[go: up one dir, main page]

CN110362512B - SCA and SDR-oriented rapid system reconstruction method - Google Patents

SCA and SDR-oriented rapid system reconstruction method Download PDF

Info

Publication number
CN110362512B
CN110362512B CN201910649819.3A CN201910649819A CN110362512B CN 110362512 B CN110362512 B CN 110362512B CN 201910649819 A CN201910649819 A CN 201910649819A CN 110362512 B CN110362512 B CN 110362512B
Authority
CN
China
Prior art keywords
axi
interrupt
data
pcie
module
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.)
Active
Application number
CN201910649819.3A
Other languages
Chinese (zh)
Other versions
CN110362512A (en
Inventor
许忠文
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.)
Chengdu Xieying Technology Co ltd
Original Assignee
Chengdu Xieying Technology 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 Chengdu Xieying Technology Co ltd filed Critical Chengdu Xieying Technology Co ltd
Priority to CN201910649819.3A priority Critical patent/CN110362512B/en
Publication of CN110362512A publication Critical patent/CN110362512A/en
Application granted granted Critical
Publication of CN110362512B publication Critical patent/CN110362512B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING OR CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00Arrangements for program control, e.g. control units
    • G06F9/06Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/44Arrangements for executing specific programs
    • G06F9/4401Bootstrapping
    • G06F9/4411Configuring for operating with peripheral devices; Loading of device drivers
    • GPHYSICS
    • G06COMPUTING OR CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F13/00Interconnection of, or transfer of information or other signals between, memories, input/output devices or central processing units
    • G06F13/10Program control for peripheral devices
    • G06F13/102Program control for peripheral devices where the programme performs an interfacing function, e.g. device driver
    • GPHYSICS
    • G06COMPUTING OR CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F2213/00Indexing scheme relating to interconnection of, or transfer of information or other signals between, memories, input/output devices or central processing units
    • G06F2213/0026PCI express
    • YGENERAL 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
    • Y02TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02DCLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
    • Y02D10/00Energy efficient computing, e.g. low power processors, power management or thermal management

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)
  • Information Transfer Systems (AREA)

Abstract

The invention discloses a rapid system reconstruction method facing SCA and SDR, the system is composed of a main control module and a waveform module, the main control module and the waveform module are interconnected and communicated through a VPX bus of a backboard, a PCIE X45.0GT/s interface is adopted, an X86 hardware platform of the main control module is used as a root component RC of the PCIE interface, a ZYNQ chip of the waveform module is used as PCIE interface terminal equipment EP, wherein the main control module comprises an X86 hardware platform, an operating system, SCA Core Framework (CF), CORBA Middleware (CORBA), SCA Device, POSIX AEP and SCA Applications, the SCA Device comprises a driver of the waveform module, the waveform module comprises ZYNQ series, FPGA7 series, DDR and other chips, the interface between the ZYNQ and the FPGA7 comprises SelectMAP, LVDS, SRIO and the like, and the external memory of the ZYNQ is a large-capacity DDR. The invention comprehensively utilizes the technologies and resources of a high-speed bus, PCIE and SRIO interfaces, DMA, an interrupt mechanism, a large-capacity DDR and the like, and realizes the rapid reconfiguration of the FPGA waveform, the rapid configuration of waveform parameters, the real-time return of waveform states and the high-speed transmission of waveform data.

Description

一种面向SCA和SDR的快速系统重构方法A Fast System Reconfiguration Method Oriented to SCA and SDR

技术领域technical field

本发明涉及软件通信体系架构和软件无线电技术领域,具体是一种面向SCA和SDR的快速系统重构方法。The invention relates to the technical field of software communication architecture and software radio, in particular to a fast system reconfiguration method oriented to SCA and SDR.

背景技术Background technique

SCA是为软件无线电SDR定义的标准体系结构,支持开发SDR通信系统使波形应用软件更容易跨无线电平台移植,SCA提供了波形软件和软件无线电的其他元素与硬件交互的独特方式。通过标准化,波形是兼容的,不仅适用于一个平台,而且适用于支持SCA的任何设备。波形软件可以用于许多不同的无线电设备,无论是手持设备还是其他通信设备。SCA is a standard architecture defined for software radio SDR, which supports the development of SDR communication systems and makes waveform application software easier to port across radio platforms. SCA provides a unique way for waveform software and other elements of software radio to interact with hardware. Through standardization, waveforms are compatible, not just for one platform, but for any device that supports SCA. Waveform software can be used with many different radios, whether handheld or other communication devices.

SCA通过定义核心框架接口和应用程序环境描述(AEP),使得波形软件具有很好的可移植性和可复用性,SCA通过一套语义复杂的XML描述文件来创建SCA域配置文件元素,用于标识SCA兼容系统中的硬件设备和软件组件的功能、属性、相互依赖关系和位置。一般的SCA系统中都包含GPP、DSP、FPGA这三种类型的芯片,当进行系统部署和波形加载时,就需要对系统中的GPP、DSP、FPGA芯片进行重新配置,以完成系统重构。SCA makes waveform software very portable and reusable by defining the core framework interface and application environment description (AEP). SCA creates SCA domain configuration file elements through a set of XML description files with complex semantics. It is used to identify the functions, attributes, interdependencies and locations of hardware devices and software components in an SCA compliant system. A general SCA system includes three types of chips: GPP, DSP, and FPGA. When system deployment and waveform loading are performed, the GPP, DSP, and FPGA chips in the system need to be reconfigured to complete system reconstruction.

随着GPP、DSP、FPGA芯片规模的不断扩大,其配置文件越来越大,完成配置所花费的时间已经严重影响SCA系统重构的性能,据目前研究的情况来看,采用单一的优化方法已不能满足用户对系统重构的时间要求,必须从架构、接口、流程上进行全面的设计来实现快速系统重构。With the continuous expansion of the scale of GPP, DSP, and FPGA chips, their configuration files are getting larger and larger, and the time it takes to complete the configuration has seriously affected the performance of SCA system reconstruction. According to the current research situation, using a single optimization method It can no longer meet the user's time requirements for system reconfiguration, and a comprehensive design must be carried out from the perspective of architecture, interface, and process to achieve rapid system reconfiguration.

发明内容Contents of the invention

为了解决上述现有技术的不足,本发明提供了一种面向SCA和SDR的快速系统重构方法,综合利用高速总线、PCIE和SRIO接口、DMA、中断机制、大容量DDR等技术与资源,实现FPGA波形快速重配、波形参数快速配置、波形状态实时回传、波形数据高速传输。In order to solve the above-mentioned deficiencies in the prior art, the present invention provides a fast system reconfiguration method facing SCA and SDR, and comprehensively utilizes technologies and resources such as high-speed bus, PCIE and SRIO interface, DMA, interrupt mechanism, and large-capacity DDR to realize Fast reconfiguration of FPGA waveforms, quick configuration of waveform parameters, real-time return of waveform status, and high-speed transmission of waveform data.

为了达到上述目的,本发明采用的技术方案为:In order to achieve the above object, the technical scheme adopted in the present invention is:

一种面向SCA和SDR的快速系统重构方法,系统由主控模块和波形模块组成,主控模块与波形模块通过背板的VPX总线互联互通,接口采用PCIE X4 5.0GT/s,主控模块的X86硬件平台作为PCIE接口的根组件RC,波形模块的ZYNQ芯片作为PCIE接口终端设备EP,其中主控模块包含X86硬件平台、操作系统、SCA Core Framework(CF)、CORBA Middleware(CORBA)、SCA Device、POSIX AEP、SCA Applications,SCA Device包含波形模块的驱动程序,其中波形模块包含ZYNQ系列、FPGA7系列、DDR等芯片,ZYNQ与FPGA7之间接口包含SelectMAP、LVDS、SRIO等,ZYNQ的外部存储器为大容量DDR。A fast system reconfiguration method for SCA and SDR. The system is composed of a main control module and a waveform module. The main control module and the waveform module are interconnected through the VPX bus on the backplane. The X86 hardware platform is used as the root component RC of the PCIE interface, and the ZYNQ chip of the waveform module is used as the terminal equipment EP of the PCIE interface. The main control module includes the X86 hardware platform, operating system, SCA Core Framework (CF), CORBA Middleware (CORBA), SCA Device, POSIX AEP, SCA Applications, SCA Device includes the driver program of the waveform module. The waveform module includes ZYNQ series, FPGA7 series, DDR and other chips. The interface between ZYNQ and FPGA7 includes SelectMAP, LVDS, SRIO, etc. The external memory of ZYNQ is Large capacity DDR.

