CN118394703B - On-chip interconnection system and on-chip communication method - Google Patents
On-chip interconnection system and on-chip communication method Download PDFInfo
- Publication number
- CN118394703B CN118394703B CN202410823066.4A CN202410823066A CN118394703B CN 118394703 B CN118394703 B CN 118394703B CN 202410823066 A CN202410823066 A CN 202410823066A CN 118394703 B CN118394703 B CN 118394703B
- Authority
- CN
- China
- Prior art keywords
- module
- chip
- data
- processing
- transmission
- 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
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F15/00—Digital computers in general; Data processing equipment in general
- G06F15/16—Combinations of two or more digital computers each having at least an arithmetic unit, a program unit and a register, e.g. for a simultaneous processing of several programs
- G06F15/163—Interprocessor communication
- G06F15/173—Interprocessor communication using an interconnection network, e.g. matrix, shuffle, pyramid, star, snowflake
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F15/00—Digital computers in general; Data processing equipment in general
- G06F15/76—Architectures of general purpose stored program computers
- G06F15/78—Architectures of general purpose stored program computers comprising a single central processing unit
- G06F15/7807—System on chip, i.e. computer system on a single chip; System in package, i.e. computer system on one or more chips in a single package
-
- 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
- Engineering & Computer Science (AREA)
- Computer Hardware Design (AREA)
- Theoretical Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Computing Systems (AREA)
- Microelectronics & Electronic Packaging (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Communication Control (AREA)
Abstract
本发明公开了一种片内互联系统及片内通信方法,涉及芯片技术领域,该片内互联系统包括:两个接口通信组件以及中间传输组件;一个接口通信组件连接片内模块,另一个接口通信组件连接其它片内模块或者其它片内互联系统;接口通信组件分别通过时钟域转换模块、校验执行模块、协议转换模块和位频转换模块,对待传输数据进行时钟域转换处理、校验处理、通信协议转换处理、位宽转换处理和频率转换处理;中间传输组件通过缓存模块实现传输数据的逐级打拍。本发明实施例的技术方案,片内互联系统不但减小了芯片的走线面积,提高了片内模块的可用面积,而且在执行高位宽数据传输任务时,可以利用较小位宽实现高频传输,确保了较高的数据传输效率。
The present invention discloses an on-chip interconnection system and an on-chip communication method, and relates to the field of chip technology. The on-chip interconnection system includes: two interface communication components and an intermediate transmission component; one interface communication component connects an on-chip module, and the other interface communication component connects other on-chip modules or other on-chip interconnection systems; the interface communication component performs clock domain conversion processing, verification processing, communication protocol conversion processing, bit width conversion processing and frequency conversion processing on the data to be transmitted through a clock domain conversion module, a verification execution module, a protocol conversion module and a bit-frequency conversion module respectively; the intermediate transmission component realizes the step-by-step beat of the transmission data through a cache module. According to the technical solution of the embodiment of the present invention, the on-chip interconnection system not only reduces the routing area of the chip and increases the available area of the on-chip module, but also can realize high-frequency transmission by using a smaller bit width when performing high-bit width data transmission tasks, thereby ensuring higher data transmission efficiency.
Description
技术领域Technical Field
本发明涉及芯片技术领域,尤其涉及一种片内互联系统及片内通信方法。The present invention relates to the field of chip technology, and in particular to an intra-chip interconnection system and an intra-chip communication method.
背景技术Background Art
随着人工智能(Artificial Intelligence,AI)技术的不断发展,AI芯片逐步向高性能、低功耗的方向发展,而上层模型的演变,使得处理大数据量和实现高带宽互联成为了AI芯片设计的重要目标。With the continuous development of artificial intelligence (AI) technology, AI chips are gradually moving towards high performance and low power consumption. The evolution of upper-level models has made processing large amounts of data and achieving high-bandwidth interconnection important goals in AI chip design.
现有技术中,AI芯片内部通常使用传统的互联网络(例如,FlexNoC结构等)来实现多个片内IP(Intellectual Property)的互联;然而这样的互联方式,伴随着处理数据量的增大,总线位宽也会变大,进而使得走线面积较大,变相压缩了功能模块的可用面积。In the prior art, AI chips usually use traditional Internet networks (for example, FlexNoC structure, etc.) to interconnect multiple on-chip IPs (Intellectual Properties). However, with this interconnection method, as the amount of data processed increases, the bus width will also increase, which will make the routing area larger and in disguise compress the available area of the functional modules.
此外,受限于传统互联网络设计本身较为复杂的路由结构,在高位宽时难以实现高频传输,无法满足高性能需求,同时,针对芯片内部多组一对一互联场景时,传统互联网络的内部冗余设计会带来更多的面积开销。In addition, due to the relatively complex routing structure of the traditional Internet design itself, it is difficult to achieve high-frequency transmission at high bit widths and cannot meet high-performance requirements. At the same time, when it comes to multiple groups of one-to-one interconnection scenarios within the chip, the internal redundant design of the traditional Internet network will bring more area overhead.
发明内容Summary of the invention
本发明提供了一种片内互联系统及片内互联方法,以解决传统布线方式走线面积较大,片内互联系统传输效率较低的问题。The present invention provides an on-chip interconnection system and an on-chip interconnection method to solve the problem that the traditional wiring method has a large wiring area and a low transmission efficiency of the on-chip interconnection system.
根据本发明的一方面,提供了一种片内互联系统,包括:According to one aspect of the present invention, there is provided an on-chip interconnection system, comprising:
两个接口通信组件以及位于两个接口通信组件之间的中间传输组件;一个接口通信组件连接片内模块,另一个接口通信组件连接其它片内模块或者其它片内互联系统;接口通信组件包括时钟域转换模块、校验执行模块、协议转换模块、位频转换模块和配置管理模块;中间传输组件包括至少一个缓存模块;Two interface communication components and an intermediate transmission component located between the two interface communication components; one interface communication component connects the on-chip module, and the other interface communication component connects other on-chip modules or other on-chip interconnection systems; the interface communication component includes a clock domain conversion module, a check execution module, a protocol conversion module, a bit-frequency conversion module and a configuration management module; the intermediate transmission component includes at least one cache module;
时钟域转换模块,连接校验执行模块,用于待传输数据的时钟域转换处理;The clock domain conversion module is connected to the verification execution module and is used for clock domain conversion processing of the data to be transmitted;
校验执行模块,连接协议转换模块,用于待传输数据的校验处理;A verification execution module, connected to a protocol conversion module, is used for verification processing of data to be transmitted;
协议转换模块,连接位频转换模块,用于待传输数据的通信协议转换处理;The protocol conversion module is connected to the bit-frequency conversion module and is used for the communication protocol conversion processing of the data to be transmitted;
位频转换模块,连接协议转换模块,用于待传输数据的位宽转换处理和频率转换处理;A bit-frequency conversion module, connected to the protocol conversion module, is used for bit width conversion and frequency conversion of data to be transmitted;
配置管理模块,用于为时钟域转换模块、校验执行模块、协议转换模块和位频转换模块,提供配置参数;A configuration management module, used for providing configuration parameters for a clock domain conversion module, a check execution module, a protocol conversion module and a bit-frequency conversion module;
中间传输组件,用于通过至少一个缓存模块执行待传输数据的打拍传输。The intermediate transmission component is used to perform the beat transmission of the data to be transmitted through at least one cache module.
根据本发明的另一方面,提供了一种片内通信方法,应用于本发明任意实施例所述的片内互联系统,包括:According to another aspect of the present invention, there is provided an intra-chip communication method, which is applied to the intra-chip interconnection system described in any embodiment of the present invention, comprising:
第一接口通信组件依次通过内部的时钟域转换模块、校验执行模块、协议转换模块和位频转换模块,对第一片内模块发出的待传输数据进行时钟域转换处理、校验处理、通信协议转换处理、位宽转换处理和频率转换处理,以将待传输数据发送给中间传输组件;The first interface communication component sequentially performs clock domain conversion processing, verification processing, communication protocol conversion processing, bit width conversion processing and frequency conversion processing on the data to be transmitted sent by the first on-chip module through the internal clock domain conversion module, verification execution module, protocol conversion module and bit-frequency conversion module, so as to send the data to be transmitted to the intermediate transmission component;
中间传输组件通过至少一个缓存模块执行待传输数据的打拍传输,以将待传输数据发送给第二接口通信组件;The intermediate transmission component performs a beat transmission of the data to be transmitted through at least one buffer module to send the data to be transmitted to the second interface communication component;
第二接口通信组件依次通过内部的位频转换模块、协议转换模块、校验执行模块和时钟域转换模块,对待传输数据进行位宽转换处理、频率转换处理、通信协议转换处理、校验处理和时钟域转换处理,以将待传输数据发送给第二片内模块。The second interface communication component sequentially performs bit width conversion processing, frequency conversion processing, communication protocol conversion processing, verification processing and clock domain conversion processing on the data to be transmitted through the internal bit-frequency conversion module, protocol conversion module, verification execution module and clock domain conversion module, so as to send the data to be transmitted to the second on-chip module.
