are used to improve execution efficiency.

[0025] According to an embodiment of the present invention, the aerospace hardware acceleration command adjustable system includes a management module for approving applications for hardware acceleration commands configured according to functional requirements, thereby preventing problems such as incompatibility of hardware acceleration commands and non-unique command opcodes.

[0026] Another aspect of the present invention provides a method for configuring aerospace hardware acceleration commands, the method comprising the following steps:

[0027] According to the functional requirements, the adjustable hardware acceleration instruction module configures the several hardware acceleration instructions corresponding to the functional requirements.

[0028] The storage configuration includes several hardware acceleration instructions;

[0029] The hardware acceleration instructions call the corresponding data and acceleration cores used to perform the same function for accelerated processing.

[0030] According to an embodiment of the present invention, configuring the plurality of hardware acceleration instructions corresponding to the functional requirements through the adjustable hardware acceleration instruction module according to the functional requirements specifically includes:

[0031] Configure the hardware acceleration instruction name according to functional requirements and following the pre-defined rules of hardware acceleration instructions;

[0032] Configure a unique instruction opcode corresponding to the hardware acceleration instruction, following the pre-defined rules of the hardware acceleration instruction.

[0033] Follow the pre-defined rules of hardware acceleration instructions to configure instruction operation type, instruction acceleration mode, instruction priority, instruction lifecycle, instruction operand type, instruction operand quantity, and instruction operands.

[0034] Another aspect of the present invention provides an aerospace acceleration processing system, which includes an FPGA acceleration terminal and a main control terminal connected to the FPGA acceleration terminal, characterized in that the main control terminal configures a plurality of hardware acceleration instructions corresponding to the functional requirements through the adjustable hardware acceleration instruction module according to the functional requirements.

[0035] The configured hardware acceleration instructions are stored in the instruction library of the main control terminal;

[0036] The main control terminal sends hardware acceleration instructions and / or data to the FPGA acceleration terminal; the hardware acceleration instructions are stored in the instruction queue buffer.

[0037] The instruction processing unit of the FPGA acceleration end queries the instruction queue buffer. When an instruction is found, it reads and parses the instruction. Based on several parameters configured in the hardware acceleration instruction, it calls the corresponding data and the acceleration core used to achieve the same function to execute the corresponding action.

[0038] A final aspect of the invention is to provide a readable storage medium. The readable storage medium stores a processing program, which, when executed by a processor, implements the aerospace hardware acceleration command adjustable method according to an embodiment of the invention.

[0039] Compared with the prior art, the present invention has the following beneficial effects:

[0040] This application utilizes an adjustable aerospace hardware acceleration instruction architecture, enabling the storage and retrieval of customized hardware acceleration instructions based on data processing requirements. This enhances the scalability of onboard real-time information processing in the aerospace field and improves program design convenience. The instruction length is flexible, dynamically scaling within a certain range according to hardware acceleration needs. The instruction set can be executed sequentially, perfectly aligning with FPGA-accelerated parallel computing and improving execution efficiency. Hardware acceleration instructions have priority settings, further enhancing execution flexibility. Each hardware acceleration instruction has a lifecycle; instructions that require delayed execution due to occupied or unprepared acceleration cores are delayed. The lifecycle decreases by one for each delayed execution; when the lifecycle reaches zero, the instruction is discarded, reducing the resource consumption of invalid instruction codes in the database. Attached Figure Description

[0041] Figure 1 According to an embodiment of the present invention, a schematic diagram of an adjustable aerospace hardware acceleration command system is shown;

[0042] Figure 2 According to an embodiment of the present invention, a rule for configuring several adjustable parameters of hardware acceleration instructions is shown.