对本发明进一步的描述,波形模块ZYNQ芯片内部程序包括ZYNQ7 ProcessingSystem(PS)、3个AXI Interconnect、3个AXI BRAM Controller、AXI Memory Mapped PCIExpress(M2PCIE)、AXI Central Direct Memory Access(CDMA)、AXI SelectMAP、DDR、BRAM、数据读写控制逻辑、中断控制逻辑功能模块,其中ZYNQ7 Processing System(PS)通过M_AXI_GPO接口与AXI Interconnect1互联访问M2PCIE、CDMA、SelectMAP、AXI BRAMController1、数据读写控制逻辑的地址空间,其中AXI Memory Mapped PCI Express(M2PCIE)通过AXI总线与AXI Interconnect3互联访问DDR、AXI BRAM Controller2、AXIBRAM Controller3的地址空间,其中AXI Central Direct Memory Access(CDMA)通过AXI总线与AXI Interconnect2互联,并与AXI Interconnect1和AXI Interconnect3级联访问M2PCIE、SelectMAP、AXI BRAM Controller1、AXI BRAM Controller3、数据读写控制逻辑的地址空间,并通过ZYNQ7Processing System(PS)的S_AXI_HPO接口实现对DDR地址空间的访问,其中AXI BRAM Controller1和AXI BRAM Controller2分别连接BRAM的两个端口,主控模块通过PCIE接口、PS及AXI总线访问同一个BRAM的地址空间。To further describe the present invention, the waveform module ZYNQ chip internal program includes ZYNQ7 ProcessingSystem (PS), 3 AXI Interconnects, 3 AXI BRAM Controllers, AXI Memory Mapped PCIExpress (M2PCIE), AXI Central Direct Memory Access (CDMA), AXI SelectMAP, DDR, BRAM, data read and write control logic, and interrupt control logic function modules, among which ZYNQ7 Processing System (PS) accesses the address space of M2PCIE, CDMA, SelectMAP, AXI BRAMController1, and data read and write control logic through the M_AXI_GPO interface and AXI Interconnect1 interconnection. AXI Memory Mapped PCI Express (M2PCIE) is interconnected with AXI Interconnect3 through AXI bus to access the address space of DDR, AXI BRAM Controller2, and AXIBRAM Controller3. Among them, AXI Central Direct Memory Access (CDMA) is interconnected with AXI Interconnect2 through AXI bus, and is connected with AXI Interconnect1 and AXI Interconnect3 cascade accesses the address space of M2PCIE, SelectMAP, AXI BRAM Controller1, AXI BRAM Controller3, and data read and write control logic, and realizes access to the DDR address space through the S_AXI_HPO interface of ZYNQ7Processing System (PS), where AXI BRAM Controller1 and AXI BRAM Controller2 is connected to two ports of BRAM respectively, and the main control module accesses the address space of the same BRAM through PCIE interface, PS and AXI bus.

对本发明进一步的描述,AXI Memory Mapped PCI Express(M2PCIE)的BAR地址分别映射DDR、AXI BRAM Controller2和AXI BRAM Controller3的地址空间,通过主控模块通过PCIE的BAR地址直接访问上述三个地址空间。To further describe the present invention, the BAR address of AXI Memory Mapped PCI Express (M2PCIE) maps the address spaces of DDR, AXI BRAM Controller2 and AXI BRAM Controller3 respectively, and directly accesses the above-mentioned three address spaces by the BAR address of PCIE by the main control module.

对本发明进一步的描述,AXI Memory Mapped PCI Express(M2PCIE)的PCIE接口与主控模块的PCIE接口互联,AXI SelectMAP与FPGA7的SelectMAP接口互联,AXI BRAMController3与FPGA7的LVDS接口互联,数据读写控制逻辑与FPGA7的SRIO接口互联,PS对数据读写控制逻辑的寄存器进行配置。To further describe the present invention, the PCIE interface of AXI Memory Mapped PCI Express (M2PCIE) is interconnected with the PCIE interface of the main control module, the SelectMAP interface of AXI SelectMAP and FPGA7 is interconnected, the LVDS interface of AXI BRAMController3 and FPGA7 is interconnected, and the data read and write control logic and The SRIO interface of FPGA7 is interconnected, and the PS configures the registers of the data read and write control logic.

对本发明进一步的描述,包括两种中断机制,分别是主控模块PCIE MSI中断机制和PS中断控制逻辑,当ZYNQ通过PCIE接口向主控模块发送特定的TLP包将触发主控模块PCIE的MSI中断,相应的中断服务程序将清除该中断,当主控模块通过PCIE接口向ZYNQ的AXI BRAM Controller2的特定地址写数据时,PS中断控制逻辑将触发PS中断,当PS通过AXI总线从AXI BRAM Controller1的特定地址读数据时,PS中断控制逻辑将清除PS中断。The further description of the present invention includes two kinds of interrupt mechanisms, namely main control module PCIE MSI interrupt mechanism and PS interrupt control logic, when ZYNQ sends a specific TLP packet to the main control module through the PCIE interface, it will trigger the MSI interrupt of the main control module PCIE , the corresponding interrupt service program will clear the interrupt. When the main control module writes data to the specific address of ZYNQ’s AXI BRAM Controller2 through the PCIE interface, the PS interrupt control logic will trigger the PS interrupt. When data is read from a specific address, the PS interrupt control logic will clear the PS interrupt.

对本发明进一步的描述,包括两个数据流向,分别是从主控模块到ZYNQ或FPGA7下行数据流向及从ZYNQ或FPGA7到主控模块的上行数据流向。The further description of the present invention includes two data flow directions, namely the downstream data flow from the main control module to ZYNQ or FPGA7 and the upstream data flow from ZYNQ or FPGA7 to the main control module.

对本发明进一步的描述,BRAM包括两个特殊的32位寄存器,分别是主控的控制状态寄存器(地址为0)和PS的控制状态寄存器(地址为4),用于系统中控制、状态信息的交换,寄存器按位定义完成标志(1位)、读/写标志(1位)、控制命令(6位)、数据类型(8位)、数据长度(16位)。Further description of the present invention, BRAM comprises two special 32-bit registers, is respectively the control status register (address is 0) of main control and the control status register (address is 4) of PS, is used for control, status information in the system Exchange, the register defines completion flag (1 bit), read/write flag (1 bit), control command (6 bits), data type (8 bits), and data length (16 bits) by bit.

对本发明进一步的描述,包括4种数据类型,分别为FPGA7BIT文件、参数文件、波形参数和状态数据、波形输入和输出数据;A further description of the present invention includes 4 types of data, which are respectively FPGA7BIT files, parameter files, waveform parameters and state data, waveform input and output data;

对本发明进一步的描述,包括6种数据处理流程,分别为配置FPGA7芯片BIT文件、配置参数文件、配置波形参数、读取波形状态、波形数据输入、波形数据输出流程。The further description of the present invention includes 6 kinds of data processing flows, which are respectively configuring FPGA7 chip BIT files, configuring parameter files, configuring waveform parameters, reading waveform status, inputting waveform data, and outputting waveform data.

对本发明进一步的描述,其中FPGA7芯片BIT文件流程包括下列步骤:The present invention is described further, wherein FPGA7 chip BIT file flow process comprises the following steps:

(101):PS已经完成系统初始化处于等待中断触发的状态,主控模块的SCA设备驱动程序打开FPGA7 BIT文件,并将文件数据写入PCIE BAR0地址空间,通过M2PCIE模块的BAR地址映射将数据写入波形模块的DDR;(101): The PS has completed the system initialization and is in the state of waiting for the interrupt to be triggered. The SCA device driver of the main control module opens the FPGA7 BIT file, and writes the file data into the PCIE BAR0 address space, and writes the data through the BAR address mapping of the M2PCIE module Input the DDR of the waveform module;

(102):主控模块驱动程序将已写入数据长度、数据类型(FPGA BIT文件)、控制命令(配置FPGA)、读/写标志(写)、完成标志(Y)写入PCIE BAR1的地址0的寄存器(主控的控制状态寄存器,实际是BRAM的地址0的32位空间)中,进入等待PCIE的MSI中断触发状态;驱动程序写主控的控制状态寄存器的操作将促使中断控制逻辑模块触发PS中断,PS的中断服务程序将通过AXI BRAM Controller1读取BRAM地址0的寄存器(32位)获取文件数据长度、数据类型、控制命令、读/写标志和完成标志,同时由中断控制逻辑模块清除中断位,中断服务程序通过配置CDMA模块启动DMA方式读取DDR中的文件数据,并存到Linux文件系统中;(102): The main control module driver writes the written data length, data type (FPGA BIT file), control command (configuration FPGA), read/write flag (write), and completion flag (Y) to the address of PCIE BAR1 0 register (the control status register of the main control, which is actually the 32-bit space of address 0 of the BRAM), enters the MSI interrupt trigger state waiting for PCIE; the operation of the driver writing the control status register of the main control will prompt the interrupt control logic module Trigger the PS interrupt, and the PS interrupt service program will read the register (32 bits) of BRAM address 0 through AXI BRAM Controller1 to obtain the file data length, data type, control command, read/write flag and completion flag, and at the same time, the logic module will be controlled by the interrupt Clear the interrupt bit, the interrupt service program reads the file data in the DDR by configuring the CDMA module to start DMA, and saves it in the Linux file system;