根据本发明的另一方面,提供了一种片内通信装置,应用于本发明任意实施例所述的片内互联系统,包括:According to another aspect of the present invention, there is provided an intra-chip communication device, which is applied to the intra-chip interconnection system described in any embodiment of the present invention, comprising:
第一转换处理模块,配置于第一接口通信组件,用于依次通过内部的时钟域转换模块、校验执行模块、协议转换模块和位频转换模块,对第一片内模块发出的待传输数据进行时钟域转换处理、校验处理、通信协议转换处理、位宽转换处理和频率转换处理,以将待传输数据发送给中间传输组件;A first conversion processing module, configured in the first interface communication component, is used to sequentially perform clock domain conversion processing, verification processing, communication protocol conversion processing, bit width conversion processing and frequency conversion processing on the data to be transmitted sent by the first on-chip module through the internal clock domain conversion module, verification execution module, protocol conversion module and bit-frequency conversion module, so as to send the data to be transmitted to the intermediate transmission component;
打拍传输执行模块,配置于中间传输组件,用于通过至少一个缓存模块执行待传输数据的打拍传输,以将待传输数据发送给第二接口通信组件;A beat transmission execution module, configured in the intermediate transmission component, for executing the beat transmission of the data to be transmitted through at least one buffer module, so as to send the data to be transmitted to the second interface communication component;
第二转换处理模块,配置于第二接口通信组件,用于依次通过内部的位频转换模块、协议转换模块、校验执行模块和时钟域转换模块,对待传输数据进行位宽转换处理、频率转换处理、通信协议转换处理、校验处理和时钟域转换处理,以将待传输数据发送给第二片内模块。The second conversion processing module is configured in the second interface communication component, and is used to perform bit width conversion processing, frequency conversion processing, communication protocol conversion processing, verification processing and clock domain conversion processing on the data to be transmitted through the internal bit-frequency conversion module, protocol conversion module, verification execution module and clock domain conversion module in sequence, so as to send the data to be transmitted to the second on-chip module.
根据本发明的另一方面,提供了一种计算机可读存储介质,所述计算机可读存储介质存储有计算机指令,所述计算机指令用于使处理器执行时实现本发明任一实施例所述的片内通信方法。According to another aspect of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the intra-chip communication method described in any embodiment of the present invention when executed.
根据本发明的另一方面,提供了一种计算机程序产品,包括计算机程序,所述计算机程序在被处理器执行时实现本发明任一实施例所述的片内通信方法。According to another aspect of the present invention, a computer program product is provided, comprising a computer program, wherein when the computer program is executed by a processor, the computer program implements the intra-chip communication method according to any embodiment of the present invention.
本发明实施例的技术方案,接口通信组件分别通过时钟域转换模块、校验执行模块、协议转换模块和位频转换模块,对待传输数据进行时钟域转换处理、校验处理、通信协议转换处理、位宽转换处理和频率转换处理,实现了片内互联系统与片内模块之间的数据传输,而中间传输组件则通过缓存模块确保了数据的逐级打拍,由此片内互联系统不但减小了芯片的走线面积,扩大了片内模块的可用面积,而且在执行高位宽数据传输任务时,依然可以利用较小位宽实现高频传输,确保了较高的数据传输效率,同时针对芯片内部多组一对一的互联场景,当前片内互联系统仅需占据较小的芯片面积,避免了冗余面积开销的出现。According to the technical solution of the embodiment of the present invention, the interface communication component performs clock domain conversion processing, verification processing, communication protocol conversion processing, bit width conversion processing and frequency conversion processing on the transmission data through the clock domain conversion module, the verification execution module, the protocol conversion module and the bit-frequency conversion module, respectively, thereby realizing data transmission between the on-chip interconnection system and the on-chip modules, and the intermediate transmission component ensures the step-by-step beating of the data through the cache module. Therefore, the on-chip interconnection system not only reduces the routing area of the chip and expands the available area of the on-chip modules, but also can still use a smaller bit width to achieve high-frequency transmission when performing high-bit-width data transmission tasks, thereby ensuring higher data transmission efficiency. At the same time, for multiple groups of one-to-one interconnection scenarios within the chip, the current on-chip interconnection system only needs to occupy a smaller chip area, avoiding the emergence of redundant area overhead.
应当理解,本部分所描述的内容并非旨在标识本发明的实施例的关键或重要特征,也不用于限制本发明的范围。本发明的其它特征将通过以下的说明书而变得容易理解。It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description.
附图说明BRIEF DESCRIPTION OF THE DRAWINGS
为了更清楚地说明本发明实施例中的技术方案,下面将对实施例描述中所需要使用的附图作简单地介绍,显而易见地,下面描述中的附图仅仅是本发明的一些实施例,对于本领域普通技术人员来讲,在不付出创造性劳动的前提下,还可以根据这些附图获得其他的附图。In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.
图1是根据本发明实施例一提供的一种片内互联系统的结构示意图;FIG1 is a schematic diagram of the structure of an on-chip interconnection system provided according to a first embodiment of the present invention;
图2是根据本发明实施例一提供的一种位频转换模块的结构示意图;FIG2 is a schematic diagram of the structure of a bit-to-frequency conversion module provided according to Embodiment 1 of the present invention;
图3是根据本发明实施例一提供的一种中间传输组件的结构示意图;FIG3 is a schematic diagram of the structure of an intermediate transmission component provided according to Embodiment 1 of the present invention;
图4是根据本发明实施例一提供的一种中间传输组件的数据流向示意图;FIG4 is a schematic diagram of data flow of an intermediate transmission component provided according to Embodiment 1 of the present invention;
图5是根据本发明实施例二提供的一种接口通信组件的结构示意图;5 is a schematic diagram of the structure of an interface communication component provided according to Embodiment 2 of the present invention;
图6是根据本发明实施例二提供的一种中间传输组件关闭时钟信号的时序图;6 is a timing diagram of a clock signal shutting down of an intermediate transmission component provided according to a second embodiment of the present invention;
图7是根据本发明具体应用场景一提供的一种片内互联系统的应用场景示意图;7 is a schematic diagram of an application scenario of an on-chip interconnection system provided according to a specific application scenario 1 of the present invention;
图8是根据本发明实施例三提供的一种片内通信方法的流程图;FIG8 is a flow chart of an intra-chip communication method provided according to Embodiment 3 of the present invention;
图9是根据本发明实施例四提供的一种片内通信装置的结构示意图。FIG9 is a schematic diagram of the structure of an intra-chip communication device provided according to a fourth embodiment of the present invention.
具体实施方式DETAILED DESCRIPTION
为了使本技术领域的人员更好地理解本发明方案,下面将结合本发明实施例中的附图,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的实施例,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做出创造性劳动前提下所获得的所有其他实施例,都应当属于本发明保护的范围。In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.
需要说明的是,本发明的说明书和权利要求书及上述附图中的术语“第一”、“第二”等是用于区别类似的对象,而不必用于描述特定的顺序或先后次序。应该理解这样使用的数据在适当情况下可以互换,以便这里描述的本发明的实施例能够以除了在这里图示或描述的那些以外的顺序实施。此外,术语“包括”和“具有”以及他们的任何变形,意图在于覆盖不排他的包含,例如,包含了一系列步骤或单元的过程、方法、系统、产品或设备不必限于清楚地列出的那些步骤或单元,而是可包括没有清楚地列出的或对于这些过程、方法、产品或设备固有的其它步骤或单元。It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.
实施例一Embodiment 1
图1为本发明实施例一提供的一种片内互联系统的结构示意图,如图1所示,该片内互联系统包括两个接口通信组件100以及位于两个接口通信组件100之间的中间传输组件200,一个接口通信组件100连接片内模块,另一个接口通信组件100连接其它片内模块或者其它片内互联系统;中间传输组件200包括至少一个缓存模块210。Figure 1 is a structural schematic diagram of an on-chip interconnection system provided in Embodiment 1 of the present invention. As shown in Figure 1, the on-chip interconnection system includes two interface communication components 100 and an intermediate transmission component 200 located between the two interface communication components 100. One interface communication component 100 connects the on-chip module, and the other interface communication component 100 connects other on-chip modules or other on-chip interconnection systems; the intermediate transmission component 200 includes at least one cache module 210.
芯片内分布有不同功能的电路模块,即片内IP(Intellectual Property),其具体是指可重复利用的、已验证的集成电路模块,又称为片内模块;接口通信组件100与片内模块或者芯片上的其它片内互联系统连接;其中,上述其它片内互联系统可以是芯片上已有的传统互联系统,也可以是本发明实施例所公开的片内互联系统。Circuit modules with different functions are distributed in the chip, namely, on-chip IP (Intellectual Property), which specifically refers to reusable and verified integrated circuit modules, also called on-chip modules; the interface communication component 100 is connected to the on-chip modules or other on-chip interconnection systems on the chip; wherein, the above-mentioned other on-chip interconnection systems can be the traditional interconnection systems already existing on the chip, or can be the on-chip interconnection systems disclosed in the embodiments of the present invention.
当两个接口通信组件100分别与不同的片内模块连接时,即实现点对点(即片内模块到片内模块)的互联通信;当一个接口通信组件100连接片内模块,另一个接口通信组件100连接其它片内互联系统时,即实现了将该片内模块便捷地接入到芯片上已有的通信网络中,也即将该片内模块与芯片上原有的一个或多个片内模块进行了通信连接。When two interface communication components 100 are respectively connected to different on-chip modules, point-to-point (i.e., on-chip module to on-chip module) interconnection communication is realized; when one interface communication component 100 is connected to an on-chip module and the other interface communication component 100 is connected to other on-chip interconnection systems, the on-chip module is conveniently connected to the existing communication network on the chip, that is, the on-chip module is communicatively connected to one or more existing on-chip modules on the chip.