[0043] Figure 3 According to an embodiment of the present invention, an implementation method for configuring hardware acceleration instructions is shown;

[0044] Figure 4 According to an embodiment of the present invention, a flowchart of an adjustable aerospace hardware acceleration command method is shown;

[0045] Figure 5 According to an embodiment of the present invention, a flowchart of a hardware acceleration instruction configuration method is shown;

[0046] Figure 6 According to an embodiment of the present invention, a schematic diagram of the structure of a computer-readable storage medium is shown. Detailed Implementation

[0047] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions in the embodiments of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention. The term "comprising" and its variations used herein indicate open-ended inclusion, i.e., "including but not limited to". Other explicit and implicit definitions may also be included below. In this invention, space travel refers to the travel of artificial Earth satellites, spacecraft, etc., in space within and outside the solar system. For example, timely and rapid acquisition of the trajectory of asteroids and quick determination of their arrival time on Earth facilitates timely subsequent processing.

[0048] Embodiment 1 of the present invention provides a specific implementation of an adjustable aerospace hardware acceleration command system, see [link to implementation details]. Figure 1 The adjustable hardware acceleration command system for aerospace includes an adjustable hardware acceleration command module, a command library, and an acceleration module. The adjustable hardware acceleration command module includes several adjustable parameters; these parameters are used to configure hardware acceleration commands according to functional requirements; the command library stores the configured hardware acceleration commands; and the acceleration module includes several acceleration cores, used to call corresponding data and acceleration cores to perform accelerated processing according to the hardware acceleration commands.

[0049] in,

[0050] Several adjustable parameters used for configuration according to functional requirements follow rules pre-set by hardware acceleration instructions.

[0051] The acceleration module includes an adjustable acceleration core module, which is used to add or remove acceleration cores according to functional requirements.

[0052] Several adjustable parameters include instruction name, instruction opcode, instruction operation type, instruction acceleration mode, instruction priority, instruction lifecycle, instruction operand type, instruction operand count, and instruction parameters.

[0053] Hardware acceleration instructions include parallel execution, serial execution, and out-of-order execution, which are used to improve execution efficiency.

[0054] Among them, such as Figure 2 As shown, several adjustable parameters used to configure hardware acceleration instructions include:

[0055] The instruction name is used by the host computer in the upper-layer software to identify specific functions. A unique instruction name is configured according to different functional requirements. For example, the pre-defined rules for hardware acceleration instructions specify that the instruction name format begins with HACC_XX, corresponding to the hardware acceleration core, where XX represents the acceleration core and can include numbers or letters. The instruction name is used by the host computer in the upper-layer software and is recognized and converted into instruction opcodes by the compiler.

[0056] Instruction opcodes are configured according to pre-defined rules for hardware acceleration instructions based on functional requirements. Different functional requirements have unique corresponding instruction opcodes. The basic machine language is used, for example, set to 32 bits in size, and is generated by the compiler based on the instruction name according to the pre-defined rules for hardware acceleration instructions. Instruction opcodes correspond to hardware acceleration cores, facilitating the selection of the hardware acceleration core type. 0x0 and 0xFFFFFFFF are system-reserved instruction codes. The above instruction names and instruction opcodes have a unique correspondence, ensuring that the same functional requirement uniquely corresponds to the same instruction at different layers. That is, the instruction opcode is the unique lookup code for function A at the physical layer, and the instruction name is the unique lookup code for the same function A at the application layer.

[0057] The instruction operation type is used to set the instruction mode for this instruction. Instruction modes include the direct instruction mode, where the remote end directly calls the accelerated computing core, and the indirect instruction mode, where the local end enters the instruction sequence for unified scheduling. For example, the size is set to 1 bit. In the pre-defined rules of hardware acceleration instructions, 0 represents an indirect instruction and 1 represents a direct instruction.

[0058] The instruction acceleration mode is used to set the operating mode of the acceleration core; the operating modes include acceleration core reset, acceleration core initialization, acceleration core startup, acceleration core stop, and acceleration core data transfer. For example, the size is set to 7 bits, and in the pre-defined rules of the hardware acceleration instructions, bits 0 to 6 represent the operating modes of the acceleration core. Up to 127 operating modes are supported, including acceleration core reset, acceleration core initialization, acceleration core startup, acceleration core stop, and acceleration core data transfer.