(103):PS的程序将已读取的数据长度、读/写标志(读)、完成标志(Y)写入AXIBRAM Controller1地址为4的寄存器(PS的控制状态寄存器,实际是BRAM的地址4的32位空间),并通过M2PCIE模块向PCIE接口写MSI中断的TLP包,该包将触发主控PCIE MSI中断并运行PCIE中断服务程序,PCIE中断服务程序清除中断位,并读取PCIE BAR1的地址4的寄存器(实际是BRAM的地址4的32位空间)获取PS程序设置的相关信息,PS的程序写完寄存器后再次进入待中断触发状态;(103): The program of PS writes the read data length, read/write flag (read), completion flag (Y) into the register of AXIBRAM Controller1 address 4 (the control state register of PS is actually the address 4 of BRAM 32-bit space), and write the MSI interrupt TLP packet to the PCIE interface through the M2PCIE module, the packet will trigger the main control PCIE MSI interrupt and run the PCIE interrupt service program, the PCIE interrupt service program clears the interrupt bit, and reads the PCIE BAR1 The register at address 4 (actually the 32-bit space of address 4 of BRAM) obtains the relevant information set by the PS program, and the PS program enters the interrupt trigger state again after writing the register;

(104):主控模块驱动程序在下载BIT文件完成后,将配置FPGA的命令写入主控的控制状态寄存器,包括配置文件名、标志;同时中断控制逻辑模块触发PS中断,中断服务程序又读取主控的控制状态寄存器,获取配置FPGA7的参数,然后打开FPGA7的BIT文件,启动DMA向AXI SelectMAP模块写入文件数据,并检查FPGA配置成功的标志;(104): After the downloading of the BIT file, the main control module driver program writes the command of configuring the FPGA into the control state register of the main control, including configuration file name and sign; the interrupt control logic module triggers the PS interrupt at the same time, and the interrupt service program again Read the control status register of the main control, obtain the parameters for configuring FPGA7, then open the BIT file of FPGA7, start DMA to write file data to the AXI SelectMAP module, and check the sign of successful FPGA configuration;

(105):PS的程序将配置结果写入PS的控制状态寄存器,包括配置FPGA是否成功的标志状态参数,并通过M2PCIE模块向PCIE接口写MSI中断的TLP包触发主控PCIE MSI中断,主控的PCIE中断服务程序清除中断位并读PS的控制状态寄存器获取FPGA配置状态参数,至此配置FPGA7 BIT文件结束;(105): The program of the PS writes the configuration result into the control status register of the PS, including the flag status parameter of whether the configuration FPGA is successful, and writes the TLP packet of the MSI interrupt to the PCIE interface through the M2PCIE module to trigger the main control PCIE MSI interrupt, and the main control The PCIE interrupt service program clears the interrupt bit and reads the control status register of the PS to obtain the FPGA configuration status parameters, so far the configuration of the FPGA7 BIT file ends;

其中波形数据输入流程包括下列步骤:The waveform data input process includes the following steps:

(201):PS已经完成系统初始化处于等待中断触发的状态,主控模块的SCA设备驱动程序打开波形输入文件,并将文件数据写入PCIE BAR0地址空间,实际上数据被直接写入了波形模块的DDR中;(201): The PS has completed the system initialization and is in the state of waiting for the interrupt trigger. The SCA device driver of the main control module opens the waveform input file and writes the file data into the PCIE BAR0 address space. In fact, the data is directly written into the waveform module. in the DDR;

(202):主控模块驱动程序将已写入数据长度、数据类型(波形数据文件)、控制命令(数据输入)、读/写标志(写)、完成标志(Y)写入PCIE BAR1的地址0的寄存器(主控的控制状态寄存器)中,进入等待PCIE的MSI中断触发状态;驱动程序写主控的控制状态寄存器的操作将促使中断控制逻辑模块触发PS中断,PS的中断服务程序将通过AXI BRAMController1读取BRAM地址0的寄存器(32位)获取文件数据长度、数据类型、控制命令、读/写标志和完成标志,同时由中断控制逻辑模块清除中断位,中断服务程序设置数据读写逻辑控制模块的寄存器,设置数据方向、长度、是否循环标志之后,通过配置CDMA模块启动DMA方式读取DDR中的文件数据,送到数据读写控制逻辑模块的SRIO接口(输入到波形);(202): The main control module driver writes the written data length, data type (waveform data file), control command (data input), read/write flag (write), and completion flag (Y) into the address of PCIE BAR1 0 register (the control status register of the main control), enter the MSI interrupt trigger state waiting for PCIE; the operation of the driver program writing the control status register of the main control will prompt the interrupt control logic module to trigger the PS interrupt, and the interrupt service program of the PS will pass AXI BRAMController1 reads the register (32 bits) of BRAM address 0 to obtain the file data length, data type, control command, read/write flag and completion flag, and at the same time, the interrupt control logic module clears the interrupt bit, and the interrupt service program sets the data read and write logic Control the register of the module, after setting the data direction, length, and whether to cycle the flag, start the DMA mode by configuring the CDMA module to read the file data in the DDR, and send it to the SRIO interface of the data read and write control logic module (input to the waveform);

(203):PS的程序将已读取的数据长度、读/写标志(读)、完成标志(Y)写入AXIBRAMController1地址为4的寄存器(PS的控制状态寄存器),并通过M2PCIE模块向PCIE接口写MSI中断的TLP包,该包将触发主控PCIE MSI中断并运行PCIE中断服务程序,PCIE中断服务程序清除中断位,并读取PCIE BAR1的地址4的寄存器获取PS程序设置的相关信息;(203): The program of PS writes the data length that has read, read/write sign (reading), completion sign (Y) into the register (the control state register of PS) that AXIBRAMController1 address is 4, and sends to PCIE by M2PCIE module The interface writes the TLP packet of MSI interrupt, which will trigger the main control PCIE MSI interrupt and run the PCIE interrupt service program, the PCIE interrupt service program clears the interrupt bit, and reads the register of address 4 of PCIE BAR1 to obtain the relevant information set by the PS program;

(204):读写控制逻辑模块判断循环读写标志,如果是循环写,在写数据后又启动DMA方式读取DDR中的数据通过SRIO送到波形,直到退出循环写为止,输入波形数据进程结束;(204): The read-write control logic module judges the cyclic read-write flag. If it is cyclic writing, start the DMA mode to read the data in the DDR after writing the data and send it to the waveform through SRIO until the loop write is exited, and the input waveform data process Finish;

其中波形数据输出流程包括下列步骤:The waveform data output process includes the following steps:

(301):PS已经完成系统初始化处于等待中断触发的状态,主控模块的SCA设备驱动程序启动读取波形输出数据进程,将要读取数据长度、数据类型(波形数据文件)、控制命令(数据输出)、读/写标志(读)、完成标志(N)写入PCIE BAR1的地址0的寄存器(主控的控制状态寄存器)中,进入等待PCIE的MSI中断触发状态;(301): The PS has completed the system initialization and is in the state of waiting for the interrupt to be triggered. The SCA device driver of the main control module starts the process of reading the waveform output data, and will read the data length, data type (waveform data file), control command (data output), read/write flag (read), and completion flag (N) are written into the register of address 0 of PCIE BAR1 (the control status register of the main control), and enter the MSI interrupt trigger state waiting for PCIE;

(302):驱动程序写主控的控制状态寄存器的操作将促使中断控制逻辑模块触发PS中断,PS的中断服务程序将通过AXI BRAM Controller1读取BRAM地址0的寄存器(32位)获取文件数据长度、数据类型、控制命令、读/写标志和完成标志,同时由中断控制逻辑模块清除中断位,中断服务程序设置数据读写逻辑控制模块的寄存器,设置数据方向、长度、是否循环标志之后,启动数据读写逻辑控制从SRIO接口读数据送到DDR;(302): The operation of the driver program writing the control status register of the main control will prompt the interrupt control logic module to trigger the PS interrupt, and the interrupt service program of the PS will read the register (32 bits) of BRAM address 0 through AXI BRAM Controller1 to obtain the file data length , data type, control command, read/write flag and completion flag, and at the same time, the interrupt control logic module clears the interrupt bit, the interrupt service program sets the register of the data read and write logic control module, and after setting the data direction, length, and whether to loop the flag, start Data read and write logic control read data from SRIO interface to DDR;

(303):PS的程序将已读取的数据长度、读/写标志(读)、完成标志(Y)写入AXIBRAM Controller1地址为4的寄存器(PS的控制状态寄存器),并通过M2PCIE模块向PCIE接口写MSI中断的TLP包,该包将触发主控PCIE MSI中断并运行PCIE中断服务程序,PCIE中断服务程序清除中断位,并读取PCIE BAR1的地址4的寄存器获取PS程序设置的相关信息;(303): The program of PS writes the data length, read/write sign (reading), completion sign (Y) into AXIBRAM Controller1 address to be 4 registers (the control state register of PS) by the program of PS, and send to by M2PCIE module The PCIE interface writes the TLP packet of the MSI interrupt, which will trigger the main control PCIE MSI interrupt and run the PCIE interrupt service program, the PCIE interrupt service program clears the interrupt bit, and reads the register of address 4 of PCIE BAR1 to obtain the relevant information set by the PS program ;

(304):驱动程序启动DMA方式读PCIE BAR0(DDR)中波形输出数据,判断循环读写标志,如果是循环读,在读完数据后又启动DMA方式读取PCIE BAR0中的数据,直到退出循环读为止,输出波形数据进程结束。(304): The driver starts the DMA mode to read the waveform output data in PCIE BAR0 (DDR), and judges the cyclic read and write flag. If it is a cyclic read, it starts the DMA mode to read the data in PCIE BAR0 after reading the data, until it exits the loop Read until the process of outputting waveform data ends.

