CN117725863A - Chip signal analysis hardware module, system, processor chip and electronic equipment - Google Patents
Chip signal analysis hardware module, system, processor chip and electronic equipment Download PDFInfo
- Publication number
- CN117725863A CN117725863A CN202311698097.3A CN202311698097A CN117725863A CN 117725863 A CN117725863 A CN 117725863A CN 202311698097 A CN202311698097 A CN 202311698097A CN 117725863 A CN117725863 A CN 117725863A
- Authority
- CN
- China
- Prior art keywords
- data
- chip
- module
- signal analysis
- protocol conversion
- 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.)
- Granted
Links
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Debugging And Monitoring (AREA)
Abstract
本申请提供一种芯片信号分析硬件模块、系统、处理器芯片及电子设备,通过软件资源模块搭载或调用用于芯片信号分析的软件资源,并通过芯片数据获取模块,读取软件资源,获取目标走线上的电路数据,进而以服务器芯片的数据传输协议,将电路数据传输至总线协议转换模块,总线协议转换模块完成时序同步和协议转换流程,将电路数据根据预设的目标协议封装成封装数据,以目标协议传输至调试模块进行芯片信号分析,由此,提升了应用场景的普适性,并提升了软件交互的效率,降低了芯片信号分析过程中软件资源的占用,解决了POWER架构下的处理器芯片信号分析系统在进行硅后芯片验证流程中,软件交互效率低、软件资源占用大的问题。
This application provides a chip signal analysis hardware module, system, processor chip and electronic equipment, which carries or calls software resources for chip signal analysis through the software resource module, and reads the software resources and obtains the target through the chip data acquisition module. The circuit data on the wiring is then transmitted to the bus protocol conversion module using the data transmission protocol of the server chip. The bus protocol conversion module completes the timing synchronization and protocol conversion process, and encapsulates the circuit data into packages according to the preset target protocol. The data is transmitted to the debugging module using the target protocol for chip signal analysis. This improves the universality of the application scenario, improves the efficiency of software interaction, reduces the occupation of software resources during the chip signal analysis process, and solves the problem of POWER architecture. In the post-silicon chip verification process of the processor chip signal analysis system, the software interaction efficiency is low and the software resources are occupied.
Description
技术领域Technical field
本申请涉及电子设计技术,尤其涉及一种芯片信号分析硬件模块、系统、处理器芯片及电子设备。This application relates to electronic design technology, and in particular to a chip signal analysis hardware module, system, processor chip and electronic equipment.
背景技术Background technique
在服务器芯片设计领域,例如POWER架构的服务器芯片,硅后芯片验证方法通常采用具备调试和溯源功能的硬件模块实现,在芯片从制造厂商回归后,该硬件模块能够为芯片设计人员提供大量连续、微观的集成电路电子信号,是芯片硬件定位的关键手段。芯片设计人员可以在出现故障的位置设置触发条件,借助片上对应的硬件资源,可以提供一段时间内的电路变化情况,若出现故障事件,则可以捕获故障前后的信号及电路情况。In the field of server chip design, such as POWER architecture server chips, post-silicon chip verification methods are usually implemented using hardware modules with debugging and traceability functions. After the chip returns from the manufacturer, this hardware module can provide chip designers with a large number of continuous, Microscopic integrated circuit electronic signals are a key means of chip hardware positioning. Chip designers can set trigger conditions at the location of a fault. With the help of corresponding on-chip hardware resources, the circuit changes over a period of time can be provided. If a fault event occurs, the signals and circuit conditions before and after the fault can be captured.
但是,当前只有部分芯片具备前述的硬件模块。对于大部分的芯片制造厂商,会在芯片上保留部分接口,内部通过一些矩阵选通,外部对接逻辑分析仪,可以观察部分的时序变化情况,以实现芯片硬件定位、信号分析、故障排查等功能。当前的硅后芯片验证方法通过软件资源模块和用于芯片走线上数据寻迹的芯片数据获取模块实现,即在原有的POWER架构中,用于储存软件资源信息或调用第三方软件资源信息的软件资源模块,和数据寻迹对应的芯片数据获取模块组成了整个的技术方案,软件可以通过总线去配置芯片数据获取模块的工作模式,在芯片故障发生后,芯片控制数据寻迹模块开始、停止采集数据;软件需要在这个过程中,轮询等待,当芯片数据获取模块对目标走线上的电路数据采集完成后,软件通过总结接口依次读取存储介质中的数据。However, currently only some chips have the aforementioned hardware modules. For most chip manufacturers, some interfaces will be retained on the chip. Through some matrix strobes internally and external logic analyzers, some timing changes can be observed to achieve functions such as chip hardware positioning, signal analysis, and troubleshooting. . The current post-silicon chip verification method is implemented through the software resource module and the chip data acquisition module used for data tracing on the chip wiring. That is, in the original POWER architecture, it is used to store software resource information or call third-party software resource information. The software resource module and the chip data acquisition module corresponding to data tracing constitute the entire technical solution. The software can configure the working mode of the chip data acquisition module through the bus. After a chip failure occurs, the chip controls the start and stop of the data tracing module. Collect data; the software needs to poll and wait during this process. When the chip data acquisition module completes the collection of circuit data on the target wiring, the software reads the data in the storage medium sequentially through the summary interface.
因此,对于软件部分的资源,由于芯片资源受限,导致应用场景受限,且在硅后芯片验证等芯片信号分析过程中,软件与芯片数据交互的效率不高,软件资源占用大。综上,POWER架构下的处理器芯片信号分析系统在进行硅后芯片验证流程中,存在软件交互效率低、软件资源占用大的问题。Therefore, for the resources of the software part, due to limited chip resources, the application scenarios are limited, and during the chip signal analysis process such as post-silicon chip verification, the efficiency of software and chip data interaction is not high, and software resources are occupied. In summary, the processor chip signal analysis system under the POWER architecture has problems such as low software interaction efficiency and large software resource usage during the post-silicon chip verification process.
发明内容Contents of the invention
本申请提供一种芯片信号分析硬件模块、系统、处理器芯片及电子设备,用以解决POWER架构下的处理器芯片信号分析系统在进行硅后芯片验证流程中,软件交互效率低、软件资源占用大的问题。This application provides a chip signal analysis hardware module, system, processor chip and electronic equipment to solve the problem of low software interaction efficiency and software resource occupation during the post-silicon chip verification process of the processor chip signal analysis system under the POWER architecture. Big question.
第一方面,本申请提供一种芯片信号分析硬件模块,所述芯片信号分析硬件模块应用于POWER架构下的目标芯片,所述芯片信号分析硬件模块用于获取所述目标芯片中预设目标走线上的电路数据,以目标协议进行封装,获得封装数据并传输至预设的存储单元或调试单元;In a first aspect, this application provides a chip signal analysis hardware module. The chip signal analysis hardware module is applied to a target chip under the POWER architecture. The chip signal analysis hardware module is used to obtain the preset target path in the target chip. The circuit data on the line is encapsulated with the target protocol, the encapsulated data is obtained and transmitted to the preset storage unit or debugging unit;
其中,所述目标协议包括高级跟踪总线协议ATB,所述封装数据用于对所述目标走线的信号响应情况进行分析,获得所述目标走线的工作状态信息或故障信息。Wherein, the target protocol includes an advanced tracking bus protocol ATB, and the encapsulation data is used to analyze the signal response of the target wiring and obtain the working status information or fault information of the target wiring.
第二方面,本申请提供一种芯片信号分析系统,所述系统包括:In a second aspect, this application provides a chip signal analysis system, which includes:
软件资源模块、芯片数据获取模块、调试模块以及第一方面所述的芯片信号分析硬件模块,所述芯片信号分析硬件模块包括总线协议转换模块;Software resource module, chip data acquisition module, debugging module and the chip signal analysis hardware module described in the first aspect, the chip signal analysis hardware module includes a bus protocol conversion module;
其中,所述软件资源模块,用于搭载或调用外部进行芯片信号分析的软件资源;Wherein, the software resource module is used to carry or call external software resources for chip signal analysis;
所述芯片数据获取模块,所述芯片数据获取模块包括存储介质,所述芯片数据获取模块用于通过所述软件资源,获取所述预设目标走线上的电路数据,所述存储介质用于存取所述电路数据;The chip data acquisition module includes a storage medium. The chip data acquisition module is used to obtain circuit data on the preset target wiring line through the software resources. The storage medium is used to access said circuit data;
所述调试模块,用于获取所述封装数据并根据所述封装数据对所述目标走线进行信号分析;The debugging module is used to obtain the packaging data and perform signal analysis on the target wiring according to the packaging data;
其中,所述软件资源模块与所述芯片数据分析模块通过存储器存取实现数据交互;Wherein, the software resource module and the chip data analysis module realize data interaction through memory access;
所述芯片数据分析模块和所述总线协议转换模块连接,所述总线协议转换模块与所述调试模块连接。The chip data analysis module is connected to the bus protocol conversion module, and the bus protocol conversion module is connected to the debugging module.
作为一种可选的实施方式,所述总线协议转换模块包括读写控制单元;As an optional implementation, the bus protocol conversion module includes a read-write control unit;
其中,所述读写控制单元与所述芯片数据获取模块中的存储介质连接;Wherein, the read-write control unit is connected to the storage medium in the chip data acquisition module;
以及,所述读写控制单元用于控制读写所述电路数据的工作参数,并根据所述工作参数读写所述电路数据。And, the read-write control unit is used to control the operating parameters of reading and writing the circuit data, and read and write the circuit data according to the operating parameters.
作为一种可选的实施方式,所述读写控制单元包括触发器发送控制子单元;As an optional implementation, the read-write control unit includes a trigger sending control subunit;
所述触发器发送控制子单元用于控制所述芯片数据获取模块采集完成所述目标走线上的电路数据之后生成目标信息包,并驱动所述目标信息包传输至所述调试模块。The trigger sending control subunit is used to control the chip data acquisition module to generate a target information packet after completing the collection of circuit data on the target wiring, and drive the target information packet to be transmitted to the debugging module.