[0059] Instruction priority is used to configure the execution priority level of instructions that arrive at the acceleration core simultaneously and / or for use during instruction queue execution. For example, the size is set to 4 bits. In the pre-defined rules for hardware acceleration instructions, 0 represents the lowest priority and 0xf represents the highest priority. Instructions arriving at the acceleration core first are executed first, and those arriving simultaneously are executed sequentially according to priority from highest to lowest.

[0060] Instruction lifecycle is used to determine whether a currently received instruction needs to be executed, and to set the number of times an instruction that needs to be executed can be delayed. For example, the size is set to 4 bits. In the pre-defined rules for hardware-accelerated instructions, 0 indicates that the instruction needs to be discarded, and a non-zero value indicates that the instruction needs to be executed. The instruction lifecycle can be set according to requirements during configuration, for example, 0xf. Because during the execution of an instruction sequence, execution may be delayed if the acceleration core is not yet ready, and to prevent instruction execution from forming a loop and causing a broadcast storm, the lifecycle function is added. Each time an instruction is delayed, the lifecycle decreases by at least 1; when the lifecycle reaches 0, the instruction is discarded.

[0061] The instruction operand type is used to set the storage type of the data participating in the operation; wherein, the type includes address and variable value. For example, the size is set to 2 bits. In the pre-defined rules of hardware acceleration instructions, 0 represents the parameter pointer address, 1 represents the variable value, and 2 and 3 are reserved.

[0062] The instruction operand count is used to set the number of instructions required to process the data according to functional requirements. For example, the size is set to 6 bits, which can provide a maximum of 64 operands.

[0063] Instruction operands are used to fill in the actual address of the data to be processed. For example, the size is set to 32 bits. In the pre-defined rules of hardware acceleration instructions, the instruction operands store the address pointer of the data to be processed and the variable value, for the acceleration core to call.

[0064] The aforementioned corresponding data refers to the data pointed to by the instruction operand address pointer in the configured hardware acceleration instructions;

[0065] The aforementioned acceleration cores with the same function are acceleration cores with the same function as the instruction opcodes in the configured hardware acceleration instructions.

[0066] The aforementioned adjustable aerospace hardware acceleration instruction system also includes a management module for approving applications for hardware acceleration instructions configured according to functional requirements, preventing issues such as incompatibility of hardware acceleration instructions and non-unique instruction opcodes. Unapproved instruction sets can only be used privately.

[0067] Embodiment 2 of the present invention provides a specific implementation of a space hardware acceleration command configuration method, see [link to embodiment 2]. Figure 4 and Figure 5 The adjustable aerospace hardware acceleration command method specifically includes the following steps:

[0068] S100: Configure the plurality of hardware acceleration instructions corresponding to the functional requirements through the adjustable hardware acceleration instruction module according to the functional requirements.

[0069] S200: Several hardware acceleration instructions for storage configuration.

[0070] S300: According to the hardware acceleration instruction, call the corresponding data and the acceleration core used to achieve the same function to perform acceleration processing.

[0071] Specifically, configuring the plurality of hardware acceleration instructions corresponding to the functional requirements through the adjustable hardware acceleration instruction module according to the functional requirements includes:

[0072] S110: Configure the hardware acceleration instruction name according to functional requirements and in accordance with the pre-defined rules of hardware acceleration instructions;

[0073] S120: Follows the pre-defined rules of hardware acceleration instructions and configures a unique instruction opcode corresponding to the hardware acceleration instructions;

[0074] S130: Follows the pre-defined rules of hardware acceleration instructions, configuring instruction operation type, instruction acceleration mode, instruction priority, instruction lifecycle, instruction operand type, instruction operand quantity, and instruction operands.