中间传输组件200作为两个接口通信组件100之间通信媒介,分别连接上述两个接口通信组件100;当中间传输组件200包括多个缓存模块210时,其通过两端的缓存模块210,各自连接一个接口通信组件100;当中间传输组件200仅包括一个缓存模块210时,显然其通过该缓存模块210同时连接上述两个接口通信组件100;其中,缓存模块210可以通过寄存器执行传输数据的缓存。The intermediate transmission component 200 serves as a communication medium between the two interface communication components 100, and is respectively connected to the above-mentioned two interface communication components 100; when the intermediate transmission component 200 includes multiple cache modules 210, it is connected to each interface communication component 100 through the cache modules 210 at both ends; when the intermediate transmission component 200 includes only one cache module 210, it is obvious that it is simultaneously connected to the above-mentioned two interface communication components 100 through the cache module 210; wherein, the cache module 210 can perform caching of transmission data through a register.
接口通信组件100包括时钟域转换模块110、校验执行模块120、协议转换模块130、位频转换模块140和配置管理模块150;其中,时钟域转换模块110,连接校验执行模块120,用于待传输数据的时钟域转换处理。不同功能模块的数据传输频率可能有所不同,而片内模块和片内互联系统之间往往也存在上述跨时钟域的问题,时钟域转换模块110具体是通过寄存器作为数据存储装置,其以速率A接收上游模块发送的数据,再将数据进行本地存储,同时以速率B(与速率A的数值可以不同)将数据发送给下游模块。The interface communication component 100 includes a clock domain conversion module 110, a check execution module 120, a protocol conversion module 130, a bit-frequency conversion module 140 and a configuration management module 150; wherein the clock domain conversion module 110 is connected to the check execution module 120 and is used for clock domain conversion processing of data to be transmitted. The data transmission frequencies of different functional modules may be different, and the above-mentioned cross-clock domain problem often exists between on-chip modules and on-chip interconnection systems. The clock domain conversion module 110 specifically uses registers as data storage devices, which receive data sent by the upstream module at rate A, and then store the data locally, and send the data to the downstream module at rate B (which may be different from the value of rate A).
时钟域转换模块110具体可以通过异步FIFO(First Input First Output,先入先出队列)作为数据存储装置,以实现将时钟域A下的待传输数据发送到时钟域B下;时钟域转换模块110的数据位宽和内部FIFO的深度,均可以根据需要预先配置完成;据此时钟域转换模块110解决了片内模块与接口通信组件100之间的跨时钟域问题;同时,片内互联系统在配置生成时,为了确保数据传输的便利性,接口通信组件100与缓存模块210也采用了相同的时钟域和通信协议,因此,时钟域转换模块110实际上也解决了片内模块与中间传输组件200之间的跨时钟域问题。The clock domain conversion module 110 can specifically use an asynchronous FIFO (First Input First Output) as a data storage device to send the data to be transmitted in clock domain A to clock domain B; the data bit width and the depth of the internal FIFO of the clock domain conversion module 110 can be pre-configured as needed; accordingly, the clock domain conversion module 110 solves the cross-clock domain problem between the on-chip module and the interface communication component 100; at the same time, when the on-chip interconnection system is configured and generated, in order to ensure the convenience of data transmission, the interface communication component 100 and the cache module 210 also use the same clock domain and communication protocol. Therefore, the clock domain conversion module 110 actually also solves the cross-clock domain problem between the on-chip module and the intermediate transmission component 200.
配置管理模块150,用于为时钟域转换模块110、校验执行模块120、协议转换模块130和位频转换模块140,提供配置参数;其中,对于时钟域转换模块110而言,配置管理模块150为其提供了片内模块的时钟域信息,以及接口通信组件100的时钟域信息;当片内模块或者片内互联系统变更了时钟域信息,或者接口通信组件100变更了连接对象(即变更为与其他片内模块连接)时,均是通过配置管理模块150对记录的片内模块一侧的时钟域信息,以及接口通信组件100一侧的时钟域信息进行配置更新。The configuration management module 150 is used to provide configuration parameters for the clock domain conversion module 110, the verification execution module 120, the protocol conversion module 130 and the bit-frequency conversion module 140; wherein, for the clock domain conversion module 110, the configuration management module 150 provides it with the clock domain information of the on-chip module and the clock domain information of the interface communication component 100; when the on-chip module or the on-chip interconnection system changes the clock domain information, or the interface communication component 100 changes the connection object (i.e., changes to connect with other on-chip modules), the configuration management module 150 is used to update the recorded clock domain information on the on-chip module side and the clock domain information on the interface communication component 100 side.
可选的,在本发明实施例中,所述接口通信组件100还包括第一选择开关;第一选择开关分别连接时钟域转换模块110和校验执行模块120;配置管理模块150,还用于通过第一选择开关,建立片内模块与时钟域转换模块110或者校验执行模块120的传输通路。具体的,当片内互联组件与片内模块的时钟域信息不同时,配置管理模块150通过第一选择开关,建立片内模块与校验执行模块120的传输通道,此时时钟域转换模块110实际上处于断路状态,未参与片内模块与接口通信组件100之间的数据传输。Optionally, in an embodiment of the present invention, the interface communication component 100 further includes a first selection switch; the first selection switch is respectively connected to the clock domain conversion module 110 and the verification execution module 120; the configuration management module 150 is also used to establish a transmission path between the on-chip module and the clock domain conversion module 110 or the verification execution module 120 through the first selection switch. Specifically, when the clock domain information of the on-chip interconnection component and the on-chip module is different, the configuration management module 150 establishes a transmission channel between the on-chip module and the verification execution module 120 through the first selection switch, and at this time, the clock domain conversion module 110 is actually in an open circuit state and does not participate in the data transmission between the on-chip module and the interface communication component 100.
而当片内互联组件与片内模块的时钟域信息相同时,例如,片内互联组件连接的片内模块变更了时钟域,或者片内互联组件变更为与其他片内模块连接,此时配置管理模块150通过第一选择开关,建立片内模块与时钟域转换模块110的传输通道,由此既确保了片内模块与片内互联组件的时钟域不同时,通过时钟域转换模块110解决跨时钟域问题,又确保了片内模块与片内互联组件的时钟域相同时,时钟域转换模块110不参与数据传输,提升数据传输效率。When the clock domain information of the on-chip interconnect component is the same as that of the on-chip module, for example, the on-chip module connected to the on-chip interconnect component has changed its clock domain, or the on-chip interconnect component is changed to be connected to other on-chip modules, the configuration management module 150 establishes a transmission channel between the on-chip module and the clock domain conversion module 110 through the first selection switch, thereby ensuring that when the clock domains of the on-chip module and the on-chip interconnect component are different, the cross-clock domain problem is solved by the clock domain conversion module 110, and that when the clock domains of the on-chip module and the on-chip interconnect component are the same, the clock domain conversion module 110 does not participate in data transmission, thereby improving data transmission efficiency.
校验执行模块120,连接协议转换模块130,用于待传输数据的校验处理;其中,校验执行模块120可以根据片内模块传输数据本身的校验方式,对该传输数据进行校验处理;例如,片内模块在传输数据时,为传输数据配置了奇偶校验位(Parity),也即每发出8个数据位加上一个奇偶校验位(Parity),校验执行模块120即可根据传输数据本身的校验方式,对传输数据进行奇偶校验;配置管理模块150可以为校验执行模块120提供片内模块的数据校验方式;校验执行模块120在传输数据未通过校验时,可以以中断形式上报校验结果;其中,校验结果可以通过时钟域转换模块110上报给片内模块,也可以通过其它接口直接上报给当前芯片的控制模块。The verification execution module 120 is connected to the protocol conversion module 130, and is used for verification processing of the data to be transmitted; wherein the verification execution module 120 can verify the transmission data according to the verification method of the data transmitted by the on-chip module itself; for example, when the on-chip module transmits data, a parity check bit (Parity) is configured for the transmission data, that is, a parity check bit (Parity) is added for every 8 data bits sent, and the verification execution module 120 can perform parity check on the transmission data according to the verification method of the transmission data itself; the configuration management module 150 can provide the verification execution module 120 with the data verification method of the on-chip module; when the transmission data fails the verification, the verification execution module 120 can report the verification result in the form of an interrupt; wherein the verification result can be reported to the on-chip module through the clock domain conversion module 110, and can also be directly reported to the control module of the current chip through other interfaces.
可选的,在本发明实施例中,所述校验执行模块120,具体用于对第一片内模块发出的第一传输数据进行第一校验处理,以及将第一传输数据变换为与第二校验处理匹配的第二传输数据,并将第二传输数据发送给协议转换模块130;其中,第二校验处理的校验位数少于第一校验处理的校验位数;所述校验执行模块120,还用于对中间传输组件200发出的第三传输数据进行第三校验处理,以及将第三传输数据还原为与第四校验处理匹配的第四传输数据,并将第四传输数据发送给时钟域转换模块110;其中,第四校验处理的校验位数多于第三校验处理的校验位数。Optionally, in an embodiment of the present invention, the verification execution module 120 is specifically used to perform a first verification process on the first transmission data sent by the first on-chip module, and to transform the first transmission data into second transmission data matching the second verification process, and to send the second transmission data to the protocol conversion module 130; wherein the number of check bits in the second verification process is less than the number of check bits in the first verification process; the verification execution module 120 is also used to perform a third verification process on the third transmission data sent by the intermediate transmission component 200, and to restore the third transmission data to fourth transmission data matching the fourth verification process, and to send the fourth transmission data to the clock domain conversion module 110; wherein the number of check bits in the fourth verification process is more than the number of check bits in the third verification process.