作为一种可选的实施方式,所述读写控制单元包括动态发送控制子单元;As an optional implementation, the read-write control unit includes a dynamic sending control subunit;
所述动态发送控制子单元用于控制所述总线协议转换模块在获取所述目标走线上的电路数据的同时,向所述调试模块传输封装数据。The dynamic transmission control subunit is used to control the bus protocol conversion module to transmit encapsulated data to the debugging module while acquiring circuit data on the target wiring.
作为一种可选的实施方式,所述读写控制单元包括模式控制子单元;As an optional implementation, the read-write control unit includes a mode control subunit;
所述模式控制子单元用于控制所述总线协议转换模块的数据发送模式。The mode control subunit is used to control the data sending mode of the bus protocol conversion module.
作为一种可选的实施方式,所述读写控制单元包括数据排空控制子单元;As an optional implementation, the read-write control unit includes a data emptying control subunit;
所述数据排空控制子单元用于根据各所述目标走线对应的电路结构的时钟域,控制所述存储介质中的数据排空,发送至所述总线协议转换模块。The data emptying control subunit is used to control the data emptying in the storage medium according to the clock domain of the circuit structure corresponding to each target wiring, and send it to the bus protocol conversion module.
作为一种可选的实施方式,所述总线协议转换模块还包括优先级处理单元;As an optional implementation, the bus protocol conversion module also includes a priority processing unit;
其中,所述优先级处理单元与所述读写控制单元连接,并与所述芯片数据获取模块中的存储介质连接;Wherein, the priority processing unit is connected to the read-write control unit and connected to the storage medium in the chip data acquisition module;
以及,所述优先级处理单元用于处理所述读写控制单元的数据读写请求,并控制数据读写的优先级;And, the priority processing unit is used to process the data read and write requests of the read and write control unit, and control the priority of data reading and writing;
以及,所述总线协议转换模块还包括协议转换单元;And, the bus protocol conversion module also includes a protocol conversion unit;
其中,所述协议转换单元与所述读写控制单元连接,并与所述调试模块通过基于所述目标协议的总线连接;Wherein, the protocol conversion unit is connected to the read-write control unit and connected to the debugging module through a bus based on the target protocol;
以及,所述协议转换单元用于通过所述总线接收所述调试模块的数据写出请求,并根据所述数据写出请求将所述封装数据发送至所述调试模块。And, the protocol conversion unit is configured to receive a data write request from the debugging module through the bus, and send the encapsulated data to the debugging module according to the data write request.
第三方面,本申请提供一种处理器芯片,所述芯片包括如第二方面所述的芯片信号分析系统。In a third aspect, the present application provides a processor chip, which includes the chip signal analysis system described in the second aspect.
第四方面,本申请提供一种电子设备,所述电子设备包括如第三方面所述的处理器芯片。In a fourth aspect, the present application provides an electronic device, which includes the processor chip as described in the third aspect.
本申请提供的芯片信号分析硬件模块、系统、处理器芯片及电子设备,通过软件资源模块搭载或调用用于芯片信号分析的软件资源,通过软件资源模块搭载或调用用于芯片信号分析的软件资源,并通过芯片数据获取模块,读取软件资源,获取目标走线上的电路数据,进而以服务器芯片的数据传输协议,将电路数据传输至总线协议转换模块,总线协议转换模块完成时序同步和协议转换流程,将电路数据根据预设的目标协议封装成封装数据,以目标协议传输至调试模块进行芯片信号分析,由此,提升了应用场景的普适性,并提升了软件交互的效率,降低了芯片信号分析过程中软件资源的占用,解决了POWER架构下的处理器芯片信号分析系统在进行硅后芯片验证流程中,软件交互效率低、软件资源占用大的问题。The chip signal analysis hardware modules, systems, processor chips and electronic equipment provided by this application are equipped with or called software resources for chip signal analysis through the software resource module, and are equipped with or called software resources used for chip signal analysis through the software resource module. , and through the chip data acquisition module, read the software resources, obtain the circuit data on the target wiring, and then transmit the circuit data to the bus protocol conversion module using the data transmission protocol of the server chip. The bus protocol conversion module completes timing synchronization and protocol The conversion process encapsulates the circuit data into encapsulated data according to the preset target protocol, and transmits it to the debugging module for chip signal analysis using the target protocol. This improves the universality of the application scenario, improves the efficiency of software interaction, and reduces It eliminates the occupation of software resources during the chip signal analysis process and solves the problems of low software interaction efficiency and large software resource occupation during the post-silicon chip verification process of the processor chip signal analysis system under the POWER architecture.
附图说明Description of the drawings
此处的附图被并入说明书中并构成本说明书的一部分,示出了符合本申请的实施例,并与说明书一起用于解释本申请的原理。The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the application and together with the description, serve to explain the principles of the application.
图1是本发明实施例公开的一种芯片信号分析系统的结构示意图;Figure 1 is a schematic structural diagram of a chip signal analysis system disclosed in an embodiment of the present invention;
图2是本发明实施例公开的另一种芯片信号分析系统的结构示意图;Figure 2 is a schematic structural diagram of another chip signal analysis system disclosed in an embodiment of the present invention;
图3是本发明实施例公开的另一种芯片信号分析系统的结构示意图;Figure 3 is a schematic structural diagram of another chip signal analysis system disclosed in an embodiment of the present invention;
图4是本发明实施例公开的一种芯片信号分析系统运行逻辑的流程示意图;Figure 4 is a schematic flowchart of the operation logic of a chip signal analysis system disclosed in an embodiment of the present invention;
图5是本发明实施例公开的另一种芯片信号分析系统运行逻辑的流程示意图;Figure 5 is a schematic flowchart of the operation logic of another chip signal analysis system disclosed in the embodiment of the present invention;
附图标记:Reference signs:
11、软件资源模块;11. Software resource module;
12、芯片数据获取模块;12. Chip data acquisition module;
13、总线协议转换模块;13. Bus protocol conversion module;
14、调试模块;14. Debugging module;
Read/write Control、读写控制单元;Read/write Control, read and write control unit;
Trigger Send Control、触发器发送控制子单元;Trigger Send Control, trigger send control subunit;
Active Send Control、动态发送控制子单元;Active Send Control, dynamic send control subunit;
Work Mode Control、模式控制子单元;Work Mode Control, mode control subunit;
Flush Control、数据排空控制子单元;Flush Control, data emptying control subunit;
SP、优先级处理单元;SP, priority processing unit;
ATB Master、协议转换单元;ATB Master, protocol conversion unit;
通过上述附图,已示出本申请明确的实施例,后文中将有更详细的描述。这些附图和文字描述并不是为了通过任何方式限制本申请构思的范围,而是通过参考特定实施例为本领域技术人员说明本申请的概念。Through the above-mentioned drawings, clear embodiments of the present application have been shown, which will be described in more detail below. These drawings and text descriptions are not intended to limit the scope of the present application's concepts in any way, but are intended to illustrate the application's concepts for those skilled in the art with reference to specific embodiments.
具体实施方式Detailed ways
这里将详细地对示例性实施例进行说明,其示例表示在附图中。下面的描述涉及附图时,除非另有表示,不同附图中的相同数字表示相同或相似的要素。以下示例性实施例中所描述的实施方式并不代表与本申请相一致的所有实施方式。相反,它们仅是与如所附权利要求书中所详述的、本申请的一些方面相一致的装置和方法的例子。Exemplary embodiments will be described in detail herein, examples of which are illustrated in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings refer to the same or similar elements unless otherwise indicated. The implementations described in the following exemplary embodiments do not represent all implementations consistent with this application. Rather, they are merely examples of apparatus and methods consistent with aspects of the application as detailed in the appended claims.
在IBM POWER等服务器芯片架构中,数据寻迹和数据调试是硅后芯片验证的一个重要方法,在芯片从制造厂商回归后,能够为芯片设计人员提供大量连续、微观的集成电路电子信号,是芯片硬件定位的关键手段。芯片设计人员可以在出现故障的位置设置触发条件,借助片上用于数据寻迹和数据调试的硬件资源,可以提供一段时间内的电路变化情况,故障前、故障后的情况都能准确的捕获。In server chip architectures such as IBM POWER, data tracing and data debugging are an important method for post-silicon chip verification. After the chip returns from the manufacturer, it can provide chip designers with a large number of continuous and microscopic integrated circuit electronic signals. A key means of chip hardware positioning. Chip designers can set trigger conditions at the location of a fault. With the help of on-chip hardware resources for data tracing and data debugging, circuit changes over a period of time can be provided, and the conditions before and after the fault can be accurately captured.
目前,在超大规模集成电路中,仅部分厂商具备类似的片上芯片验证能力,业界大部分的芯片制造厂商在芯片上未设置相应的硬件资源,通常在芯片的输入输出接口上保留部分接口,内部通过一些矩阵选通,外部对接一些逻辑分析仪,可以观察部分的时序变化情况。At present, in VLSI, only some manufacturers have similar on-chip chip verification capabilities. Most chip manufacturers in the industry do not set up corresponding hardware resources on the chip. They usually reserve some interfaces on the input and output interfaces of the chip, and internally Through some matrix gating and external connection to some logic analyzers, some timing changes can be observed.