[0075] For example, see attached Figure 3As shown, functional requirements are divided according to project needs. Based on these requirements and following pre-defined rules for hardware acceleration instructions, the hardware acceleration instruction names HACC_detect, HACC_slice, and HACC_mmul are configured. HACC_detect is for target detection hardware acceleration, HACC_slice for data slicing hardware acceleration, and HACC_mmul for matrix multi-operation acceleration. In this embodiment, the instruction opcodes requested from the instruction library are 0xa0000101 for HACC_detect, 0xa0000102 for HACC_slice, and 0xa0000103 for HACC_mmul. Instruction opcodes must be unique and can be automatically managed by the instruction library or determined and managed by administrators to ensure no conflicts with other hardware acceleration instructions. They also have a certain lifespan; once expired, they are deleted from the library to reduce the resource consumption of invalid instruction codes. The HACC_detect instruction is set to 0, meaning it is an indirect instruction; the instruction acceleration mode is set to 0, meaning the acceleration core is reset (the hardware acceleration instructions pre-set the acceleration mode as follows: 0 for acceleration core reset, 1 for acceleration core initialization, 2 for acceleration core startup, 3 for acceleration core shutdown, and 4 for acceleration core data migration); the instruction priority is set to 0, meaning the lowest priority (0 represents the lowest priority, 0xf represents the highest priority); the instruction lifetime is set to 5 times, meaning the instruction needs to be executed and can be delayed for 5 times. Each time the instruction is delayed, the lifetime is decremented by 1. When the lifetime is 0, the instruction is discarded because it is not needed; the instruction operand type is set to address (0 represents the parameter pointer address, 1 represents the variable value, 2 and 3 are reserved); the number of instruction operands is set to 1, and the address pointed to by the instruction operand address pointer is 0x80000000. Therefore, the configured hardware acceleration instruction format is "HACC_detect 0x00 0x05 0x41 0x80000000". The instruction is stored in the instruction library and sent according to the upper-level host computer's trigger instruction. Other requirements for this acceleration function can also be configured. For example, if the hardware acceleration instruction "HACC_detect 0x02 0x05 0x41 0x80000000" is specified to start the detection of the target accelerated by the called acceleration core, that is, the instruction acceleration mode is set to 2; for example, if the hardware acceleration instruction "HACC_detect 0x02 0x15 0x41 0x80000000" is specified to start the detection of the target accelerated by the called acceleration core and increase the priority of the hardware acceleration instruction.When multiple identical applications require acceleration, resources can be requested simultaneously from the acceleration pool. The commands "HACC_detect 0x020x15 0x41 0x80000000" and "HACC_detect 0x02 0x15 0x41 0x90000000" can be sent simultaneously. This will activate acceleration cores with the same acceleration capabilities, allowing them to run in parallel and compute data at different addresses. The above method is used to configure the hardware acceleration command HACC_slice for data slicing acceleration and the hardware acceleration command for matrix multiplication acceleration. HACC_mmul is a two-parameter hardware acceleration command, sent in the format "HACC_mmul 0x00 0x05 0x41 0xb00000000xc0000000", indicating matrix operations on data at addresses 0xb0000000 and 0xc0000000. The remaining functions are the same as those for HACC_detect.

[0076] The aforementioned corresponding data refers to the data pointed to by the instruction operand address pointer in the configured hardware acceleration instructions;

[0077] The aforementioned acceleration cores with the same function are acceleration cores with the same function as the instruction opcodes in the configured hardware acceleration instructions.

[0078] The instruction length is flexible and can be dynamically scaled within a certain range according to the needs of hardware acceleration; the instruction set can be executed sequentially, which is very compatible with FPGA-accelerated parallel computing and improves execution efficiency; the instruction set has priority, which can improve the flexibility of execution.