与现有技术相比,本发明的优点:综合利用高速总线、PCIE和SRIO接口、DMA、中断机制、大容量DDR等技术与资源,实现FPGA波形快速重配、波形参数快速配置、波形状态实时回传、波形数据高速传输。Compared with the prior art, the present invention has the advantages of comprehensively utilizing high-speed bus, PCIE and SRIO interfaces, DMA, interrupt mechanism, large-capacity DDR and other technologies and resources to realize fast reconfiguration of FPGA waveforms, rapid configuration of waveform parameters, and real-time waveform status High-speed transmission of backhaul and waveform data.

附图说明Description of drawings

图1为本发明方法的系统架构图。Fig. 1 is a system architecture diagram of the method of the present invention.

图2为本发明方法的ZYNQ内部功能模块连接关系图。Fig. 2 is the connection diagram of ZYNQ internal function modules of the method of the present invention.

图3为配置FPGA7 BIT文件流程图。Figure 3 is a flow chart of configuring the FPGA7 BIT file.

图4为配置波形参数和读取波形状态流程图。Figure 4 is a flow chart of configuring waveform parameters and reading waveform status.

图5为波形数据输入流程图。Figure 5 is a flow chart of waveform data input.

图6为波形数据输出流程图。Figure 6 is a flow chart of waveform data output.

具体实施方式Detailed ways

下面将结合附图对本发明实施例中的技术方案进行清楚、完整地描述,显然,所描述的实施例仅仅是本发明一部分实施例,而不是全部的实施例。基于本发明中的实施例,本领域普通技术人员在没有作出创造性劳动前提下所获得的所有其他实施例,都属于本发明保护的范围。The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Apparently, the described embodiments are only some, not all, embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without creative efforts fall within the protection scope of the present invention.

实施例1:Example 1:

如图1,本发明的验证系统由主控模块和波形模块两个模块组成,通过背板的VPX总线实现互联互通,采用PCIE接口并配置为X4 5.0GT/s。As shown in Figure 1, the verification system of the present invention is composed of two modules, the main control module and the waveform module, and realizes interconnection and intercommunication through the VPX bus of the backplane, adopts PCIE interface and is configured as X4 5.0GT/s.

主控模块包含X86架构的硬件平台,运行操作系统以及SCA的核心框架CF、CORBA中间件和SCA的波形应用,POSIX兼容的应用程序环境描述AEP为SCA的波形应用提供了标准的应用程序函数接口API,SCA设备管理器控制下的Device为波形模块中FPGA7芯片提供驱动和代理功能,驱动实现对波形模块中各硬件资源的访问接口,代理实现SCA对该设备的管理接口,包括设备初始化(initializ)、加载程序(load)、运行程序(execute)、配置参数(configure)、查询状态(query)、终止程序(terminate)、卸载程序(unload)、释放对象(releaseObject)等。X86硬件平台包含PCIE接口的根组件RC,实现与波形模块的PCIE接口的通信与配置。The main control module includes the hardware platform of X86 architecture, running the operating system and SCA's core framework CF, CORBA middleware and SCA's waveform application, POSIX compatible application program environment description AEP provides a standard application program function interface for SCA's waveform application API, Device under the control of the SCA device manager provides driver and proxy functions for the FPGA7 chip in the waveform module, the driver implements the access interface to each hardware resource in the waveform module, and the agent implements the SCA management interface for the device, including device initialization (initializ ), load program (load), run program (execute), configuration parameter (configure), query status (query), terminate program (terminate), unload program (unload), release object (releaseObject), etc. The X86 hardware platform includes the root component RC of the PCIE interface, which realizes the communication and configuration with the PCIE interface of the waveform module.

波形模块主要包括ZYNQ系列和FPGA 7系列FPGA芯片、大容量DDR芯片等。ZYNQ与FPGA7之间的接口主要有SelectMAP、LVDS、SRIO等,SelectMAP接口实现ZYNQ对FPGA7芯片的波形重配置,差分LVDS接口实现ZYNQ对FPGA7芯片的波形参数配置和波形状态查询,SRIO接口作为波形数据输入输出接口,通过ZYNQ的PCIE接口终端设备EP与主控模块的PCIE接口通信。DDR作为数据和大文件的转存设备,通过ZYNQ的PCIE接口映射到主控模块PCIE接口的BAR0上。Waveform modules mainly include ZYNQ series and FPGA 7 series FPGA chips, large-capacity DDR chips, etc. The interface between ZYNQ and FPGA7 mainly includes SelectMAP, LVDS, SRIO, etc. SelectMAP interface realizes ZYNQ’s waveform reconfiguration of FPGA7 chip, differential LVDS interface realizes ZYNQ’s waveform parameter configuration and waveform status query of FPGA7 chip, and SRIO interface is used as waveform data The input and output interface communicates with the PCIE interface of the main control module through the PCIE interface terminal equipment EP of ZYNQ. DDR is used as a dump device for data and large files, and is mapped to BAR0 of the PCIE interface of the main control module through the PCIE interface of ZYNQ.

ZYNQ内部的功能模块及连接关系如图2所示,主要包括PS(ZYNQ7 ProcessingSystem)、总线互联模块(AXI Interconnect)、块内存控制器(AXI BRAM Controller)、M2PCIE内存映射模块(AXI Memory Mapped PCI Express)、CDMA模块(AXI Central DirectMemory Access)、SelectMAP接口模块(AXI SelectMAP)、DDR、BRAM、数据读写控制逻辑模块、中断控制逻辑模块。The functional modules and connection relationship inside ZYNQ are shown in Figure 2, mainly including PS (ZYNQ7 Processing System), bus interconnection module (AXI Interconnect), block memory controller (AXI BRAM Controller), M2PCIE memory mapping module (AXI Memory Mapped PCI Express ), CDMA module (AXI Central DirectMemory Access), SelectMAP interface module (AXI SelectMAP), DDR, BRAM, data read and write control logic module, interrupt control logic module.

设计时首先配置PS的系统参数,包括配置DDR的参数、设置GPO(M_AXI_GPO)接口、HPO(S_AXI_HPO)接口位宽等参数,以及中断触发等参数;其次添加各个功能模块到工程,然后连接各个功能模块的接口并配置相应的参数,本实施例中AXI总线宽度都是32位。PS通过GPO的主端口M_AXI_GPO与AXIInterconnect1的AXI总线从端口相连,通过HPO的从端口与AXI Interconnect2的AXI总线主端口相连,中断控制逻辑的输出连接PS的中断请求(对应PL到PS中断号);AXI Interconnect1的AXI总线主端口分别与CDMA、M2PCIE、AXI BRAMController1、AXI SelectMAP、数据读写控制逻辑5个模块的AXI总线从端口相连,支持PS访问5个模块地址空间。M2PCIE模块的PCIE接口通过VPX总线与主控模块的PCIE接口互联,AXI总线主端口与AXI Interconnect3的AXI总线从端口相连,AXI Interconnect3的AXI总线主端口又分别与DDR、AXI BRAM Controller2、AXI BRAM Controller3的AXI总线从端口相连,配置M2PCIE模块参数,通道为X4、带宽速率为5.0GT/s,3个BAR地址分别映射DDR、AXI BRAMController2、AXI BRAM Controller3的地址空间,实现主控模块通过PCIE接口可以直接访问DDR、AXI BRAM Controller2、AXI BRAMController3的地址空间。CDMA模块的AXI总线主端口与AXI Interconnect2的AXI总线从端口相连,AXI Interconnect2的AXI总线主端口又与AXI Interconnect1和AXI Interconnect3级联,并与PS的HPO从端口相连,实现CDMA对DDR、M2PCIE、SelectMAP、AXI BRAM Controller1、AXI BRAM Controller3、数据读写控制逻辑模块的地址空间的访问。AXI SelectMAP模块通过SelectMAP接口直接与FPGA7相连,AXIBRAM Controller3模块通过LVDS接口与FPGA7相连,数据读写控制逻辑模块通过SRIO接口与FPGA7相连。BRAM的两个端口分别连接AXI BRAM Controller1和AXI BRAM Controller2,实现主控模块通过PCIE接口、PS通过AXI总线访问同一个BRAM的地址空间,BRAM中的前8字节作为主控模块的控制状态寄存器和When designing, first configure the system parameters of the PS, including configuring DDR parameters, setting parameters such as GPO (M_AXI_GPO) interface, HPO (S_AXI_HPO) interface bit width, and interrupt triggering parameters; secondly, add each function module to the project, and then connect each function The interface of the module is configured with corresponding parameters. In this embodiment, the width of the AXI bus is 32 bits. The PS is connected to the AXI bus slave port of AXIInterconnect1 through the main port M_AXI_GPO of GPO, and connected to the AXI bus master port of AXI Interconnect2 through the slave port of HPO, and the output of the interrupt control logic is connected to the interrupt request of PS (corresponding to PL to PS interrupt number); The AXI bus master port of AXI Interconnect1 is respectively connected to the AXI bus slave ports of the 5 modules of CDMA, M2PCIE, AXI BRAMController1, AXI SelectMAP, and data read and write control logic, and supports PS access to 5 module address spaces. The PCIE interface of the M2PCIE module is interconnected with the PCIE interface of the main control module through the VPX bus, the AXI bus master port is connected with the AXI bus slave port of AXI Interconnect3, and the AXI bus master port of AXI Interconnect3 is respectively connected with DDR, AXI BRAM Controller2, and AXI BRAM Controller3 The AXI bus is connected from the port, configure the M2PCIE module parameters, the channel is X4, the bandwidth rate is 5.0GT/s, and the three BAR addresses are respectively mapped to the address spaces of DDR, AXI BRAM Controller2, and AXI BRAM Controller3, so that the main control module can be accessed through the PCIE interface. Directly access the address space of DDR, AXI BRAM Controller2, AXI BRAM Controller3. The AXI bus master port of the CDMA module is connected to the AXI bus slave port of AXI Interconnect2, and the AXI bus master port of AXI Interconnect2 is cascaded with AXI Interconnect1 and AXI Interconnect3, and is connected with the HPO slave port of PS to realize CDMA to DDR, M2PCIE, SelectMAP, AXI BRAM Controller1, AXI BRAM Controller3, data read and write control access to the address space of the logic module. The AXI SelectMAP module is directly connected to the FPGA7 through the SelectMAP interface, the AXIBRAM Controller3 module is connected to the FPGA7 through the LVDS interface, and the data read and write control logic module is connected to the FPGA7 through the SRIO interface. The two ports of BRAM are respectively connected to AXI BRAM Controller1 and AXI BRAM Controller2, so that the main control module accesses the address space of the same BRAM through the PCIE interface and the PS through the AXI bus, and the first 8 bytes in the BRAM are used as the control status register of the main control module and