具体的,当A片内模块发出传输数据(即第一传输数据)后,A接口通信组件100中的A校验执行模块120根据第一传输数据的校验方式(即第一校验方式),对其进行第一校验处理,同时,由于第一校验方式的数据位数较多,数据传输效率较慢,校验执行模块120需要将第一传输数据变换为数据校验位数较少的第二传输数据。Specifically, when the A chip module sends out transmission data (i.e., the first transmission data), the A verification execution module 120 in the A interface communication component 100 performs a first verification process on the first transmission data according to the verification method of the first transmission data (i.e., the first verification method). At the same time, since the first verification method has more data bits and the data transmission efficiency is slower, the verification execution module 120 needs to convert the first transmission data into the second transmission data with fewer data verification bits.
以上述技术方案为例,奇偶校验需要每8个数据位加上一个奇偶校验位,那么在发出1024位的数据时,需要配置128位的奇偶校验位,而错误检查和纠正(Error Checkingand Correcting,ECC)校验,仅需要为1024位的数据配置12个数据位即可,因此,A校验执行模块120将1024数据位+128校验位的第一传输数据,变更为1024数据位+12校验位的第二传输数据;之后A校验执行模块120将第二传输数据通过协议转换模块130、位频转换模块140和中间传输组件200,传输给另一端的B接口通信组件100。Taking the above technical solution as an example, the parity check needs to add a parity check bit for every 8 data bits. Therefore, when sending 1024 bits of data, 128 parity check bits need to be configured, and the error checking and correction (ECC) check only needs to configure 12 data bits for the 1024 bits of data. Therefore, the A check execution module 120 changes the first transmission data of 1024 data bits + 128 check bits into the second transmission data of 1024 data bits + 12 check bits; then the A check execution module 120 transmits the second transmission data to the B interface communication component 100 at the other end through the protocol conversion module 130, the bit-frequency conversion module 140 and the intermediate transmission component 200.
同样的,当B片内模块发出传输数据(即第四传输数据)后,B接口通信组件100中的B校验执行模块120根据第四传输数据的校验方式(即第四校验方式),对其进行第四校验处理,同时,由于第四校验方式的数据位数较多,数据传输效率较慢,B校验执行模块120需要将第四传输数据变换为数据校验位数较少的第三传输数据。Similarly, when the B chip module sends out transmission data (i.e., the fourth transmission data), the B verification execution module 120 in the B interface communication component 100 performs a fourth verification process on the fourth transmission data according to the verification method of the fourth transmission data (i.e., the fourth verification method). At the same time, since the fourth verification method has a large number of data bits and the data transmission efficiency is slow, the B verification execution module 120 needs to convert the fourth transmission data into the third transmission data with a smaller number of data verification bits.
以上述技术方案为例,第四校验处理可以为奇偶校验,第三校验处理可以为ECC校验,对于A校验执行模块120而言,其获取到B校验执行模块120发出的第三传输数据,先对其进行ECC校验,然后将1024数据位+12校验位的第三传输数据,还原为1024数据位+128校验位的第四传输数据,进而通过A时钟域转换模块110发送给A片内模块,以确保片内互联组件的发出数据与接收数据完全相同。由此在实现多类型数据校验的同时,提高了片内互联系统的数据传输效率,且确保了一端片内模块的发出数据,与另一端片内模块的接收数据完全相同。Taking the above technical solution as an example, the fourth check process can be a parity check, and the third check process can be an ECC check. For the A check execution module 120, it obtains the third transmission data sent by the B check execution module 120, first performs an ECC check on it, and then restores the third transmission data of 1024 data bits + 12 check bits to the fourth transmission data of 1024 data bits + 128 check bits, and then sends it to the A on-chip module through the A clock domain conversion module 110 to ensure that the data sent by the on-chip interconnection component is exactly the same as the received data. In this way, while realizing multi-type data verification, the data transmission efficiency of the on-chip interconnection system is improved, and it is ensured that the data sent by the on-chip module at one end is exactly the same as the received data of the on-chip module at the other end.
协议转换模块130,连接位频转换模块140,用于待传输数据的通信协议转换处理;协议转换模块130具体完成片内模块支持的通信协议,与片内互联系统支持的通信协议之间的数据转换;其中,配置管理模块150为协议转换模块130提供了片内模块的通信协议类型,以及片内互联系统的通信协议类型;当片内模块或者片内互联系统变更了传输协议类型,或者接口通信组件100变更了连接对象时,均是通过配置管理模块150对记录的片内模块一侧的通信协议,以及接口通信组件100一侧的通信协议进行配置更新。The protocol conversion module 130 is connected to the bit-frequency conversion module 140 and is used for the communication protocol conversion processing of the data to be transmitted; the protocol conversion module 130 specifically completes the data conversion between the communication protocol supported by the on-chip module and the communication protocol supported by the on-chip interconnection system; wherein, the configuration management module 150 provides the protocol conversion module 130 with the communication protocol type of the on-chip module and the communication protocol type of the on-chip interconnection system; when the on-chip module or the on-chip interconnection system changes the transmission protocol type, or the interface communication component 100 changes the connection object, the communication protocol on the recorded on-chip module side and the communication protocol on the interface communication component 100 side are configured and updated through the configuration management module 150.
位频转换模块140,连接协议转换模块130,用于待传输数据的位宽转换处理和频率转换处理。为了降低片内互联系统占据的芯片面积,需要将中间传输组件200的数据位宽配置为较小数值,而片内模块发出的数据通常数据位宽较大,因此,位频转换模块140需要对大位宽数据进行倍频处理,以适应中间传输组件200的传输需求,例如,若获取到片内模块以2G赫兹频率发出的1024位的数据,位频转换模块140将该数据转换为以4G赫兹频率发出的512位的数据;而在获取到中间传输组件200发送的小位宽数据时,如果作为数据接收方的片内模块需要的数据位宽较大,位频转换模块140还需要对小位宽数据进行分频处理;例如,若获取到中间传输组件200以4G赫兹频率发出512位的数据,位频转换模块140将该数据转换为以2G赫兹频率发出的1024位的数据。The bit-frequency conversion module 140 is connected to the protocol conversion module 130 and is used for bit width conversion and frequency conversion of data to be transmitted. In order to reduce the chip area occupied by the on-chip interconnection system, the data bit width of the intermediate transmission component 200 needs to be configured to a smaller value, and the data sent by the on-chip module usually has a larger data bit width. Therefore, the bit-frequency conversion module 140 needs to perform frequency doubling processing on the large bit width data to meet the transmission requirements of the intermediate transmission component 200. For example, if 1024 bits of data sent by the on-chip module at a frequency of 2 GHz is obtained, the bit-frequency conversion module 140 converts the data into 512 bits of data sent at a frequency of 4 GHz; and when the small bit width data sent by the intermediate transmission component 200 is obtained, if the data bit width required by the on-chip module as the data receiver is larger, the bit-frequency conversion module 140 also needs to perform frequency division processing on the small bit width data; for example, if 512 bits of data sent by the intermediate transmission component 200 at a frequency of 4 GHz is obtained, the bit-frequency conversion module 140 converts the data into 1024 bits of data sent at a frequency of 2 GHz.
可选的,在本发明实施例中,所述位频转换模块140,具体用于通过同步双时钟先入先出队列,执行位宽转换处理和频率转换处理。如图2所示,位频转换模块140的源时钟单元142和目的时钟单元143将两侧接口的握手协议(即vld/rdy协议)转化为先入先出队列141的读写使能,位于中间的同步双时钟下的先入先出队列141,可以实现分频数据转换或者倍频数据转换;其中,同步双时钟意味着时钟信号a与时钟信号b的时钟域相同;而读指针单元144用于指向同步双时钟的先入先出队列141的数据读取位置,写指针单元145用于指向同步双时钟的先入先出队列141的数据写入位置,读判断单元146用于判断先入先出队列141是否已为空,其影响下游模块的数据读取,写判断单元147,用于判断先入先出队列141是否已写满,其影响上游模块的数据写入。Optionally, in an embodiment of the present invention, the bit-frequency conversion module 140 is specifically configured to perform bit width conversion processing and frequency conversion processing through a synchronous dual-clock first-in-first-out queue. As shown in FIG2 , the source clock unit 142 and the destination clock unit 143 of the bit-frequency conversion module 140 convert the handshake protocol (i.e., the vld/rdy protocol) of the interfaces on both sides into the read and write enable of the first-in-first-out queue 141. The first-in-first-out queue 141 under the synchronous dual clock in the middle can realize frequency division data conversion or frequency multiplication data conversion; wherein, the synchronous dual clock means that the clock domain of the clock signal a and the clock signal b are the same; and the read pointer unit 144 is used to point to the data reading position of the first-in-first-out queue 141 of the synchronous dual clock, and the write pointer unit 145 is used to point to the data writing position of the first-in-first-out queue 141 of the synchronous dual clock. The read judgment unit 146 is used to judge whether the first-in-first-out queue 141 is empty, which affects the data reading of the downstream module, and the write judgment unit 147 is used to judge whether the first-in-first-out queue 141 is full, which affects the data writing of the upstream module.