[0079] Embodiment 3 of the present invention provides a specific implementation of a space acceleration processing system, which includes:

[0080] The system consists of an FPGA acceleration terminal and a main control terminal connected to the FPGA acceleration terminal. The main control terminal, based on project needs, divides functional requirements, compiles corresponding hardware acceleration instructions according to these requirements and following pre-defined rules, and then sends these instructions, along with the data, to the FPGA acceleration terminal.

[0081] The main control unit configures the plurality of hardware acceleration instructions corresponding to the functional requirements through the adjustable hardware acceleration instruction module according to the functional requirements.

[0082] The configured hardware acceleration instructions are stored in the instruction library of the main control terminal;

[0083] The main control terminal sends hardware acceleration instructions and / or data to the FPGA acceleration terminal; the hardware acceleration instructions are stored in the instruction queue buffer.

[0084] The instruction processing unit of the FPGA acceleration end queries the instruction queue buffer. When an instruction is found, it reads and parses the instruction. Based on several parameters configured in the hardware acceleration instruction, it calls the corresponding data and the acceleration core used to achieve the same function to execute the corresponding action.

[0085] The corresponding data represents the data stored at the address pointed to by the address pointer in the hardware acceleration instruction; the execution of the corresponding actions includes: controlling the computing acceleration core to accelerate data processing; instructions that fail to execute normally are stored in a delayed instruction buffer and retried at a later time. Instruction opcodes and hardware acceleration resource cores are mapped according to certain rules, and new instruction sets can be requested to be added when needed.

[0086] It is understood that various aspects of the technical solution of the present invention can be implemented as a system, method, or program product. Therefore, various aspects of the technical solution of the present invention can be specifically implemented in the following forms: a completely hardware implementation, a completely software implementation (including firmware, microcode, etc.), or an implementation combining hardware and software aspects, which can be collectively referred to here as a "circuit", "unit", or "platform".

[0087] Those skilled in the art should understand that the various units or steps of the present invention described above can be implemented using general-purpose computing devices. They can be centralized on a single computing device or distributed across a network of multiple computing devices. Optionally, they can be implemented using computing device-executable program code, thereby storing them in a storage medium for execution by the computing device. In some cases, the steps shown or described can be performed in a different order than those described herein, or they can be fabricated as separate integrated circuit units, or multiple units or steps can be fabricated as a single integrated circuit unit.

[0088] In some embodiments of the present invention, a computer-readable storage medium is also provided, on which a computer program is stored, and when executed by a processor, the computer program is able to implement the various functional units in the above-disclosed aerospace acceleration processing system.

[0089] Although other specific implementation methods are not listed exhaustively in this embodiment, in some possible implementation methods, various aspects of the technical solution described in this invention can also be implemented in the form of a program product, which includes program code. When the program product is run on a terminal device, the program code is used to cause the terminal device to execute the steps of the various embodiments of the technical solution described in the aerospace acceleration processing method area of ​​this invention.

[0090] Figure 6A schematic diagram of the structure of a computer-readable storage medium is shown according to some embodiments of the present invention. For example... Figure 6 As shown, a program product 800 for implementing the above-described method according to an embodiment of the technical solution of the present invention is described. This product may employ a portable compact disc read-only memory (CD-ROM) and include program code, and can run on a terminal device, such as a personal computer. Of course, the program product generated according to this embodiment is not limited to this. In the technical solution of the present invention, the readable storage medium may be any tangible medium containing or storing a program, which may be used by or in conjunction with an instruction execution system, apparatus, or device.

[0091] The program product may employ any combination of one or more readable media. A readable medium may be a readable signal medium or a readable storage medium. A readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples (a non-exhaustive list) of readable storage media include: electrical connections having one or more wires, portable disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0092] Computer-readable storage media may include data signals propagated in baseband or as a carrier wave region, carrying readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A readable storage medium may also be any readable medium other than a readable storage medium that can transmit, propagate, or transfer a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the readable storage medium may be transmitted using any suitable medium, including but not limited to wireless, wired, optical fiber, RF, etc., or any suitable combination thereof.