通过SelectMAP模块配置FPGA7BIT文件分为两个阶段,第一阶段是下载BIT文件,第二阶段是配置FPGA;整个配置过程需主控模块的SCA设备驱动程序与PS中的程序配合进行,详细步骤如图3所示:Configuring the FPGA7BIT file through the SelectMAP module is divided into two stages. The first stage is to download the BIT file, and the second stage is to configure the FPGA; the entire configuration process requires the cooperation of the SCA device driver of the main control module and the program in the PS. The detailed steps are as follows: As shown in Figure 3:

第一步:PS已经完成系统初始化处于等待中断触发的状态,主控模块的SCA设备驱动程序打开FPGA7BIT文件,并将文件数据写入PCIE BAR0地址空间,实际上通过M2PCIE模块的BAR地址映射,数据被直接写入了波形模块的DDR中;Step 1: PS has completed the system initialization and is in the state of waiting for interrupt triggering. The SCA device driver of the main control module opens the FPGA7BIT file and writes the file data into the PCIE BAR0 address space. In fact, through the BAR address mapping of the M2PCIE module, the data It is directly written into the DDR of the waveform module;

第二步:主控模块驱动程序将已写入数据长度、数据类型(FPGA BIT文件)、控制命令(配置FPGA)、读/写标志(写)、完成标志(Y)写入PCIE BAR1的地址0的寄存器(主控的控制状态寄存器,实际是BRAM的地址0的32位空间)中,进入等待PCIE的MSI中断触发状态;驱动程序写主控的控制状态寄存器的操作将促使中断控制逻辑模块触发PS中断,PS的中断服务程序将通过AXI BRAM Controller1读取BRAM地址0的寄存器(32位)获取文件数据长度、数据类型、控制命令、读/写标志和完成标志等,同时由中断控制逻辑模块清除中断位,中断服务程序通过配置CDMA模块启动DMA方式读取DDR中的文件数据,并存到Linux文件系统中;Step 2: The main control module driver writes the written data length, data type (FPGA BIT file), control command (configure FPGA), read/write flag (write), and completion flag (Y) to the address of PCIE BAR1 0 register (the control status register of the main control, which is actually the 32-bit space of address 0 of the BRAM), enters the MSI interrupt trigger state waiting for PCIE; the operation of the driver writing the control status register of the main control will prompt the interrupt control logic module Trigger the PS interrupt, the PS interrupt service program will read the register (32 bits) of BRAM address 0 through AXI BRAM Controller1 to obtain the file data length, data type, control command, read/write flag and completion flag, etc., and the interrupt control logic The module clears the interrupt bit, and the interrupt service program reads the file data in the DDR by configuring the CDMA module to start DMA, and saves it in the Linux file system;

第三步:PS的程序将已读取的数据长度、读/写标志(读)、完成标志(Y)等写入AXIBRAM Controller1地址为4的寄存器(PS的控制状态寄存器,实际是BRAM的地址4的32位空间),并通过M2PCIE模块向PCIE接口写MSI中断的TLP包,该包将触发主控PCIE MSI中断并运行PCIE中断服务程序,PCIE中断服务程序清除中断位,并读取PCIE BAR1的地址4的寄存器(实际是BRAM的地址4的32位空间)获取PS程序设置的相关信息,PS的程序写完寄存器后再次进入等待中断触发状态;Step 3: The PS program writes the read data length, read/write flag (read), completion flag (Y), etc. into the register of AXIBRAM Controller1 whose address is 4 (the control status register of PS is actually the address of BRAM 4’s 32-bit space), and write the MSI interrupt TLP packet to the PCIE interface through the M2PCIE module, the packet will trigger the main control PCIE MSI interrupt and run the PCIE interrupt service program, the PCIE interrupt service program clears the interrupt bit, and reads the PCIE BAR1 The register of address 4 (actually the 32-bit space of address 4 of BRAM) obtains the relevant information set by the PS program, and the PS program enters the waiting interrupt trigger state again after writing the register;

第四步:主控模块驱动程序在下载BIT文件完成后,将配置FPGA的命令写入主控的控制状态寄存器,包括配置文件名、标志等;同时中断控制逻辑模块触发PS中断,中断服务程序又读取主控的控制状态寄存器,获取配置FPGA7的参数,然后打开FPGA7的BIT文件,启动DMA向AXI SelectMAP模块写入文件数据,并检查FPGA配置成功的标志;Step 4: After the main control module driver finishes downloading the BIT file, write the command to configure the FPGA into the control status register of the main control, including the configuration file name, flag, etc.; at the same time, the interrupt control logic module triggers the PS interrupt, and the interrupt service program Read the control status register of the main control again, obtain the parameters for configuring FPGA7, then open the BIT file of FPGA7, start DMA to write file data to the AXI SelectMAP module, and check the sign of FPGA configuration success;

第五步:PS的程序将配置结果写入PS的控制状态寄存器,包括配置FPGA是否成功的标志状态等参数,并通过M2PCIE模块向PCIE接口写MSI中断的TLP包触发主控PCIE MSI中断,主控的PCIE中断服务程序清除中断位并读PS的控制状态寄存器获取FPGA配置状态等参数,至此配置FPGA7BIT文件结束。Step 5: The program of the PS writes the configuration result into the control status register of the PS, including parameters such as the flag status of whether the FPGA configuration is successful, and writes the MSI interrupt TLP packet to the PCIE interface through the M2PCIE module to trigger the main control PCIE MSI interrupt. The PCIE interrupt service program of the control clears the interrupt bit and reads the control status register of the PS to obtain parameters such as the FPGA configuration status, and the configuration of the FPGA7BIT file ends.

主控模块PCIE的BAR2通过M2PCIE模块映射AXI BRAM Controller3的地址空间,通过LVDS接口实现配置波形参数和读取波形形态,详细流程如图4所示:The BAR2 of the main control module PCIE maps the address space of the AXI BRAM Controller3 through the M2PCIE module, and configures the waveform parameters and reads the waveform form through the LVDS interface. The detailed process is shown in Figure 4:

主控模块SCA设备驱动程序首先打开波形参数文件,然后通过PCIE接口按地址写入BAR2地址空间,AXI BRAM Controller3模块随后输出时钟、地址、参数数据、写使能信号到LVDS接口,驱动程序完成配置波形参数;The SCA device driver of the main control module first opens the waveform parameter file, and then writes the address into the BAR2 address space through the PCIE interface. The AXI BRAM Controller3 module then outputs the clock, address, parameter data, and write enable signal to the LVDS interface, and the driver completes the configuration. Waveform parameters;

主控模块SCA设备驱动程序读取波形形态,首先通过PCIE接口按地址读取BAR2地址空间,AXI BRAM Controller3模块随后输出时钟、地址、读使能信号到LVDS接口,在下一个时钟上升沿到来时返回参数地址对应的值给PCIE接口,驱动程序完成读取波形状态。The main control module SCA device driver reads the waveform shape, first reads the BAR2 address space through the PCIE interface according to the address, and the AXI BRAM Controller3 module then outputs the clock, address, and read enable signal to the LVDS interface, and returns when the next rising edge of the clock arrives The value corresponding to the parameter address is given to the PCIE interface, and the driver completes reading the waveform state.