位频转换模块140配置的同步双时钟先入先出队列,由于不存在跨时钟域问题,不但控制逻辑较为简单,而且相比于传统的异步双时钟先入先出队列,避免了当前芯片制作工艺下的频率限制(通常被限制在2G赫兹以下),提高了数据传输频率(可以提升到5G赫兹以上),同时经过分频处理后的传输数据,极大地压缩了总线位宽,节省了芯片资源。The synchronous dual-clock first-in-first-out queue configured by the bit-frequency conversion module 140 has a relatively simple control logic because there is no cross-clock domain problem. Compared with the traditional asynchronous dual-clock first-in-first-out queue, it avoids the frequency limitation under the current chip manufacturing process (usually limited to below 2 GHz), increases the data transmission frequency (can be increased to above 5 GHz), and at the same time, the transmission data after frequency division processing greatly compresses the bus bit width, saving chip resources.
中间传输组件200,用于通过至少一个缓存模块210执行待传输数据的打拍传输。中间传输组件200由一个或多个缓存模块210组成,具体数量可以根据打拍需求预先配置完成;数据打拍传输实际上是通过在传输路径上添加缓存模块210,对传输数据进行延拍,以消除亚稳态的影响。The intermediate transmission component 200 is used to perform the beat transmission of the data to be transmitted through at least one cache module 210. The intermediate transmission component 200 is composed of one or more cache modules 210, and the specific number can be pre-configured according to the beat requirement; data beat transmission is actually to delay the transmission data by adding a cache module 210 on the transmission path to eliminate the influence of metastable state.
可选的,在本发明实施例中,所述中间传输组件200包括至少一个第二选择开关;所述中间传输组件200,还用于通过所述至少一个第二选择开关,管理传输通道中的缓存模块210数量。具体的,中间传输组件200中可以配置一个或多个第二选择接开关,其可以根据具体传输需求的不同,例如,根据传输数据的数据类型或者片内模块的类型,选择将所有缓存模块210中的一个或多个加入传输通道,将剩余的缓存模块210隔离在传输通道之外,以实现传输通道中缓存模块210的灵活配置。Optionally, in the embodiment of the present invention, the intermediate transmission component 200 includes at least one second selection switch; the intermediate transmission component 200 is further used to manage the number of cache modules 210 in the transmission channel through the at least one second selection switch. Specifically, one or more second selection switches can be configured in the intermediate transmission component 200, which can select one or more of all cache modules 210 to be added to the transmission channel according to different specific transmission requirements, for example, according to the data type of the transmitted data or the type of the on-chip module, and isolate the remaining cache modules 210 outside the transmission channel, so as to realize flexible configuration of the cache modules 210 in the transmission channel.
例如,可以通过将首个缓存模块210与其它缓存模块210分别通过第二选择开关连接,再将除首个缓存模块210之外的其它缓存模块210依次串联,最后将首个缓存模块210也通过第二选择开关与距离较远的接口通信组件100连接;以图3为例,若中间传输组件200包括四个缓存模块210,即A缓存模块、B缓存模块、C缓存模块和D缓存模块,可以将A缓存模块分别通过不同的第二选择开关,分别与B缓存模块、C缓存模块和D缓存模块连接,再将B缓存模块、C缓存模块和D缓存模块串联连接,最后将A缓存模块与距离较远的接口通信组件100也通过第二选择开关连接,由此中间传输组件200可以将任意数量的缓存模块210加入传输通道,实现了任意打拍级数的管控,提升了中间传输组件200适用的数据传输范围。For example, the first cache module 210 can be connected to other cache modules 210 through the second selection switch respectively, and then the other cache modules 210 except the first cache module 210 are connected in series in sequence, and finally the first cache module 210 is also connected to the interface communication component 100 at a longer distance through the second selection switch; taking Figure 3 as an example, if the intermediate transmission component 200 includes four cache modules 210, namely, cache module A, cache module B, cache module C and cache module D, the cache module A can be connected to the cache module B, cache module C and cache module D respectively through different second selection switches, and then the cache module B, cache module C and cache module D are connected in series, and finally the cache module A is also connected to the interface communication component 100 at a longer distance through the second selection switch, so that the intermediate transmission component 200 can add any number of cache modules 210 to the transmission channel, realize the control of any beat level, and improve the data transmission range applicable to the intermediate transmission component 200.
可选的,在本发明实施例中,所述中间传输组件200,还用于将待传输数据拆分为多个分组传输数据,以及将握手通信信号拆分为多个分组握手信号,以使分组握手信号与分组传输数据一一匹配。具体的,如图4所示,大数据位宽传输时往往存在扇出(fanout)过高的问题,导致时序收敛困难,因此中间传输组件200可以根据数据位数,将传输数据拆分为多个分组传输数据(即date0至dateN)的同时,将握手协议信号(即vld/rdy协议信号)同样拆分为多组,即多个分组握手信号(即vld0至vldN,rdy0至rdyN),每个分组握手信号分别驱动数据总线的部分位域,由此确保了中间传输组件200的时序收敛。Optionally, in an embodiment of the present invention, the intermediate transmission component 200 is also used to split the data to be transmitted into multiple packet transmission data, and split the handshake communication signal into multiple packet handshake signals, so that the packet handshake signal matches the packet transmission data one by one. Specifically, as shown in FIG4, there is often a problem of too high fanout when transmitting large data bit widths, which leads to difficulty in timing convergence. Therefore, the intermediate transmission component 200 can split the transmission data into multiple packet transmission data (i.e., date0 to dateN) according to the number of data bits, and also split the handshake protocol signal (i.e., vld/rdy protocol signal) into multiple groups, i.e., multiple packet handshake signals (i.e., vld0 to vldN, rdy0 to rdyN), and each packet handshake signal drives part of the bit field of the data bus respectively, thereby ensuring the timing convergence of the intermediate transmission component 200.
特别的,本发明实施例中的接口通信组件100和缓存模块210可以在开发完成后,预先存储于数据库中,计算机根据任务模板选择一个或多个缓存模块210组成中间传输组件200,进而将两个接口通信组件100与上述配置完成的中间传输组件200,组成片内互联系统,并根据参数配置表格中的配置参数(例如,中间传输组件的位宽和频率,接口通信组件连接的片内模块的位宽和频率等)对上述片内互联系统进行参数配置,由此完成定制化的片内互联系统开发。In particular, the interface communication component 100 and the cache module 210 in the embodiment of the present invention can be pre-stored in the database after development is completed. The computer selects one or more cache modules 210 to form the intermediate transmission component 200 according to the task template, and then the two interface communication components 100 and the intermediate transmission component 200 configured as above are combined into an on-chip interconnection system, and the on-chip interconnection system is parameterized according to the configuration parameters in the parameter configuration table (for example, the bit width and frequency of the intermediate transmission component, the bit width and frequency of the on-chip module connected to the interface communication component, etc.), thereby completing the development of a customized on-chip interconnection system.
本发明实施例的技术方案,接口通信组件分别通过时钟域转换模块、校验执行模块、协议转换模块和位频转换模块,对待传输数据进行时钟域转换处理、校验处理、通信协议转换处理、位宽转换处理和频率转换处理,实现了片内互联系统与片内模块之间的数据传输,而中间传输组件则通过缓存模块确保了数据的逐级打拍,由此片内互联系统不但减小了芯片的走线面积,扩大了片内模块的可用面积,而且在执行高位宽数据传输任务时,依然可以利用较小位宽实现高频传输,确保了较高的数据传输效率,同时针对芯片内部多组一对一的互联场景,当前片内互联系统仅需占据较小的芯片面积,避免了冗余面积开销的出现。According to the technical solution of the embodiment of the present invention, the interface communication component performs clock domain conversion processing, verification processing, communication protocol conversion processing, bit width conversion processing and frequency conversion processing on the transmission data through the clock domain conversion module, the verification execution module, the protocol conversion module and the bit-frequency conversion module, respectively, thereby realizing data transmission between the on-chip interconnection system and the on-chip modules, and the intermediate transmission component ensures the step-by-step beating of the data through the cache module. Therefore, the on-chip interconnection system not only reduces the routing area of the chip and expands the available area of the on-chip modules, but also can still use a smaller bit width to achieve high-frequency transmission when performing high-bit-width data transmission tasks, thereby ensuring higher data transmission efficiency. At the same time, for multiple groups of one-to-one interconnection scenarios within the chip, the current on-chip interconnection system only needs to occupy a smaller chip area, avoiding the emergence of redundant area overhead.
实施例二Embodiment 2
图5为本发明实施例二提供的一种接口通信组件的结构示意图,如图5所示,所述接口通信组件100还包括功耗控制模块160;功耗控制模块160,连接时钟域转换模块110、校验执行模块120、协议转换模块130和位频转换模块140,用于控制接口通信组件100的时钟启停,以及向下游的缓存模块发送时钟启停信号,以及获取上游的缓存模块发送的时钟启停信号。Figure 5 is a structural diagram of an interface communication component provided in Embodiment 2 of the present invention. As shown in Figure 5, the interface communication component 100 also includes a power consumption control module 160; the power consumption control module 160 is connected to the clock domain conversion module 110, the verification execution module 120, the protocol conversion module 130 and the bit-frequency conversion module 140, and is used to control the clock start and stop of the interface communication component 100, and send clock start and stop signals to the downstream cache module, and obtain the clock start and stop signals sent by the upstream cache module.