[0093] Program code for executing the technical solutions of this invention can be written in any combination of one or more programming languages. These programming languages ​​include object-oriented programming languages—such as Java and C++—as well as conventional procedural programming languages—such as C or similar languages. The program code can execute entirely on the user's computing device, locally on the user's device, as a standalone software package, locally on the user's computing device, locally on a remote computing device, or entirely on a remote computing device or server. In cases involving remote computing devices, the remote computing device can be connected to the user's computing device via any type of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computing device (e.g., via the Internet using an Internet service provider).

[0094] In summary, the technical solution proposed in this invention enables the storage and retrieval of customized hardware acceleration instructions based on data processing requirements, enhancing the scalability of onboard real-time information processing in the aerospace field and improving the convenience of program design. The instruction length is flexible, dynamically scaling within a certain range according to hardware acceleration needs; the instruction set can be executed sequentially, perfectly matching the parallel computing accelerated by FPGAs, thus improving execution efficiency; the hardware acceleration instructions have priorities, further enhancing execution flexibility. Hardware acceleration instructions have an instruction lifecycle; for hardware acceleration instructions that need to be delayed due to occupied or unprepared acceleration cores, the lifecycle decreases by one for each delayed execution. When the lifecycle reaches 0, the hardware acceleration instruction is discarded and no longer executed, reducing the resource consumption of invalid instruction codes in the information base.

[0095] The above description is merely a description of a preferred embodiment of the technical solution of the present invention and is not intended to limit the scope of the technical solution of the present invention in any way. Any changes or modifications made by those skilled in the art based on the above disclosure shall fall within the protection scope of the claims.

Claims

1. A tunable space hardware acceleration instruction system, characterized by, The adjustable aerospace hardware acceleration command system includes an adjustable hardware acceleration command module, a command library, and an acceleration module; The adjustable hardware acceleration instruction module includes several adjustable parameters; The adjustable parameters are used to configure hardware acceleration instructions according to functional requirements. The adjustable parameters include instruction lifecycle, which is used to set whether the currently received instruction needs to be executed and the number of times the instruction to be executed can be delayed. Hardware acceleration instructions that need to be delayed because the acceleration core is occupied or not ready are delayed. At the same time, to prevent instruction execution from forming a loop and causing a broadcast storm, a lifecycle function is added. Each time the instruction is delayed, the lifecycle is reduced by 1. When the lifecycle is 0, the hardware acceleration instruction is no longer executed and is discarded. The instruction library is used to store pre-set rules and configured hardware acceleration instructions; the acceleration module includes several acceleration cores, which are used to call corresponding data and acceleration cores that perform the same function according to the hardware acceleration instructions to perform acceleration processing; the acceleration cores with the same function are acceleration cores with the same function corresponding to the instruction opcode in the configured hardware acceleration instructions, wherein the corresponding data is the data pointed to by the instruction operand address pointer in the configured hardware acceleration instructions.

2. An adjustable space hardware acceleration instruction system according to claim 1, wherein, The adjustable parameters used for configuration according to functional requirements follow the rules preset by the hardware acceleration instructions. The acceleration module includes an adjustable acceleration core module, which is used to add or delete acceleration cores according to functional requirements.