主控模块PCIE的BAR0通过M2PCIE模块映射DDR地址空间,SCA设备驱动程序通过向PCIE接口BAR0地址空间写来实现波形数据输入,详细流程如图5所示:The BAR0 of the main control module PCIE maps the DDR address space through the M2PCIE module, and the SCA device driver implements waveform data input by writing to the BAR0 address space of the PCIE interface. The detailed process is shown in Figure 5:

第一步:PS已经完成系统初始化处于等待中断触发的状态,主控模块的SCA设备驱动程序打开波形输入文件,并将文件数据写入PCIE BAR0地址空间,实际上数据被直接写入了波形模块的DDR中;Step 1: PS has completed the system initialization and is waiting for an interrupt trigger. The SCA device driver of the main control module opens the waveform input file and writes the file data into the PCIE BAR0 address space. In fact, the data is directly written into the waveform module. in the DDR;

第二步:主控模块驱动程序将已写入数据长度、数据类型(波形数据文件)、控制命令(数据输入)、读/写标志(写)、完成标志(Y)写入PCIE BAR1的地址0的寄存器(主控的控制状态寄存器)中,进入等待PCIE的MSI中断触发状态;驱动程序写主控的控制状态寄存器的操作将促使中断控制逻辑模块触发PS中断,PS的中断服务程序将通过AXI BRAMController1读取BRAM地址0的寄存器(32位)获取文件数据长度、数据类型、控制命令、读/写标志和完成标志等,同时由中断控制逻辑模块清除中断位,中断服务程序设置数据读写逻辑控制模块的寄存器,设置数据方向、长度、是否循环等标志之后,通过配置CDMA模块启动DMA方式读取DDR中的文件数据,送到数据读写控制逻辑模块的SRIO接口(输入到波形);Step 2: The main control module driver writes the written data length, data type (waveform data file), control command (data input), read/write flag (write), and completion flag (Y) into the address of PCIE BAR1 0 register (the control status register of the main control), enter the MSI interrupt trigger state waiting for PCIE; the operation of the driver program writing the control status register of the main control will prompt the interrupt control logic module to trigger the PS interrupt, and the interrupt service program of the PS will pass AXI BRAMController1 reads the register (32 bits) of BRAM address 0 to obtain the file data length, data type, control command, read/write flag and completion flag, etc. At the same time, the interrupt control logic module clears the interrupt bit, and the interrupt service program sets the data read and write The register of the logic control module, after setting the data direction, length, whether to cycle and other signs, read the file data in the DDR by configuring the CDMA module to start the DMA mode, and send it to the SRIO interface of the data read and write control logic module (input to the waveform);

第三步:PS的程序将已读取的数据长度、读/写标志(读)、完成标志(Y)等写入AXIBRAM Controller1地址为4的寄存器(PS的控制状态寄存器),并通过M2PCIE模块向PCIE接口写MSI中断的TLP包,该包将触发主控PCIE MSI中断并运行PCIE中断服务程序,PCIE中断服务程序清除中断位,并读取PCIE BAR1的地址4的寄存器获取PS程序设置的相关信息;Step 3: The PS program writes the read data length, read/write flag (read), completion flag (Y), etc. into the AXIBRAM Controller1 address 4 register (PS control status register), and through the M2PCIE module Write the TLP package of MSI interrupt to the PCIE interface, the package will trigger the main control PCIE MSI interrupt and run the PCIE interrupt service program, the PCIE interrupt service program clears the interrupt bit, and reads the register of address 4 of PCIE BAR1 to obtain the relevant information set by the PS program information;

第四步:读写控制逻辑模块判断循环读写标志,如果是循环写,在写数据后又启动DMA方式读取DDR中的数据通过SRIO送到波形,直到退出循环写为止,输入波形数据进程结束。Step 4: The read-write control logic module judges the cyclic read-write flag. If it is cyclic writing, start the DMA mode to read the data in the DDR after writing the data and send it to the waveform through SRIO until the cyclic writing is exited, and the input waveform data process Finish.

SCA设备驱动程序通过从PCIE接口BAR0地址空间读来实现波形数据输出,详细流程如图6所示:The SCA device driver implements waveform data output by reading from the PCIE interface BAR0 address space. The detailed process is shown in Figure 6:

第一步:PS已经完成系统初始化处于等待中断触发的状态,主控模块的SCA设备驱动程序启动读取波形输出数据进程,将要读取数据长度、数据类型(波形数据文件)、控制命令(数据输出)、读/写标志(读)、完成标志(N)写入PCIE BAR1的地址0的寄存器(主控的控制状态寄存器)中,进入等待PCIE的MSI中断触发状态;Step 1: The PS has completed the system initialization and is in the state of waiting for the interrupt trigger. The SCA device driver of the main control module starts the process of reading the waveform output data, and the data length, data type (waveform data file), control command (data output), read/write flag (read), and completion flag (N) are written into the register of address 0 of PCIE BAR1 (the control status register of the main control), and enter the MSI interrupt trigger state waiting for PCIE;

第二步:驱动程序写主控的控制状态寄存器的操作将促使中断控制逻辑模块触发PS中断,PS的中断服务程序将通过AXI BRAM Controller1读取BRAM地址0的寄存器(32位)获取文件数据长度、数据类型、控制命令、读/写标志和完成标志等,同时由中断控制逻辑模块清除中断位,中断服务程序设置数据读写逻辑控制模块的寄存器,设置数据方向、长度、是否循环等标志之后,启动数据读写逻辑控制从SRIO接口读数据送到DDR;Step 2: The operation of the driver program to write the control status register of the main control will prompt the interrupt control logic module to trigger the PS interrupt, and the interrupt service program of the PS will read the register (32 bits) of BRAM address 0 through AXI BRAM Controller1 to obtain the file data length , data type, control command, read/write flag and completion flag, etc., and the interrupt control logic module clears the interrupt bit at the same time. , Start the data read and write logic control to read data from the SRIO interface and send it to DDR;

第三步:PS的程序将已读取的数据长度、读/写标志(读)、完成标志(Y)等写入AXIBRAM Controller1地址为4的寄存器(PS的控制状态寄存器),并通过M2PCIE模块向PCIE接口写MSI中断的TLP包,该包将触发主控PCIE MSI中断并运行PCIE中断服务程序,PCIE中断服务程序清除中断位,并读取PCIE BAR1的地址4的寄存器获取PS程序设置的相关信息;Step 3: The PS program writes the read data length, read/write flag (read), completion flag (Y), etc. into the AXIBRAM Controller1 address 4 register (PS control status register), and through the M2PCIE module Write the TLP package of MSI interrupt to the PCIE interface, the package will trigger the main control PCIE MSI interrupt and run the PCIE interrupt service program, the PCIE interrupt service program clears the interrupt bit, and reads the register of address 4 of PCIE BAR1 to obtain the relevant information set by the PS program information;

第四步:驱动程序启动DMA方式读PCIE BAR0(DDR)中波形输出数据,判断循环读写标志,如果是循环读,在读完数据后又启动DMA方式读取PCIE BAR0中的数据,直到退出循环读为止,输出波形数据进程结束。Step 4: The driver starts the DMA mode to read the waveform output data in PCIE BAR0 (DDR), and judges the cyclic read and write flag. If it is a cyclic read, start the DMA mode to read the data in PCIE BAR0 after reading the data, until it exits the loop Read until the process of outputting waveform data ends.

在本发明中,除非另有明确的规定和限定,第一特征在第二特征之“上”或之“下”可以包括第一和第二特征直接接触,也可以包括第一和第二特征不是直接接触而是通过它们之间的另外的特征接触。而且,第一特征在第二特征“之上'“上方”和“上面”包括第一特征在第二特征正上方和斜上方,或仅仅表示第一特征水平高度高于第二特征,第一特征在第二特征“之下'“下方”和“下面”包括第一特征在第二特征正上方和斜上方,或仅仅表示第一特征水平高度小于第一特征。In the present invention, unless otherwise clearly specified and limited, a first feature being "on" or "under" a second feature may include direct contact between the first and second features, and may also include the first and second features Not in direct contact but through another characteristic contact between them. Moreover, "above", "above" and "above" the first feature on the second feature include that the first feature is directly above and obliquely above the second feature, or simply means that the level of the first feature is higher than that of the second feature, the first "Below" and "under" the second feature include that the first feature is directly above and obliquely above the second feature, or simply means that the first feature is less horizontally than the first feature.

在本说明书的描述中,参考术语“一个实施例'“一些实施例'“示例'“具体示例'或“一些示例”等的描述意指结合该实施例或示例描述的具体特征、结构、材料或者特点包含于本发明的至少一个实施例或示例中,在本说明书中,对上述术语的示意性表述不一定指的是相同的实施例或示例。而且,描述的具体特征、结构、材料或者特点可以在任何的一个或多个实施例或示例中以合适的方式结合。In the description of this specification, descriptions referring to the terms "one embodiment", "some embodiments", "examples", "specific examples" or "some examples" mean specific features, structures, materials described in connection with the embodiments or examples. Or features are included in at least one embodiment or example of the invention, in this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.