请参阅图1,图1是本发明实施例公开的一种芯片信号分析系统的结构示意图,如图1所示,作为一种示例性的实施方式,在POWER架构中,用于储存软件资源信息或调用第三方软件资源信息的软件资源模块,和数据寻迹对应的芯片数据获取模块组成了整个的技术方案,软件可以通过总线去配置芯片数据获取模块的工作模式,在芯片故障发生后,芯片控制数据寻迹模块开始、停止采集数据;软件需要在这个过程中,轮询等待,当芯片数据获取模块对目标走线上的电路数据采集完成后,软件通过总结接口依次读取存储介质中的数据。Please refer to Figure 1. Figure 1 is a schematic structural diagram of a chip signal analysis system disclosed in an embodiment of the present invention. As shown in Figure 1, as an exemplary implementation, in the POWER architecture, it is used to store software resource information. Or the software resource module that calls third-party software resource information, and the chip data acquisition module corresponding to data tracing form the entire technical solution. The software can configure the working mode of the chip data acquisition module through the bus. After a chip failure occurs, the chip Control the data tracing module to start and stop collecting data; the software needs to poll and wait during this process. When the chip data acquisition module completes the collection of circuit data on the target wiring, the software sequentially reads the data in the storage medium through the summary interface. data.
当前的芯片信号分析系统在应用于POWER架构的服务器芯片的硅后芯片验证过程中,存在如下问题:The current chip signal analysis system has the following problems in the post-silicon chip verification process applied to POWER architecture server chips:
第一,由于芯片资源的原因,存储容量有限,每个目标路径仅包含有限个接口,即使有数据压缩的算法,存储的信息也比较局限。另外,加上时间戳、压缩的时间差等信息,有效数据进一步减少。提升硅上的存储容量,只能少许改善,但会增加巨额的成本。First, due to chip resources, the storage capacity is limited, and each target path only contains a limited number of interfaces. Even with data compression algorithms, the stored information is relatively limited. In addition, with the addition of timestamps, compressed time differences and other information, the effective data is further reduced. Increasing the storage capacity on silicon will only provide a small improvement, but will increase the cost by a huge amount.
第二,在IBM POWER的架构中,片上存储的数据,软件是可以通过总线访问的方式获取,但是效率较低。Second, in the IBM POWER architecture, the software can obtain the data stored on the chip through bus access, but the efficiency is low.
第三,软件消耗较大,在芯片信号分析过程中需要软件不停的轮询、获取片上的数据。Third, the software consumes a lot of money. During the chip signal analysis process, the software needs to continuously poll and obtain on-chip data.
基于此,本申请提供了一种芯片信号分析硬件模块,芯片信号分析硬件模块应用于POWER架构下的目标芯片,芯片信号分析硬件模块用于获取目标芯片中预设目标走线上的电路数据,以目标协议进行封装,获得封装数据并传输至预设的存储单元或调试单元;Based on this, this application provides a chip signal analysis hardware module. The chip signal analysis hardware module is applied to the target chip under the POWER architecture. The chip signal analysis hardware module is used to obtain circuit data on the preset target wiring in the target chip. Encapsulate with the target protocol, obtain the encapsulated data and transmit it to the preset storage unit or debugging unit;
其中,目标协议包括高级跟踪总线协议ATB,封装数据用于对目标走线的信号响应情况进行分析,获得目标走线的工作状态信息或故障信息。Among them, the target protocol includes the advanced tracking bus protocol ATB, and the encapsulated data is used to analyze the signal response of the target wiring and obtain the working status information or fault information of the target wiring.
由此,通过该芯片信号分析硬件模块,完成时序同步和协议转换流程,将电路数据根据预设的目标协议封装成封装数据,实现了数据的交互和转换,提升了芯片信号分析硬件模块在应用场景上的普适性,并提升了软件交互的效率,降低了芯片信号分析过程中软件资源的占用,解决了POWER架构下的处理器芯片信号分析系统在进行硅后芯片验证流程中,软件交互效率低、软件资源占用大的问题。As a result, the chip signal analysis hardware module completes the timing synchronization and protocol conversion process, and encapsulates the circuit data into encapsulated data according to the preset target protocol, realizing data interaction and conversion, and improving the application of the chip signal analysis hardware module. It has universal applicability in scenarios, improves the efficiency of software interaction, reduces the occupation of software resources during the chip signal analysis process, and solves the problem of software interaction during the post-silicon chip verification process of the processor chip signal analysis system under the POWER architecture. Problems of low efficiency and large software resource usage.
本申请的技术构思在于,通过用于协议转换的总线协议转换模块,完成数据的打包和交互,本申请提供的系统结构主要分为软件(软件资源模块)、POWER Trace(芯片数据获取模块)、Trace to atb tranfer(总线协议转换模块)、DDR/Debugger(调试模块)等四个部分。软件可以是运行在芯片驱动核心上的软件,也可以是第三方的软件,第三方的软件则需要借助芯片的外设。POWER Trace是IBM POWER架构中原有的硬件结构,支持压缩的算法,可以把一段时间内变化不频繁的信号进行压缩;支持非压缩的算法,存储介质有限的情况下,观察的信号范围较小;可以根据观察信号设置触发器,控制目标路径上采集若干个数据。总线协议转换模块是本技术方案的核心,用于解决当前存在的技术问题。DDR是大容量的硬件存储设备,Debugger是第三方的软硬件调试盒子,根据硬件系统实际应用场景的不同,选用对应的实现方式,以实现软件调试功能。The technical concept of this application is to complete the packaging and interaction of data through the bus protocol conversion module for protocol conversion. The system structure provided by this application is mainly divided into software (software resource module), POWER Trace (chip data acquisition module), Trace to atb transfer (bus protocol conversion module), DDR/Debugger (debugging module) and other four parts. The software can be software running on the chip driver core, or third-party software. Third-party software requires the use of chip peripherals. POWER Trace is the original hardware structure in the IBM POWER architecture. It supports compression algorithms and can compress signals that change infrequently within a period of time. It supports non-compression algorithms. When the storage medium is limited, the observed signal range is smaller; Triggers can be set based on the observation signal to control the collection of several data on the target path. The bus protocol conversion module is the core of this technical solution and is used to solve current technical problems. DDR is a large-capacity hardware storage device, and Debugger is a third-party software and hardware debugging box. According to the actual application scenarios of the hardware system, the corresponding implementation method is selected to realize the software debugging function.
具体的,本申请通过软件资源模块搭载或调用用于芯片信号分析的软件资源,通过软件资源模块搭载或调用用于芯片信号分析的软件资源,并通过芯片数据获取模块,读取软件资源,获取目标走线上的电路数据,进而以服务器芯片的数据传输协议,将电路数据传输至总线协议转换模块,总线协议转换模块完成时序同步和协议转换流程,将电路数据根据预设的目标协议封装成封装数据,以目标协议传输至调试模块进行芯片信号分析,由此,提升了应用场景的普适性,并提升了软件交互的效率,降低了芯片信号分析过程中软件资源的占用,解决了POWER架构下的处理器芯片信号分析系统在进行硅后芯片验证流程中,软件交互效率低、软件资源占用大的问题。Specifically, this application carries or calls software resources for chip signal analysis through the software resource module, loads or calls software resources for chip signal analysis through the software resource module, and reads the software resources through the chip data acquisition module to obtain The circuit data on the target wiring is then transmitted to the bus protocol conversion module using the data transmission protocol of the server chip. The bus protocol conversion module completes the timing synchronization and protocol conversion process, and encapsulates the circuit data according to the preset target protocol. The data is encapsulated and transmitted to the debugging module using the target protocol for chip signal analysis. This improves the universality of the application scenario, improves the efficiency of software interaction, reduces the occupation of software resources during the chip signal analysis process, and solves the problem of POWER In the post-silicon chip verification process of the processor chip signal analysis system under the architecture, the software interaction efficiency is low and the software resources are occupied.
请参阅图2,图2是本发明实施例公开的另一种芯片信号分析系统的结构示意图,系统包括芯片信号分析硬件模块,芯片信号分析硬件模块应用于POWER架构下的目标芯片,具体的,如图2所示,系统包括:Please refer to Figure 2. Figure 2 is a schematic structural diagram of another chip signal analysis system disclosed in an embodiment of the present invention. The system includes a chip signal analysis hardware module. The chip signal analysis hardware module is applied to the target chip under the POWER architecture. Specifically, As shown in Figure 2, the system includes:
软件资源模块11、芯片数据获取模块12、调试模块14以及如前述实施方式中的芯片信号分析硬件模块,芯片信号分析硬件模块包括总线协议转换模块13。The software resource module 11 , the chip data acquisition module 12 , the debugging module 14 and the chip signal analysis hardware module as in the aforementioned embodiments. The chip signal analysis hardware module includes a bus protocol conversion module 13 .
软件资源模块11,用于搭载或调用进行芯片信号分析的软件资源;Software resource module 11, used to carry or call software resources for chip signal analysis;
软件资源模块用于搭载芯片信号分析过程中所可能涉及的所有软件资源,在具体的实现方式中,可以理解为一种用于存储软件资源的后端数据库,根据前端发起的相应指令进行调用,对于之后提到的调试模块,其中一种实施方式可以是,用于调试的软件资源也通过软件资源模块进行获取。The software resource module is used to carry all software resources that may be involved in the chip signal analysis process. In a specific implementation, it can be understood as a back-end database for storing software resources, which is called according to the corresponding instructions initiated by the front-end. For the debugging module mentioned later, one implementation may be that software resources used for debugging are also obtained through the software resource module.
此外,需要说明的是,软件资源模块还可用于指示获取外部软件资源的接口,即该软件资源模块可以不仅仅用于软件资源的搭载,还可作为与第三方软件数据进行数据交互的软硬件接口,以实现软件资源获取功能。In addition, it should be noted that the software resource module can also be used to indicate the interface for obtaining external software resources. That is, the software resource module can not only be used to carry software resources, but can also be used as software and hardware for data interaction with third-party software data. Interface to implement software resource acquisition function.
芯片数据获取模块12,芯片数据获取模块12包括存储介质,芯片数据获取模块12用于通过软件资源,获取目标走线上的电路数据,存储介质用于存取电路数据;Chip data acquisition module 12. The chip data acquisition module 12 includes a storage medium. The chip data acquisition module 12 is used to obtain circuit data on the target wiring through software resources. The storage medium is used to access circuit data;
该部分结构用于实现传统服务器芯片架构中的Trace流程,即对目标走线上的电子器件进行寻踪及获取相应电参数的流程。This part of the structure is used to implement the Trace process in the traditional server chip architecture, that is, the process of tracking electronic devices on the target wiring and obtaining the corresponding electrical parameters.