具体的,功耗控制模块160实时监测各个通道内的传输数据,如果在预设数量的时钟周期内,未检测到传输数据通过某个传输通道,那么将当前接口通信组件100中该传输通道的时钟信号进行关闭,并向当前接口通信组件100连接的缓存模块210发出时钟停止信号。缓存模块,还用于将获取到的时钟启停信号发送给下游的接口通信组件100或者下游的其它缓存模块。Specifically, the power consumption control module 160 monitors the transmission data in each channel in real time. If no transmission data is detected passing through a transmission channel within a preset number of clock cycles, the clock signal of the transmission channel in the current interface communication component 100 is turned off, and a clock stop signal is sent to the cache module 210 connected to the current interface communication component 100. The cache module is also used to send the acquired clock start and stop signal to the downstream interface communication component 100 or other downstream cache modules.
缓存模块在获取到时钟停止信号后,关闭当前通道的时钟信号,再将其继续向下游的缓存模块发送,直至发送到末端缓存模块时,末端缓存模块在关闭当前通道的时钟信号后,继续向其连接的接口通信组件100发出时钟停止信号;该接口通信组件100的功耗控制模块160在获取到缓存模块传输的时钟停止信号后,同样关闭该通道的时钟信号,由此完成时钟信号的逐级关闭。After obtaining the clock stop signal, the cache module turns off the clock signal of the current channel, and then continues to send it to the downstream cache module until it is sent to the end cache module. After turning off the clock signal of the current channel, the end cache module continues to send a clock stop signal to the interface communication component 100 connected to it; after obtaining the clock stop signal transmitted by the cache module, the power consumption control module 160 of the interface communication component 100 also turns off the clock signal of the channel, thereby completing the step-by-step shutdown of the clock signal.
以图6为例,假设缓存模块的各个传输通道的数据传输方向均相同,均是由第一片内模块传输给第二片内模块,如果中间传输组件包括四个缓存模块,即A缓存模块、B缓存模块、C缓存模块和D缓存模块;那么在T1阶段,每个缓存模块的时钟信号均为使能状态,即各个缓存模块均处于工作状态;到了T2阶段时,根据与第一片内模块连接的接口通信组件100,即左侧位置的接口通信组件100,其内部的功耗控制模块160的计算,会关闭A缓存模块的时钟信号,此时B缓存模块、C缓存模块和D缓存模块仍然可以继续工作。Taking Figure 6 as an example, assuming that the data transmission directions of each transmission channel of the cache module are the same, and are all transmitted from the first on-chip module to the second on-chip module, if the intermediate transmission component includes four cache modules, namely, cache module A, cache module B, cache module C and cache module D; then in the T1 stage, the clock signal of each cache module is enabled, that is, each cache module is in a working state; when it comes to the T2 stage, according to the interface communication component 100 connected to the first on-chip module, that is, the interface communication component 100 on the left side, the power consumption control module 160 inside it calculates that the clock signal of cache module A will be turned off, and at this time, cache module B, cache module C and cache module D can still continue to work.
到了T3阶段时,A缓存模块和B缓存模块的时钟信号均已被关闭,C缓存模块和D缓存模块仍处于工作状态;到了T4阶段时,A缓存模块、B缓存模块和C缓存模块的时钟信号均已被关闭,D缓存模块仍处于工作状;到了T5阶段,A缓存模块、B缓存模块、C缓存模块和D缓存模块的时钟信号均已被关闭,各个缓存模块均进入空闲状态,由此实现了片内互联系统的功耗监测及时钟停止信号的逐级传输,极大地降低了片内互联系统的时钟功耗。At the T3 stage, the clock signals of cache modules A and B have been turned off, while cache modules C and D are still in working state; at the T4 stage, the clock signals of cache modules A, B and C have been turned off, while cache module D is still in working state; at the T5 stage, the clock signals of cache modules A, B, C and D have been turned off, and each cache module enters an idle state, thereby realizing power consumption monitoring of the intra-chip interconnection system and step-by-step transmission of the clock stop signal, greatly reducing the clock power consumption of the intra-chip interconnection system.
特别的,如果各个传输通道的数据传输方向不同,那么需要从传输通道维度,对各个传输通道的时钟信号分别进行关闭。此外,功耗控制模块160在关闭当前接口通信组件100的时钟信号时,可以保留时钟域转换模块110的时钟信号,而将校验执行模块120、协议转换模块130、位频转换模块140和配置管理模块150的时钟信号进行关闭,以便于片内模块发出传输数据时,时钟域转换模块110可以及时获取到,避免出现数据丢失现象。In particular, if the data transmission directions of each transmission channel are different, then it is necessary to shut down the clock signals of each transmission channel separately from the transmission channel dimension. In addition, when the power consumption control module 160 shuts down the clock signal of the current interface communication component 100, it can retain the clock signal of the clock domain conversion module 110, and shut down the clock signals of the verification execution module 120, the protocol conversion module 130, the bit-frequency conversion module 140 and the configuration management module 150, so that when the on-chip module sends transmission data, the clock domain conversion module 110 can obtain it in time to avoid data loss.
同样的,在接口通信组件100关闭时钟信号后,若功耗控制模块160监测到时钟域转换模块110接收到传输数据,功耗控制模块160开启校验执行模块120、协议转换模块130、位频转换模块140和配置管理模块150的时钟信号,并向当前接口通信组件100连接的缓存模块发出时钟启动信号,缓存模块在获取到时钟启动信号后,开启当前通道的时钟信号,再将其继续向下游的缓存模块发送,直至发送到末端缓存模块时,末端缓存模块在开启当前通道的时钟信号后,继续向其连接的接口通信组件100发出时钟启动信号;该接口通信组件100的功耗控制模块160在获取到缓存模块传输的时钟启动信号后,同样开启该通道的时钟信号,由此完成时钟启动信号的逐级传输。Similarly, after the interface communication component 100 turns off the clock signal, if the power consumption control module 160 detects that the clock domain conversion module 110 has received the transmission data, the power consumption control module 160 turns on the clock signals of the verification execution module 120, the protocol conversion module 130, the bit-frequency conversion module 140 and the configuration management module 150, and sends a clock start signal to the cache module connected to the current interface communication component 100. After obtaining the clock start signal, the cache module turns on the clock signal of the current channel, and then continues to send it to the downstream cache module until it is sent to the end cache module. After turning on the clock signal of the current channel, the end cache module continues to send a clock start signal to the interface communication component 100 connected to it; after obtaining the clock start signal transmitted by the cache module, the power consumption control module 160 of the interface communication component 100 also turns on the clock signal of the channel, thereby completing the step-by-step transmission of the clock start signal.
本发明实施例的技术方案,功耗控制模块,通过控制接口通信组件的时钟启停,以及向下游的缓存模块发送时钟启停信号,以及获取上游的缓存模块发送的时钟启停信号,实现了接口通信组件的时钟启停管理及时钟启停信号传输,缓存模块则根据获取到的时钟启停信号,控制时钟启停,并将时钟启停信号发送给下游的接口通信组件或者下游的其它缓存模块,由此极大地降低了片内互联系统的时钟功耗。The technical solution of the embodiment of the present invention is a power consumption control module, which realizes the clock start and stop management and clock start and stop signal transmission of the interface communication component by controlling the clock start and stop of the interface communication component, sending the clock start and stop signal to the downstream cache module, and obtaining the clock start and stop signal sent by the upstream cache module. The cache module controls the clock start and stop according to the obtained clock start and stop signal, and sends the clock start and stop signal to the downstream interface communication component or other downstream cache modules, thereby greatly reducing the clock power consumption of the on-chip interconnection system.
具体应用场景一Specific application scenario 1
图7为本发明具体应用场景一提供的片内互联系统的应用场景示意图,如图7所示,芯片上原本已规划完成四种原始片内模块,即原始片内模块1、原始片内模块2、原始片内模块3和原始片内模块4,而上述四种原始片内模块的中间区域则通过传统互联系统搭建完成,如果此时还需要在芯片上新增四个片内模块,即片内模块A、片内模块B、片内模块C和片内模块D,由于上述四个原始片内模块以及传统互联系统已占据了较大面积,剩余芯片面积无法满足传统的走线方式,此时可以在剩余区域的不同位置,分别配置片内模块A、片内模块B、片内模块C和片内模块D,再为片内模块A、片内模块B、片内模块C和片内模块D,分别配置一个片内互联系统,通过该片内互联系统将对应的片内模块接入传统互联系统,由此在不改变已有的原始片内模块和传统互联系统的基础上,利用本发明实施例中的片内互联系统,仅需要占据较小的芯片面积,即可将新增片内模块加入布局完成的芯片中,并完成新增片内模块的互联。FIG7 is a schematic diagram of an application scenario of an on-chip interconnection system provided by a specific application scenario 1 of the present invention. As shown in FIG7 , four original on-chip modules have been planned on the chip, namely, original on-chip module 1, original on-chip module 2, original on-chip module 3 and original on-chip module 4, and the middle area of the above four original on-chip modules is built by a traditional interconnection system. If four additional on-chip modules are required on the chip at this time, namely, on-chip module A, on-chip module B, on-chip module C and on-chip module D, since the above four original on-chip modules and the traditional interconnection system have occupied a large area, the remaining chip area cannot meet the traditional interconnection system. In this case, the on-chip module A, the on-chip module B, the on-chip module C and the on-chip module D can be respectively configured at different positions of the remaining area, and then an on-chip interconnection system can be respectively configured for the on-chip module A, the on-chip module B, the on-chip module C and the on-chip module D, and the corresponding on-chip module can be connected to the traditional interconnection system through the on-chip interconnection system. Therefore, on the basis of not changing the existing original on-chip module and the traditional interconnection system, the on-chip interconnection system in the embodiment of the present invention only needs to occupy a small chip area, and the newly added on-chip module can be added to the chip with the layout completed, and the interconnection of the newly added on-chip module can be completed.