尽管上面已经示出和描述了本发明的实施例,可以理解的是,上述实施例是示例性的,不能理解为对本发明的限制,本领域的普通技术人员在不脱离本发明的原理和宗旨的情况下在本发明的范围内可以对上述实施例进行变化、修改、替换和变型。Although the embodiments of the present invention have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be construed as limitations to the present invention. Variations, modifications, substitutions, and modifications to the above-described embodiments are possible within the scope of the present invention.

Claims (6)

1. A rapid system reconstruction method facing SCA and SDR is characterized in that the system is composed of a main control module and a waveform module, the main control module and the waveform module are interconnected and communicated through a VPX bus of a backboard, an interface adopts PCIE X4.0 GT/S, an X86 hardware platform of the main control module is used as a root component RC of a PCIE interface, a ZYNQ chip of the waveform module is used as PCIE interface terminal equipment EP, the main control module comprises SCA Device components, each SCA Device component is a logic abstraction of one waveform module and comprises a driving program of the waveform module, the internal program of the waveform module ZYNQ chip comprises ZYNQ7Processing System, 3 AXI interfaces, 3 AXI BRAM Controller, AXI Memory Mapped PCI Express, AXI Central Direct Memory Access, AXI select MAP, DDR, BRAM, data read-write control logic and interrupt control logic function modules, the ZYNQ7Processing System is interconnected with AXI Internect 1 through an M_AXI_GPO interface to access AXI Memory Mapped PCI Express, AXI Central Direct Memory Access, AXI SelectMAP, AXI BRAM Controller1 and address space of data read-write control logic, AXI Memory Mapped PCI Express is interconnected with AXI Internect 3 through an AXI bus to access DDR, AXI BRAM Controller2 and AXI BRAM Controller3, AXI Central Direct Memory Access is interconnected with AXI Internect 2 through an AXI bus and is connected with AXI Internect 1 and AXI Internect 3 in cascade to access AXI Memory Mapped PCI Express, AXI SelectMAP, AXI BRAM Controller1, AXI BRAM Controller3 and address space of data read-write control logic, access to DDR address space is realized through an S_AXI_HPO interface of ZYNQ7Processing System, AXI BRAM Controller and AXI BRAM Controller are respectively connected with two ports of BRAM, and the master control module is connected with address space of the same BRAM through a PCIE interface, ZYNQ7Processing System and the AXI bus; also included are 4 data types that are of a type, respectively an FPGA7BIT file, a parameter file, waveform parameters, state data and waveform input and output data; the 6 data processing flows are respectively a flow for configuring a BIT file, a configuration parameter file, a configuration waveform parameter, a waveform reading state, waveform data input and waveform data output of the FPGA7 chip;
the process for configuring the BIT file of the FPGA7 chip comprises the following steps of:
(101): ZYNQ7PROCESSING SYSTEM has completed the system initialization in the state of waiting for interrupt triggering, the SCA device driver of the main control module opens the FPGA7BIT file, writes the file data into PCIE BAR0 address space, and writes the data into the DDR of the waveform module through the BAR address mapping of the AXI Memory Mapped PCI Express module;
(102): the master control module driver writes the written data length, the data type of the FPGA BIT file, the control command of the configuration FPGA, the writing mark and the finishing mark of 'Y' into a master control state register of which the address is 0 of PCIE BAR1, wherein the register is actually positioned in a 32-BIT space of address 0 of BRAM and enters an MSI interrupt triggering state waiting for PCIE; the operation of the control state register of the driver program writing master control causes the interrupt control logic module to trigger ZYNQ7PROCESSING SYSTEM to interrupt, the interrupt service program of ZYNQ7PROCESSING SYSTEM obtains the file data length, the data type, the control command, the read/write mark and the completion mark by reading the 32-bit register of BRAM address 0 through AXI BRAM Controller1, meanwhile, the interrupt control logic module clears the interrupt bit, and the interrupt service program reads the file data in DDR through a DMA mode started by the configuration AXI Central Direct Memory Access module and stores the file data in a Linux file system;
(103): the program of ZYNQ7PROCESSING SYSTEM writes the read data length, the read flag and the completion flag representing 'Y' into a control state register of ZYNQ7PROCESSING SYSTEM with address 4 of AXI BRAM Controller, wherein the register is actually located in a 32-bit space of address 4 of BRAM, writes a TLP packet of MSI interrupt to a PCIE interface through AXI Memory Mapped PCI Express module, the packet triggers PCIE MSI interrupt and runs a PCIE interrupt service routine, the PCIE interrupt service routine clears interrupt bit, reads a control state register of ZYNQ7PROCESSING SYSTEM with address 4 of PCIE BAR1, is actually located in a 32-bit space of address 4 of BRAM, acquires relevant information set by the ZYNQ7PROCESSING SYSTEM, and enters a trigger state to be interrupted again after the program of ZYNQ7PROCESSING SYSTEM writes the register;
(104): after the host module driver finishes downloading the BIT file, writing a command for configuring the FPGA into a control state register of the host, wherein the control state register comprises a configuration file name and a mark; meanwhile, the interrupt control logic module triggers ZYNQ7PROCESSING SYSTEM to interrupt, an interrupt service routine reads a control state register of the main control to acquire parameters configuring the FPGA7, then a BIT file of the FPGA7 is opened, DMA is started to write file data into the AXI select MAP module, and a mark that the configuration of the FPGA is successful is checked;
(105): the program of ZYNQ7PROCESSING SYSTEM writes configuration results into a control state register of ZYNQ7PROCESSING SYSTEM, wherein the configuration results comprise flag state parameters of whether the configuration of the FPGA is successful or not, a TLP packet of MSI interrupt is written into a PCIE interface through a AXI Memory Mapped PCI Express module to trigger PCIE MSI interrupt to be controlled, a PCIE interrupt service routine of the main control clears interrupt BITs and reads the control state register of ZYNQ7PROCESSING SYSTEM to obtain the configuration state parameters of the FPGA, and the configuration of the FPGA7BIT file is finished;
the waveform data input flow comprises the following steps:
(201): ZYNQ7PROCESSING SYSTEM has completed the system initialization in the state of waiting for interrupt triggering, the SCA device driver of the main control module opens the waveform input file and writes the file data into PCIE BAR0 address space, in fact, the data is directly written into DDR of the waveform module;
(202): the master control module driver writes the written data length, the data type representing the waveform data file, the control command representing the data input, the writing mark and the finishing mark representing 'Y' into the master control state register of address 0 of PCIE BAR1, and enters the MSI interrupt triggering state waiting for PCIE; the operation of the control state register of the driver program writing master control causes the interrupt control logic module to trigger ZYNQ7PROCESSING SYSTEM to interrupt, the interrupt service program of ZYNQ7PROCESSING SYSTEM obtains the file data length, the data type, the control command, the read/write mark and the completion mark by reading the 32-bit register of BRAM address 0 through AXI BRAM Controller1, meanwhile, the interrupt control logic module clears the interrupt bit, the interrupt service program sets the register of the data read/write logic control module, after the data direction, the length and the cycle mark are set, the data read/write logic module is started to read the file data in the DDR through the configuration AXI Central Direct Memory Access module, and the data is sent to the SRIO interface of the data read/write control logic module to realize the data input to waveforms;
(203): the program of ZYNQ7PROCESSING SYSTEM writes the read data length, the read flag and the completion flag representing 'Y' into a AXI BRAM Controller control status register of ZYNQ7PROCESSING SYSTEM with address 4, writes a TLP packet of MSI interrupt to a PCIE interface through a AXI Memory Mapped PCI Express module, triggers PCIE MSI interrupt to be controlled and runs a PCIE interrupt service routine, clears interrupt bit by the PCIE interrupt service routine, reads a register with address 4 of PCIE BAR1 and acquires related information set by the ZYNQ7PROCESSING SYSTEM program;
(204): the read-write control logic module judges a cyclic read-write mark, if the cyclic write mark is cyclic write, the DMA mode is started to read data in the DDR after the data is written, the data is sent to the waveform through the SRIO until the cyclic write is exited, and the process of inputting the waveform data is ended;
the waveform data output flow comprises the following steps:
(301): ZYNQ7PROCESSING SYSTEM is in a state of waiting for interrupt triggering after completing system initialization, the SCA device driver of the main control module starts a waveform output data reading process, writes a to-be-read data length, a waveform data file data type, a data output control command, a reading mark and an 'N' completion mark into a control state register of the main control of address 0 of PCIE BAR1, and enters an MSI interrupt triggering state of waiting for PCIE;
(302): the operation of the control state register of the driver write master control causes the interrupt control logic module to trigger ZYNQ7PROCESSING SYSTEM to interrupt, the interrupt service program of ZYNQ7PROCESSING SYSTEM acquires the file data length, the data type, the control command, the read/write mark and the completion mark by reading the 32-bit register of BRAM address 0 through AXI BRAM Controller1, meanwhile, the interrupt control logic module clears the interrupt bit, the interrupt service program sets the register of the data read/write logic control module, and after the data direction, the length and the cycle mark are set, the data read/write logic control is started to read data from the SRIO interface to DDR;
(303): the program of ZYNQ7PROCESSING SYSTEM writes the read data length, the read flag and the completion flag representing 'Y' into a AXI BRAM Controller control status register of ZYNQ7PROCESSING SYSTEM with address 4, writes a TLP packet of MSI interrupt to a PCIE interface through a AXI Memory Mapped PCI Express module, triggers PCIE MSI interrupt to be controlled and runs a PCIE interrupt service routine, clears interrupt bit by the PCIE interrupt service routine, reads a register with address 4 of PCIE BAR1 and acquires related information set by the ZYNQ7PROCESSING SYSTEM program;
(304): the driver program starts the DMA mode to read PCIE BAR0, namely waveform output data in DDR, judges a cycle read-write mark, if the cycle read is completed, starts the DMA mode to read the data in PCIE BAR0 again after the data is read, and ends the process of outputting waveform data until the cycle read is exited.
2. The rapid system reconstruction method for SCA and SDR of claim 1 wherein the BAR addresses of AXI Memory Mapped PCI Express map the address spaces of DDR, AXI BRAM Controller, and AXI BRAM Controller3, respectively, and three address spaces of DDR, AXI BRAM Controller2, and AXI BRAM Controller3 are accessed directly by the master control module through the BAR addresses of PCIE.
3. The rapid system reconstruction method for SCA and SDR according to claim 1, wherein a PCIE interface of AXI Memory Mapped PCI Express is interconnected with a PCIE interface of a main control module, AXI SelectMAP is interconnected with an AXI SelectMAP interface of FPGA7, AXI BRAM Controller3 is interconnected with an LVDS interface of FPGA7, data read-write control logic is interconnected with an SRIO interface of FPGA7, and ZYNQ7PROCESSING SYSTEM configures registers of the data read-write control logic.
4. The rapid system reconfiguration method for SCA and SDR according to claim 1, comprising two interrupt mechanisms, namely a main control module PCIE MSI interrupt mechanism and a ZYNQ7PROCESSING SYSTEM interrupt control logic, wherein when the ZYNQ sends a specific TLP packet to the main control module through a PCIE interface, the MSI interrupt of the main control module PCIE will be triggered, the corresponding interrupt service routine will clear the interrupt, when the main control module writes data to a specific address of AXI BRAM Controller of the ZYNQ through the PCIE interface, the ZYNQ7PROCESSING SYSTEM interrupt control logic will trigger the ZYNQ7PROCESSING SYSTEM interrupt, and when the ZYNQ7PROCESSING SYSTEM reads data from a specific address of AXI BRAM Controller1 through an AXI bus, the ZYNQ7PROCESSING SYSTEM interrupt control logic will clear the ZYNQ7PROCESSING SYSTEM interrupt.
5. The rapid system reconstruction method for SCA and SDR according to claim 1, comprising two data streams, namely downstream from the master control module to the ZYNQ or FPGA7 and upstream from the ZYNQ or FPGA7 to the master control module.
6. The rapid system reconfiguration method for SCA and SDR according to claim 1, wherein BRAM includes two special 32-bit registers, a master control status register and a control status register of ZYNQ7PROCESSING SYSTEM, respectively, for control in the system, exchange of status information, the registers being defined by bits: a 1-bit completion flag, a 1-bit read/write flag, a 6-bit control command, an 8-bit data type, a 16-bit data length.
CN201910649819.3A 2019-07-18 2019-07-18 SCA and SDR-oriented rapid system reconstruction method Active CN110362512B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201910649819.3A CN110362512B (en) 2019-07-18 2019-07-18 SCA and SDR-oriented rapid system reconstruction method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201910649819.3A CN110362512B (en) 2019-07-18 2019-07-18 SCA and SDR-oriented rapid system reconstruction method