3. An adjustable space hardware acceleration instruction system according to claim 1, wherein, The adjustable parameters include instruction name, instruction opcode, instruction operation type, instruction acceleration mode, instruction priority, instruction operand type, instruction operand quantity, and instruction operands. The instruction name is used by the host computer of the upper-layer software to identify specific functions, and a unique corresponding instruction name is configured according to different functional requirements; The instruction opcode is used to configure according to the pre-set rules of hardware acceleration instructions based on functional requirements. Different functional requirements have a unique corresponding instruction opcode. The instruction operation type is used to set the instruction mode of this instruction; the instruction mode includes the direct instruction mode, which allows the remote end to directly call the accelerated computing core, and the indirect instruction mode, which allows the local end to enter the instruction sequence for unified scheduling. The instruction acceleration mode is used to set the operating mode of the control acceleration core; the operating mode includes acceleration core reset, acceleration core initialization, acceleration core start, acceleration core stop, and acceleration core data transfer; The instruction priority is used to configure the priority execution level of instructions that arrive at the acceleration core simultaneously; in; The instruction operand type is used to set the type of data required for accelerated processing; The instruction operand count is used to set the number of instructions required to process the data according to functional requirements; The instruction operands are used to represent the data required for accelerated processing.

4. An adjustable space hardware acceleration instruction system according to claim 3, wherein, The instruction name and the instruction opcode have a unique correspondence, which is used to ensure that the same functional requirement corresponds to the same instruction at different layers.

5. The adjustable aerospace hardware acceleration command system according to claim 1, characterized in that, The hardware acceleration instructions include parallel execution, serial execution, and out-of-order execution, which are used to improve execution efficiency.

6. The adjustable aerospace hardware acceleration command system according to claim 1, characterized in that, It includes a management module for approving applications for hardware acceleration instructions configured according to functional requirements, preventing issues such as incompatibility of hardware acceleration instructions and non-unique instruction opcodes.

7. A method for configuring aerospace hardware acceleration commands, implemented based on an adjustable aerospace hardware acceleration command system as described in any one of claims 1-6, characterized in that, Includes the following steps: According to the functional requirements, the adjustable hardware acceleration instruction module configures the several hardware acceleration instructions corresponding to the functional requirements. The storage configuration includes several hardware acceleration instructions; The hardware acceleration instructions call the corresponding data and acceleration cores used to perform the same function for accelerated processing.

8. A method for configuring aerospace hardware acceleration commands according to claim 7, characterized in that, The configuration of the plurality of hardware acceleration instructions corresponding to the functional requirements through the adjustable hardware acceleration instruction module according to the functional requirements specifically includes: Configure the hardware acceleration instruction name according to functional requirements and following the pre-defined rules of hardware acceleration instructions; Configure a unique instruction opcode corresponding to the hardware acceleration instruction, following the pre-defined rules of the hardware acceleration instruction. Follow the pre-defined rules of hardware acceleration instructions to configure instruction operation type, instruction acceleration mode, instruction priority, instruction lifecycle, instruction operand type, instruction operand quantity, and instruction operands.

9. An aerospace hardware acceleration processing system employing the aerospace hardware acceleration command configuration method as described in any one of claims 7-8, comprising an FPGA acceleration terminal and a main control terminal connected to the FPGA acceleration terminal, characterized in that, The main control unit configures the plurality of hardware acceleration instructions corresponding to the functional requirements through the adjustable hardware acceleration instruction module according to the functional requirements. The configured hardware acceleration instructions are stored in the instruction library of the main control terminal; The main control terminal sends hardware acceleration instructions and / or data to the FPGA acceleration terminal; the hardware acceleration instructions are stored in the instruction queue buffer. The instruction processing unit of the FPGA acceleration end queries the instruction queue buffer. When an instruction is found, it reads and parses the instruction. Based on several parameters configured in the hardware acceleration instruction, it calls the corresponding data and the acceleration core used to achieve the same function to execute the corresponding action.

10. A readable storage medium, characterized in that, The readable storage medium stores a processing program, which, when executed by a processor, implements the aerospace hardware acceleration instruction configuration method as described in any one of claims 7 to 8.

Citation Information

Patent Citations

  • Instruction processing method and device

    CN111666103A

  • FPGA acceleration system and method, electronic equipment and computer readable storage medium

    CN112506087A

  • Task processing method, processor, equipment and readable storage medium

    CN113407352A