实施例三Embodiment 3
图8为本发明实施例三提供的一种片内通信方法的流程图,该方法可以由片内通信装置来执行,该片内通信装置可以采用硬件和/或软件的形式实现,该片内通信装置可配置于本发明任意实施例中的片内互联系统。如图8所示,该方法包括:FIG8 is a flow chart of an intra-chip communication method provided in Embodiment 3 of the present invention. The method can be executed by an intra-chip communication device. The intra-chip communication device can be implemented in the form of hardware and/or software. The intra-chip communication device can be configured in the intra-chip interconnection system in any embodiment of the present invention. As shown in FIG8 , the method includes:
S301、第一接口通信组件依次通过内部的时钟域转换模块、校验执行模块、协议转换模块和位频转换模块,对第一片内模块发出的待传输数据进行时钟域转换处理、校验处理、通信协议转换处理、位宽转换处理和频率转换处理,以将待传输数据发送给中间传输组件。S301, the first interface communication component sequentially performs clock domain conversion processing, verification processing, communication protocol conversion processing, bit width conversion processing and frequency conversion processing on the data to be transmitted sent by the first on-chip module through the internal clock domain conversion module, verification execution module, protocol conversion module and bit-frequency conversion module, so as to send the data to be transmitted to the intermediate transmission component.
S302、中间传输组件通过至少一个缓存模块执行待传输数据的打拍传输,以将待传输数据发送给第二接口通信组件。S302. The intermediate transmission component performs beat transmission of the data to be transmitted through at least one cache module to send the data to be transmitted to the second interface communication component.
S303、第二接口通信组件依次通过内部的位频转换模块、协议转换模块、校验执行模块和时钟域转换模块,对待传输数据进行位宽转换处理、频率转换处理、通信协议转换处理、校验处理和时钟域转换处理,以将待传输数据发送给第二片内模块。S303, the second interface communication component sequentially performs bit width conversion processing, frequency conversion processing, communication protocol conversion processing, verification processing and clock domain conversion processing on the data to be transmitted through the internal bit-frequency conversion module, protocol conversion module, verification execution module and clock domain conversion module, so as to send the data to be transmitted to the second on-chip module.
本发明实施例的技术方案,接口通信组件分别通过时钟域转换模块、校验执行模块、协议转换模块和位频转换模块,对待传输数据进行时钟域转换处理、校验处理、通信协议转换处理、位宽转换处理和频率转换处理,实现了片内互联系统与片内模块之间的数据传输,而中间传输组件则通过缓存模块确保了数据的逐级打拍,由此片内互联系统不但减小了芯片的走线面积,扩大了片内模块的可用面积,而且在执行高位宽数据传输任务时,依然可以利用较小位宽实现高频传输,确保了较高的数据传输效率,同时针对芯片内部多组一对一的互联场景,当前片内互联系统仅需占据较小的芯片面积,避免了冗余面积开销的出现。According to the technical solution of the embodiment of the present invention, the interface communication component performs clock domain conversion processing, verification processing, communication protocol conversion processing, bit width conversion processing and frequency conversion processing on the transmission data through the clock domain conversion module, the verification execution module, the protocol conversion module and the bit-frequency conversion module, respectively, thereby realizing data transmission between the on-chip interconnection system and the on-chip modules, and the intermediate transmission component ensures the step-by-step beating of the data through the cache module. Therefore, the on-chip interconnection system not only reduces the routing area of the chip and expands the available area of the on-chip modules, but also can still use a smaller bit width to achieve high-frequency transmission when performing high-bit-width data transmission tasks, thereby ensuring higher data transmission efficiency. At the same time, for multiple groups of one-to-one interconnection scenarios within the chip, the current on-chip interconnection system only needs to occupy a smaller chip area, avoiding the emergence of redundant area overhead.
实施例四Embodiment 4
图9是本发明实施例四所提供的一种片内通信装置的结构框图,该装置具体包括:FIG9 is a structural block diagram of an intra-chip communication device provided by Embodiment 4 of the present invention, the device specifically comprising:
第一转换处理模块401,配置于第一接口通信组件,用于依次通过内部的时钟域转换模块、校验执行模块、协议转换模块和位频转换模块,对第一片内模块发出的待传输数据进行时钟域转换处理、校验处理、通信协议转换处理、位宽转换处理和频率转换处理,以将待传输数据发送给中间传输组件;The first conversion processing module 401 is configured in the first interface communication component, and is used to perform clock domain conversion processing, verification processing, communication protocol conversion processing, bit width conversion processing and frequency conversion processing on the data to be transmitted sent by the first on-chip module through the internal clock domain conversion module, verification execution module, protocol conversion module and bit-frequency conversion module in sequence, so as to send the data to be transmitted to the intermediate transmission component;
打拍传输执行模块402,配置于中间传输组件,用于通过至少一个缓存模块执行待传输数据的打拍传输,以将待传输数据发送给第二接口通信组件;A beat transmission execution module 402, configured in the intermediate transmission component, is used to execute the beat transmission of the data to be transmitted through at least one buffer module to send the data to be transmitted to the second interface communication component;
第二转换处理模块403,配置于第二接口通信组件,用于依次通过内部的位频转换模块、协议转换模块、校验执行模块和时钟域转换模块,对待传输数据进行位宽转换处理、频率转换处理、通信协议转换处理、校验处理和时钟域转换处理,以将待传输数据发送给第二片内模块。The second conversion processing module 403 is configured in the second interface communication component, and is used to perform bit width conversion processing, frequency conversion processing, communication protocol conversion processing, verification processing and clock domain conversion processing on the data to be transmitted through the internal bit-frequency conversion module, protocol conversion module, verification execution module and clock domain conversion module in sequence, so as to send the data to be transmitted to the second on-chip module.
本发明实施例的技术方案,接口通信组件分别通过时钟域转换模块、校验执行模块、协议转换模块和位频转换模块,对待传输数据进行时钟域转换处理、校验处理、通信协议转换处理、位宽转换处理和频率转换处理,实现了片内互联系统与片内模块之间的数据传输,而中间传输组件则通过缓存模块确保了数据的逐级打拍,由此片内互联系统不但减小了芯片的走线面积,扩大了片内模块的可用面积,而且在执行高位宽数据传输任务时,依然可以利用较小位宽实现高频传输,确保了较高的数据传输效率,同时针对芯片内部多组一对一的互联场景,当前片内互联系统仅需占据较小的芯片面积,避免了冗余面积开销的出现。According to the technical solution of the embodiment of the present invention, the interface communication component performs clock domain conversion processing, verification processing, communication protocol conversion processing, bit width conversion processing and frequency conversion processing on the transmission data through the clock domain conversion module, the verification execution module, the protocol conversion module and the bit-frequency conversion module, respectively, thereby realizing data transmission between the on-chip interconnection system and the on-chip modules, and the intermediate transmission component ensures the step-by-step beating of the data through the cache module. Therefore, the on-chip interconnection system not only reduces the routing area of the chip and expands the available area of the on-chip modules, but also can still use a smaller bit width to achieve high-frequency transmission when performing high-bit-width data transmission tasks, thereby ensuring higher data transmission efficiency. At the same time, for multiple groups of one-to-one interconnection scenarios within the chip, the current on-chip interconnection system only needs to occupy a smaller chip area, avoiding the emergence of redundant area overhead.
上述装置可执行本发明任意实施例所提供的片内通信方法,具备执行方法相应的功能模块和有益效果。未在本实施例中详尽描述的技术细节,可参见本发明任意实施例提供的片内通信方法。The above device can execute the intra-chip communication method provided by any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method. For technical details not described in detail in this embodiment, please refer to the intra-chip communication method provided by any embodiment of the present invention.
实施例五Embodiment 5
在一些实施例中,片内通信方法可被实现为计算机程序,其被有形地包含于计算机可读存储介质,例如存储单元。在一些实施例中,计算机程序的部分或者全部可以经由ROM 和/或通信单元而被载入和/或安装到异构硬件加速器上。当计算机程序加载到RAM 并由处理器执行时,可以执行上文描述的片内通信方法的一个或多个步骤。备选地,在其他实施例中,处理器可以通过其他任何适当的方式(例如,借助于固件)而被配置为执行片内通信方法。In some embodiments, the intra-chip communication method may be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as a storage unit. In some embodiments, part or all of the computer program may be loaded and/or installed on a heterogeneous hardware accelerator via a ROM and/or a communication unit. When the computer program is loaded into RAM and executed by a processor, one or more steps of the intra-chip communication method described above may be performed. Alternatively, in other embodiments, the processor may be configured to perform the intra-chip communication method in any other appropriate manner (e.g., by means of firmware).