Publications (2)

Publication Number Publication Date
CN110362512A CN110362512A (en) 2019-10-22
CN110362512B true CN110362512B (en) 2023-04-25

Family

ID=68220184

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201910649819.3A Active CN110362512B (en) 2019-07-18 2019-07-18 SCA and SDR-oriented rapid system reconstruction method

Country Status (1)

Country Link
CN (1) CN110362512B (en)

Families Citing this family (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111262742B (en) * 2020-01-20 2022-04-26 中国人民解放军国防科技大学 SCA-based dynamic partially reconfigurable equipment resource virtualization and waveform deployment method
CN111416631B (en) * 2020-03-04 2020-11-10 上海介方信息技术有限公司 Waveform deployment implementation method based on waveform multistage nesting
CN111353235B (en) * 2020-03-13 2021-05-04 上海介方信息技术有限公司 Component port modeling method based on model drive and terminal
CN111427823B (en) * 2020-03-30 2023-05-02 天津光电通信技术有限公司 Driving design method supporting PC and FPGA to communicate through PCIE
CN111857839B (en) * 2020-06-05 2023-06-27 北京航天测控技术有限公司 Linux-based PXI/PXIe bus device driving system
CN112131176B (en) * 2020-09-29 2023-12-12 中国船舶集团有限公司第七二四研究所 FPGA (field programmable Gate array) quick local reconstruction method based on PCIE (peripheral component interface express)
CN113297112B (en) * 2021-04-15 2022-05-17 上海安路信息科技股份有限公司 PCIe bus data transmission method and system and electronic equipment
CN113890782B (en) * 2021-09-24 2023-06-06 哈尔滨工程大学 Software radio platform, method and device for task program loading
CN114443170B (en) * 2022-01-29 2023-10-24 中国航空无线电电子研究所 FPGA dynamic parallel loading and unloading system
CN115757222B (en) * 2022-09-28 2023-06-20 航天恒星科技有限公司 Data transmission method and device based on PL and PS
CN115328832B (en) * 2022-10-11 2023-01-17 三未信安科技股份有限公司 Data scheduling system and method based on PCIE DMA
CN119003434B (en) * 2024-10-24 2024-12-27 北京坤驰科技有限公司 FPGA-based data transmission method

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7404074B2 (en) * 2002-07-12 2008-07-22 Sca Technica, Inc. Self-booting software defined radio module
KR101366747B1 (en) * 2008-12-01 2014-02-24 한국전자통신연구원 SDR terminal apparatus and reconstruction method of SDR terminal apparatus
CN105515858B (en) * 2015-12-07 2018-10-23 中国电子科技集团公司第五十研究所 Waveform components dynamic address based on SCA configures system and method
CN109547367B (en) * 2018-11-09 2021-07-16 中国航空无线电电子研究所 Software radio platform based on SCA

Also Published As

Publication number Publication date
CN110362512A (en) 2019-10-22

Similar Documents

Publication Publication Date Title
CN110362512B (en) SCA and SDR-oriented rapid system reconstruction method
US9311437B2 (en) Modeling a bus for a system design incorporating one or more programmable processors
Vipin et al. DyRACT: A partial reconfiguration enabled accelerator and test platform
CN115146568B (en) A UVM-based chip verification system and verification method
US20040141518A1 (en) Flexible multimode chip design for storage and networking
CN103412834B (en) The multiplexing method of a kind of single SOC and single SOC multi-operation mode
KR101035832B1 (en) Integrated endpoint device, integrated PCI Express endpoint device and PCI Express communication system
JP2002259157A (en) In-circuit emulation device, chip design method thereof, and in-circuit emulation system
CN112286746A (en) Universal verification platform and method for AXI slave device interface
KR20130050383A (en) Incorporating an independent logic block in a system-on-a-chip
CN113270137B (en) DDR2 testing method based on FPGA embedded soft core
CN114444423B (en) Data processing method and system based on verification platform and electronic equipment
CN101699415A (en) Core logic circuit, computer system and peripheral equipment initialization method
CN118586052A (en) Bus interconnection verification method, device, electronic device, storage medium and program
CN111506461B (en) Back pressure module based on bus and used for testing and implementation method thereof
US20050144436A1 (en) Multitasking system level platform for HW/SW co-verification
US10409935B2 (en) Modeling a bus for a system design incorporating one or more programmable processors
CN101232522A (en) A kind of network card driving method
CN112232004B (en) System-on-chip design scheme test method and system-on-chip
CN107329863B (en) A COMe-based Universal Hardware Platform for Measuring Instruments
CN117194135A (en) Verification method, system and computer equipment of universal asynchronous receiving and transmitting device
CN117077221A (en) Hard disk screen printing distribution method and device, computer equipment and medium
CN114297982A (en) UVM-based SPI bus verification method and system
CN115481078A (en) Master-slave communication system and method
CN106326172A (en) APB bus slave interface expansion circuit and use method thereof

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant
PE01 Entry into force of the registration of the contract for pledge of patent right

Denomination of invention: A Fast System Refactoring Method for SCA and SDR

Granted publication date: 20230425

Pledgee: Agricultural Bank of China Limited by Share Ltd. Chengdu high tech Industrial Development Zone Branch

Pledgor: Chengdu Xieying Technology Co.,Ltd.

Registration number: Y2024510000088

PE01 Entry into force of the registration of the contract for pledge of patent right
PC01 Cancellation of the registration of the contract for pledge of patent right

Granted publication date: 20230425

Pledgee: Agricultural Bank of China Limited by Share Ltd. Chengdu high tech Industrial Development Zone Branch

Pledgor: Chengdu Xieying Technology Co.,Ltd.

Registration number: Y2024510000088

PC01 Cancellation of the registration of the contract for pledge of patent right