总线协议转换模块13,用于根据目标走线上的电路数据,以目标协议进行封装获得封装数据并进行传输;The bus protocol conversion module 13 is used to encapsulate the circuit data on the target wiring according to the target protocol to obtain the encapsulated data and transmit it;
本申请中,相对于传统的服务器芯片信号分析系统架构,所增添的总线协议转换模块主要用于数据的封装及协议的转换,具体的,在总线协议转换模块中执行不同时钟域的硬件设备的时序同步及封装数据的协议转换流程,将电路数据根据预设的目标协议封装成封装数据,以目标协议传输至调试模块进行芯片信号分析。In this application, compared with the traditional server chip signal analysis system architecture, the added bus protocol conversion module is mainly used for data encapsulation and protocol conversion. Specifically, the bus protocol conversion module executes the processing of hardware devices in different clock domains. The protocol conversion process of timing synchronization and encapsulated data encapsulates the circuit data into encapsulated data according to the preset target protocol, and transmits it to the debugging module using the target protocol for chip signal analysis.
其中,总线协议转换模块13为前述本申请提供的一种芯片信号分析硬件模块的具体实现方式,芯片信号分析硬件模块应用于POWER架构下的目标芯片,芯片信号分析硬件模块包括总线协议转换模块13;Among them, the bus protocol conversion module 13 is a specific implementation method of the chip signal analysis hardware module provided by the present application. The chip signal analysis hardware module is applied to the target chip under the POWER architecture. The chip signal analysis hardware module includes the bus protocol conversion module 13 ;
其中,总线协议转换模块13,用于获取目标芯片中预设目标走线上的电路数据,以目标协议进行封装,获得封装数据并传输至预设的存储单元或调试单元;Among them, the bus protocol conversion module 13 is used to obtain the circuit data on the preset target wiring in the target chip, package it with the target protocol, obtain the packaged data and transmit it to the preset storage unit or debugging unit;
其中,目标协议包括高级跟踪总线协议ATB,封装数据用于对目标走线的信号响应情况进行分析,获得目标走线的工作状态信息或故障信息。Among them, the target protocol includes the advanced tracking bus protocol ATB, and the encapsulated data is used to analyze the signal response of the target wiring and obtain the working status information or fault information of the target wiring.
调试模块14,用于获取封装数据并根据封装数据对目标走线进行信号分析;The debugging module 14 is used to obtain packaging data and perform signal analysis on the target wiring according to the packaging data;
其中,软件资源模块11与芯片数据分析模块通过存储器存取实现数据交互;Among them, the software resource module 11 and the chip data analysis module realize data interaction through memory access;
芯片数据分析模块和总线协议转换模块13连接,总线协议转换模块13与调试模块14连接。The chip data analysis module is connected to the bus protocol conversion module 13 , and the bus protocol conversion module 13 is connected to the debugging module 14 .
调试模块14作为硅后芯片验证的软硬件分析模块,硬件部分可以以大容量存储装置实现,软件部分搭载于该存储装置上,或者如前由软件资源模块获取。The debugging module 14 serves as a software and hardware analysis module for post-silicon chip verification. The hardware part can be implemented as a large-capacity storage device, and the software part is loaded on the storage device, or obtained from the software resource module as before.
通过软件资源模块搭载或调用用于芯片信号分析的软件资源,并通过芯片数据获取模块,读取软件资源,获取目标走线上的电路数据,进而以服务器芯片的数据传输协议,将电路数据传输至总线协议转换模块,总线协议转换模块完成时序同步和协议转换流程,将电路数据根据预设的目标协议封装成封装数据,以目标协议传输至调试模块进行芯片信号分析,由此,提升了应用场景的普适性,并提升了软件交互的效率,降低了芯片信号分析过程中软件资源的占用,解决了POWER架构下的处理器芯片信号分析系统在进行硅后芯片验证流程中,软件交互效率低、软件资源占用大的问题。The software resource module is used to load or call software resources for chip signal analysis, and through the chip data acquisition module, the software resources are read and the circuit data on the target wiring is obtained, and then the circuit data is transmitted using the data transmission protocol of the server chip. To the bus protocol conversion module, the bus protocol conversion module completes the timing synchronization and protocol conversion process, encapsulates the circuit data into encapsulated data according to the preset target protocol, and transmits it to the debugging module for chip signal analysis using the target protocol. This improves the application The universality of the scenario improves the efficiency of software interaction, reduces the occupation of software resources during the chip signal analysis process, and solves the problem of software interaction efficiency in the post-silicon chip verification process of the processor chip signal analysis system under the POWER architecture. The problem is that it is low and takes up a lot of software resources.
请参阅图3,图3是本发明实施例公开的另一种芯片信号分析系统的结构示意图,如图3所示,总线协议转换模块13包括读写控制单元;Please refer to Figure 3. Figure 3 is a schematic structural diagram of another chip signal analysis system disclosed in an embodiment of the present invention. As shown in Figure 3, the bus protocol conversion module 13 includes a read and write control unit;
需要说明的是,在图3的描述中,以下概念是一致的,在具体描述中应当共通理解:It should be noted that in the description of Figure 3, the following concepts are consistent and should be commonly understood in the specific description:
Read/write Control、读写控制单元;Trigger Send Control、触发器发送控制子单元;Active Send Control、动态发送控制子单元;Work Mode Control、模式控制子单元;Flush Control、数据排空控制子单元;SP、优先级处理单元;ATB Master、协议转换单元;Read/write Control, read and write control unit; Trigger Send Control, trigger send control subunit; Active Send Control, dynamic send control subunit; Work Mode Control, mode control subunit; Flush Control, data emptying control subunit; SP, priority processing unit; ATB Master, protocol conversion unit;
以及,在图3中,软件资源模块为左上角软件字样的方块,通过存储器存取与芯片数据获取模块实现数据交互,芯片数据获取模块为右上角标注POWER8 Trace的方块,通过存储介质及优先级处理单元与总线协议转换模块进行数据交互,总线协议转换模块为右下角标注Trace to atb的方块,通过ATB总线与左下角标注DDR或者Debugger的调试模块进行数据交互。And, in Figure 3, the software resource module is the box with the word "software" in the upper left corner. It realizes data interaction with the chip data acquisition module through memory access. The chip data acquisition module is the box with the word "POWER8 Trace" in the upper right corner. Through the storage medium and priority The processing unit interacts with the bus protocol conversion module for data. The bus protocol conversion module is the box marked Trace to atb in the lower right corner. It interacts with the debugging module marked DDR or Debugger in the lower left corner through the ATB bus.
其中,读写控制单元与芯片数据获取模块12中的存储介质连接;Among them, the read-write control unit is connected to the storage medium in the chip data acquisition module 12;
以及,读写控制单元用于控制读写电路数据的工作参数,并根据工作参数读写电路数据。And, the reading and writing control unit is used to control the working parameters of reading and writing circuit data, and to read and write the circuit data according to the working parameters.
通过读写控制单元,可确定芯片数据获取模块与总线协议转换模块13之间数据交互的工作参数,以及确定数据交互模式等信息,以及执行根据确定出的工作参数进行电路数据读写的流程,由此,提升了应用场景的普适性,并提升了软件交互的效率,降低了芯片信号分析过程中软件资源的占用。Through the read-write control unit, the working parameters of data interaction between the chip data acquisition module and the bus protocol conversion module 13 can be determined, information such as the data interaction mode can be determined, and the process of reading and writing circuit data according to the determined working parameters can be executed. This improves the universality of application scenarios, improves the efficiency of software interaction, and reduces the occupation of software resources during chip signal analysis.
在一种可行的架构中,软件资源模块11和芯片数据获取模块12组成了整个的技术方案,软件可以通过总线去配置芯片数据获取模块12的工作模式,在芯片故障发生后,芯片控制目标路径开始、停止采集数据;软件需要在这个过程中,轮询等待,当目标路径数据采集完成后,软件通过总结接口依次读取存储介质中的数据。In a feasible architecture, the software resource module 11 and the chip data acquisition module 12 form the entire technical solution. The software can configure the working mode of the chip data acquisition module 12 through the bus. After a chip failure occurs, the chip controls the target path. Start and stop data collection; the software needs to poll and wait during this process. When the target path data collection is completed, the software reads the data in the storage medium sequentially through the summary interface.
相对于此,本申请提供的实施方式中还增加了Trace to atb的硬件模块,能够将存储介质中的数据封装成ARM ATB协议格式的数据包,然后通过相应的硬件结构,将数据发送至DDR,也可以直接发给专用的Debugger。In contrast, the implementation provided by this application also adds a Trace to ATB hardware module, which can encapsulate the data in the storage medium into a data packet in the ARM ATB protocol format, and then send the data to the DDR through the corresponding hardware structure. , or you can send it directly to a dedicated Debugger.
作为一种可选的实施方式,读写控制单元包括触发器发送控制子单元;As an optional implementation manner, the reading and writing control unit includes a trigger sending control subunit;
触发器发送控制子单元用于控制芯片数据获取模块12采集完成目标走线上的电路数据之后生成目标信息包,并驱动目标信息包传输至调试模块14。The trigger sending control subunit is used to control the chip data acquisition module 12 to generate a target information packet after completing the collection of circuit data on the target wiring, and drive the target information packet to be transmitted to the debugging module 14 .
具体的,对于触发器发送控制子单元的Trigger Send Control逻辑,在Trace数据采集完成后,主动向外部推送数据,数据推送完毕后,可以通过中断、特色的packet主动告知Debugger,节省软件开销,进而不再需要软件一直轮询等待。Specifically, for the Trigger Send Control logic of the trigger send control sub-unit, after the Trace data collection is completed, the data is actively pushed to the outside. After the data push is completed, the Debugger can be actively notified through interrupts and special packets to save software overhead and thereby There is no need for software to keep polling and waiting.