本文中以上描述的系统和技术的各种实施方式可以在数字电子电路系统、集成电路系统、现场可编程门阵列(FPGA)、专用集成电路(ASIC)、专用标准产品(ASSP)、芯片上的系统(SOC)、负载可编程逻辑设备(CPLD)、计算机硬件、固件、软件、和/或它们的组合中实现。这些各种实施方式可以包括:实施在一个或者多个计算机程序中,该一个或者多个计算机程序可在包括至少一个可编程处理器的可编程系统上执行和/或解释,该可编程处理器可以是专用或者通用可编程处理器,可以从存储系统、至少一个输入装置、和至少一个输出装置接收数据和指令,并且将数据和指令传输至该存储系统、该至少一个输入装置、和该至少一个输出装置。Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and/or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and/or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.
用于实施本发明的方法的计算机程序可以采用一个或多个编程语言的任何组合来编写。这些计算机程序可以提供给通用计算机、专用计算机或其他可编程数据处理装置的处理器,使得计算机程序当由处理器执行时使流程图和/或框图中所规定的功能/操作被实施。计算机程序可以完全在机器上执行、部分地在机器上执行,作为独立软件包部分地在机器上执行且部分地在远程机器上执行或完全在远程机器或服务器上执行。Computer programs for implementing the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that when the computer program is executed by the processor, the functions/operations specified in the flow chart and/or block diagram are implemented. The computer program may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.
在本发明的上下文中,计算机可读存储介质可以是有形的介质,其可以包含或存储以供指令执行系统、装置或设备使用或与指令执行系统、装置或设备结合地使用的计算机程序。计算机可读存储介质可以包括但不限于电子的、磁性的、光学的、电磁的、红外的、或半导体系统、装置或设备,或者上述内容的任何合适组合。备选地,计算机可读存储介质可以是机器可读信号介质。机器可读存储介质的更具体示例会包括基于一个或多个线的电气连接、便携式计算机盘、硬盘、随机存取存储器(RAM)、只读存储器(ROM)、可擦除可编程只读存储器(EPROM或快闪存储器)、光纤、便捷式紧凑盘只读存储器(CD-ROM)、光学储存设备、磁储存设备、或上述内容的任何合适组合。In the context of the present invention, a computer-readable storage medium may be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, device, or equipment. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or equipment, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
为了提供与用户的交互,可以在异构硬件加速器上实施此处描述的系统和技术,该异构硬件加速器具有:用于向用户显示信息的显示装置(例如,CRT(阴极射线管)或者LCD(液晶显示器)监视器);以及键盘和指向装置(例如,鼠标或者轨迹球),用户可以通过该键盘和该指向装置来将输入提供给异构硬件加速器。其它种类的装置还可以用于提供与用户的交互;例如,提供给用户的反馈可以是任何形式的传感反馈(例如,视觉反馈、听觉反馈、或者触觉反馈);并且可以用任何形式(包括声输入、语音输入或者、触觉输入)来接收来自用户的输入。To provide interaction with a user, the systems and techniques described herein may be implemented on a heterogeneous hardware accelerator having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to a user; and a keyboard and a pointing device (e.g., a mouse or trackball) through which a user can provide input to the heterogeneous hardware accelerator. Other types of devices may also be used to provide interaction with a user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, voice input, or tactile input).
可以将此处描述的系统和技术实施在包括后台部件的计算系统(例如,作为数据服务器)、或者包括中间件部件的计算系统(例如,应用服务器)、或者包括前端部件的计算系统(例如,具有图形用户界面或者网络浏览器的用户计算机,用户可以通过该图形用户界面或者该网络浏览器来与此处描述的系统和技术的实施方式交互)、或者包括这种后台部件、中间件部件、或者前端部件的任何组合的计算系统中。可以通过任何形式或者介质的数字数据通信(例如,通信网络)来将系统的部件相互连接。通信网络的示例包括:局域网(LAN)、广域网(WAN)、区块链网络和互联网。The systems and techniques described herein may be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.
计算系统可以包括客户端和服务器。客户端和服务器一般远离彼此并且通常通过通信网络进行交互。通过在相应的计算机上运行并且彼此具有客户端-服务器关系的计算机程序来产生客户端和服务器的关系。服务器可以是云服务器,又称为云计算服务器或云主机,是云计算服务体系中的一项主机产品,以解决了传统物理主机与VPS服务中,存在的管理难度大,业务扩展性弱的缺陷。A computing system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The client and server relationship is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services.
应该理解,可以使用上面所示的各种形式的流程,重新排序、增加或删除步骤。例如,本发明中记载的各步骤可以并行地执行也可以顺序地执行也可以不同的次序执行,只要能够实现本发明的技术方案所期望的结果,本文在此不进行限制。It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of the present invention can be achieved, and this document does not limit this.
上述具体实施方式,并不构成对本发明保护范围的限制。本领域技术人员应该明白的是,根据设计要求和其他因素,可以进行各种修改、组合、子组合和替代。任何在本发明的精神和原则之内所作的修改、等同替换和改进等,均应包含在本发明保护范围之内。The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (8)
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410823066.4A CN118394703B (en) | 2024-06-25 | 2024-06-25 | On-chip interconnection system and on-chip communication method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410823066.4A CN118394703B (en) | 2024-06-25 | 2024-06-25 | On-chip interconnection system and on-chip communication method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN118394703A CN118394703A (en) | 2024-07-26 |
| CN118394703B true CN118394703B (en) | 2024-09-10 |
Family
ID=91989220
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410823066.4A Active CN118394703B (en) | 2024-06-25 | 2024-06-25 | On-chip interconnection system and on-chip communication method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118394703B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP1035705A2 (en) * | 1999-03-12 | 2000-09-13 | Nokia Mobile Phones Ltd. | Communication system and data adapter |
| CN113726693A (en) * | 2021-09-01 | 2021-11-30 | 安徽芯纪元科技有限公司 | Low-speed parallel asynchronous communication method and communication system between FPGA (field programmable Gate array) chips |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5802278A (en) * | 1995-05-10 | 1998-09-01 | 3Com Corporation | Bridge/router architecture for high performance scalable networking |
| US7032119B2 (en) * | 2000-09-27 | 2006-04-18 | Amphus, Inc. | Dynamic power and workload management for multi-server system |
| US7475153B2 (en) * | 2004-07-16 | 2009-01-06 | International Business Machines Corporation | Method for enabling communication between nodes |
| US8051231B2 (en) * | 2008-11-06 | 2011-11-01 | International Business Machines Corporation | Data communications among electronic devices within a computer |
| US9625980B2 (en) * | 2014-12-16 | 2017-04-18 | Nxp Usa, Inc. | Low power configuration for USB (Universal Serial Bus) devices |
-
2024
- 2024-06-25 CN CN202410823066.4A patent/CN118394703B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP1035705A2 (en) * | 1999-03-12 | 2000-09-13 | Nokia Mobile Phones Ltd. | Communication system and data adapter |
| CN113726693A (en) * | 2021-09-01 | 2021-11-30 | 安徽芯纪元科技有限公司 | Low-speed parallel asynchronous communication method and communication system between FPGA (field programmable Gate array) chips |
Also Published As
| Publication number | Publication date |
|---|---|
| CN118394703A (en) | 2024-07-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2021129689A1 (en) | Data bit width conversion method and device | |
| CN110222001B (en) | Feedback control system and feedback control method based on PXIe chassis | |
| CN111901164A (en) | Adaptive control method, device, equipment and system for OCP NIC network card | |
| WO2016192217A1 (en) | Apb bus bridge | |
| WO2025103479A1 (en) | Optical module access method, apparatus and system, electronic device, and nonvolatile readable storage medium | |
| WO2024212744A1 (en) | Server, heterogeneous device, and data processing apparatus thereof | |
| US10223310B2 (en) | Multi-source flow management using elastic FIFO structures | |
| CN117827726B (en) | Method and device for realizing non-transparent bridge transmission, electronic equipment and storage medium | |
| CN111736867A (en) | A kind of FPGA updating device, method and storage medium | |
| Slogsnat et al. | An open-source hypertransport core | |
| CN103222286B (en) | Route switching device, network switching system and route switching method | |
| CN116974636B (en) | Multi-path interconnection system and bus interface initialization method and device thereof | |
| CN118394193A (en) | Power supply control method, power supply control device, server, and storage medium | |
| CN118394703B (en) | On-chip interconnection system and on-chip communication method | |
| CN115580365A (en) | A clock signal transmission method, device, equipment and medium | |
| CN120046562A (en) | SSD controller prototype verification method and device based on FPGA | |
| US10958595B2 (en) | Cut-through bridge error isolation | |
| CN119882935A (en) | Timing sequence convergence system, method, device and storage medium | |
| Gibiluka et al. | BAT-Hermes: a transition-signaling bundled-data NoC router | |
| CN118656221A (en) | Microservice merging method, device, equipment and storage medium | |
| CN117970994A (en) | A cross-clock domain processing device, processing method and chiplet integration system | |
| CN116306417A (en) | Clock delay method, clock delay device, electronic equipment and storage medium | |
| CN116089026A (en) | Centralized control interrupt control method, interrupt control system and chip | |
| CN111475572B (en) | Block generation method, device, equipment and medium | |
| Li et al. | Architecture Design of Protocol Controller Based on Traffic-Driven Software Defined Interconnection |
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 |