通过触发器发送控制子单元,可在电路数据采集完成后主动向外部推送数据,并通过预设的目标信息包告知调试模块当前的数据状态,无需软件重复轮询等待,由此提升了读写控制单元数据交互的效率,并提升了应用场景的普适性,并提升了软件交互的效率,降低了芯片信号分析过程中软件资源的占用。The control subunit is sent through a trigger, which can actively push data to the outside after the circuit data collection is completed, and inform the debugging module of the current data status through the preset target information packet, eliminating the need for repeated polling and waiting by the software, thereby improving read and write performance. It controls the efficiency of data interaction between units, improves the universality of application scenarios, improves the efficiency of software interaction, and reduces the occupation of software resources during the chip signal analysis process.
作为一种可选的实施方式,读写控制单元包括动态发送控制子单元;As an optional implementation manner, the reading and writing control unit includes a dynamic sending control subunit;
动态发送控制子单元用于控制总线协议转换模块13在获取目标走线上的电路数据的同时,向调试模块14传输封装数据。The dynamic transmission control subunit is used to control the bus protocol conversion module 13 to transmit encapsulated data to the debugging module 14 while acquiring the circuit data on the target wiring.
对于动态发送控制子单元的Active Send Control逻辑,在数据采集的同时,能够向外发送数据,在数据率变化不大的情况下,通过压缩的算法,可以达到更长时间的观测,理想情况下,可以观测到整个芯片运行周期的信息。For the Active Send Control logic of the dynamic send control subunit, data can be sent out while data is being collected. When the data rate does not change much, longer observations can be achieved through the compression algorithm. Ideally, , the information of the entire chip operation cycle can be observed.
通过动态发送控制子单元,可实现读写控制单元在数据采集的同时向外发送数据的功能,通过压缩算法可达成更长时间周期的数据观测,提升了应用场景的普适性,并提升了软件交互的效率,降低了芯片信号分析过程中软件资源的占用。Through the dynamic sending control sub-unit, the read-write control unit can realize the function of sending data out while collecting data. The compression algorithm can achieve longer period of data observation, which improves the universality of application scenarios and improves the The efficiency of software interaction reduces the occupation of software resources during chip signal analysis.
作为一种可选的实施方式,读写控制单元包括模式控制子单元;As an optional implementation manner, the reading and writing control unit includes a mode control subunit;
模式控制子单元用于控制总线协议转换模块13的数据发送模式。The mode control subunit is used to control the data sending mode of the bus protocol conversion module 13.
对于模式控制子单元Work Mode Control,不仅维持工作的运行,而且负责协调各种不同的工作模式切换。For the mode control sub-unit Work Mode Control, it not only maintains the operation of the work, but is also responsible for coordinating the switching of various different working modes.
通过模式控制子单元维持读写控制单元工作的正常运行,并协调总线协议转换模块的数据发送模式,提升了数据交互的效率,提升了应用场景的普适性,降低了芯片信号分析过程中软件资源的占用。The mode control sub-unit maintains the normal operation of the read-write control unit and coordinates the data transmission mode of the bus protocol conversion module, which improves the efficiency of data interaction, improves the universality of application scenarios, and reduces the software cost in the chip signal analysis process. Resource occupation.
作为一种可选的实施方式,读写控制单元包括数据排空控制子单元;As an optional implementation manner, the read/write control unit includes a data emptying control subunit;
数据排空控制子单元用于根据各目标走线对应的电路结构的时钟域,控制存储介质中的数据排空,发送至总线协议转换模块13。The data emptying control subunit is used to control the data emptying in the storage medium according to the clock domain of the circuit structure corresponding to each target wiring, and sends it to the bus protocol conversion module 13 .
对于数据排空控制子单元Flush Control相关的逻辑,主要完成存储介质中数据的排空,保证工作在不同时钟域的所有硬件模块,均能够有效、准确的排空数据,保证软件拿到足够的数据。For the logic related to the data flushing control subunit Flush Control, it mainly completes the flushing of data in the storage medium to ensure that all hardware modules working in different clock domains can flush data effectively and accurately to ensure that the software gets enough data.
通过数据排空控制子单元,可完成存储介质中数据的排空,确保工作在不同时钟域的硬件模块可有效排空数据,使得软件侧可获取足量数据,提升了数据交互的效率,提升了应用场景的普适性,降低了芯片信号分析过程中软件资源的占用。Through the data emptying control subunit, data emptying in the storage medium can be completed, ensuring that hardware modules working in different clock domains can effectively empty data, so that the software side can obtain sufficient data, improving the efficiency of data interaction and improving It improves the universality of application scenarios and reduces the occupation of software resources during chip signal analysis.
作为一种可选的实施方式,总线协议转换模块13还包括优先级处理单元;As an optional implementation manner, the bus protocol conversion module 13 also includes a priority processing unit;
其中,优先级处理单元与读写控制单元连接,并与芯片数据获取模块12中的存储介质连接;Among them, the priority processing unit is connected to the read-write control unit and connected to the storage medium in the chip data acquisition module 12;
以及,优先级处理单元用于处理读写控制单元的数据读写请求,并控制数据读写的优先级。And, the priority processing unit is used to process the data reading and writing requests of the reading and writing control unit, and control the priority of data reading and writing.
通过优先级处理单元,可用于处理读写控制单元的数据读写请求,并控制数据读写的优先级,提升了数据交互的效率,提升了应用场景的普适性,降低了芯片信号分析过程中软件资源的占用。The priority processing unit can be used to process data read and write requests from the read and write control unit, and control the priority of data read and write, which improves the efficiency of data interaction, improves the universality of application scenarios, and reduces the chip signal analysis process. usage of software resources.
作为一种可选的实施方式,目标协议包括高级跟踪总线协议ATB;As an optional implementation manner, the target protocol includes Advanced Trace Bus Protocol ATB;
总线协议转换模块13还包括协议转换单元;The bus protocol conversion module 13 also includes a protocol conversion unit;
其中,协议转换单元与读写控制单元连接,并与调试模块14通过基于目标协议的总线连接;Among them, the protocol conversion unit is connected to the read-write control unit and connected to the debugging module 14 through a bus based on the target protocol;
以及,协议转换单元用于通过总线接收调试模块14的数据写出请求,并根据数据写出请求将封装数据发送至调试模块14。And, the protocol conversion unit is configured to receive the data write request from the debugging module 14 through the bus, and send the encapsulated data to the debugging module 14 according to the data write request.
对于协议转换单元ATB Master的相关逻辑,该单元是最终完成协议转换的单元,能够向下发送数据,能够接收下游的Flush请求、完成相关的数据排空。Regarding the relevant logic of the protocol conversion unit ATB Master, this unit is the unit that finally completes the protocol conversion. It can send data downward, receive downstream Flush requests, and complete related data emptying.
根据预设的ATB总线协议,通过协议转换单元完成协议的转换和数据的排空,提升了数据交互的效率,提升了应用场景的普适性,降低了芯片信号分析过程中软件资源的占用。According to the preset ATB bus protocol, protocol conversion and data emptying are completed through the protocol conversion unit, which improves the efficiency of data interaction, improves the universality of application scenarios, and reduces the occupation of software resources during chip signal analysis.
对于图中的其余逻辑部分,例如第三方读写控制接口,用于根据工作模式发起对应的读请求,接收读数据;SP即前述的优先级处理单元,用于控制数据读写的时序,通常需要确保写优先级最高;数据打包逻辑部分,主要插入状态和标记信息;AFIFO部分,负责完成不同时钟域的异步转换。For the remaining logical parts in the figure, such as the third-party read and write control interface, it is used to initiate corresponding read requests and receive read data according to the working mode; SP is the aforementioned priority processing unit, used to control the timing of data reading and writing, usually It is necessary to ensure that the writing priority is the highest; the data packaging logic part mainly inserts status and tag information; the AFIFO part is responsible for completing asynchronous conversion of different clock domains.
综上,本申请通过总线协议转换模块13的设计,实现了数据的封装、协议转化及数据交互,由此提升了应用场景的普适性,并提升了软件交互的效率,降低了芯片信号分析过程中软件资源的占用,解决了POWER架构下的处理器芯片信号分析系统在进行硅后芯片验证流程中,软件交互效率低、软件资源占用大的问题。对于如前的各实施方式可以结合实施,在一个具体的应用场景下,整个技术方案的系统结构主要分为软件、Power Trace、Trace to atb tranfer、DDR/Debugger等四个部分。软件可以是运行在Power Core上的软件,也可以是第三方的软件,第三方的软件则需要借助芯片的外设,总体而言,可通过软件资源模块的形式,实现软件资源的搭载或外部交互。Power Trace是IBM Power架构中原有的硬件结构,支持压缩的算法,可以把一段时间内变化不频繁的信号进行压缩;支持非压缩的算法,存储介质有限的情况下,观察的信号范围较小;可以根据观察信号设置触发条件及触发信号,控制芯片数据获取模块12在目标走线上采集若干个数据。总线协议转换模块13的设计以图示Trace to atb的硬件结构实现,实现了数据的封装、协议转化及数据交互,由此提升了应用场景的普适性,并提升了软件交互的效率,降低了芯片信号分析过程中软件资源的占用,解决了POWER架构下的处理器芯片信号分析系统在进行硅后芯片验证流程中,软件交互效率低、软件资源占用大的问题。此外,DDR是大容量的硬件存储设备,Debugger是第三方的软硬件调试盒子,为调试模块14的具体软硬件实现方式,具体可参考前述的相关具体描述。In summary, this application realizes data encapsulation, protocol conversion and data interaction through the design of the bus protocol conversion module 13, thereby improving the universality of application scenarios, improving the efficiency of software interaction, and reducing the cost of chip signal analysis. The occupation of software resources during the process solves the problem of low software interaction efficiency and large software resource occupation during the post-silicon chip verification process of the processor chip signal analysis system under the POWER architecture. The previous implementation methods can be combined and implemented. In a specific application scenario, the system structure of the entire technical solution is mainly divided into four parts: software, Power Trace, Trace to ATB transfer, and DDR/Debugger. The software can be software running on the Power Core, or it can be third-party software. Third-party software needs to use the peripherals of the chip. Generally speaking, software resources can be carried or externalized in the form of software resource modules. Interaction. Power Trace is the original hardware structure in the IBM Power architecture. It supports compression algorithms and can compress signals that change infrequently within a period of time. It supports non-compression algorithms. When the storage medium is limited, the observed signal range is smaller; The trigger conditions and trigger signals can be set according to the observation signal, and the chip data acquisition module 12 is controlled to collect several data on the target wiring. The design of the bus protocol conversion module 13 is implemented with the hardware structure of Trace to ATB as shown in the figure, which realizes data encapsulation, protocol conversion and data interaction, thus improving the universality of application scenarios, improving the efficiency of software interaction and reducing It eliminates the occupation of software resources during the chip signal analysis process and solves the problems of low software interaction efficiency and large software resource occupation during the post-silicon chip verification process of the processor chip signal analysis system under the POWER architecture. In addition, DDR is a large-capacity hardware storage device, and Debugger is a third-party software and hardware debugging box, which is the specific software and hardware implementation of the debugging module 14. For details, please refer to the relevant descriptions mentioned above.
具体的,在Trace to atb硬件模块中,包括如下的各逻辑单元,如前文所提到的硬件模块名称描述的一致性不变,仍可参考前述关于图3的相应描述。Specifically, the Trace to atb hardware module includes the following logical units. As mentioned above, the consistency of the hardware module name description remains unchanged. You can still refer to the corresponding description of Figure 3.
软件资源模块11搭载或通过接口实时获取第三方的软件资源,通过存储记忆通道与POWER架构下内置的数据寻踪的芯片实现数据方面的通信连接,即前述的芯片数据获取模块12,该模块直接用于获取目标走线上的数据,以实现后续的硅后芯片验证功能。其中,Power8 Trace模块中通过存储介质与本申请提供的Trace to atb硬件模块实现数据交互,通过串行寄存器访问接口,实现与软件资源模块的数据交互,通过数据读写控制单元实现软件时序的控制,以及数据存储、数据捕获模式设定、数据压缩等数据操作,并集成打包数据,向总线协议转换模块13发起数据读写请求。The software resource module 11 is equipped with or obtains third-party software resources in real time through the interface, and realizes data communication connection through the storage memory channel and the built-in data tracking chip under the POWER architecture, that is, the aforementioned chip data acquisition module 12, which directly Used to obtain data on the target trace to achieve subsequent post-silicon chip verification functions. Among them, the Power8 Trace module realizes data interaction with the Trace to ATB hardware module provided by this application through the storage medium, realizes data interaction with the software resource module through the serial register access interface, and realizes software timing control through the data read and write control unit. , as well as data operations such as data storage, data capture mode setting, data compression, etc., and integrates packaged data, and initiates data read and write requests to the bus protocol conversion module 13.
总线协议转换模块13的主要单元包括读写控制单元和协议转换单元的相关结构,对于图示的Trace to atb的硬件模块,能够将数据获取模块12存储介质中的数据封装成ARM ATB协议格式的封装数据,然后通过协议转换单元完成数据协议的转换和时序匹配等流程,通过ATB总线,将数据发送至DDR,也可以直接发给专用的Debugger,实现调试功能。The main units of the bus protocol conversion module 13 include the relevant structures of the read-write control unit and the protocol conversion unit. For the Trace to ATB hardware module shown in the figure, the data in the storage medium of the data acquisition module 12 can be encapsulated into the ARM ATB protocol format. The data is encapsulated, and then the data protocol conversion and timing matching are completed through the protocol conversion unit. The data is sent to the DDR through the ATB bus, or directly to a dedicated Debugger to implement the debugging function.
在读写控制单元中,可以包括前文提到的四个子逻辑单元,这四个子单元可以共同实施,设计于读写控制单元中,并与第三方读写控制接口连接,进一步的连接外部的优先级处理、数据标记及协议转换等功能单元,在硅后芯片验证流程中控制相应单元的工作模式或数据交互方式。The read-write control unit can include the four sub-logic units mentioned above. These four sub-units can be implemented together, designed in the read-write control unit, and connected to a third-party read-write control interface to further connect to external priority Functional units such as level processing, data marking, and protocol conversion control the working mode or data interaction mode of the corresponding unit in the post-silicon chip verification process.
具体的,Trigger Send Control逻辑单元,在数据获取模块对目标走线上的电子数据采集完成后,可自动化的传输及推送数据,数据推送完毕后,可以通过具体应用场景下设计的,带有中断功能的封装数据主动告知Debugger,节省软件开销,进而不再需要软件一直轮询等待,降低软件资源的消耗。Active Send Control逻辑单元,在控制数据获取模块12进行数据采集的同时,能够向总线协议转换模块13等外部结构发送数据,实现数据读写的并行执行,在数据率变化不大的情况下,通过压缩的算法,可以达到更长时间的观测,理想情况下,可以观测到整个芯片运行周期的信息,提升了硅后芯片验证时观测的数据量,进而提升了数据观测的效率。Work Mode Control单元,不仅维持工作的运行,而且负责协调各种不同的工作模式切换,主要用于对目标走线进行数据获取以及各模块间数据交互过程的控制,确保数据交互的效率和有效性。Flush Control相关的逻辑单元,主要完成存储介质中数据的排空,保证工作在不同时钟域的所有硬件模块,均能够有效、准确的排空数据,保证调试模块14中可获取到足够的数据,以进行数据调试。读写控制模块通过控制工作模式及数据交互方式,提升软件交互的效率,并降低软件资源的占用。Specifically, the Trigger Send Control logic unit can automatically transmit and push data after the data acquisition module completes the electronic data collection on the target wiring. After the data push is completed, it can be designed under specific application scenarios with interrupts. The encapsulated data of the function is actively notified to the Debugger, saving software overhead. This eliminates the need for the software to poll and wait all the time, reducing the consumption of software resources. The Active Send Control logic unit, while controlling the data acquisition module 12 to collect data, can send data to external structures such as the bus protocol conversion module 13 to achieve parallel execution of data reading and writing. When the data rate changes little, through The compressed algorithm can achieve longer-term observations. Ideally, information on the entire chip operation cycle can be observed, which increases the amount of data observed during post-silicon chip verification, thereby improving the efficiency of data observation. The Work Mode Control unit not only maintains the operation of the work, but is also responsible for coordinating the switching of various working modes. It is mainly used to obtain data from the target wiring and control the data interaction process between modules to ensure the efficiency and effectiveness of data interaction. . The logic unit related to Flush Control mainly completes the emptying of data in the storage medium, ensuring that all hardware modules working in different clock domains can effectively and accurately empty data, and ensuring that sufficient data can be obtained in the debugging module 14. for data debugging. The read-write control module improves the efficiency of software interaction and reduces the occupation of software resources by controlling the working mode and data interaction method.
另一个重要单元,即协议转换单元及其周边结构,可用于提升应用场景的普适性,ATB Master的相关逻辑单元,是最终完成协议转换的单元,能够向下发送数据,接收下游的Flush请求、完成相关的数据排空。Another important unit, the protocol conversion unit and its surrounding structures, can be used to improve the universality of application scenarios. The related logical unit of the ATB Master is the unit that ultimately completes the protocol conversion and can send data downwards and receive downstream Flush requests. , complete related data emptying.
对于总线协议转换模块13其余逻辑单元部分,第三方读写控制接口,根据读写控制单元确定出的工作模式及数据交互模式,向上游的芯片数据获取模块12发起读写请求并获取数据;SP是绝对优先级处理模块,在具体应用场景下,通常需要确保写优先级最高;Package部分,主要插入Flag、status等状态及标记信息,以丰富打包数据中的信息量,便于其他模块根据打包数据获取相关的数据状态;AFIFO部分,则用于完成不同时钟域的异步转换,其转换逻辑可以参考下文的图4和图5的相关描述。For the remaining logical units of the bus protocol conversion module 13, the third-party read-write control interface initiates a read-write request to the upstream chip data acquisition module 12 and obtains data according to the working mode and data interaction mode determined by the read-write control unit; SP It is an absolute priority processing module. In specific application scenarios, it is usually necessary to ensure that the writing priority is the highest; in the Package part, status and mark information such as Flag and status are mainly inserted to enrich the amount of information in the packaged data and facilitate other modules based on the packaged data. Obtain the relevant data status; the AFIFO part is used to complete asynchronous conversion of different clock domains. For its conversion logic, please refer to the relevant descriptions in Figure 4 and Figure 5 below.
请参阅图4,图4是本发明实施例公开的一种芯片信号分析系统运行逻辑的流程示意图,如图4所示,数据排空控制子单元面对不同时钟域的硬件模块时,其电平变换方式及执行流程可以参考图中所示的变换方向及文字描述。Please refer to Figure 4. Figure 4 is a schematic flow chart of the operation logic of a chip signal analysis system disclosed in an embodiment of the present invention. As shown in Figure 4, when the data emptying control subunit faces hardware modules in different clock domains, its electrical For the flat transformation method and execution process, please refer to the transformation direction and text description shown in the figure.
具体的,图示的运行逻辑用于指示处理器芯片核心的时钟域与ATB协议下时钟域通过软件对硬件结构的信号进行交互和转换的流程,用以最终实现协议的转换。Specifically, the running logic shown in the diagram is used to instruct the process of interaction and conversion of the clock domain of the processor chip core and the clock domain under the ATB protocol through software to the signals of the hardware structure to ultimately realize the protocol conversion.
其中,该时序图左侧用于指示处理器芯片核心本身的时钟域core clock domain,右侧用于指示ATB协议下的目标时钟域atb clock domain,flush请求指示软件中用于控制数据流的请求信息,对缓冲区的数据进行处理,达成按需获取或控制数据流的效果,即图示中的数据排空。flush_req和flush_ack为时序转换过程中发起的指示信号。根据这两个信号的转换方式及条件,即可组成实现协议的转换的逻辑。Among them, the left side of the timing diagram is used to indicate the core clock domain of the processor chip core itself, the right side is used to indicate the target clock domain atb clock domain under the ATB protocol, and the flush request indicates the request used to control the data flow in the software. Information, process the data in the buffer to achieve the effect of obtaining or controlling the data flow on demand, that is, emptying the data in the illustration. flush_req and flush_ack are indication signals initiated during the timing conversion process. According to the conversion methods and conditions of these two signals, the logic for realizing the conversion of the protocol can be formed.
atb clock domain收到flush的请求,向core clock domain发起一个高电平的flush_req,即拉升atb2core方向的flush_req信号,core clock domain检测到atb clockdomain的请求,开始读取Trace arrary的数据,直到所有flush的数据均读完,当package中的数据也排空后,拉起一个ACK信号,即拉高core2atb_flush_ack信号,atb clock domain收到core clock domain的ACK信号,开始等待AFIFO、ATB master中的所有数据向箭头下游方向排空,排空完成后,拉低flush_req,即拉低atb2core方向flush_req信号,core clockdomain检测到atb clock domain的flush_req拉低,进入IDLE状态,同时拉低ACK信号,即拉低core2atb方向flush_ack信号,最终,atb clock domain检测到ACK信号拉低,同时也进入IDLE状态。The atb clock domain receives the flush request and initiates a high-level flush_req to the core clock domain, that is, the flush_req signal in the direction of atb2core is raised. The core clock domain detects the request of the atb clockdomain and begins to read the data of the Trace array until all After all the flush data is read, when the data in the package is also emptied, an ACK signal is pulled up, that is, the core2atb_flush_ack signal is pulled high. The atb clock domain receives the ACK signal from the core clock domain and begins to wait for all data in AFIFO and ATB master. The data is drained in the downstream direction of the arrow. After the emptying is completed, flush_req is pulled low, that is, the flush_req signal in the atb2core direction is pulled low. The core clockdomain detects that the flush_req of the atb clock domain is pulled low, and enters the IDLE state. At the same time, the ACK signal is pulled low, that is, the flush_req signal is pulled low. core2atb direction flush_ack signal, finally, atb clock domain detects the ACK signal and pulls it low, and also enters the IDLE state.
此外,请参阅图5,图5是本发明实施例公开的另一种芯片信号分析系统运行逻辑的流程示意图,如图5所示,数据排空控制子单元在运行过程中,其电平变换的条件及对应程序命令可以参考图中所示的相关描述。In addition, please refer to Figure 5. Figure 5 is a schematic flowchart of the operation logic of another chip signal analysis system disclosed in an embodiment of the present invention. As shown in Figure 5, during the operation of the data emptying control subunit, its level changes For conditions and corresponding program commands, please refer to the relevant descriptions shown in the figure.
同样,该时序图左侧用于指示处理器芯片核心本身的时钟域,右侧用于指示ATB协议下的目标时钟域,在一个实现方式下,core clock domain在时序图所示的流程中包括四个flush相关的数据状态,即Flush_IDLE、Flush_ARR、FLUSH_PKT和ATB_FLUSH,而atb clockdomain也包括了四个flush相关的数据状态,即IDLE、Trace_FLUSH、ATB_FLUSH和FLUSH_END。IDLE至Trace_FLUSH的过程用于指示atb clock domain收到flush的请求,满足coresight_aflush_req||software_trig&trig_mode||trace_run_stop_trig&trig_mode命令时,向core clock domain发起一个高电平的flush_req,即atb2core_flush_req_d3命令,触发core clock domain从IDLE状态转换成声明flush的状态,通过arr_has_rd_out命令触发flush相关数据的打包封装,当出现pkt_flush_finish状态时确定出封装完成,向atb clock domain侧发起信号,即前述的core clock domain检测到atb clock domain的请求,开始读取Trace arrary的数据,直到所有flush的数据均读完,当package中的数据也排空后,拉起一个ACK信号,拉高core2atb_flush_ack信号,即core2atb_flush_ack_d3命令,atb clock domain收到core clock domain的ACK信号,开始等待AFIFO、ATB master中的所有数据向箭头下游方向排空,排空完成后,发起atb_domain_empty命令,以触发拉低atb2core方向flush_req信号的动作,core clock domain检测到atb clock domain的flush_req拉低,即!atb2core_flush_req_d3命令后,进入IDLE状态,同时拉低ACK信号,即拉低core2atb方向flush_ack信号,最终,atb clock domain检测到ACK信号拉低,即!core2atb_flush_ack_d3命令后,同时也进入IDLE状态。Similarly, the left side of the timing diagram is used to indicate the clock domain of the processor chip core itself, and the right side is used to indicate the target clock domain under the ATB protocol. In one implementation, the core clock domain is included in the process shown in the timing diagram. There are four flush-related data states, namely Flush_IDLE, Flush_ARR, FLUSH_PKT and ATB_FLUSH, and the atb clockdomain also includes four flush-related data states, namely IDLE, Trace_FLUSH, ATB_FLUSH and FLUSH_END. The process from IDLE to Trace_FLUSH is used to instruct the atb clock domain to receive a flush request and when the coresight_aflush_req||software_trig&trig_mode||trace_run_stop_trig&trig_mode command is met, a high-level flush_req is initiated to the core clock domain, that is, the atb2core_flush_req_d3 command, triggering the core clock domain to start from IDLE The state is converted to a state where flush is declared, and the packaging and encapsulation of flush-related data is triggered through the arr_has_rd_out command. When the pkt_flush_finish state appears, it is determined that the encapsulation is completed, and a signal is sent to the atb clock domain side, that is, the aforementioned core clock domain detects the request of the atb clock domain. , start reading the data of the Trace array until all the flush data is read. When the data in the package is also emptied, pull up an ACK signal and pull up the core2atb_flush_ack signal, that is, the core2atb_flush_ack_d3 command. The atb clock domain receives the core clock Domain's ACK signal starts waiting for all data in AFIFO and ATB master to be emptied in the downstream direction of the arrow. After the emptying is completed, the atb_domain_empty command is initiated to trigger the action of pulling down the flush_req signal in the atb2core direction. The core clock domain detects the atb clock domain The flush_req is pulled low, that is! After the atb2core_flush_req_d3 command, it enters the IDLE state and pulls down the ACK signal at the same time, that is, pulls down the flush_ack signal in the core2atb direction. Finally, the atb clock domain detects that the ACK signal is pulled down, that is! After the core2atb_flush_ack_d3 command, it also enters the IDLE state.
其中,关于flush请求相关的命名方式、数据命令及声明语句,可参考具体的开发语言进行设置,其逻辑功能可结合图4进行理解。Among them, regarding the naming method, data commands and declaration statements related to the flush request, you can refer to the specific development language for settings, and its logical functions can be understood in conjunction with Figure 4.
综上,本申请提供的实施方式,可节省软件开销,数据推送完毕后,可以告知Debugger,进而不再需要软件一直轮询等待,并提高工作效率,主动通过专用的ATB总线推送数据,工作频率更高、总线带宽更大,以及在存储介质不变的情况下,数据变化较慢的场景下,可以采集更长时间范围内的数据,理想情况下,可以观测整个芯片运行周期内的数据。In summary, the implementation provided by this application can save software overhead. After the data is pushed, the Debugger can be notified, so that the software no longer needs to poll and wait all the time, and improves work efficiency. It actively pushes data through the dedicated ATB bus, and the working frequency Higher, larger bus bandwidth, and in scenarios where the storage medium remains unchanged and the data changes slowly, data can be collected over a longer period of time. Ideally, data within the entire chip operating cycle can be observed.
通过软件资源模块搭载或调用用于芯片信号分析的软件资源,并通过芯片数据获取模块,读取软件资源,获取目标走线上的电路数据,进而以服务器芯片的数据传输协议,将电路数据传输至总线协议转换模块,总线协议转换模块完成时序同步和协议转换流程,将电路数据根据预设的目标协议封装成封装数据,以目标协议传输至调试模块进行芯片信号分析,由此,提升了应用场景的普适性,并提升了软件交互的效率,降低了芯片信号分析过程中软件资源的占用,解决了POWER架构下的处理器芯片信号分析系统在进行硅后芯片验证流程中,软件交互效率低、软件资源占用大的问题。The software resource module is used to load or call software resources for chip signal analysis, and through the chip data acquisition module, the software resources are read and the circuit data on the target wiring is obtained, and then the circuit data is transmitted using the data transmission protocol of the server chip. To the bus protocol conversion module, the bus protocol conversion module completes the timing synchronization and protocol conversion process, encapsulates the circuit data into encapsulated data according to the preset target protocol, and transmits it to the debugging module for chip signal analysis using the target protocol. This improves the application The universality of the scenario improves the efficiency of software interaction, reduces the occupation of software resources during the chip signal analysis process, and solves the problem of software interaction efficiency in the post-silicon chip verification process of the processor chip signal analysis system under the POWER architecture. The problem is that it is low and takes up a lot of software resources.
此外,本申请提供一种处理器芯片,处理器芯片包括如任一实施例的芯片信号分析系统。In addition, the present application provides a processor chip, which includes the chip signal analysis system as in any embodiment.
以及,本申请提供一种电子设备,电子设备包括如任一实施例的处理器芯片。And, the present application provides an electronic device, which includes the processor chip as in any embodiment.
以上所描述的装置实施例仅是示意性的,其中作为分离部件说明的模块可以是或者也可以不是物理上分开的,作为模块显示的部件可以是或者也可以不是物理模块,即可以位于一个地方,或者也可以分布到多个网络模块上。The device embodiments described above are only illustrative. The modules described as separate components may or may not be physically separated. The components shown as modules may or may not be physical modules, that is, they may be located in one place. , or it can be distributed to multiple network modules.
可以根据实际的需要选择其中的部分或者全部模块来实现本实施例方案的目的。本领域普通技术人员在不付出创造性的劳动的情况下,即可以理解并实施。Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. Persons of ordinary skill in the art can understand and implement the method without any creative effort.
本领域技术人员在考虑说明书及实践这里公开的发明后,将容易想到本申请的其它实施方案。本申请旨在涵盖本申请的任何变型、用途或者适应性变化,这些变型、用途或者适应性变化遵循本申请的一般性原理并包括本申请未公开的本技术领域中的公知常识或惯用技术手段。说明书和实施例仅被视为示例性的,本申请的真正范围和精神由下面的权利要求书指出。Other embodiments of the present application will be readily apparent to those skilled in the art from consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of this application that follow the general principles of this application and include common knowledge or customary technical means in the technical field that are not disclosed in this application. . It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the application being indicated by the following claims.
应当理解的是,本申请并不局限于上面已经描述并在附图中示出的精确结构,并且可以在不脱离其范围进行各种修改和改变。本申请的范围仅由所附的权利要求书来限制。It is to be understood that the present application is not limited to the precise structures described above and illustrated in the accompanying drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the application is limited only by the appended claims.
Claims (10)
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311698097.3A CN117725863B (en) | 2023-12-11 | 2023-12-11 | Chip signal analysis device, system, processor chip and electronic equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311698097.3A CN117725863B (en) | 2023-12-11 | 2023-12-11 | Chip signal analysis device, system, processor chip and electronic equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117725863A true CN117725863A (en) | 2024-03-19 |
| CN117725863B CN117725863B (en) | 2024-09-10 |
Family
ID=90208128
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311698097.3A Active CN117725863B (en) | 2023-12-11 | 2023-12-11 | Chip signal analysis device, system, processor chip and electronic equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117725863B (en) |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6523076B1 (en) * | 1999-11-08 | 2003-02-18 | International Business Machines Corporation | Method and apparatus for synchronizing multiple bus arbiters on separate chips to give simultaneous grants for the purpose of breaking livelocks |
| US6691257B1 (en) * | 2000-04-13 | 2004-02-10 | Stratus Technologies Bermuda Ltd. | Fault-tolerant maintenance bus protocol and method for using the same |
| CN1964232A (en) * | 2006-09-30 | 2007-05-16 | 厦门大学 | A serial communication card of optical fiber |
| CN101315555A (en) * | 2007-10-24 | 2008-12-03 | 天津理工大学 | USB Communication System and Working Method of Excitation Controller Based on DSP Chip |
| US20100023682A1 (en) * | 2007-10-11 | 2010-01-28 | Super Talent Electronics Inc. | Flash-Memory System with Enhanced Smart-Storage Switch and Packed Meta-Data Cache for Mitigating Write Amplification by Delaying and Merging Writes until a Host Read |
| CN101969393A (en) * | 2010-10-27 | 2011-02-09 | 四川省绵阳西南自动化研究所 | Protocol converter for FlexRay bus and PC104 bus |
| CN105654993A (en) * | 2016-02-22 | 2016-06-08 | 深圳市同创国芯电子有限公司 | Function verification method and platform for DDR3 SDRAM (double data rate 3 synchronous dynamic random access memory) controller |
| US20180293123A1 (en) * | 2017-04-10 | 2018-10-11 | Western Digital Technologies, Inc. | Detecting Memory Failures in the Runtime Environment |
| CN112562121A (en) * | 2020-12-30 | 2021-03-26 | 洛阳伟信电子科技有限公司 | Storage data and rapid downloading method based on SRIO protocol |
| CN115587058A (en) * | 2022-12-05 | 2023-01-10 | 苏州浪潮智能科技有限公司 | Data processing system, method, equipment and medium |
| CN116483319A (en) * | 2022-12-27 | 2023-07-25 | 清华大学 | Operator processing method, device, equipment and medium for software-defined chips |
-
2023
- 2023-12-11 CN CN202311698097.3A patent/CN117725863B/en active Active
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6523076B1 (en) * | 1999-11-08 | 2003-02-18 | International Business Machines Corporation | Method and apparatus for synchronizing multiple bus arbiters on separate chips to give simultaneous grants for the purpose of breaking livelocks |
| US6691257B1 (en) * | 2000-04-13 | 2004-02-10 | Stratus Technologies Bermuda Ltd. | Fault-tolerant maintenance bus protocol and method for using the same |
| CN1964232A (en) * | 2006-09-30 | 2007-05-16 | 厦门大学 | A serial communication card of optical fiber |
| US20100023682A1 (en) * | 2007-10-11 | 2010-01-28 | Super Talent Electronics Inc. | Flash-Memory System with Enhanced Smart-Storage Switch and Packed Meta-Data Cache for Mitigating Write Amplification by Delaying and Merging Writes until a Host Read |
| CN101315555A (en) * | 2007-10-24 | 2008-12-03 | 天津理工大学 | USB Communication System and Working Method of Excitation Controller Based on DSP Chip |
| CN101969393A (en) * | 2010-10-27 | 2011-02-09 | 四川省绵阳西南自动化研究所 | Protocol converter for FlexRay bus and PC104 bus |
| CN105654993A (en) * | 2016-02-22 | 2016-06-08 | 深圳市同创国芯电子有限公司 | Function verification method and platform for DDR3 SDRAM (double data rate 3 synchronous dynamic random access memory) controller |
| US20180293123A1 (en) * | 2017-04-10 | 2018-10-11 | Western Digital Technologies, Inc. | Detecting Memory Failures in the Runtime Environment |
| CN112562121A (en) * | 2020-12-30 | 2021-03-26 | 洛阳伟信电子科技有限公司 | Storage data and rapid downloading method based on SRIO protocol |
| CN115587058A (en) * | 2022-12-05 | 2023-01-10 | 苏州浪潮智能科技有限公司 | Data processing system, method, equipment and medium |
| CN116483319A (en) * | 2022-12-27 | 2023-07-25 | 清华大学 | Operator processing method, device, equipment and medium for software-defined chips |
Non-Patent Citations (5)
| Title |
|---|
| HH CHUANG, CJ HSU AND ETC, .: "A Broadband Chip-Level Power-Bus Model Feasible for Power Integrity Chip-Package Codesign in High-speed memory circuits", IEEE TRANSACTIONS ON ELECTROMAGNETIC COMPATIBILITY, 18 February 2010 (2010-02-18) * |
| 徐元哲: "Fs88x6系列加密芯片烧录器的设计与实现", 中国优秀硕士学位论文全文数据库, no. 01, 15 January 2011 (2011-01-15) * |
| 武甲东: "基于龙芯SoC的USB主机控制器的设计研究", 中国优秀博硕士学位论文全文数据库 (硕士), no. 07, 15 November 2005 (2005-11-15) * |
| 潘勇: "超高频RFID读写器基带信号处理SoC系统设计", 中国优秀硕士学位论文全文数据库, no. 04, 15 April 2012 (2012-04-15) * |
| 钟丽;刘彦;余思洋;谢中;: "嵌入式系统芯片中SM2算法软硬件协同设计与实现", 计算机应用, no. 05, 10 May 2015 (2015-05-10) * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117725863B (en) | 2024-09-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103513650B (en) | The record ripple control method of converter valve opertaing device | |
| US5586279A (en) | Data processing system and method for testing a data processor having a cache memory | |
| CN113127302B (en) | A monitoring method and device for board GPIO | |
| CN101539902A (en) | DMA device for nodes in multi-computer system and communication method | |
| CN118363914B (en) | Data processing method, solid state disk device and host | |
| CN116301627A (en) | A kind of NVMe controller and its initialization, data reading and writing method | |
| CN118733516A (en) | Inter-chip sideband transmission serial communication method, device and electronic device | |
| CN115794705A (en) | A CHI bus memory for fast function verification of chip prototype | |
| CN104714907A (en) | Design method for converting PCI bus into ISA bus or APB bus | |
| CN114756494A (en) | Conversion interface of standard communication protocol and on-chip packet transmission protocol of multi-die interconnection | |
| CN117472815A (en) | Storage module conversion interface under AXI protocol and conversion method thereof | |
| CN119311627B (en) | Descriptor-based PCIe bus DMA high-speed data transmission method and device | |
| US10088523B2 (en) | Debug adapter | |
| CN113946524B (en) | A dual-port RAM system and method based on FPGA | |
| CN117725863B (en) | Chip signal analysis device, system, processor chip and electronic equipment | |
| CN118860921A (en) | Dynamic mapping method for multi-bank access, electronic device and storage medium | |
| CN109726149B (en) | Method and device for accessing NAND FLASH through AXI bus | |
| CN108897696B (en) | Large-capacity FIFO controller based on DDRx memory | |
| CN117472669A (en) | Chip debugging structure based on multi-core system | |
| CN115328832B (en) | Data scheduling system and method based on PCIE DMA | |
| CN117539805A (en) | Sideband access method, memory controller, core particle and computer equipment | |
| CN116841458A (en) | Memory read and write control method, system, terminal and storage medium | |
| CN208027343U (en) | A remote dynamic self-updating loading system for FPGA | |
| CN221175403U (en) | Reading device with built-in DMA control module | |
| CN119473956B (en) | Controller of storage device and instruction scheduling method |
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 | ||
| PP01 | Preservation of patent right |
Effective date of registration: 20250728 Granted publication date: 20240910 |
|
| PP01 | Preservation of patent right |